We recently received an RTX 4070 Founders Edition from Nvidia and we have been testing it for the past week by using 11 PC games plus hybrid benchmarks. The Founders Edition will be available at $599 starting tomorrow. Nvidia is touting this GPU as a great upgrade to the seven-year-old 10 series cards – and while it is true that its value proposition is undeniable – you will have to decide if it is worth the price of admission.

Thankfully, Nvidia has kept pricing sane as the RTX 4070 arrives with multiple new features, including DLSS 3, which brings an incredible performance uplift. For this reason alone, this new entry in the 40 series is extremely exciting. The higher-priced RTX 3080 is hard to find new in stock so we will compare them to find the best value.

The RTX 4070 further cements its new value role as it supports impressive new features including Displaced Micro-Meshes and Nvidia DLSS 3 technology, all while using less power than the RTX 3070 Ti. Additionally, for broadcasters, the RTX 4070 supports AV1 encoding which in and of itself is a good enough reason to upgrade from older generations of cards. 

The RTX 4070 Feature set

Right out of the gate, many will be concerned about being bound by the RTX 4070’s 12GB VRAM capacity as modern games skyrocket in their usage. While it may not age the best in 5-6 years, its VRAM is more than enough for current needs. Compared with buying a used RTX 3080, the RTX 4070 is a new card with a warranty, and it includes the latest leaps in DLSS/AI technology.

Comparing generation-to-generation there is a giant leap thanks to TGP/power improvements. There is also no price category competition from AMD, and FSR 2.0 is well behind DLSS 3 in performance. Again, the value proposition in comparison to the rest of the rather expensive 40 series is obvious, but the RTX 4070 may not offer the raw GPU power upgrade that those with RTX 3080s are looking for at this price range.

Today, there are over 400 games and applications that have RTX support which includes over 50 announced titles with DLSS 3. According to Nvidia, DLSS 3 has the fastest adoption of any of their technologies to date, with DLSS 3 being adopted 7x faster than DLSS 2. DLSS 3 is featured in some of the hottest current and upcoming titles including Forza Horizon 5, Diablo IV, Redfall, and The Finals. This value alone may be enough to move the needle for potential buyers.

GeForce RTX 4070 Founders Edition Board

The GeForce RTX 4070 Founders Edition graphics card incorporates many of the new design elements that are also found on NVIDIA’s GeForce RTX 4070 Ti, RTX 4080, and RTX 4090 Founders Edition boards. The RTX 4070 Founders Edition board is crafted with premium materials and components including a die-cast aluminum body and dual axial fans. It looks fantastic in our PC and especially for ITX enthusiasts, this may be a go-to card with its perfect compact size.

The RTX 4070’s new design also provides about a 20% increase improvement in airflow compared to the RTX 3070 Ti Founders Edition.

Our Plan

Nvidia is in an interesting spot with the RTX 4070 delivering outstandingly efficient performance. For this review, we are planning to benchmark the RTX 4080, RTX 4070 Ti, RTX 4070, RTX 3070, and RTX 3080 within an average build that most gamers today would be able to complete on a reasonable budget.

We think comparing the RTX 4070 performance against current family and iconic cards from Nvidia’s previous RTX generation of GPUs may be worth it for Turing or Ampere users considering an upgrade to Ada Lovelace.

The GeForce RTX 4070 is based on the AD104 GPU and equipped with 5,888 CUDA Cores providing 29 FP32 Shader-TFLOPS of power for rendering traditional rasterized graphics. It also boasts 184 Fourth Generation Tensor Cores offering 466 Tensor-TFLOPS (with Sparsity) for AI processing and DLSS, 46 Third Generation Ada RT Cores capable of 67 RT-TFLOPS for powering next-generation ray-traced graphics, and 12GB of GDDR6X memory.

We want to test if it really can hold up with 12GB of VRAM to deliver a potential amazing value proposition compared to the RTX 3080 launched at $699 two years ago and the RTX 2080 Ti which launched at $99 MSRP four years ago. We want to especially compare the $599 RTX 4070 with the $849 ASUS Tuff Gaming 4070 Ti OC to determine its place in the current lineup.

Features & Specifications

The RTX 4070 technological innovations include:

• New Streaming Multiprocessors (SM) – The new SM delivers up to 2x performance and power efficiency
• 4th Generation Tensor Cores and Optical Flow – Enable and accelerate transformative AI
  technologies, including the new frame rate multiplying Nvidia DLSS 3
• 3rd Generation RT Cores – Up to 2x ray tracing performance, delivering incredibly detailed
  virtual worlds like never before
• Shader Execution Reordering (SER) – SER improves ray tracing operations by 2x, boosting FPS up to 44% in Cyberpunk with RT: Overdrive Mode
• DLSS 3 – A revolutionary breakthrough in AI-powered graphics that massively boosts
  performance using AI to generate additional high-quality frames
• Nvidia Studio – Unmatched performance in 3D rendering, video editing, and live streaming
• AV1 Encoders – The 8th generation Nvidia Encoder (NVENC) with AV1 is 40% more efficient than H.264, enabling new possibilities for streamers, broadcasters, and video callers

Key Features from ASUS

• Dedicated 3rd generation ray tracing cores (46)
• Dedicated 4th generation Tensor cores (184)
• Nvidia DLSS 3 support
• Game Ready and Nvidia Studio drivers
• Nvidia GeForce Experience
• Nvidia Broadcast
• Nvidia G-Sync
• Nvidia GPU Boost
• PCI Express Gen 4
• Microsoft DirectX 12 Ultimate support
• Support for Vulkan RT APIs, Vulkan 1.3, and OpenGL 4.6
• HDCP 2.3 support
• DisplayPort 1.4 support: up to 4K at 240Hz or 8K at 60Hz with DSC, HDR
• HDMI 2.1 support: up to 4K 240Hz, Gaming VRR, HDR

Specifications

Nvidia’s MSRP price for RTX 4070 Ti GPUs is $799, and ASUS’s price for the RTX 4070 Ti TUF GAMING OC Edition is $849.99. So it’s exciting to get back down to Nvidia’s MSRP for the RTX 4070 FE at just $599.

Here is the RTX 4070 in GPU-Z:

According to GPU-Z, the RTX 4070 has the default GPU clock of 1920Mhz compared to the RTX 4070 Ti TUF GAMING OC’s 2310MHz, about an 18.4% difference, and a boost GPU clock of 2475Mhz compared to the RTX 4070 Ti TUF GAMING OC’s 2730MHz which is around a 10% difference. So for $350 less, you will have a very competitive performance with the latest 40 series feature set.

Below is the advanced general information on the RTX 4070 FE as reported by the GPU-Z tool.

As you can see from the GPU-Z screenshots, you can even increase both power and temperature limits to some degree, and while there is little room for overclocking, there is some additional potential performance for gamers looking to get more from their builds. RTX 4070 partner boards should be interesting to see based on their cooling methods and pricing.

Our Benchmarking PC

We benchmark using FrameView on a recent install of Windows 11 Pro Edition 22H2, at 3440×1440 using an Intel Core i7-12700K with stock clocks and 32GB of DDR4 Corsair Vengeance RGB PRO 3600MHz memory on an ASUS PRIME Z690-P D4 motherboard. All games and benchmarks are the latest versions, and we use the latest GeForce 531.42 press drivers for our testing. The games tested, display driver, settings, and hardware are identical except for the GPUs we compare.

Fair warning – We lost our RTX 4080 during benchmarking due to a faulty power cable and PSU when swapping out video cards, so this is why the card is missing from some tests. We are in the process of getting them replaced.

Let’s unbox and take a closer look at this graphics card.

A Closer Look at the RTX 4070 Founder’s Edition

Packaging

The box cover highlights a sleek approach to the packaging. I am a huge fan of the presentation Nvidia has been providing for its FE line. It offers a very premium unboxing experience. The graphics card image for the recent RTX cards is iconic and that shape is also shown on the box.

Accessories

As we open the box, it folds out beautifully displaying the brand new RTX 4070 FE. Beneath the card are the new standard 16 PIN (12VHPWR) PCIe connector and an installation guide with a QR code to visit Nvidia’s website for more guides and information if needed.

The Card

The RTX 4070 FE is a small dual-fan graphics card with classic RTX aesthetics that are still refined and look great. The logo sadly does not light up on the RTX 4070 FE so that is one downside of its lower price range.

For comparison here are some photos of our 4070 Ti:

The IO panel connectors are 3 DisplayPorts and 1 HDMI connection.

Inside the case

The RTX 4070 FE looks subtle and classic inside a case. I would personally love to build an SFF PC using this card. It is the perfect size offering a powerful small gaming rig at a reasonable price compared to the sky-high prices we have seen these last few years. We like that it is small and discrete but some may not like the unlit logo

Next is our testing configuration, methodology, and more.

Test Configuration

Benching Methodology

Test Configuration – Hardware

• 12th Gen Intel Core i7-12700K (Hyper-Threading/Turbo boost on; stock settings)
• ASUS PRIME Z690-P D4 motherboard (Intel Z690 chipset, v.1008 BIOS)
• CORSAIR Vengeance RGB PRO 32GB DDR4 (2×16GB, dual-channel at 3600 MHz XMP)
• RTX 4070 FE 12GB, stock clocks; supplied by Nvidia
• ASUS TUF GAMING GeForce RTX 4070 Ti OC Edition 12GB, stock clocks; supplied by Nvidia
• RTX 3080 FE 10GB, stock clocks
• RTX 3070 FE, stock clocks
• 1 x TeamGroup 1 TB NVMe M.2 SSD
• 2 x WD Blue 1TB SATA SSD
• Corsair RM850x, 850W 80PLUS Gold power supply unit
• ALIENWARE 34″ CURVED QD-OLED GAMING MONITOR – AW3423DWF

Test Configuration – Software

• NVIDIA GeForce 531.42 game-ready press drivers; ‘Prefer maximum performance’ (on a per-game profile basis); Shader Cache Size ‘Unlimited’ (globally); fixed refresh rate (globally).
• We enable Resizable BAR
• ‘V-Sync application controlled’ in the control panel; V-Sync off in-game.
• We note and specify the main in-game display, graphics, AA, and scaling settings in the performance summary charts.
• Windows 11 64-bit Pro edition, latest updates v22H2, High-performance power plan, HAGS & Game Mode are enabled, Game DVR & Game Bar features off, Control Flow Guard (CFG) off on a per-game basis, Hypervisor and Virtualization-based security are disabled.
• We do not install ASUS tools.
• Latest DirectX
• All games are patched to their latest versions at the time of publication.
• 3DMark suite, the latest version
• RivaTuner Statistics Server (RTSS), the latest version
• FrameView, the latest version
• Display Driver Uninstaller (DDU), the latest version; always uninstall drivers using DDU in safe mode, clean, and restart.
• ISLC (Purge Standby List) before each benchmark.

GeForce Driver Suite-related

• We use DCH Game Ready drivers.
• The display driver is installed.
• We install the latest version of PhysX.

Hybrid & Non-Synthetic Tests-related

• Single run per test.

Game Benchmarks-related

• We use the corresponding built-in or custom benchmark sequence.

Frametimes Capture

• We use FrameView for capturing frame times and analyzing the relevant performance numbers obtained from each recorded built-in or custom benchmark sequence.

Benchmark Suite:

PC Games

DX11 Games

• Total War: Warhammer III (DX11)

DX12 Games

• Chernoblyte (DX12)
• Call of Duty: Modern Warfare 2 (DX12)
• Cyberpunk 2077 (DX12)
• F1 2022 (DX12)
• Far Cry 6 (DX12)
• Resident Evil 4 (DX12)
• Red Dead Redemption 2(DX12)
• Horizon Zero Dawn (DX12)
• Metro Exodus PC Enhanced Edition (DX12)
• Dirt 5 (DX 12)

Vulkan Games

• DOOM Eternal (VK)

Hybrid Tests (3DMark)

• DLSS Feature Test
• Fire Strike Extreme
• Fire Strike Ultra
• Time Spy
• Time Spy Extreme

Nvidia Control Panel settings

Here are the global Nvidia Control Panel settings:

Noise, Temperatures, and Power Consumption

Unfortunately, we did not have time to check out the overclocking potential, but temperatures were controlled and the RTX 4070 runs very cool.

The RTX 4070 FE is quiet, and its fans never spin up at idle, even under a heavy or full load to be irritating or noticeable. When gaming it spins up to full often under very heavy loads – like Metro Exodus with all features turned on. It is as silent as the RTX 4070 Ti we tested previously.

Below is our thermal and approximate power consumption analysis on idle (Windows desktop, no user interactions.

Idle conditions

We ran into issues testing under load (our PSU died during near the end of our testing) but throughout our testing overall, the GPU was very efficient and in-line with our expectations. There is no heater meme to be found here.

Nvidia Performance Graphs and information

The biggest selling point for the RTX 4070 FE at $599 has to be DLSS 3. This is a gigantic leap – while some purists may not like frame generation – the performance increase and quality retention is simply stunning. On our QD-OLED display, we could barely see any image quality differences but performance increased dramatically. BTR plans on doing a full DLSS 3 gaming suite test in the near future.

In our testing, the RTX 4070 FE with DLSS 3 is almost 3 times faster than the RTX 2070 Super and it nearly doubles the performance of the RTX 3070 Ti while using less power.

Let’s head to the performance charts to compare the graphics performance of the RTX 4070 FE.

Gaming Performance Charts

Main Performance Gaming Summary Charts

Here are BTR’s summary charts of 19 games, six hybrid, and three non-synthetic tests. We note and specify the main in-game display, graphics, AA, and scaling settings on the performance summary charts below. The benches were run at 3440×1440.

FireStrike Graphics Score:
3070: 16807
4070: 21121
Timespy Graphics Score:
3070: 6611
4070: 17881
4070Ti: 22467
Averaged Framerates & Relative GPU Performance

Averaged Game Framerates

We averaged the aggregated FPS of all games and testing scenarios, and we represented the total game average FPS for each tested graphics card by the following chart:

On average, the RTX 4070 is faster than the RTX 3080 at 3340×1440. However, owners of this card can likely hold off on an upgrade unless they want lower power consumption, DLSS 3, and better ray tracing performance at a reasonable price. DLSS 3 is a game-changer for the games that have it. For games that do not and that are on DLSS 2 the RTX 3080 loses narrowly or wins in some older games in rare cases.

The 4070 Ti is a clear improvement but at a significantly higher buy-in price. In comparison to the rest of the 40 series lineup, the RTX 4070’s performance is unmatched for budget/value gamers.

Final Thoughts & Verdict

This has been an enjoyable exploration evaluating the new RTX 4070 FE. Overall, it is the best 40 series value for your money if the current adoption rate of DLSS and DLSS 3 continues. The performance gain is stunning. Without this technology, we can compare the RTX 4070 to the RTX 3080 from the previous generation with very distinct differences. Nvidia has improved performance with supplemental tech like DLSS 3, the architecture, cooling, and most of all, the pricing.

The RTX 4070 is compact and amazingly efficient compared to the RTX 30 series and its 40 series brothers. The idle fan stop is huge for us, and support for AV1 encoding is stellar for a lot of streamers at this price.

The RTX 4070 performed above the RTX 3080 in most cases and well above the RTX RTX 3070. The gap widened significantly with frame generation/DLSS 3 – So much so that this is a no-brainer. However, this is not a “wow” with the raster performance jump over the previous generation. Instead, the RTX 4070 is more efficient, more compact, and has much better features especially if you are still on a 10-series card. This is a worthy point in time with a card that is finally available at a reasonable price as a poster child for the generational leaps Nvidia is making with its technology and DLSS 3.

There is no early adopter woe here as there are many AAA titles to enjoy – right now – with DLSS 3 enabled, unlike with ray tracing at its launch. However, If DLSS 3 means little to you, we would hesitate to recommend upgrading from an RTX 3080 to an RTX 4070. However, the RTX 3070 user base will see enough significant performance gains to be able to make this a worthwhile consideration.

For a couple hundred dollars more you could buy an RTX 4070 Ti or a current AMD offering – but there is no card currently in this class and price that comes close to competing. The value of your dollar here will make any gamer happy. Especially at 1080p and 1440p, this card is a beast ready to serve your needs.

We do implore you to look at our upcoming DLSS 3 comparison of the current generation. This technology is finally allowing Nvidia to realize the dream that has been ray tracing. We can now maintain great performance while having the full suite of RTX features on an mid-level card. Safe to say, we give the RTX 4070 Founders Edition our Editor’s Choice award.

–Happy gaming! Per Audacia Ad Astra. Rest in Peace to Lance Reddick – our Commander Zavala.

BTR received a $1199 RTX 4080 Founders Edition (FE) from Nvidia but are late because we got a bad case of COVID-19 after the AMD Event in Las Vegas over two weeks ago. This Part 1 RTX 4080 review is a 10-VR game performance analysis versus the RTX 3080 Ti, the RTX 4090, and the RX 6900 XT using the Valve Index. Although the RTX 4080 is not a workstation card, we have included workstation SPEC benches and selected popular creative and synthetic apps. Next week, Part 2 will feature the performance of more than 40 games and we will include DLSS 3.0 results.

We will focus on raw VR performance to consider whether the new RTX 4080 Founders Edition at $1199 delivers a good value as a compelling upgrade from the last generation Ampere RTX 3080 Ti which also launched at $1199. We will also compare performance with Nvidia’s current flagship, the $1599 RTX 4090, and the former AMD flagship, the RX 6900 XT, which launched at $999. In addition to gaming, VR, and SPECworkstation3 GPU results, we have added creative results using Geekbench, the Blender 3.3.0 benchmark, and complete Sandra 2020 and AIDA64 GPGPU benchmark results plus some pro applications including Blender rendering and OTOY OctaneRender.

We benchmark using FCAT VR and Windows 11 Pro Edition 2H22 with Intel’s Core i9-13900K that we just upgraded from i9-12900K, and 32GB of T-FORCE DELTA RGB 6400MHz CL40 DDR5 2x16GB memory on an ASUS Prime-A Wi-Fi Z790 motherboard. All games and benchmarks are patched to their latest versions, and we use GeForce Game Ready 526.98 drivers for GeForce cards and Adrenalin 22.11.1 for the RX 6900 XT.

Let’s first take a quick look at the RTX 4080 Founders Edition before we go to the test configuration

The RTX 4080 FE

The RTX 4080 and RTX 4090 are externally physically identical twin cards. Refer to the RTX 4090 unboxing as the features and sizes are the same.

The primary differences lie inside in the RTX 4080’s cut down GPU chip capabilities with about 40% fewer CUDA cores and its lesser power draw requirements. Nvidia recommends a 850W minimum PSU for the 450W TDP RTX 4090 and includes a 4X Molex cable adapter while the 320W TDP RTX 4080 only needs a 750W minimum PSU and a 3X Molex cable adapter is included in that box.

Newer PSUs may offer the new PCIe Gen5 single cable connector instead of using a bulky quad or tri cable adapter. Nvidia advises that the adapter should be firmly inserted into the card before installing it in the PC as an improperly seated connection may cause overheating and melting of the connector.

Both the RTX 4090 and RTX 4080 Founders Edition cards are beautifully designed using a very unique industrial style, and they are each much larger than the RTX 3080 Ti which is itself an imposing card. However, these new 40×0 FE cards tend not to heat up like the smaller last gen cards and they are also much quieter under full load. Either card looks great installed inside a case.

Disassembly appears to be difficult and should give pause to any enthusiast who may have custom watercooling in mind. In fact, we think that watercooling is a waste for the RTX 4080 Founders Edition as it doesn’t have any thermal issues.

Let’s check out our test configuration.

Test Configuration

Test Configuration – Hardware

• Intel Core i9-13900KF (HyperThreading and Turbo boost at stock settings)
• ASUS Prime-A Z790 LGA1700 motherboard (Intel Z790 chipset, latest BIOS, PCIe 5.0, DDR5)
• T-Force Delta RGB PC5-51200 6400MHz DDR5 CL40 2x16GB kit, supplied by TeamGroup
• Valve Index, 90Hz / 100% SteamVR Render Resolution
• RTX 4080 16GB Founders Edition, stock clocks, supplied by Nvidia
• Gigabyte RX 6900 XT GAMING OC, GV-R69XTGAMING OC-16GD 16GB, factory clocks
• RTX 3090 24GB Founders Edition, factory clocks, supplied by Nvidia
• RTX 4090 24GB Founders Edition, stock clocks, supplied by Nvidia
• 2 x T-Force Cardea Ceramic C440; 2TB PCIe Gen 4 x4 NVMe SSDs (one for AMD/one for Nvidia)
• T-Force M200 4TB USB 3.2 Gen2x2 Type-C external SSD, supplied by TeamGroup
• Super Flower LedEx, 1200W Platinum 80+ power supply unit
• MSI MAG Series CORELIQUID 360R (AIO) 360mm liquid CPU cooler
• Corsair 5000D ATX mid-tower (plus 1 x 140mm fan & 2 x 120mm Noctua fans)
• BenQ EW3270U 32? 4K HDR 60Hz
• LG C1 48″ 4K OLED HDR 120Hz display

Test Configuration – Software

• GeForce 526.98 drivers for Nvidia cards, and Adrenalin 22.11.1 for the RX 6900 XT.
• High Quality, prefer maximum performance, single display, set in the Nvidia control panel.
• High Quality textures, all optimizations off in the Adrenalin control panel
• VSync is off in the control panel and disabled for each game
• AA enabled as noted in games; all in-game settings are Ultra Preset or highest with 16xAF always applied – no upscaling is used except for five DLSS games tested using the Quality preset.
• Highest quality sound (stereo) used in all games
• All games have been patched to their latest versions
• VR charts use frametimes in ms where lower is better, but we also compare “unconstrained framerates” which shows what a video card could deliver (headroom; higher is better)
• Windows 11 Pro edition; 22H2 clean install for GeForce and Radeon cards using separate identical NVMe SSDs.
• Latest DirectX
• SteamVR latest beta

Games

VR Games

• Assetto Corsa: Competizione
• Elite Dangerous
• F1 2022
• Kayak Mirage
• Moss: Book II
• No Man’s Sky
• Project CARS 2
• Skyrim
• Sniper Elite
• The Walking Dead: Saints & Sinners

Synthetic

• Time Spy & Time Spy Extreme (DX12)
• 3DMark FireStrike – Ultra & Extreme
• Superposition
• VRMark Blue Room
• AIDA64 GPGPU benchmarks
• Blender 3.3.0 benchmark
• Geekbench
• Sandra 2020 GPGPU Benchmarks
• SPECworkstation3
• SPECviewperfect 2020
• Octanebench
• FCAT VR benching tool
• OpenVR Benchmark tool

NVIDIA Control Panel settings

Here are the NVIDIA Control Panel settings.

Unfortunately, we also did not have time to check out overclocking, but temperatures and noise levels are much lower than the RTX 3090 or RTX 3080 Ti FEs. We plan to follow up with a 40-plus pancake gaming review including DLSS 3.

Let’s check out performance using 10 VR games plus Workstation and creative benches.

First, we look at VR performance.

10 VR Games

For this review, we benchmarked the Valve Index using FCAT VR and set the SteamVR render resolution to 100% (2016×2240) which uses a factor of 1.4X (the native resolution is 1440×1600) to compensate for lens distortion and to increase clarity. The Index is still considered one of the best overall headsets due to its outstanding tracking and solid feature set, and we are going to compare the performance of the RX 4080 versus the RX 4090, the RX 3080 Ti, and against the RX 6900 XT at each game’s Ultra/Highest settings.

IMPORTANT: BTR’s charts use frametimes in ms where lower is better, but we also compare “unconstrained framerates” which shows what a video card could deliver (headroom) if it wasn’t locked to either 90 FPS or to 45 FPS by the HMD. In the case of unconstrained FPS, measuring just one important performance metric, faster is better.

We had planned to include Star Wars: Squadrons and MS Flight Simulator 2020, but neither game records properly with FCAT VR. So let’s individually look at our 10 sim-heavy VR games’ performance using FCAT VR.

First up, Assetto Corsa: Competizione.

Assetto Corsa: Competizione (ACC)

BTR’s sim/racing editor, Sean Kaldahl created the replay benchmark run that we use for both the pancake game and the VR game. It is run at night with 20 cars, lots of geometry, and the lighting effects of the headlights, tail lights, and everything around the track looks spectacular.

Just like with Project CARS, you can save a replay after a race. Fortunately, the CPU usage is the same between a race and its replay so it is a reasonably accurate benchmark using the Circuit de Spa-Francorchamps.

iRacing may be more accurate or realistic, but Assetto Corsa: Competizione has some appeal because it feels more real than many other racing sims. It delivers the sensation of handling a highly-tuned racing machine driven to its edge.

Here are the ACC FCAT VR frametimes using VR Ultra using the RTX 3080 Ti, the RTX 4080, and the RTX 4090. Unfortunately, the latest ACC patch made it impossible to bench the RTX 6900 XT.

Here are the details are reported by FCAT-VR:

The RTX 3080 Ti managed 91.83 unconstrained FPS with 4649 (36%) synthesized frames with 1 dropped frame and 1 Warp miss.

The RTX 4080 delivered 118.42 unconstrained FPS with 207 (2%) synthesized frames with 1 dropped frame and 1 Warp miss.

The RTX 4090 achieved 164.03 unconstrained FPS together with 1 synthetic frame but with no dropped frames nor Warp misses.

The ACC racing experience is best with the RTX 4090 although the RTX 4080 delivers a nearly constant 90 FPS on the Epic VR preset unlike the RTX 3080 Ti which requires more than one-third of its frames to be synthesized. Only the RTX 4090 has the performance headroom to increase the render resolution to above 100% or use the Index’ faster refresh rate (120Hz/144Hz) or even to use a higher resolution headset like the Reverb G2.

Next, we check out Elite Dangerous.

Elite Dangerous (ED)

Elite Dangerous is a popular space sim built using the COBRA engine. It is hard to find a repeatable benchmark outside of the training missions.

A player will probably spend a lot of time piloting his space cruiser while completing a multitude of tasks as well as visiting space stations and orbiting a multitude of different planets. Elite Dangerous is also co-op and multiplayer with a dedicated following of players.

We picked the Ultra Preset and we set the Field of View to its maximum. The RX 6900 XT wouldn’t run with the latest driver according to the error message.

Here are the frametimes.

Here are the details as reported by FCAT-VR:

The RTX 3080 Ti managed 182.93 unconstrained FPS with 2 synthesized frames plus 1 dropped frame and 1 Warp miss.

The RTX 4080 delivered 230.98 unconstrained FPS with 1 synthesized frame and 1 dropped frame and 1 Warp miss.

The RTX 4090 brings 296.16 unconstrained FPS together with 2 synthetic frames but with 2 dropped frames and 2 Warp misses.

The experience playing Elite Dangerous at Ultra settings is not perceptibly different on any tested video card but the RTX 4090 has a lot more performance headroom to increase the render resolution or to use a higher resolution headset like the Reverb G2 or the Vive Pro 2.

Let’s look at our newest VR sim, F1 2022.

F1 2022

Codemasters has captured the entire Formula 1 2021 season racing in F1 2022, and the VR immersion is good. The graphics are customizeable and solid, handling and physics are good, the AI is acceptable, the scenery is outstanding, and the experience ticks many of the necessary boxes for a racing sim.

Here is the frametime plot for F1 2022.

Here are the details as reported by FCAT-VR.

The RX 6900 XT delivered 132.21 unconstrained FPS with no synthesized or dropped frames nor Warp misses.

The RTX 3080 Ti managed 152.67 unconstrained FPS with no synthesized or dropped frames nor Warp misses.

The RTX 4080 achieved 200.24 unconstrained FPS with no synthesized or dropped frames nor Warp misses.

The RTX 4090 delivered 254.72 unconstrained FPS together with 3 synthetic frames plus with 3 dropped frames and 3 Warp misses.

The experience playing F1 2022 using the Ultra preset is not very different on any of these video cards but the RTX 4090 and RTX 4080 have considerably more performance headroom to use 120Hz/144Hz or to use a higher resolution headset.

Kayak VR: Mirage

The outstanding near-photorealistic visual fidelity really sets Kayak VR: Mirage apart from other simulators. It boasts a wide range of locales with day/night/sunset options offering tropical, icy, desert, and even stormy scenarios with trips to Costa Rica, Antarctica, Norway, and Australia and occasional interactions with wildlife. It can be played as a relaxing sim or as a strenous workout with competitive time trials which offer asynchronous multiplayer and ranking on global leaderboards.

We benchmark at 100% resolution with the highest “Cinematic” in-game settings but do not use DLSS or FSR.

Here is the frametime plot for Kayak VR: Mirage.

Here are the FCAT-VR details.

The RX 6900 XT delivered 189.78 unconstrained FPS with no synthesized frames or dropped frames nor Warp misses.

The RTX 3080 Ti managed 201.89 unconstrained FPS with 1 synthetic frame plus 1 dropped frame and 1 Warp miss.

The RTX 4080 delivered 257.16 unconstrained FPS with 1 synthesized and 1 dropped frame and 1 Warp miss.

The RTX 4090 got 329.35 unconstrained FPS together with 1 synthetic frame and 1 dropped frame plus 1 Warp miss.

Kayak VR: Mirage looks fantastic at 100% resolution with maximum settings and would be well-suited for play on the Reverb G2 with any of our test cards.

Next, we look at Moss: Book II.

Moss: Book II

Moss: Book II is an amazing VR experience with much better graphics than the original game. It’s a 3rd person puzzle adventure game played seated that offers a direct physical interaction between you (the Reader) and your avatar, Quill, a mouse that bring real depth to the story. Extreme attention has been paid to the tiniest details with overall great art composition and outstanding lighting that make this game a must-play for gamers of all ages.

Moss II boasts very good visuals and we use the in-game highest settings.

Here are the frametimes plots of our four cards.

Here are the details are reported by FCAT-VR:

The RX 6900 XT delivered 260.40 unconstrained FPS with no synthesized or dropped frames nor Warp misses.

The RTX 3080 Ti managed 242.83 unconstrained FPS with no synthesized or dropped frames nor Warp misses.

The RTX 4080 delivered 308.44 unconstrained FPS with 1 synthetic and 1 dropped frame and 1 Warp miss.

The RTX 4090 achieved 436.34 unconstrained FPS no synthesized or dropped frames nor Warp misses.

The experience playing Moss II on the Valve Index is identical between cards. It is not demanding so it can be played on a high resolution headset like the Reverb G2 or the Vive Pro 2 and supersampling may also be used.

Next, we will check out another demanding VR game, No Man’s Sky.

No Man’s Sky (NMS)

No Man’s Sky is an action-adventure survival single and multiplayer game that emphasizes survival, exploration, fighting, and trading. It is set in a procedurally generated deterministic open universe, which includes over 18 quintillion unique planets using its own custom game engine.

The player takes the role of a Traveller in an uncharted universe by starting on a random planet with a damaged spacecraft equipped with only a jetpack-equipped exosuit and a versatile multi-tool that can also be used for defense. The player is encouraged to find resources to repair his spacecraft allowing for intra- and inter-planetary travel, and to interact with other players.

Here is the No Man’s Sky Frametime plot. We set the settings to Maximum which is a step over Ultra including setting the anisotropic filtering to 16X and upgrading to FXAA+TAA. Since DLSS is available for RTX cards and the Quality setting improves performance without impacting image quality, we used it. Updated: We did not use any upscaling method.

Here are the FCAT-VR details of our comparative runs.

The RX 6900 XT brought 104.13 unconstrained FPS with 1601 (23%) synthesized frames plus 9 dropped frames and 9 Warp misses.

The RTX 3080 Ti managed 119.88 unconstrained FPS with 14 synthetic frames and with 1 dropped frame and 1 Warp miss.

The RTX 4080 delivered 159.10 unconstrained FPS with 2 synthesized frames but no dropped frames nor Warp misses.

The RTX 4090 achieved 201.96 unconstrained FPS together with 17 synthetic frames but with no dropped frames nor Warp misses.

The experience playing No Man’s Sky using the highest settings is not very different using the three GeForce video cards but RX 6900 XT gamers may want to lower some individual settings to remain above 90 FPS. The RTX 4080 and RTX 4090 have enough performance headroom to increase the refresh rate, render resolution, or to perhaps use a higher resolution headset.

Let’s continue with another VR game, Project CARS 2, that we still like better than its successor even though it is no longer available for online play.

Project CARS 2 (PC2)

There is still a sense of immersion that comes from playing Project CARS 2 in VR using a wheel and pedals. It uses its in-house Madness engine, and the physics implementation is outstanding.

Project CARS 2 offers many performance options and settings and we prefer playing with SMAA Ultra rather than to use MSAA.

We used maximum settings including for Motion Blur.

Here is the frametime plot.

Here are the FCAT-VR details.

The RX 6900 XT delivered 163.67 unconstrained FPS with 1 synthesized and 1 dropped frames plus 1 Warp miss.

The RTX 3080 Ti managed 142.16 unconstrained FPS with 2 synthetic frames plus 2 dropped frames and 2 Warp misses.

The RTX 4080 got 200.88 unconstrained FPS with no synthesized frames nor dropped frames and no Warp misses.

The RTX 4090 achieved 253.50 unconstrained FPS together with 3 synthetic frames plus 2 dropped frames and 2 Warp misses.

The experience playing Project CARS 2 using maximum settings is similar for all four video cards but the RTX 4090 and RTX 4080 have far more performance headroom to increase the frequency to 120Hz or to use a higher resolution headset.

Next we will check out a classic VR game, Skyrim VR.

Skyrim VR

Skyrim VR is an older game that is no longer supported by Bethesda, but fortunately the modding community has adopted it. It is not as demanding as many of the newer VR ports so its performance is still very good on maxed-out settings using its Creation engine.

We benchmarked vanilla Skyrim using its highest settings plus we increased the in-game Supersample option to maximum.

Here are the frametime results.

Here are the details of our comparative runs as reported by FCAT-VR.

The RX 6900 XT provided 162.13 unconstrained FPS with no synthesized or dropped frames nor Warp misses.

The RTX 3080 Ti managed 194.63 unconstrained FPS with no synthesized or dropped frames nor Warp misses.

The RTX 4080 achieved 239.08 unconstrained FPS with 2 synthetic frames plus 2 dropped frames and 1 Warp miss.

The RTX 4090 delivered 337.76 unconstrained FPS together with 2 synthetic frame and with 2 dropped frames plus 1 Warp miss.

All cards deliver an identical vanilla Skyrim VR experience with a ton of extra performance headroom to add mods and, in addition, to raise the render resolution using the two faster cards.

Next we check out Sniper Elite VR.

Sniper Elite VR

Sniper Elite VR’s visuals are decent with good texture work that is well-realized. The building architecture and panoramas look good, explosions are convincing and the weapons convey a sense of weight, although not achieving realism. It is primarily an arcade style sniping game featuring its signature X-Ray kill cam, but it offers multiple ways to achieve goals including with explosives and by using three other main weapon choices besides your rifle.

We benchmarked using the Highest settings.

Here is the frametime plot.

Here are the details:

The RX 6900 XT delivered 222.93 unconstrained FPS with no synthesized or dropped frames nor Warp misses.

The RTX 3080 Ti got 239.07 unconstrained FPS with no synthesized or dropped frames nor Warp misses.

The RTX 4080 delivered 223.33 unconstrained FPS with no synthesized or dropped frames nor Warp misses.

The RTX 4090 brought 318.03 unconstrained FPS together with 1 synthetic and 1 dropped frames and 1 Warp miss.

All four cards deliver a similar playing experience on High with the faster two cards offering more performance headroom. We recommend that any performance headroom be used for increasing the SteamVR render resolution. Since the RTX 4080 only matches the RX 6900 XT and is behind the RTX 3080 Ti, we suspect it may be a driver issue.

Last up, The Walking Dead: Saints & Sinners.

The Walking Dead: Saints & Sinners

The Walking Dead: Saints & Sinner is the last of BTR’s 10 VR game benching suite. It is a first person survival horror adventure RPG with a strong emphasis on crafting. Its visuals using the Unreal 4 engine are very good and it makes good use of physics for interactions.

We benchmarked Saints and Sinners using its High preset and we left the Pixel Density at 100%. Here is the frametime chart.

Here are the details as reported by FCAT-VR.

The RX 6900 XT delivered 240.00 unconstrained FPS with 1 synthetic frames and 1 dropped frames and 1 Warp miss.

The RTX 3080 Ti managed 222.18 unconstrained FPS with 1 synthetic frames and 1 dropped frames and 1 Warp miss.

The RTX 4080 got 260.94 unconstrained FPS with 1 synthetic frames and 1 dropped frames and 1 Warp miss.

The RTX 4090 achieved 366.41 unconstrained FPS together with 6 synthetic frames and with 4 dropped frames and 4 Warp misses.

Playing The Walking Dead: Saints & Sinners using the in-game Ultra preset and 100% Pixel Density is similar for all 4 cards but there is significantly more performance headroom for the RTX 4090 and to a lesser extent the RTX 4080 for increasing the render resolution or using a higher resolution headset.

Let’s check out synthetic VR tests and unconstrained framerates.

Unconstrained Framerates & Synthetic VR Benchmarks

The following chart summarizes the overall Unconstrained Framerates (the performance headroom) of our two cards using our 10 VR test games. The preset is listed on the chart and higher is better. In addition, we present three synthetic VR benchmarks.

The RX 4080 FE averages close to one-third higher unconstrained frames for many VR benchmarks and sits in between the RTX 3080 Ti and the RTX 4090 FE in this important performance metric. However, unconstrained framerates are just one metric that has to be taken together with the frametime plots to have real meaning.

It is clear that the RTX 4090 and the RTX 4080 are ready for higher resolution headsets than the Valve Index. In many cases, either card may be able to use the higher refresh rates of 120Hz/144Hz for fast-paced and/or sim racing games.

Creative, Pro & Workstation Apps

Let’s look at Creative applications next to see if the RTX 4080 is a good upgrade from the RTX 3080 Ti or RX 6900 XT. We test starting with Geekbench.

GeekBench

GeekBench is an excellent CPU/GPU benchmarking program which runs a series of tests and times how long a GPU (in this case) takes to complete its tasks. It benchmarks OpenCL, Vulkan, and CUDA performance

OpenCL, Vulkan, and CUDA Performance

The RTX 4080 OpenCL, Vulkan, and CUDA performance are charted below.

Next we test the summary charts below show the overall comparative performance scores.

Again, the RTX 4090 performance is outstanding.

Next up, Blender benchmark.

Blender 3.3.0 Benchmark

Blender is a very popular open source 3D content creation suite. It supports every aspect of 3D development with a complete range of tools for professional 3D creation.

For the following chart, higher is better as the benchmark renders a scene multiple times and gives the results in samples per minute.

Blender’s benchmark performance is highest using the RTX 4090, and often the amount of time saved is substantial over using the next fastest card, the RTX 4080.

Next, we look at the OctaneBench.

OTOY Octane Bench

OctaneBench allows you to benchmark your GPU using OctaneRender. The hardware and software requirements to run OctaneBench are the same as for OctaneRender Standalone.

We run OctaneBenc 2020.1.5 for Windows and here are the RTX 4080’s complete results and overall score of 946.30

Here is the summary comparing the three cards that can run this render benchmark.

The RTX 4090 is a beast of a card when used for rendering and the RTX 4080 sits in between, above the RTX 3080 Ti.

Next, we move on to AIDA64 GPGPU benchmarks.

AIDA64

AIDA64 is an important industry tool for benchmarkers. Its GPGPU benchmarks measure performance and give scores to compare against other popular video cards.

AIDA64’s benchmark code methods are written in Assembly language, and they are well-optimized for every popular AMD, Intel, NVIDIA and VIA processor by utilizing the appropriate instruction set extensions. We use the Engineer’s full version of AIDA64 courtesy of FinalWire. AIDA64 is free to to try and use for 30 days.

Here is the chart summary of the AIDA64 GPGPU benchmarks with the RTX 4090, the RTX 4080, the RTX 3080 Ti, and the RX 6900 XT side-by-side.

Generally the RTX 4090 is faster at almost all of AIDA64’s GPGPU benchmarks than the other cards with the RTX 4080 in second place. So let’s look at Sandra 2020 next.

SiSoft Sandra 2020

To see where the CPU, GPU, and motherboard performance results differ, there is no better tool than SiSoft’s Sandra 2020. SiSoftware SANDRA (the System ANalyser, Diagnostic and Reporting Assistant) is a excellent information & diagnostic utility in a complete package. It is able to provide all the information about your hardware, software, and other devices for diagnosis and for benchmarking. Sandra is derived from a Greek name that implies “defender” or “helper”.

There are several versions of Sandra, including a free version of Sandra Lite that anyone can download and use. 20/21-R16a is the latest version, and we are using the full engineer suite courtesy of SiSoft. Sandra 2020 features continuous multiple monthly incremental improvements over earlier versions of Sandra. It will benchmark and analyze all of the important PC subsystems and even rank your PC while giving recommendations for improvement.

With the above in mind, we ran Sandra’s intensive GPGPU benchmarks and charted the results summarizing them. The performance results of the RTX 4080 and RTX 4090 are compared with the performance results of the RTX 3080 Ti, and the RX 6900 XT.

Second only to the RTX 4090, the RTX 4080 is faster than the RTX 3080 Ti and it distinguishes itself in every area – Processing, Cryptography, Financial and Scientific Analysis, Image Processing, and Bandwidth.

Next up, SPEC benchmarks.

SPECworkstation3.1 Benchmarks

All the SPECworkstation 3 benchmarks are based on professional applications, most of which are in the CAD/CAM or media and entertainment fields. All of these benchmarks are free except for vendors of computer-related products and/or services.

The most comprehensive workstation benchmark is SPECworkstation 3. It’s a free-standing benchmark which does not require ancillary software. It measures GPU, CPU, storage and all other major aspects of workstation performance based on actual applications and representative workloads. We only tested the GPU-related workstation performance as checked in the image above.

Here are our RTX 4080 raw SPECworkstation 3.1 raw scores:

Below are the SPECworkstation 3.1 RTX 4090 results summarized in a chart along with the three competing cards, the RTX 4090, the RTX 3080 Ti, and the RTX 6900 XT. Higher is better since we are comparing scores.

The RTX 4090 is not a workstation card, yet it uses brute force to win most of the benches against the other cards with the RTX 4080 in second place.

Finally, we benchmark using SPECviewperfect GPU benches.

SPECviewperf 2020 GPU Benches

The SPEC Graphics Performance Characterization Group (SPECgpc) has released a 2020-22 version of its SPECviewperf benchmark that features updated viewsets, new models, support for up to 4K display resolutions, and improved set-up and results management. We use the highest available 3800×2120 display resolution for high end cards.

Here are SPECviewperf 2020 GPU RTX 4080 benchmarks summarized in a chart together with our three competing cards.

Although we see three architectures with different strengths and weaknesses, the RTX 4090 is a beast in SPEC benchmarks followed by the RTX 4080 in a solid second place.

After seeing the totality of the benches, creative users may choose to upgrade their existing systems with a new RTX 4080 or 4090 based on the performance increases and the associated increases in productivity that they require. The question to buy a RTX 4080 should be based on the workflow and requirements of each user as well as budget. Time is money to a professional depending on how these apps are used.

Let’s head to our conclusion.

Final Thoughts

Besides suffering with COVID-19, this has been an enjoyable exploration evaluating the new Ada Lovelace RTX 4080 FE versus the RTX 4090 FE, the RTX 3080 TI FE, and Gigabyte RTX 6900 XT Gaming OC. The RTX 4080 performed brilliantly performance-wise, blowing away its other two former flagship competitors and slotting in almost one-third slower than the RTX 4090.

The RTX 4080 at $1199 is a decent upgrade from the $1199 RTX 3080 Ti although it is not as impressive as upgrading from a RTX 3090 ($1499) to a RTX 4090 ($1599). If a VR enthusiast wants the very fastest card, then the RTX 4090 is the best choice for high resolution VR headsets and for creative/workstation apps. For $400 less, the RTX 4080 is a solid choice and it will provide a performance uplift over the RTX 3080 Ti, but not as dramatically as the RTX 4090.

In the last Ampere generation, the $699 RTX 3080 provided the best value, but almost unbelievably, the RTX 4090 provides the best performance bang for buck in Nvidia’s new Ada Lovelace lineup so far. We also notice that there is plenty of performance room for a possible future RTX 4080 Ti to slot in between the RTX 4080 and the RTX 4090.

We are very impressed with the RTX 4080 raw performance after testing it over the past few days in VR. It currently stands as the second fastest video card in the world and would be a solid choice for a high resolution VR headset.

The Founders Edition of the RTX 4080 is well-built, solid, good-looking, and it stays cool and quiet even when overclocked – the big card does not get hot like the RTX 3090 and under load it is quieter than the RTX 3080 Ti FE. The RTX 4080 Founders Edition offers a big performance improvement over the previous $1199 RTX 3080 FE for VR and pro apps although it doesn’t give as great a value as when the RTX 3080 FE launched for $699 two years ago.

However, we cannot give any final verdict now as there is a lot more RTX 4080 benchmarking results on the way next week. We will more extensively test the RTX 4080, RTX 4090, RTX 3080 Ti, and RX 6900 XT using 40-plus pancake games including testing DLSS 3 to determine overall value. Future reviews will hopefully test AMD’s upcoming 7000 series cards and we also plan to test the role of the CPU for VR performance. Stay tuned to BTR!

Happy Gaming!

BTR recently received a RTX 4090 Founders Edition (FE) from Nvidia and we have been testing it for the past ten days by using 45 pancake and VR games plus GPGPU benchmarks. In addition, although the RTX 4090 is not a workstation card, we have added workstation SPEC benches and selected popular creative and synthetic apps. Although this new flagship Ada Lovelace card arrives with multiple new features including DLSS 3, this review will focus on testing raw performance, not upscaling.

The RTX 4090 is touted as a beast in every way by Nvidia as the fastest video card for gaming with up to 4X the performance of the Ampere flagship or up to 2X the performance without using DLSS 3. The RTX 4090 boasts 83 Shader-TFLOPS, 191 RT-TFLOPS, and 1.3 Tensor petaFLOPS. It achieves over 1 TB/s memory bandwidth using 24 GB of GDDR6X memory. Its 72 MB L2 cache offers a 12X improvement over the RTX 3090 Ti.

We will focus on RTX 4090 raw performance as well as consider whether the new RTX 4090 Founders Edition at $1599 delivers a good value as a compelling upgrade from the RTX 3090 which launched at $1499 two years ago. We will also compare performance with the RX 6900 XT. In addition to gaming, VR, and SPECworkstation3 GPU results, we have added creative results using Geekbench, the Blender 3.3.0 benchmark, and complete Sandra 2020 and AIDA64 GPGPU benchmark results plus some pro applications including Blender rendering and OTOY OctaneRender.

We benchmark using Windows 11 Pro Edition 2H22 at 2560×1440 and at 3840×2160 using Intel’s Core i9-12900K and 32GB of T-FORCE DELTA RGB 6400MHz CL40 DDR5 2x16GB memory on an ASUS ROG Maximus Apex motherboard. All games and benchmarks are the latest versions, and we use the latest GeForce Game Ready press drivers for games and for testing pro apps, and Adrenalin 22.10.1 for the RX 6900 XT benching.

Let’s first take a quick look at the RTX 4090 Founders Edition before we go to the test configuration

The RTX 4090 Founders Edition Unboxing

Just like as with all RTX 3000 Founders Edition cards, the RTX 4090 comes in a similar “shoebox” style where the card inside lays flat at an slight incline for display. However, the RTX 4090 box is much thicker and probably close to 50% larger.

The system requirements, contents, and warranty information are printed on the bottom of each box. The RTX 4090 requires an 850W power supply unit – 100W more than the RTX 3090 – and the case must have space for a 304mm (L) x 137mm (W) x 61mm (H) three-slot card. However, it easily fits in our Corsair 5000D ATX mid-tower. The extra thick packing of the box protects the card. Inside the box and beneath the card are warnings, a quick start guide, and warranty information, plus the PCIe Gen5 power cable adapter to quad PCIe 8-pin cables that will be required to connect the RTX 4090 to most PSUs.

The RTX 4090 Founders Edition exudes a premium and solid feel from its industrial design. It is a very heavy 3-slot card and we use two thumbscrews to lock it down, taking care not to damage our PCIe slot.

The power connector adapter requires three or four molex cables from the PSU to operate; the fourth cable is for overclocking. Newer PSUs may offer the new PCIe Gen5 single cable connector instead of a bulky quad cable.

Turning the card over, we see a similar design of the Ada Lovelace FEs to the Ampere cards with a fan also on the other side.

The IO panel has a very large air vent and four connectors. The connectors are similar to the Founders Edition of the RTX 3090. Three DisplayPort 1.4 connectors are included, and the HDMI 2.1 connector allows for 4K/120Hz or 8K/60Hz over a single HDMI cable.

The other end offers a removable plate for enthusiasts to use a support bracket.

The RTX 4090 Founders Edition is a beautiful card with a very unique industrial style, and it’s larger than the RTX 3090 which is itself an imposing card. However, the larger Ada Lovelace card tends not to heat up like the RTX 3090 and it is also much quieter under full load. It looks great installed inside a case.

Disassembly appears to be very difficult and should give pause to any enthusiast who may have custom watercooling in mind. In fact, we think that watercooling is a waste for the Founders Edition as it doesn’t have any thermal issues.

Let’s check out our test configuration.

Test Configuration

Test Configuration – Hardware

• Intel Core i9-12900KF (HyperThreading and Turbo boost at stock settings)..
• ASUS ROG Maximus Z690 Apex LGA1700 motherboard (Intel Z690 chipset, latest BIOS, PCIe 5.0, DDR5)
• T-Force Delta RGB PC5-51200 6400MHz DDR5 CL40 2x16GB kit, supplied by TeamGroup
• Valve Index, 90Hz
• Gigabyte RX 6900 XT GAMING OC, GV-R69XTGAMING OC-16GD 16GB, factory clocks
• RTX 3090 24GB Founders Edition, factory clocks, supplied by Nvidia
• RTX 4090 24GB Founders Edition, stock clocks, supplied by Nvidia
• 2 x T-Force Cardea Ceramic C440; 2TB PCIe Gen 4 x4 NVMe SSD (one for AMD/one for Nvidia)
• T-Force M200 4TB USB 3.2 Gen2x2 Type-C Portable SSD, supplied by TeamGroup
• Super Flower LedEx, 1200W Platinum 80+ power supply unit
• MSI MAG Series CORELIQUID 360R (AIO) 360mm liquid CPU cooler
• Corsair 5000D ATX mid-tower (plus 1 x 140mm fan; 2 x 120mm Noctua fans)
• BenQ EW3270U 32? 4K HDR 60Hz FreeSync monitor
• LG C1 48″ 4K OLED HDR 120Hz display/TV

Test Configuration – Software

• GeForce press drivers for the RTX 3090 and RTX 4090, and Adrenalin 22.10.1 for the RX 6900 XT.
• High Quality, prefer maximum performance, single display, set in the Nvidia control panel.
• High Quality textures, all optimizations off in the Adrenalin control panel
• VSync is off in the control panel and disabled for each game
• AA enabled as noted in games; all in-game settings are Ultra Preset or highest with 16xAF always applied – no upscaling is used except for five DLSS games tested using the Quality preset.
• Highest quality sound (stereo) used in all games
• All games have been patched to their latest versions
• Gaming results show average frame rates in bold including minimum frame rates shown on the chart next to the averages in a smaller italics font where higher is better. The minimums are expressed by 1% lows (99th-percentile) in FPS
• Windows 11 Pro edition; 22H2 clean install for GeForce and Radeon cards on separate identical NVMe SSDs. DX11 titles are run under the DX11 render path. DX12 titles are run under DX12, and seven games use the Vulkan API.
• Latest DirectX

Games

Vulkan

• Sniper Elite
• DOOM Eternal
• Red Dead Redemption 2
• Ghost Recon: Breakpoint
• Wolfenstein Youngblood
• World War Z
• Strange Brigade

DX12

• Spiderman: Remastered
• F1 2022
• Ghostwire: Tokyo
• Elden Ring
• God of War
• Dying Light 2
• Forza Horizon 5
• Call of Duty: Vanguard
• Marvel’s Guardians of the Galaxy
• Far Cry 6
• DEATHLOOP
• Chernobylite
• Resident Evil Village
• Metro Exodus Enhanced Edition
• Hitman 3
• Godfall
• DiRT 5
• Assassin’s Creed Valhalla
• Cyberpunk 2077
• Watch Dogs: Legions
• Horizon Zero Dawn
• Death Stranding
• Borderlands 3
• Tom Clancy’s The Division 2
• Civilization VI – Gathering Storm Expansion
• Battlefield V
• Shadow of the Tomb Raider

DX11

• Overwatch 2
• Total War: Warhammer III
• Days Gone
• Crysis Remastered
• Destiny 2 Shadowkeep
• Total War: Three Kingdoms
• Grand Theft Auto V

VR Games

• Assetto Corsa: Competizione
• Elite Dangerous
• F1 2022
• No Man’s Sky
• ProjectCARS 2

Synthetic

• TimeSpy & Time Spy Extreme (DX12)
• 3DMark FireStrike – Ultra & Extreme
• Superposition
• VRMark Blue Room
• AIDA64 GPGPU benchmarks
• Blender 3.3.0 benchmark
• Geekbench
• Sandra 2020 GPGPU Benchmarks
• SPECworkstation3
• SPECviewperfect 2020
• Octanebench
• FrameView benching tool
• OCAT benching tool

NVIDIA Control Panel settings

Here are the NVIDIA Control Panel settings.

Unfortunately, we did not have time to check out overclocking, but temperatures and noise levels are lower than the RTX 3090 FE. We plan to follow this review up with a VR review, an overclocking review, and a DLSS 3 review.

Let’s check out performance using 41 pancake and 5 VR games plus Workstation and creative benches on the next page.

Performance summary charts & graphs

Main Performance Gaming Summary Charts

Here are the summary charts of 41 games and 6 synthetic tests. The highest settings are always chosen, ray tracing is enabled for all games that offer it, and the settings are listed on the chart. The benches were run at 2560×1440 and at 3840×2160 as it is pointless to test at 1920×1080 with such a powerful card. In fact, we see CPU bottlenecking at 1440P for certain games.

Three cards are compared and they are listed in order starting with the RTX 4090, the RX 6900 XT, and the RTX 3090. We do not have a RX 6950 XT or a RTX 3090 Ti and no other cards are in this class.

All results, except for synthetic scores, show average framerates, and higher is better. Minimum framerates are next to the averages in italics and in a slightly smaller font. Minimum framerates are expressed by the 99th-percentile (1% lows) and higher is better.

The RTX 4090 offers an overall 160% to 180% improvement over the RTX 3090’s baseline performance (at 100%) depending on the resolution and individual game, and in several examples it offers nearly a 200% improvement!

All of the games that we tested ran well on the RTX 4090. Although some games show less of a performance increase than others due to being CPU bound even at 1440P, it is a blowout and the RTX 4090 FE wins every game benchmark over the RTX 3090 and RX 6900 XT – it crushes the former Ampere and RDNA2 flagships. This is achieved with no upscaling whatsoever!

The RTX 4090 is the first single-GPU card that is truly suitable for 4K/60+ FPS using ultra/maxed-out ray traced settings for most modern demanding games without any upscaling, and it’s probably even solid for 4K/120 FPS using Quality DLSS which has equivalent or better visuals than the native image.

Next we look at five RTX/DLSS enabled games, each using maximum ray traced settings and the highest Quality DLSS.

RTX/DLSS Benchmarks

The RTX 4090 FE maintains its performance dominance over the RTX 3090 FE and pulls even further away when Quality DLSS is enabled.

Using Quality DLSS, we can see that the RTX 4090 will take advantage of an LG C1 4K/120Hz panel using the most demanding ray traced modern games. From testing DLSS 2 exhaustively, we note that the Quality setting at 4K is visually equal to or better than the native image.

We only had a little time to check out DLSS 3 which upscales far better than DLSS 2 and looks just as good. We believe that DLSS 3 will prove especially advantageous for the less powerful upcoming Ada Lovelace cards and will devote an upcoming review to it.

Next, we look at VR performance.

VR Games

For this review we benchmarked the Valve Index and set the SteamVR render resolution to 150% (2758×2740) which is considered ideal, if overkill, to compensate for lens distortion, and it’s well above our usual benchmarking render resolution at 100%. This higher render resolution gives the visuals exceptional clarity. The Index is still considered one of the best overall headsets due to its outstanding tracking and solid feature set, and we are going to compare the performance of the RX 4090 versus the RX 3090 at each game’s Ultra/Highest settings.

IMPORTANT: BTR’s charts use frametimes in ms where lower is better, but we also compare “unconstrained framerates” which shows what a video card could deliver (headroom) if it wasn’t locked to either 90 FPS or to 45 FPS by the HMD. In the case of unconstrained FPS, measuring just one important performance metric, faster is better.

Let’s individually look at our 5 sim-heavy VR games’ performance using FCAT VR. All of these games were benchmarked at 150% SteamVR resolution as we compare the stock-clocked RTX 4090 FE with the stock RTX 3090 FE using the Valve Index and FCAT VR.

First up, Assetto Corsa Competizione.

Assetto Corsa: Competizione (ACC)

BTR’s sim/racing editor, Sean Kaldahl created the replay benchmark run that we use for both the pancake game and the VR game. It is run at night with 20 cars, lots of geometry, and the lighting effects of the headlights, tail lights, and everything around the track looks spectacular.

Just like with Project CARS, you can save a replay after a race. Fortunately, the CPU usage is the same between a race and its replay so it is a reasonably accurate benchmark using the Circuit de Spa-Francorchamps.

iRacing may be more accurate or realistic, but Assetto Corsa: Competizione has some appeal because it feels more real than many other racing sims. It delivers the sensation of handling a highly-tuned racing machine driven to its edge. We test the VR Ultra preset.

VR Ultra

Here are the ACC FCAT VR frametimes using VR Ultra.

Here are the details are reported by FCAT-VR:

The RTX 3090 delivered 46.26 unconstrained FPS with 7884 (62%) synthesized frames with 1 dropped frame and 1 Warp miss.

The RTX 4090 delivered 78.63 unconstrained FPS together with 6346 (50%) synthetic frames but with no dropped frames nor Warp misses.

The ACC racing experience is better with the RTX 4090. However, the only way that the RTX 4090 can play on VR Epic is to lower the SteamVR render resolution to 100% as it is best to have no synthesized frames.

Next, we check out Elite Dangerous next.

Elite Dangerous (ED)

Elite Dangerous is a popular space sim built using the COBRA engine. It is hard to find a repeatable benchmark outside of the training missions.

A player will probably spend a lot of time piloting his space cruiser while completing a multitude of tasks as well as visiting space stations and orbiting a multitude of different planets. Elite Dangerous is also co-op and multiplayer with a dedicated following of players.

We picked the Ultra Preset and we set the Field of View to its maximum. Here are the frametimes.

Here are the details as reported by FCAT-VR:

The RTX 3090 delivered 137.33 unconstrained FPS with 1 synthesized frames with 1 dropped frame and 1 Warp miss.

The RTX 4090 delivered 236.17 unconstrained FPS together with no synthetic frames and with no dropped frames nor Warp misses.

The experience playing Elite Dangerous at Ultra settings is not perceptibly different on either video card but the RTX 4090 has a lot more performance headroom to increase the render resolution or to use a higher resolution headset like the Reverb G2 or the Vive Pro 2.

Let’s look at our newest VR sim, F1 2022.

F1 2022

Codemasters has captured the entire Formula 1 2021 season racing in F1 2022, and the VR immersion is good. The graphics are customizeable and solid, handling and physics are good, the AI is acceptable, the scenery is outstanding, and the experience ticks many of the necessary boxes for a racing sim.

Here is the frametime plot for F1 2022.

Here are the details as reported by FCAT-VR.

The RTX 3090 delivered 117.16 unconstrained FPS with 2 synthesized frames but no dropped frames or Warp misses.

The RTX 4090 delivered 192.66 unconstrained FPS together with no synthetic frames and with no dropped frames nor Warp misses.

The experience playing F1 2022 using the Ultra preset is not very different on either video card but the RTX 4090 has more performance headroom to increase the framerate to 120Hz or to use a higher resolution headset.

Next we check out No Man’s Sky.

No Man’s Sky (NMS)

No Man’s Sky is an action-adventure survival single and multiplayer game that emphasizes survival, exploration, fighting, and trading. It is set in a procedurally generated deterministic open universe, which includes over 18 quintillion unique planets using its own custom game engine.

The player takes the role of a Traveller in an uncharted universe by starting on a random planet with a damaged spacecraft equipped with only a jetpack-equipped exosuit and a versatile multi-tool that can also be used for defense. The player is encouraged to find resources to repair his spacecraft allowing for intra- and inter-planetary travel, and to interact with other players.

Here is the No Man’s Sky Frametime plot. We set the settings to Maximum which is a step over Ultra including setting the anisotropic filtering to 16X and upgrading to FXAA+TAA. Since DLSS is available for RTX cards and the Quality setting improves performance without impacting image quality, we used it.

Here are the FCAT-VR details of our comparative runs.

The RTX 3090 delivered 109.88 unconstrained FPS with 198 (3%) synthesized frames but no dropped frames or Warp misses.

The RTX 4090 delivered 183.68 unconstrained FPS together with 4 synthetic frames and with no dropped frames nor Warp misses.

The experience playing No Man’s Sky using the highest settings is not very different on either video card but the RTX 4090 has far more performance headroom.

Let’s continue with another VR game, ProjectCARS 2, that we still like better than its successor even though it is no longer available for online play.

Project CARS 2 (PC2)

There is a real sense of immersion that comes from playing Project CARS 2 in VR using a wheel and pedals. It uses its in-house Madness engine, and the physics implementation is outstanding. We are disappointed with Project CARS 3, and will continue to use the older game instead for VR benching.

Project CARS 2 offers many performance options and settings and we prefer playing with SMAA Ultra rather than to use MSAA.

We used maximum settings including for Motion Blur.

Here is the frametime plot.

Here are the FCAT-VR details.

The RTX 3090 delivered 113.60 unconstrained FPS with no synthesized frames and no dropped frames or Warp misses.

The RTX 4090 delivered 209.53 unconstrained FPS together with no synthetic frames and with no dropped frames nor Warp misses.

The experience playing Project CARS 2 using maximum settings is similar for both video cards but the RTX 4090 has far more performance headroom to increase the framerate to 120Hz or to use a higher resolution headset like the Vive Pro 2 or Reverb G2.

Amazing. Although all of these maxed out VR benchmarks were run at SteamVR’s 150% render resolution (2758×2740), the RTRX 4090 only broke a sweat playing ACC.

Unconstrained Framerates

The following chart summarizes the overall Unconstrained Framerates (the performance headroom) of our two cards using our 5 VR test games. The preset is listed on the chart and higher is better.

The RX 4090 FE delivers far higher unconstrained frames for all VR benchmarks over the RTX 3090 FE in this important performance metric. However, unconstrained framerates are just one metric that has to be taken together with the frametime plots to have any meaning. It is clear that the RTX 4090 is ready for higher resolution headsets than the Valve Index. We’ll follow up this review with an expanded VR review using the Reverb G2 and Vive Pro.

Creative, Pro & Workstation Apps

Let’s look at Creative applications next to see if the RTX 4090 is a good upgrade from the RTX 3090 or RX 6900 XT. We test starting with Geekbench.

GeekBench

GeekBench is an excellent CPU/GPU benchmarking program which runs a series of tests and times how long a GPU (in this case) takes to complete its tasks. It benchmarks OpenCL, Vulcan, and CUDA performance

OpenCL, Vulcan, and CUDA Performance

RTX 6900 XT

First, OpenCL performance.

Next we test the RTX 6900 XT using Vulcan.

The 6900 XT does not run CUDA, so we move on to RTX 3090 performance.

RX 3090

First, OpenCL performance.

Next we test the RX 3090 using Vulcan.

Finally, RTX 3090 CUDA performance.

We move on to RTX 4090 performance

RX 4090

First, OpenCL performance.

Next we test the RX 4090 using Vulcan.

Finally, RTX 4090 CUDA performance.

The summary charts below show the comparative performance scores.

Again, the RTX 4090 performance is outstanding.

Next up, Blender benchmark.

Blender 3.3.0 Benchmark

Blender is a very popular open source 3D content creation suite. It supports every aspect of 3D development with a complete range of tools for professional 3D creation.

For the following chart, higher is better as the benchmark renders a scene multiple times and gives the results in samples per minute.

Blender’s benchmark performance is highest using the RTX4090, and often the amount of time saved is substantial over using the next fastest card, the RTX 3090.

Next, we look at the OctaneBench.

OTOY Octane Bench

OctaneBench allows you to benchmark your GPU using OctaneRender. The hardware and software requirements to run OctaneBench are the same as for OctaneRender Standalone.

We run OctaneBenc 2020.1.5 for Windows and here are the RTX 3090’s complete results and overall score of 671.17

We compare the score and results for the RTX 4090 – almost double the RTX 3090 score with 1261.64

Here is the summary chart comparing the RTX 4090 with the RTX 3090 overall scores.

The RTX 4090 is a beast of a card when used for rendering.

Next, we move on to AIDA64 GPGPU benchmarks.

AIDA64

AIDA64 is an important industry tool for benchmarkers. Its GPGPU benchmarks measure performance and give scores to compare against other popular video cards.

AIDA64’s benchmark code methods are written in Assembly language, and they are well-optimized for every popular AMD, Intel, NVIDIA and VIA processor by utilizing the appropriate instruction set extensions. We use the Engineer’s full version of AIDA64 courtesy of FinalWire. AIDA64 is free to to try and use for 30 days. CPU results are also shown for comparison with the RTX 4090 GPGPU benchmarks.

Here is the chart summary of the AIDA64 GPGPU benchmarks with the RTX 4090, the RTX 3090 and the RX 6900 XT side-by-side.

Generally the RTX 3090 is faster at almost all of AIDA64’s GPGPU benchmarks than the other cards. So let’s look at Sandra 2020 next.

SiSoft Sandra 2020

To see where the CPU, GPU, and motherboard performance results differ, there is no better tool than SiSoft’s Sandra 2020. SiSoftware SANDRA (the System ANalyser, Diagnostic and Reporting Assistant) is a excellent information & diagnostic utility in a complete package. It is able to provide all the information about your hardware, software, and other devices for diagnosis and for benchmarking. Sandra is derived from a Greek name that implies “defender” or “helper”.

There are several versions of Sandra, including a free version of Sandra Lite that anyone can download and use. 20/21-R16a is the latest version, and we are using the full engineer suite courtesy of SiSoft. Sandra 2020 features continuous multiple monthly incremental improvements over earlier versions of Sandra. It will benchmark and analyze all of the important PC subsystems and even rank your PC while giving recommendations for improvement.

With the above in mind, we ran Sandra’s intensive GPGPU benchmarks and charted the results summarizing them. The performance results of the RTX 4090 are compared with the performance results of the RTX 3080, and the RX 6900 XT.

In Sandra GPGPU benchmarks, the RTX 4090 is much faster than the RTX 3090 and it distinguishes itself in every area – Processing, Cryptography, Financial and Scientific Analysis, Image Processing, and Bandwidth.

Next up, SPEC benchmarks.

SPECworkstation3.1 Benchmarks

All the SPECworkstation 3 benchmarks are based on professional applications, most of which are in the CAD/CAM or media and entertainment fields. All of these benchmarks are free except for vendors of computer-related products and/or services.

The most comprehensive workstation benchmark is SPECworkstation 3. It’s a free-standing benchmark which does not require ancillary software. It measures GPU, CPU, storage and all other major aspects of workstation performance based on actual applications and representative workloads. We only tested the GPU-related workstation performance as checked in the image above.

Here are our raw SPECworkstation 3.1 raw scores for the RX 6900 XT:

Here are our raw SPECworkstation 3.1 raw scores for the RTX 3090:

Finally, here are our SPECworkstation 3.1 raw scores for the RTX 4090:

Below are the SPECworkstation 3.1 RTX 4090 results summarized in a chart along with the two competing cards, the RTX 3090, and the RTX 6900 XT. Higher is better since we are comparing scores.

The RTX 4090 is not a workstation card, yet it uses brute force to win most of the benches against the other two cards. The Radeon scores unbelievably high in snx-03, however, and we have no explanation for this result.

Finally, we benchmark using SPECviewperfect GPU benches.

SPECviewperf 2020 GPU Benches

The SPEC Graphics Performance Characterization Group (SPECgpc) has released a 2020-22 version of its SPECviewperf benchmark that features updated viewsets, new models, support for up to 4K display resolutions, and improved set-up and results management. We use the highest available 3800×2120 display resolution for highend cards.

Here are SPECviewperf 2020 GPU RTX 4090 benchmarks summarized in a chart together with our two competing cards.

Although we see different architectures with different strengths and weaknesses, the RTX 4090 is a beast in these SPEC benchmarks.

The RTX 4090 doesn’t offer any certifications for professional applications and it is not expected. It is likely that in workstation specific benchmarks, there will be cases where a Quadro board will outperform the RTX 4090 GeForce card. This is why professionals pay much more for Quadro than for any GeForce with otherwise equivalent raw performance.

After seeing the totality of the benches, many creative users will probably upgrade their existing systems with a new RTX 4090 series card based on the performance increases and the associated increases in productivity that they require. The question to buy the RTX 4090 or the RTX 3090 should be based on the workflow and requirements of each user as well as budget. Time is money to a professional depending on how these apps are used. Hopefully the benchmarks that we ran may help you decide.

Let’s head to our conclusion.

Final Thoughts

This has been a very enjoyable exploration evaluating the new Ada Lovelace RTX 4090 FE versus the RTX 3090 FE and Gigabyte RTX 6900 XT Gaming OC. The RTX 4090 performed brilliantly performance-wise. It totally blows away its other competitors as it is much faster. The RTX 4090 at $1599 is the upgrade from the $1499 RTX 3090 since the RTX 4090 gives at least a 160% (1.6X) improvement over its baseline performance. If a gaming enthusiast wants the very fastest card – just as the RTX 3090 was for the past two years (until the up to 10% faster RTX 3090 Ti was released), and doesn’t mind the $100 price increase – then the RTX 4090 is the only choice for intensive gaming and high resolution VR headsets.

The RTX 4090 is the flagship gaming card that can also run intensive creative apps very well, especially by virtue of its huge 24GB framebuffer. But it is still not a Quadro. These cards cost a lot more and are optimized specifically for workstations and also for professional and creative apps.

For RTX 3090 gamers who paid $1499 and who have disposable cash for their hobby, the RTX 4090 Founders Edition which costs $1599 is the card to maximize their upgrade. And for high-end gamers who also use creative apps, this card may become a very good value. Hobbies are very expensive to maintain, and the expense of PC gaming pales in comparison to what golfers, skiers, audiophiles, and many other hobbyists pay for their entertainment.

We cannot call the $1600 RTX 4090 a “good value” generally for gamers as it is a halo card although it provides more than 1.6X the performance of a RTX 3090. Of course, a RTX 3090 can be currently found at many etailers for under $1000 and a RTX 6900 XT for less than $700. Value is in the eye of the beholder, and the RTX 4090 delivers on its raw performance promises.

In addition, DLSS 3 brings a great future value to the new 4000 series as it has already received support from many of the world’s leading game developers, with more than 35 games and applications announcing support including game engines, including Unity, Unreal, and Frostbite Engine. If a game already uses DLSS 2 Super Resolution, upgrading to DLSS 3 is a relatively simple process that will make both Super Resolution and Frame Generation available. DLSS 3 leverages the same integration points as DLSS 2 (color buffer, depth buffer, engine motion vectors, and output buffers) and Nvidia Reflex, making upgrades from these existing SDKs easy for devs using the DLSS 3 Streamline plugin.

We will follow up with a DLSS 3 review since what we have experienced so far is extremely promising especially for upcoming less powerful Ada Lovelace cards.

Conclusion

We are very impressed with the RTX 4090 raw performance after spending more than 100 hours testing it over the past few days. It offers exceptional performance at Ultra 4K and and it even supports smooth playable gaming at 4K/120Hz using Quality DLSS and may be used for 8K gaming. It currently stands alone as the fastest video card in the world.

The Founders Edition of the RTX 4090 is well-built, solid, good-looking, and it stays cool and quiet even when overclocked – the card does not get hot like the RTX 3090 and it is much quieter under load. The RTX 4090 Founders Edition offers a big performance improvement over any previous Founders Editions in every metric.

Pros

• The RTX 4090 is the fastest video card in the world
• The RTX 4090 at $100 more than the RTX 3090 launched at is at least a 1.6X+ jump in raw performance
• The RTX 4090 offers an overall 160% to 180% improvement over the RTX 3090’s baseline performance (at 100%) depending on the resolution and individual game, and in several examples it offers nearly a 200% improvement!
• 24GB of fast vRAM and a fast core allow for 4K/120Hz gaming and it’s also very useful for intensive creative apps
• Ray tracing is a game changer in every way and DLSS 2 is pure magic. DLSS 3 looks to be even more of a game changer
• Ada Lovelace improves over Ampere with AI/deep learning and ray tracing to improve visuals while also increasing performance with DLSS 2 and especially with DLSS 3
• The RTX 4090 Founders Edition design cooling is quiet and efficient; the GPU in a well-ventilated case stays much cooler and quieter than the RTX 3090
• The industrial design is eye-catching and it is solidly built

Con

• Price. At $1600, the RTX 4090 is not a good value for gaming except as a multi-purpose halo card or for bragging rights

The Verdict:

If you are a gamer who also uses creative apps where saving time is important, you may do yourself a favor by upgrading to a RTX 4090. For high-end gamers with disposable income, the RTX 4090 is a true 4K/120Hz video card for most modern games and it offers the highest performance as an upgrade from a RTX 3090 delivering from 160% to nearly 200% improvement in raw performance power.

Stay tuned, there is a lot more on the way from BTR. Next up, we will more extensively test the RTX 4090, RTX 3080, and RX 6900 XT in VR using the Vive Pro 2 and Reverb G2 with an ETA of early next week. We also plan to follow up with an RTX 4090 overclocking review and a DLSS 3 deep dive. Stay tuned to BTR!

Happy Gaming!

UPDATE: I shared my data with John Peddie Research and they posted their own take on the RTX 4090 focusing on averages and easy-to-read charts including their famous Pmark overall comparison.

This overclocking showdown is the final follow-up to the two Red Devil RX 6900 XT reviews last week versus the RTX 3090 Founders Edition (FE). Today, we have optimized our overclocks with all performance options set to their highest limits to get the most performance from each card.

At stock, the Red Devil $1139/1179 RX 6900 XT trades blows with the $1499 RTX 3090 FE in rasterized games although it is slower overall. The Red Devil 6900 XT also provided a better VR experience than the RX 6800 XT although its performance again fell short of the RTX 3090 FE’s VR performance. This time, we will overclock the each card as far as they will go to see where they stand in relation to each other when manually overclocked.

Overclocking the Red Devil RX 6900 XT

The Red Devil version of the RX 6900 XT is factory overclocked 90MHz higher than the reference version at 2250MHz using the OC BIOS. According to its specifications, the Red Devil RX 6900 XT boost can clock up to 2340MHz out of the box. We benched it using out of the box settings at its default Power Limit setting of 280W and looped Heaven 4.0 in a 1440P window at max settings to load the GPU to 98%.

We generally see it boosting even higher than its rated specs, and it generally settles in above 2380MHz with peaks above 2400MHz. Increasing the Power Limit to its maximum (320W) we see a small increase in performance.

It will boost about 35MHz higher with a maxed out Power Limit slider, and we tested its 320W performance compared with out of the box 280W performance using synthetic benches which are sensitive to overclocking and also very accurate.

There is approximately a 1-3% performance increase with 40W added, so some enthusiasts may prefer to undervolt rather than overvolt their Red Devil depending on their preferences.

This isn’t much performance additional performance to be gained either with the RTX 3090 that gains about 2-5% by increasing its power limit. But we want to see how a maximum core and memory overclock will increase the Red Devil’s performance without consideration for the power usage.

We maxed the Power Limit, added the maximum voltage, maxed the memory OC slider to the right to add 7% more frequency, and moved the Max Frequency slider to 2750MHz after testing each +25MHz increment between 2600 and 2800MHz. Each increase in the maximum core frequency increased performance slightly which we confirmed as we looped Heaven 4.0. We were stopped dead at 2775MHz when the PC locked up.

We were disappointed with the performance increases until we also increased the fan profile to keep the GPU below 74C. Although, the fans started to become more audible, the performance stabilized and the GPU no longer throttled. With a cooler running GPU, the core clocks stayed relatively steady just below 2700MHz which is about 250MHz above the default Red Devil core clocks and the voltage pegged at 1175mV as the card consistently drew more than 300W, peaking above 320W.

Before we compare the Red Devil’s overclocked performance with the RTX 3090 FE, let’s look at its chief competitor’s overclock.

Overclocking the RTX 3090 FE

We devoted a separate evaluation to the RTX 3090 which you can read here. After testing multiple combinations further, our RTX 3090’s final stable overclock for this review to achieve the highest overall performance adds +55 MHz offset to the core and +600 MHz to the memory to perform above 1950MHz

The RTX 3090 FE is power-limited, and to achieve a higher overclock will take more voltage than what adding .1mV can deliver. But before we compare the RTX 3090 at stock and overclocked with the Red Devil RX 6900 XT, please check out our testing platform and configuration.

Test Configuration – Hardware

• Intel Core i9-10900K (HyperThreading/Turbo boost On; All cores overclocked to 5.1GHz/5.0Ghz. Comet Lake DX11 CPU graphics)
• EVGA Z490 FTW motherboard (Intel Z490 chipset, v1.9 BIOS, PCIe 3.0/3.1/3.2 specification, CrossFire/SLI 8x+8x), supplied by EVGA
• T-FORCE DARK Z 32GB DDR4 (2x16GB, dual channel at 3600MHz), supplied by Team Group
• Red Devil RX 6900 XT 16GB, stock and overclocked, on short term loan from PowerColor
• RTX 3090 Founders Edition 24GB, stock and overclock, on loan from NVIDIA
• 1TB Team Group MP33 NVMe2 PCIe SSD for C: drive
• 1.92TB San Disk enterprise class SATA III SSD (storage)
• 2TB Micron 1100 SATA III SSD (storage)
• 1TB Team Group GX2 SATA III SSD (storage)
• 500GB T-FORCE Vulcan SSD (storage), supplied by Team Group
• ANTEC HCG1000 Extreme, 1000W gold power supply unit
• BenQ EW3270U 32? 4K HDR 60Hz FreeSync monitor
• Samsung G7 Odyssey (LC27G75TQSNXZA) 27? 2560×1440/240Hz/1ms/G-SYNC/HDR600 monitor
• DEEPCOOL Castle 360EX AIO 360mm liquid CPU cooler
• Phanteks Eclipse P400 ATX mid-tower (plus 1 Noctua 140mm fan) – All benchmarking and overclocking performed with the case closed

Test Configuration – Software

• GeForce 461.09 for the RTX 3090
• Adrenalin 2020 Edition 20.12.2 drivers used for the Red Devil RX 6900 XT
• High Quality, prefer maximum performance, single display, set in the NVIDIA control panel.
• VSync is off in the control panel and disabled for each game
• AA enabled as noted in games; all in-game settings are specified with 16xAF always applied
• Highest quality sound (stereo) used in all games
• All games have been patched to their latest versions
• Gaming results show average frame rates in bold including minimum frame rates shown on the chart next to the averages in a smaller italics font where higher is better. Games benched with OCAT show average framerates but the minimums are expressed by frametimes (99th-percentile) in ms where lower numbers are better.
• Windows 10 64-bit Pro edition; latest updates v2004. DX11 titles are run under the DX11 render path. DX12 titles are generally run under DX12, and multiple games use the Vulkan API.
• Latest DirectX
• Precision X1
• Wattman

Games

Vulkan

• Ghost Recon: Breakpoint
• Wolfenstein Youngblood
• World War Z
• Strange Brigade
• Rainbow 6 Siege

DX12

• Horizon Zero Dawn
• Tom Clancy’s The Division 2
• Metro Exodus

DX11

• Total War: Three Kingdoms
• Far Cry New Dawn

Synthetic

• TimeSpy (DX12)
• 3DMark FireStrike – Ultra & Extreme
• Superposition
• Heaven 4.0 benchmark

NVIDIA Control Panel settings

Here are the NVIDIA Control Panel settings.

We used Precision X1 to set the GeForce card’s power and temperature limits to maximum and used Wattman for the Radeon.

AMD Adrenalin Control Center Settings

All AMD settings are configured so that all optimizations are off, Vsync is forced off, Texture filtering is set to High, and Tessellation uses application settings. All Navi and Navi 2 cards are capable of high Tessellation unlike earlier generations of Radeons.

Anisotropic Filtering is disabled by default but we always use 16X for all game benchmarks.

Let’s overclock both of our cards to their maximum and check out their performance.

Performance summary charts

Below is the summary chart of 10 games and 3 synthetic tests. The highest settings are always chosen and DX12 was picked above DX11 where available. Specific settings are listed on the performance charts. The benches were run at 1920×1080, 2560×1440, and at 3840×2160.

All results, except for the synthetic benchmarks show average frame rates and higher is always better. Minimum frame rates are shown when they are available next to the averages but they are in italics and in a slightly smaller font. The Red Devil RX 6900 XT’s stock results are in the first column and its overclocked results are in the second column, while the third column represents the overclocked RTX 3090’s results with its stock-clocked results in the fourth column.

There isn’t much change in the overall ranking by overclocking both cards although each card scales decently. Horizon Zero Dawn gave inconsistent benching results at 4K (between 77 and 84 FPS) which may still indicate some throttling for the Red Devil. Although both cards are constrained by their respective power limits and tiny voltage increases, the Red Devil is an aftermarket card that should not be so power-limited.

PowerColor claims that their Red Devil can handle 400W, but is limited to 320W which means a lot of performance is left on the table by AMD’s lockdown. It is possible to circumvent this power limit with overclocking tools like MorePowerTools, but we have not tested it as our sample had to be returned to PowerColor in working condition after two weeks of intensive benchmarking.

Let’s head to our conclusion.

Conclusion

This has been quite an interesting exploration for us, evaluating the overclocked $1139/$1179 Red Devil RX 6900 XT versus the overclocked $1499 RTX 3090 FE. The Red Devil is a slower card, but it also costs about $350 less. We are very disappointed that the Red Devil’s voltage and power limit has been locked down, but the blame for this lays with AMD for an illogical decision for their partner cards – not with PowerColor who would love to be able to allow enthusiasts to push their card much harder.

PowerColor has taken the RX 6900 XT GPU and over-engineered it with a 16 Phase design versus the 11+2 Phase VRM design on the reference design. Red Devils are over-spec’d in order to deliver the best stability and overclocking headroom. It’s not only capable of using well over 400W, and by having the best VRMs, it will run cooler and last longer than any reference version

The Red Devil is the equivalent of a Ferrari – and AMD has forcibly installed a limiter on it. This is a poor decision by AMD foisted on their partners and we respectfully disagree with them. We really like the Red Devil and will continue to recommend it as among the very best of the AIB RX 6900 XTs, and a great choice for AMD gamers.

Next week, we will return to benching Virtual Reality (VR). Sean has sent us a Reverb G1 Pro for evaluation to compare with a Reverb G2 that Hewlett Packard is also sending us for evaluation versus the Vive Pro. And Sean is working on his Assetto Corsa: Competizione review also.

Rodrigo’s next review is a GeForce 461.09 driver performance analysis using Ampere and Turing, and his following feature review will cover Windows 10 ‘Game Mode’, On vs. Off. Stay tuned to BTR!

Happy Gaming!

The RTX 3060 Ti Founders Edition (FE) has arrived at $399 and we want to see how this upper midrange video card performs in VR compared with the new Ampere and Navi 2 cards. This review presents a VR performance showdown featuring the RTX 3060 Ti FE, the reference RX 6800 versus the RTX 3070 FE, and the reference RX 6800 XT versus the RTX 3080 FE using the Vive Pro and FCAT VR.

In pancake games, the $399 RTX 3060 Ti is faster than the Turing GeForce RTX 2080 SUPER which launched originally at $699, and on average 40% faster than its predecessor, the GeForce RTX 2060 Super, the new GeForce RTX 3060 Ti is the brand new addition to the GeForce RTX 30 Series. It delivers 16.2 shader TFLOPs,31.6 RT TFLOPs, and 129.6 Tensor TFLOPs over a 256-bit memory interface using 8GB of GDDR6 and it is based on the same GA104 chip used in the $499 RTX 3070 but with less processing units.

We want to first see where the RTX 3060 Ti fits regarding VR performance and if it is a worthy upgrade from an older video card. Although a fast CPU is important for geometry and other processing, smooth VR delivery depends mostly on the video card. An underpowered video card can even cause reprojecting and artifacting for a substandard playing experience that can even lead to VR sickness.

Since we posted our original review, we have benchmarked more than 20 VR games for our follow-up reviews over the past nearly four years. We have also compared FCAT-VR with our own video benchmarks using a camera to capture images directly from our HMD’s lenses. For BTR’s VR testing methodology, please refer to this evaluation.

We currently benchmark fifteen VR games using the Vive Pro that features 2880×1600 resolution (1440×1600 pixels per eye), the same as the Valve Index. BTR’s testing platform is an overclocked Intel Core i9-10900K, an EVGA Z490 FTW motherboard, and 32 GB of Vulcan Dark Z DDR4 at 3600MHz on a clean install of Windows 10 64-bit Pro Edition using GeForce Game Ready Driver, 457.51, and Adrenalin Software 20.11.3 drivers.

It is important to be aware of VR performance since poorly delivered frames can make a VR experience unpleasant. It’s also important to understand how we accurately benchmark VR games using FCAT-VR as explained here. But before we benchmark our VR games, check out our Test Configuration below.

Test Configuration – Hardware

• Intel Core i9-10900K (HyperThreading/Turbo boost On; All cores overclocked to 5.1GHz/5.0Ghz. Comet Lake DX11 CPU graphics)
• EVGA Z490 FTW motherboard (Intel Z490 chipset, v1.3 BIOS, PCIe 3.0/3.1/3.2 specification, CrossFire/SLI 8x+8x), supplied by EVGA
• T-FORCE DARK Z 32GB DDR4 (2x16GB, dual channel at 3600MHz), supplied by Team Group
• Vive Pro, on loan from HTC/Vive; the Wireless Adapter is not used for benchmarking
• RTX 3060 Ti Founders Edition 8GB, stock clocks, on loan from NVIDIA
• RTX 3070 Founders Edition 10GB, stock clocks, on loan from NVIDIA
• RTX 3080 Founders Edition 10GB, stock clocks, on loan from NVIDIA
• RX 6800 Reference Edition 16GB, stock clocks, on loan from AMD.
• RX 6800 XT Reference Edition 16GB, stock clocks, on loan from AMD.
• 1TB Team Group MP33 NVMe2 PCIe SSD for C: drive
• 1.92TB San Disk enterprise class SATA III SSD (storage)
• 2TB Micron 1100 SATA III SSD (storage)
• 1TB Team Group GX2 SATA III SSD (storage)
• 500GB T-FORCE Vulcan SSD (storage), supplied by Team Group
• ANTEC HCG1000 Extreme, 1000W gold power supply unit
• BenQ EW3270U 32? 4K HDR 60Hz FreeSync monitor
• Samsung G7 Odyssey (LC27G75TQSNXZA) 27? 2560×1440/240Hz/1ms/G-SYNC/HDR600 monitor
• DEEPCOOL Castle 360EX AIO 360mm liquid CPU cooler
• Phanteks Eclipse P400 ATX mid-tower (plus 1 Noctua 140mm fan)

Test Configuration – Software

• GeForce 457.30 (RTX 3080) and 457.51 (RTX 3070/RTX 3060 Ti) Game Ready drivers – High Quality, prefer maximum performance, single display, no optimizations, Vsync is off as set in the NVIDIA control panel
• Adrenalin Software 20.11.2 (RX 6800 XT) and 20.11.3 (RX 6800) – All optimizations are off, Vsync is forced off, Texture filtering is set to High, and Tessellation uses application settings
• Windows 10 64-bit Pro edition; latest updates v2004.
• Latest DirectX
• MSI’s Afterburner, 4.6.3 beta to set the cards’ power and temperature limits to their maximums
• All 15 VR games are patched to their latest versions at time of publication
• FCAT-VR Capture (latest Beta 03/04/20)
• FCAT-VR Beta 18
• SteamVR – at 100% resolution unless specified

15 VR Game benchmark suite & 4 synthetic tests

Synthetic

• VRMark Cyan Room
• VRMark Blue Room
• Unigine Superposition VR Benchmark
• OpenVR Benchmark

SteamVR /Epic Platform Games

• ARK: Park
• Assetto Corsa Competizione
• Boneworks
• Elite Dangerous
• Fallout 4
• Half Life: Alyx
• Hellblade: Senua’s Sacrifice
• Into the Radius
• No Man’s Sky
• Obduction
• Project CARS 2
• Skyrim
• Subnautica
• The Vanishing of Ethan Carter
• The Walking Dead: Saints & Sinners

The Unreal 4 engine is a very popular engine for VR development, and seven of our fifteen test games are created with it. The Creation and Unity engines are each used for two games, while the COBRA, No Man’s Sky, Source, and Madness engines are each represented by one game.

It is important to remember that BTR’s charts use frametimes in ms where lower is better, but we also compare “unconstrained framerates” which shows what a video card could deliver (headroom) if it wasn’t locked to either 90 FPS or to 45 FPS by the HMD. In the case of unconstrained FPS which measures just one important performance metric, faster is better.

Let’s individually look at our fifteen VR games’ performance using FCAT VR. All of our games were benchmarked at 100% SteamVR resolution or higher (as noted), although some games were also set in-game to a higher resolution or increased pixel density as described for each game, as we benchmark the RTX 3060 Ti to see where it stands in relation to the other 4 new video cards.

First up, ARK Park.

BTR’s sim/racing editor, Sean Kaldahl created the replay benchmark run used for both the pancake game and the VR game. It is run at night with lots of geometry, and the lighting effects of the headlights, tail lights, and everything around the track adds to the feel of racing.

The race was cut short on the RTX 3070 at 71 seconds instead of the usual 107 seconds. Here are the VR High frametimes.

Here are the details are reported by FCAT-VR:

The RTX 3060 Ti managed to deliver 92.67 unconstrained FPS with no dropped frame or Warp misses, but 3597 (37%) frames were synthesized.

In runs accidentally cut short, the RTX 3070 achieved 103.85 unconstrained FPS with 7 dropped and 668 (10%) synthetic frames together with 7 Warp misses. If the runs were completed, the unconstrained FPS may have been higher as the first part of the course is more demanding.

The RX 6800 delivered 102.45 unconstrained FPS with 17 dropped frames and 743 (8%) frames were synthesized. It suffered 17 Warp misses.

The RX 6800 XT delivered 126.67 unconstrained FPS with no dropped frames or Warp misses but 125 frames were synthesized.

The RTX 3080 achieved 139.01 unconstrained FPS with 2 dropped and 181 synthetic frames together with 2 warp misses.

The RTX 3060 Ti would perform better using lowered settings as 37% synthesized frames are unacceptable for a high quality VR sim experience. The RTX 3070 and the RX 6800 also suffer many Warp misses so these three cards are next tested on VR Low.

Here are the VR Low details as reported by FCAT-VR:

The RTX 3060 Ti delivered 157.08 unconstrained FPS with 2 dropped frames and 2 Warp misses, but only 2 frames were synthesized using the lower preset.

In runs accidentally cut short, the RTX 3070 achieved 175.89 unconstrained FPS with 1 dropped and 4 synthetic frames together with 1 Warp miss. If the runs were completed, the unconstrained FPS would probably have been higher.

The RX 6800 delivered 186.66 unconstrained FPS with 4 dropped frames and no frames were synthesized. It suffered no Warp misses on the lower preset.

VR Low shows a noticeable drop in visuals from VR High, and we would suggest lowering individual settings instead of dropping from the High to Low preset to stay out of reprojection. Unfortunately, there is no “VR Medium” preset.

Here are the details at highest settings/150%.

The RTX 3060 Ti produced 81.41 unconstrained FPS with 3 dropped frames and 3 Warp misses, but 3186 (50%) frames were still being synthesized. We would suggest dropping the resolution to around 130% or experiment with the in-game settings to achieve a steady 90 FPS smooth framerate delivery to the HMD.

The RTX 3070 managed 93.86 unconstrained FPS with 5 dropped frames and 5 Warp misses, but 2238 (35%) frames were synthesized. Drop the resolution by 10% or experiment with lowered in-game settings.

The RX 6800 delivered 100.30 unconstrained FPS with 15 dropped frames and no Warp misses or synthesized frames.

Next we look at Into the Radius.

Into the Radius

Into the Radius is a hardcore single player stealth survival adventure/exploration game, and it pays homage to the S.T.A.L.K.E.R. series by being set in the dangerous large open world of the Pechorsk Exclusion Zone.

Into the Radius has decent visuals using Unreal Engine 4 that really add to its spooky and intense atmosphere, and its performance demands on the video card are also higher since its a huge open world game with a decent draw distance.

There are no user options for changing individual graphics settings other than High, Medium, and Low presets. There is also a slider to drop or subsample the resolution down to 65%, or to increase the resolution to 110%. We benchmarked our cards at High Quality and with 110% resolution.

Here is the frametime plot for Into the Radius.

Here are the FCAT-VR details.

The RTX 3060 Ti managed 102.27 unconstrained FPS with 2 dropped frames and 2 Warp misses, but 665 (8%) frames were synthesized.

The RTX 3070 produced 147.67 unconstrained FPS with no dropped frames and no Warp misses nor synthetic frames.

The RX 6800 delivered 104.25 unconstrained FPS with 15 dropped frames and 15 Warp misses, but with 601 (7%) synthesized frames.

The RX 6800 XT delivered 135.89 unconstrained FPS with 11 dropped frames, but no warp misses were reported and no synthetic frames needed to be synthesized.

The RTX 3080 achieved 151.01 unconstrained FPS with 3 dropped and 7 synthetic frames. However, it suffered 3 Warp Misses.

The RTX 3060 Ti provided a similar VR experience playing Into the Radius as using a RX 6800 whereas the RTX 3070 offers more performance headroom over the RX 6800 XT. Surprisingly, the RTX 3080 is just slightly faster than the RTX 3070.

Into the Radius becomes demanding on the High preset when the resolution is set at 100% or 110%.

Next, we will check out another demanding VR game, No Man’s Sky.

No Man’s Sky

No Man’s Sky is an action-adventure survival single and multiplayer game that emphasizes survival, exploration, fighting, and trading. It is set in a procedurally generated deterministic open universe, which includes over 18 quintillion unique planets using its own custom game engine.

The player takes the role of a Traveller in an uncharted universe by starting on a random planet with a damaged spacecraft equipped with only a jetpack-equipped exosuit and a versatile multi-tool that can also be used for defense. The player is encouraged to find resources to repair his spacecraft allowing for intra- and inter-planetary travel, and to interact with other players.

Here is the No Man’s Sky Frametime plot. We set the settings to High which is above Enhanced but below Ultra, and we also set the anisotropic filtering to 16X and upgraded to FXAA+TAA.

Here are the FCAT-VR details of our comparative runs.

The RTX 3060 Ti delivered 83.06 unconstrained FPS with no dropped frames or Warp misses, but 3202 (50%) frames were synthesized.

The RTX 3070 produced 93.83 unconstrained FPS with no dropped frames nor any Warp misses, but it required 2678 (42%) synthetic frames.

The RX 6800 managed 91.57 unconstrained FPS with no dropped frames nor Warp misses, but with 3215 (50%) synthesized frames.

The RX 6800 XT delivered 108.90 unconstrained FPS with 4 warp misses and 187 (3%) dropped frames probably because none of its frames were synthesized.

The RTX 3080 achieved 126.48 unconstrained FPS with no dropped frames and no warp misses, and it only required 39 synthetic frames.

The RTX 3080 edges out the RX 6800 XT and delivers more performance headroom and a smoother experience. We do not recommend using Ultra settings for either card, however, and the Enhanced Preset may be better suited for the 3 lesser cards.

Next, we will check out another VR game, Obduction.

Obduction

Obduction is considered to be the spiritual successor to Myst and Riven. It is an adventure game developed by Cyan Worlds using the Unreal 4 engine and it has very good visuals. There is an emphasis on puzzle solving which get more and more difficult as a player progresses.

Obduction’s visuals are both graphically impressive and relatively intensive. There are adjustable menu sliders that will allow the player to dial down the settings but we picked Epic and increased the in-game resolution to 140% plus 75% texture memory. Lowering the settings means the world textures are less detailed and the player will see them pop into view rather suddenly, breaking VR immersion.

Here is Obduction’s frametime plot.

Here are the details.

The RTX 3060 Ti delivered 92.22 unconstrained FPS with 5 dropped frames and 5 Warp misses, but 2591 (40%) frames were synthesized. Dropping the resolution to 100% or 110% should keep it from needing Motion Smoothing.

The RTX 3070 delivered 107.61 unconstrained FPS with 4 dropped frames and 4 Warp misses but 617 (10%) synthetic frames were generated. Again, lower the in-game resolution down from 140%.

The RX 6800 delivered 93.67 unconstrained FPS with 634 (10%) dropped instead of synthesized frames and 8 Warp misses. Lowering the resolution to 120% or 130% would probably be a good starting choice to avoid reprojected, synthesized, or dropped frames.

The RX 6800 XT produced 112.83 unconstrained FPS with 3 dropped frames, 191 synthetic frames (3%), and 3 warp misses.

The RTX 308 0 delivered 130.54 unconstrained FPS and with 3 dropped frames, 10 synthetic frames and 3 warp misses.

Next we will check out another demanding VR game, Project CARS 2 that we like better than Project CARS 3.

Project CARS 2

There is a real sense of immersion that comes from playing Project CARS 2 in VR using a wheel and pedals. It uses its in-house Madness engine, and the physics implementation is outstanding. We are disappointed with Project CARS 3, and will continue to use the older game instead for VR benching.

Project CARS 2 offers many performance options and settings and we prefer playing with SMAA Ultra.

We used maximum settings including for Motion Blur although it looks best to us on Low or Medium. For lesser cards, we would also recommend lowering grass and reflections to maximize framerate delivery as motion smoothing or reprojection tends to cause visible artifacting.

Here is the frametime plot.

Here are the FCAT-VR details.

The RTX 3060 Ti managed 77.45 unconstrained FPS with no dropped frames but 4814 (50%) frames were synthesized. It suffered no Warp Misses.

The RTX 3070 achieved 87.61 unconstrained FPS with 4 dropped and 4581 (47%) synthetic frames together with 4 Warp misses.

The RX 6800 delivered 106.35 unconstrained FPS with 19 dropped frames and 498 (5%) frames were synthesized. It suffered 19 Warp misses.

The RX 6800 XT produced 123.10 unconstrained FPS, with 5 dropped frames, 5 warp misses but 210 of its frames (2%) had to be synthesized.

The RTX 3080 delivered 117.70 unconstrained FPS with 2 dropped frames, 2 warp misses, but it only required 18 synthetic frames.

The experience playing Project CARS 2 on our maximum VR settings is similar on the top three cards, but we would recommend lowering settings for the RTX 3070 and RTX 3060 Ti.

Let’s benchmark Skyrim VR.

Skyrim VR

Skyrim VR is an older game that is no longer supported by Bethesda, but fortunately the modding community has adopted it. It is not as demanding as many of the newer VR ports so its performance is still very good on maxed-out settings using its Creation engine.

We benchmarked Skyrim VR using its highest settings, and we also increased the resolution to its in-game maximum.

Here are the frametime results.

Here are the details of our comparative runs as reported by FCAT-VR.

The RTX 3060 Ti delivered 124.01 unconstrained FPS with 8 dropped frames and 8 Warp misses, and 57 frames were synthesized.

The RTX 3070 managed 131.56 unconstrained FPS with 3 dropped frames, 19 synthetic frames, and 3 Warp misses.

The RX 6800 achieved 118.83 unconstrained FPS with no dropped frames or synthesized frames and no Warp misses.

The RX 6800 XT produced 134.93 unconstrained FPS with no dropped frames, no synthetic frames, and no warp misses.

The Red Devil RX 6800 XT arrived at BTR for evaluation on a short-term loan from PowerColor on Wednesday, the same day the card launched for sale with very limited supply and with no manufacturer recommended pricing although it has been listed out of stock for $799.99 at Newegg. We have been benching it versus the $699 RTX 3080 Founders Edition (FE) and the $649 RX 6800 XT reference card that we got that same day from AMD using GPGPU, workstation, SPEC, 37 games and synthetic benchmarks. We concluded from our preliminary 9-game PC and PCVR 15-game review relative to the RTX 3080, that the reference RX 6800 XT is probably faster at pancake gaming than at VR gaming.

We will also compare the performance of these competing cards with the RX 5700 XT Anniversary Edition (AE) and the GTX 1080 Ti FE to see how older cards fare, and we also include all of the GeForce Turing Super cards and the Ampere cards to complete BTR’s 10-card Big Picture.

The Red Devil RX 6800 XT is factory clocked 90MHz higher than the reference version using the OC BIOS. According to its specifications, the Red Devil RX 6800 XT boost can clock up to 2340MHz out of the box. It also looks different from older generation classic Red Devils, arriving in a more neutral gray color instead of in all red and black. The Red Devil RX 6800 XT features a RGB mode whose LEDs default to a bright red which may be customized by PowerColor’s DevilZone software.

The Red Devil RX 6800 XT Features & Specifications

Here are the Red Devil RX 6800 XT specifications according to PowerColor:

Specifications

Features

Here are the Red Devil RX 6800 XT features.

Additional Information from PowerColor

• The card has 2 modes, OC and Silent. 281W / 255W Power target. There’s a bios switch on the side of the card. Even on performance mode it’s considerably quieter than reference board but the silent mode is truly whisper quiet, with a normal case with a optimal airflow, you most likely see the card run around 1000 rpm.
• The board has 16 Phase VS the 11+2 Phase VRM design on the reference design meaning is over spec’d in order to deliver the best stability and overclock headroom,not only capable of well over 400w but by having such VRM it will run cooler and last longer.
• DrMos and high-polymer Caps are used with no compromises.
• The cooler features 2 x 100mm with a center 1x90mm fan, all with two ball bearing fans with 7 heat pipes (3X8Φ and 4X6Φ heatpipes) across a high density heatsink with a copper base. The PCB is shorter than the cooler.
• RGB is enhanced, Red Devil now connects to the motherboard aRGB (5v 3 pin connector).
• Red Devil has Mute fan technology, fans stop under 60C.
• The ports are LED illuminated. Now you can see in the dark where to plug.
• The card back plate does not have thermal pads but instead there are openings across the backplate for the PCB to breathe.
• Red Devil RX 6800 XT Graphics Card Limited Edition provides the unique and high quality crafted Red Devil keycap to make your keyboard look Devilish.
• Buyers or Red Devil Limited edition will be able to join exclusive giveaway as well access to the Devil Club website. A membership club for Devil users only which gives them access to News, Competitions, Downloads and most important instant support via Live chat.

The Big Navi 2 Radeon 6000 family

The Radeon 6800 competes with the RTX 3070 and is priced a little higher at $579 while the RTX 6800 XT at $649 competes with the RTX 3080 at $699. Next week, the RTX 6900 XT releases at $999 to compete with the $1499 RTX 3090.

Here is a die shot of the GPU powering the Radeon 6000 series courtesy of AMD

AMD has their own ecosystem for gamers and many unique new features for the Radeon 6000 series.

The Test Bed

BTR’s test bed consists of 37 games and 3 synthetic game benchmarks at 1920×1080, 2560×1440, and at 3840×2160 as well as SPEC, Workstation, and GPGPU benchmarks. Our latest games include Watch Dogs: Legions, Call of Duty Black Ops: Cold War and Assassin’s Creed: Valhalla. The testing platform uses a clean installation of Windows 10 64-bit Pro Edition, and our CPU is an i9-10900K which turbos all 10 cores to 5.1/5.0GHz, an EVGA Z490 FTW motherboard, and 32GB of T-FORCE Dark Z DDR4 3600MHz. The games, settings, and hardware are identical except for the cards being compared.

First, let’s take a closer look at the new PowerColor Red Devil RX 6800 XT.

A Closer Look at the PowerColor Red Devil RX 6800 XT

Although the Red Devil RX 6800 XT advertises itself as a premium 7nm card on AMD’s RDNA 2 architecture which features FidelityFX, FreeSync 2 HDR and PCIe 4.0, the cover of the box uses almost no text in favor of stylized imagery.

The back of the box touts key features which now include HDMI 2.1 VRR, ray tracing technology, and VR Ready Premium as well as states it’s 850W power and system requirements although there are a lot of blank and unused space on the box. AMD’s technology features are highlighted, but the box does not even mention PowerColor’s custom cooling solution, Dual-BIOSes, RGB software and output LEDs and backplate.

Opening its very well-padded box, we now see advertising that instead probably should have been on the box’s outside. Also inside are a quick installation guide, RGB LED cable, and an invitation to join PowerColor’s Devil’s Club. In addition, a couple of key caps are included which could prove useful for benchmarking while wearing a HMD. PowerColor’s is a nicer presentation than AMD’s reference RX 6800 XT.

The Red Devil RX 6800 XT is a large tri-fan card in a three slot design which is quite handsome with PowerColor’s colors and even more striking with the RGB on. Here is the Red Devil next to a RTX 3080 FE to show how much larger and beefier a card it is.

It uses two 1×8-pin PCIe connections. Above is the reference RX 6800 XT backplate.

The PowerColor Red Devil RX 6800 XT’s sturdy backplate features a stylized custom devil symbol that lights up in the color of your choice if synced, red being the default. This card is number 41 out of a 1000 limited edition set. We do not know what this means. There is also a switch to choose between the default overclock (OC) BIOS and the Silent BIOS. We didn’t bother with the Silent BIOS but it is good to have in case a flash goes bad.

The Red Devil’s RX 6800 XT’s connectors include 2 DisplayPorts, 1 HDMI connection, and a USB Type C connector. There is an LED that illuminates this panel for making easier connections in the dark.

The specifications look good and the card itself looks great with its default RGB bright red contrasting with the black backplate and its aggressively lit-up end perhaps stylistically reminiscent of an automotive grill.

Let’s check out its performance after we look over our test configuration and more on the next page.

Test Configuration – Hardware

• Intel Core i9-10900K (HyperThreading/Turbo boost On; All cores overclocked to 5.1GHz/5.0Ghz. Comet Lake DX11 CPU graphics)
• EVGA Z490 FTW motherboard (Intel Z490 chipset, v1.9 BIOS, PCIe 3.0/3.1/3.2 specification, CrossFire/SLI 8x+8x), supplied by EVGA
• T-FORCE DARK Z 32GB DDR4 (2x16GB, dual channel at 3600MHz), supplied by Team Group
• Radeon RX 6800 XT Reference version 16GB, stock settings, on loan from AMD
• Red Devil RX 6800 XT 16GB, stock and overclocked, on short term loan from PowerColor
• RTX 3080 Founders Edition 10GB, stock, on loan from NVIDIA
• Radeon RX 5700 XT 8GB Anniversary Edition, stock AE clocks.
• RTX 3090 Founders Edition 24GB, stock clocks, on loan from NVIDIA
• RTX 3070 Founders Edition 8GB, stock clocks, on loan from NVIDIA
• RTX 2080 Ti Founders Edition 11GB, stock clocks, on loan from NVIDIA
• RTX 2080 SUPER Founders Edition 8GB, stock clocks, on loan from NVIDIA
• RTX 2070 Ti Founders Edition 8GB, stock clocks, on loan from NVIDIA
• GTX 1080 Ti Founders Edition 11GB, stock clocks, on loan from NVIDIA
• 1TB Team Group MP33 NVMe2 PCIe SSD for C: drive
• 1.92TB San Disk enterprise class SATA III SSD (storage)
• 2TB Micron 1100 SATA III SSD (storage)
• 1TB Team Group GX2 SATA III SSD (storage)
• 500GB T-FORCE Vulcan SSD (storage), supplied by Team Group
• ANTEC HCG1000 Extreme, 1000W gold power supply unit
• BenQ EW3270U 32″ 4K HDR 60Hz FreeSync monitor
• Samsung G7 Odyssey (LC27G75TQSNXZA) 27″ 2560×1440/240Hz/1ms/G-SYNC/HDR600 monitor
• DEEPCOOL Castle 360EX AIO 360mm liquid CPU cooler
• Phanteks Eclipse P400 ATX mid-tower (plus 1 Noctua 140mm fan) – All benchmarking and overclocking performed with the case closed

Test Configuration – Software

• GeForce 456.96 for the RTX 3070, the RTX 2080 Ti, and the RTX 2070/2080 SUPER; and GeForce 456.16 Press drivers and GeForce 456.38 public drivers (functionally identical) are used for the other GeForce cards. GeForce GRD 457.30 is used for games released in late October and November although otherwise there were no general game performance driver improvements since Ampere launched.
• Adrenalin 2020 Edition 20.11.2 public launch drivers used for the RX 6800 XT reference and Red Devil editions at their factory clocks and the Red Devil was also overclocked. Adrenalin 2020 Edition 20.10.1 drivers used for the RX 5700 XT Anniversary Edition (AE) at AE clocks.
• High Quality, prefer maximum performance, single display, set in the NVIDIA control panel.
• VSync is off in the control panel and disabled for each game
• AA enabled as noted in games; all in-game settings are specified with 16xAF always applied
• Highest quality sound (stereo) used in all games
• All games have been patched to their latest versions
• Gaming results show average frame rates in bold including minimum frame rates shown on the chart next to the averages in a smaller italics font where higher is better. Games benched with OCAT show average framerates but the minimums are expressed by frametimes (99th-percentile) in ms where lower numbers are better.
• Windows 10 64-bit Pro edition; latest updates v2004. DX11 titles are run under the DX11 render path. DX12 titles are generally run under DX12, and multiple games use the Vulkan API.
• Latest DirectX
• MSI’s Afterburner, 4.6.3 beta to set the RTX 3070’s power and temperature limits to their maximums

Games

Vulkan

• DOOM Eternal
• Red Dead Redemption 2
• Ghost Recon: Breakpoint
• Wolfenstein Youngblood
• World War Z
• Strange Brigade
• Rainbow 6 Siege

DX12

• Call of Duty Black Ops: Cold War
• Assassin’s Creed: Valhalla
• Watch Dogs: Legion
• Horizon Zero Dawn
• Death Stranding
• F1 2020
• Mech Warrior 5: Mercenaries
• Call of Duty Modern Warfare
• Gears 5
• Anno 1800
• Tom Clancy’s The Division 2
• Metro Exodus
• Civilization VI – Gathering Storm Expansion
• Battlefield V
• Assetto Corsa: Competitione
• Shadow of the Tomb Raider
• Project CARS 2
• Forza 7

DX11

• Crysis Remastered
• A Total War Saga: Troy
• Star Wars: Jedi Fallen Order
• The Outer Worlds
• Destiny 2 Shadowkeep
• Borderlands 3
• Total War: Three Kingdoms
• Far Cry New Dawn
• Assassin’s Creed Odyssey
• Monster Hunter: World
• Overwatch
• Grand Theft Auto V

Synthetic

• TimeSpy (DX12)
• 3DMark FireStrike – Ultra & Extreme
• Superposition
• Heaven 4.0 benchmark
• AIDA64 GPGPU benchmarks
• Blender 2.90 benchmark
• Sandra 2020 GPGPU Benchmarks
• SPECworkstation3
• SPECviewperf 2020

NVIDIA Control Panel settings

Here are the NVIDIA Control Panel settings.

We used MSI’s Afterburner to set all video cards’ power and temperature limits to maximum.

AMD Adrenalin Control Center Settings

All AMD settings are set so that all optimizations are off, Vsync is forced off, Texture filtering is set to High, and Tessellation uses application settings. All Navi cards are capable of high Tessellation unlike earlier generations of Radeons.

Anisotropic Filtering is disabled by default but we always use 16X for all game benchmarks.

Let’s check out overclocking, temperatures and noise next.

BTR received and benchmarked a Red Dragon RX 5600 XT 6GB ($289.99 MSRP) review sample from PowerColor versus the ASUS RX 5600 XT EVO OC 6GB ($309.99), and versus the RTX 2060 6GB Founders Edition ($299.99). Although these cards are designed for Ultra 1080P, 50 game benchmarks were also run at 2560×1440 to stress them beyond their limits.

We posted our Tuff Gaming X3 ASUS RX 5600 XT EVOC OC 6GB review in January where it solidly beat the GTX 1660 Ti and the Super, and nothing has changed since then. The new vBIOS updated the memory from 12Gbps to 14Gbps, and every Red Dragon RX 5600 XT for sale already comes with the faster 14GBps memory. Although its MSRP is $289, Amazon and Newegg are each selling the Red Dragon RX 5600 XT for $299. Newegg has the better deal since they include a game bundle with RE3, Monster Hunter World, and 3-month XBox PC game pass. The GeForce competitor to the RX 5600 XTs ($279-$309) are the entry level RTX 2060 ($299) video cards.

The PowerColor Red Dragon RX 5600 XT 8GB

Although there is no AMD reference design, the Red Dragon RX 5600 XT 6GB shares the same specifications including the maximum Boost clock, but its game/Boost clock is set higher at 1560MHz/1620MHz which is also 10MHz faster than the ASUS RX 5600 XT EVO OC.

The Red Dragon specs look great and it should be a good match up versus the RTX 2060 Founders Edition, and versus the ASUS RX 5600 XT EVO OC, a bigger and a more expensive card.

Unboxing the Red Dragon RX 5600 XT 8GB

The Red Dragon RX 5600 XT 6GB comes in an medium box that advertise its features. The Red Dragon is an 6GB RX 5600 XT built on 7nm using AMD’s latest RDNA architecture and it features 4.0 PCIe support, Fidelity FX and FreeSync 2. “Memory clock at 14Gbps” makes it clear that it already uses the fastest memory.

The features and the minimum system requirements including the need for a 500W power supply are detailed on the back of the box. VR readiness, power efficiency, up to 8K video streaming, image sharpening, FreeSync 2 HDR, and other Radeon key features are listed.

Mute fan technology means the dual axial fans will come to a dead stop below 60℃ under a light gaming load providing silent gaming while reducing power consumption. PowerColor stresses that lower temperatures mean better performance. The cooling fan is equipped with two-ball bearings which increases longevity.

Opening the box we see a well-packed card and a quick start guide.

The Red Dragon RX 5600 XT 6GB is a handsome basic black non-RGB card with a dragon logo on the fan as well as one on the back. One 8-pin PCIe connector is required.

Turning it over, we see the Red Dragon logo on a solid metal 1.5mm backplate that protects the PCB against flexing and prevents damaging the traces.

There is a dual-BIOS switch that either automatically overclocks or downclocks slightly depending on user noise/performance preferences. We used the OC BIOS for maximum performance, and the card remained very quiet even under load.

Looking at the Red Dragon RX 5600 XT 6GB from one edge we see the large heatsinks with many dense fins that extend for much of the length of the PCB with the heatpipes connecting the heatsinks.

On one end, the connectors consist of three DisplayPorts and a HDMI connector. We like this setup better for SteamVR platforms than those with a DVI connector.

The Red Dragon RX 5600 XT 6GB looks good from any angle.

Before we explore overclocking and then performance testing, let’s take a closer look at our test configuration.

Test Configuration, Settings, Benching Suite, Overclocking & Noise

Test Configuration – Hardware

• Intel Core i7-8700K (HyperThreading and Turbo boost is on to 4.8GHz for all cores; Coffee Lake DX11 CPU graphics).
• EVGA Z370 FTW motherboard (Intel Z370 chipset, latest BIOS, PCIe 3.0/3.1 specification, CrossFire/SLI 8x+8x), supplied by EVGA
• T-FORCE 16GB DDR4 (2x8GB, dual channel at 3866 MHz), supplied by Team Group
• Red Dragon RX 5600 XT 6GB at Red Dragon clocks, on loan from PowerColor
• TUF Gaming X3 ASUS RX 5600 XT EVO OC 6GB, at factory settings, on loan from ASUS/AMD
• RTX 2060 Founders Edition 6GB, at FE clocks, on loan from NVIDIA
• EVGA GTX 1660 Ti XC 6GB, at XC clocks, on loan from EVGA
• EVGA GTX 1660 SUPER XC 6GB at EVGA factory settings, on loan from EVGA
• EVGA GTX 1660 XC 6GB at EVGA XC factory settings, on loan from EVGA
• EVGA GTX 1060 SC 6GB, factory SC clocks, on loan from EVGA
• Sapphire RX 5500 XT Pulse OC 4GB on loan from Sapphire
• PowerColor Red Devil RX 5500 XT 8GB on loan from PowerColor
• PowerColor Red Devil RX 590 8GB on loan from PowerColor
• Red Devil RX 570 4GB, at Red Devil factory overclocked settings, on loan from PowerColor
• 2 x 480GB Team Group SSDs – one for AMD, and one for NVIDIA
• 1.92TB San Disk enterprise class SSD
• 2TB Micron 1100 enterprise class SSD
• T-FORCE 500GB Vulkan SSD, supplied by Team Group
• EVGA 1000G 1000W Gold power supply unit
• Cooler Master 240mm CPU water cooler
• EVGA Nu Audio PCIe soundcard, supplied by EVGA
• Edifier R1320T Active speakers
• EVGA DG-77, mid-tower case supplied by EVGA
• Monoprice Crystal Pro 4K

Test Configuration – Software

• GeForce 445.87 used for the RTX 2060. GeForce 441.87 used for all other NVIDIA cards except 441.41 used for the GTX 1660 and 436.48 drivers used for the GTX 1060 SC. Game Ready 441.41 drivers are used for the GTX 1660. See NVIDIA Control Panel image below.
• AMD Adrenalin Software 20.4.2 used for both RX 5600 XTs. 20.1.2 is used for the RX Vega 56 and 19.12.2 is used for the RX 5500 XTs and the Red Devils RX 590 and RX 570. See the AMD Control Panel image below.
• VSync is forced off.
• AA enabled as noted in games; all in-game settings are specified with 16xAF always applied
• Gaming results show average frame rates in bold including minimum frame rates shown on the chart next to the averages in a smaller italics font where higher is better. Games benched with OCAT show average framerates but the minimums are expressed by the 99th percentile frametime in ms where lower numbers are better.
• Highest quality sound (stereo) used in all games.
• Windows 10 64-bit Home edition. DX11 titles are run under DX11 render paths. DX12 titles are generally run under the DX12 render path unless performance is lower than with DX11; and Borderlands 3, Total War Warhammer II, and Hitman 2 are tested on DX11 and on DX12. Seven games use the Vulkan API.
• Latest DirectX
• All 50 games are patched to their latest versions at time of posting.
• The Adrenalin 2020 control panel is used to set Radeon options
• The NVIDIA control panel is used for GeForce settings
• OCAT, latest version
• Fraps, latest version
• Afterburner, latest non-beta version
• Unigine Heaven 4.0 benchmark

50 PC Game benchmark suite & 3 synthetic tests

Vulkan Games

• DOOM Eternal
• Red Dead Redemption 2
• Wolfenstein: Youngblood
• Star Control Origins
• World War Z
• Strange Brigade
• Doom

DX12 Games

• Call of Duty Modern Warfare
• Borderlands 3
• Gears 5
• Control
• F1 2019
• Anno 1800
• Tom Clancy’s The Division 2
• Metro Exodus
• Battlefield V
• Shadow of the Tomb Raider
• Warhammer: Vermintide 2
• Total War: Warhammer II
• Forza 7
• Sniper Elite 4
• Civilization VI
• Ashes of the Singularity Escalation

DX11 Games

• Mech Warrior 5: Mercenaries
• Star Wars: Jedi Fallen Order
• Ghost Recon Breakpoint
• Destiny 2 Shadowkeep
• Borderlands 3
• Devil May Cry 5
• FarCry New Dawn
• Anthem
• Resident Evil 2
• Just Cause 4
• Hitman 2
• Assassin’s Creed: Odyssey
• Conan Exiles
• Far Cry 5
• Final Fantasy XV
• Kingdom Come: Deliverance
• Monster Hunter: World
• Star Wars: Battlefront II
• Middle Earth: Shadow of War
• Total Wars: Warhammer II
• Project CARS 2
• Prey
• ARK: Survival Evolved
• Mass Effect: Andromeda
• For Honor
• Overwatch
• Rainbow Six Siege
• Fallout 4
• The Witcher 3
• Grand Theft Auto V

Synthetic

• Firestrike – Basic & Extreme
• Time Spy DX12
• Superposition

AMD Adrenalin Control Center Settings

All AMD settings are set so as to be apples-to-apples when compared to NVIDIA’s control panel settings – all optimizations are off, Vsync is forced off, Texture filtering is set to High, and Tessellation uses application settings.

We pick the above Custom Global Performance settings for our benchmarking which has to be precisely repeatable, so Surface Format and other optimizations are disabled. Anisotropic Filtering is disabled by default but we always use 16X for all game benchmarks.

NVIDIA Control Panel settings

Here are the NVIDIA Control Panel settings that match AMD’s settings.

Let’s check out overclocking, temperatures, and noise next.

Overclocking, temperatures & noise

The Red Dragon RX 5600 XT 6GB is a low-power and quiet card even when overclocked. At stock settings, we could not hear it over the fans of our PC even when it ramps up, unlike with the Red Devil RX 590 or even the RX 570.

Above default tuning settings used maxed-out Heaven 4.0 running at stock clocks which allowed the Boost clocks to stay right around 1739MHz which is well above AMD’s typical clocks and also 10MHz higher than the ASUS RX 5600 XT EVO OC’s average 1729MHz clock.

The Red Dragon stayed cool generally at or below 75℃ even under a heavy load so its dual fans never became intrusive although it is slightly louder than the ASUS card. The RTX 2060 is another quiet card and it’s hard to pick based on (lack of) noise. At stock fan settings, the Red Dragon RX 5500 XT 8GB runs about 15℃ warmer than the ASUS card’s temperatures but we couldn’t hear either card over our PC’s cooling fans.

Overclocking is locked down to +2% on the core (1787MHz) and a max memory speed of 1860MHz. However, maxing out the overclock uses 25W more power and the temps go up.

If the power limit is set to maximum, performance increases by a small fraction, but the extra power used generates extra heat. For each user, we would recommend finding their GPU’s sweet spot for undervolting with max performance.

We will check performance compared with two other competing cards using 50 games, and then head for our conclusion.

Performance Summary Charts & Conclusion

Here are the performance results of 50 games and 3 synthetic tests comparing the PowerColor Red Dragon RX 5600 XT 6GB with the ASUS RX 5600 EVO OC 6GB, and versus the RX 2060 Founders Edition 6GB. Although the Red Dragon RX 5500 XT is designed for Ultra 1080P, the benches were also run at 2560×1440 to push the card beyond its limits.

Most gaming results show average framerates in bold text, and higher is better. Minimum framerates are next to the averages in italics and in a slightly smaller font. The games benched with OCAT show average framerates but the .1 minimums are expressed by frametimes in ms where lower numbers are better.

The first column gives the ($299) RTX 2060 FE performance results versus the Red Dragon RX 5600 XT ($289) in the second, and versus the ASUS RX 5600 XT EVO OC ($309) in the last. Performance “wins” are given by yellow text; yellow is also used for ties.

The Red Dragon RX 5600 XT is consistently faster than the more expensive ASUS EVO OC XT, probably due to its higher game/boost clocks. And it also does very well against the more expensive RTX 2060 where they trade blows.

Of course, the bigger picture adds the less expensive GTX 1660 Super and Ti cards into the mix. However, the Ti was almost made instantly irrelevant by the Super card which was not only fractionally slower, but much less expensive

The Big Picture

In this larger view, we add a few more cards – the EVGA GTX 1660 Super XC as well as a Red Devil RX Vega 56. In addition, we benchmark a Red Devil RX 590 and a vanilla EVGA RTX 1660 XC and a EVGA GTX 1060 SC 6GB as well as the lower end Pulse and Red Dragon RX 5500 XTs and the Red Devil RX 570.

We can see that the ASUS RX 5600 XT EVO OC is much faster than either the GTX 1660 Super or the Ti winning most of the benches. Although the GTX 1660 Ti has never justified its $50 more expensive price tag against the Super, the RX 5600 XT gives much more performance for its $289 price tag. The Red Dragon RX 5600 XT trades blows with the RTX 2060. Considering that it is priced $10 cheaper than the RTX 2060 and $20 cheaper than the slower ASUS RTX 5600 XT, we’d say it is a good value.

Let’s check out our conclusion.

Conclusion

The Red Dragon RX 5600 XT 6GB at $289/$299 (PowerColor’s MSRP is $289; we find it for sale at $299) brings an incrementally higher level of performance than the $309 ASUS RX 5600 T EVO OC, and it does very well against the RTX 2060. Any of these three cards would be an excellent upgrade from a RX 560/570 level of card. For playing older games, they are OK for 1440P gaming although they may require lowering settings for some modern titles.

AMD brings some great new features with Adrenalin 2020 software that turns it into an all-in-one platform for launching games, streaming, and sharing. And there is currently a great gaming bundle that really adds to the Red Dragon’s value at Newegg. Bundled RE3, Monster Hunter World, and a 3 month Xbox PC Pass bundle are very tempting free add-ons. The RTX 2060 that are priced at $299 do not come with any game bundles. For some gamers, their choice may come down to deciding if ray tracing effects with an entry level card are worth it now for the few games that implement some of these features coupled with a performance hit.

The PowerColor Red Dragon RX 5600 XT 6GB is a very handsome and solid modern Navi card with improving performance potential for AMD gamers that no longer happens with the older Polaris and Vega cards.

Red Dragon RX 5600 XT 6GB Pros

• The Red Dragon RX 5600 XT 6GB brings great new features at $289 and it’s faster than the more expensive ASUS RX 5600 XT EVO OC
• 14Gbps memory out of the box with no need to update the BIOS
• New RDNA architecture brings higher performance per clock and per watt
• Twenty new or updated features include anti-lag technology and Radeon Boost for competitive gamers and image sharpening for everyone with Adrenalin 2020 drivers
• The Red Dragon RX 5600 XT 6GB is a fast card for ultra 1080P gaming, high 1440P gaming, and a performance and power improvement over the Polaris series
• The Red Dragon RX 5600 XT trades performance blows with the RTX 2060
• Comes bundled with RE3, Monster Hunter World, and with the XBox PC Pass by some resellers

Cons: None

Nitpicks: No hardware ray tracing; overclocking locked down

The Verdict: Good Value

• 7nm hardware and new architecture brings 1080P RX 5600 XT AMD graphic cards to a higher playing field. RDNA will power several more years of Radeon PC graphics
• The RX 5600 XT brings a solid improvement over AMD’s last Polaris generation, and Adrenalin 2020 Edition software brings welcome new and improved features for AMD gamers

Extra new choices arriving with the RX 5600 XT benefit 1080P/1440P gamers looking for an upgrade. PowerColor makes an exceptionally nice offering with the Red Dragon, and together with its bundled games at Newegg for a limited time sweeten the deal.

We are going to take a short break to set up a Z390 motherboard with a clean installation of Win 10 in anticipation of an upcoming T-Force 2x32GB DDR4 memory kit review. Mario will post an Autonomous L-Shaped standing desk review in the meantime.

Happy Gaming!

The RTX 2060 6GB launched just over a year ago when BTR compared it versus the RX Vega 56 8GB using 39 games. We concluded this last January:

The RTX 2060 Founders Edition is priced at a reasonable $349 with no price premium over other partner RTX 2060s, and it is faster than … the more expensive premium factory overclocked RX Vega 56.

Since then, AMD has released a faster RX 5700 to replace Vega 56 which has dropped into the $265 price range if they still can be found, and NVIDIA has dropped the price of the entry-level and Founders Editions of the RTX 2060 to $299. We also saw that AMD’s Adrenalin 2020 drivers and NVIDIA’s CES drivers each bring new features and performance improvements. What about today’s performance using the latest drivers for each card?

For this look-back evaluation, we are going to benchmark more than 60 games at 2560×1440 and at 1920×1080 to test the PowerColor Red Devil RX Vega against the RTX 2060 6GB Founders Edition (FE) to see if anything has changed. BTR’s game benchmark suite is continually evolving as we always update our games to include the latest and most popular titles including Mech Warrior 5: Mercenaries and Jedi: Fallen Order.

Let’s check out the cards’ performance after we look over our test configuration.

Test Configuration – Hardware

• Intel Core i7-8700K (HyperThreading and Turbo boost is on to 4.8GHz for all cores; Coffee Lake DX11 CPU graphics).
• EVGA Z370 FTW motherboard (Intel Z370 chipset, latest BIOS, PCIe 3.0/3.1 specification, CrossFire/SLI 8x+8x), supplied by EVGA
• T-FORCE 16GB DDR4 (2x8GB, dual channel at 3866 MHz), supplied by Team Group
• Red Devil RX Vega 56 8GB at Red Devil clocks, on loan from PowerColor
• RTX 2060 Founders Edition at Founders edition factory settings, on loan from NVIDIA
• 2 x 480GB Team Group SSDs – one for AMD, and one for NVIDIA
• 1.92TB San Disk enterprise class SSD
• 2TB Micron 1100 enterprise class SSD
• 500GB Vulkan SSD, supplied by Team Group
• EVGA 1000G 1000W Gold power supply unit
• Cooler Master 240mm CPU water cooler
• EVGA Nu Audio PCIe soundcard, supplied by EVGA
• Edifier R1320T Active speakers
• EVGA DG-77, mid-tower case supplied by EVGA
• LG 43″ HDR 4K TV
• Monoprice Crystal Pro 4K

Test Configuration – Software

• GeForce 417.54 drivers (January, 2019) and 441.41 drivers (January, 2020). See NVIDIA Control Panel image below.
• AMD Adrenalin Software 18.12.3 (January, 2019) and 20.1.2 (January 2020). See the AMD Control Panel image below.
• VSync is forced off.
• AA enabled as noted in games; all in-game settings are specified with 16xAF always applied
• Gaming results show average frame rates in bold including minimum frame rates shown on the chart next to the averages in a smaller italics font where higher is better. Games benched with OCAT show average framerates but the minimums are expressed by the 99th percentile frametime in ms where lower numbers are better.
• Highest quality sound (stereo) used in all games.
• Windows 10 64-bit Home edition. DX11 titles are run under DX11 render paths. DX12 titles are generally run under the DX12 render path unless performance is lower than with DX11; and Borderlands 3, Total War Warhammer II, and Hitman 2 are tested on DX11 and on DX12. Seven games use the Vulkan API.
• Latest DirectX
• All 60+ games are patched to their latest versions at time of posting.
• The Adrenalin 2020 control panel is used to set Radeon options
• The NVIDIA control panel is used for GeForce settings
• OCAT, latest version
• Fraps, latest version
• Unigine Heaven 4.0 benchmark

60+ PC Game benchmark suite & 3 synthetic tests

Synthetic

• Firestrike – Basic & Extreme
• Time Spy DX12
• Superposition

DX11 Games

• Grand Theft Auto V
• The Witcher 3
• Fallout 4
• Rainbow Six Siege
• Battlefield 1
• Resident Evil 7
• Overwatch
• For Honor
• Mass Effect: Andromeda
• Ghost Recon Wildlands
• ARK: Survival Evolved
• Prey
• Hellblade: Senua’s Sacrifice
• Project CARS 2
• Total Wars: Warhammer II
• Middle Earth: Shadow of War
• Star Wars: Battlefront II
• Monster Hunter: World
• Kingdom Come: Deliverance
• Final Fantasy XV
• Far Cry 5
• The Crew 2
• F1 2018
• Conan Exiles
• Assassin’s Creed: Odyssey
• Call of Duty: Black Ops 4
• Hitman 2
• Just Cause 4
• Resident Evil 2
• Anthem
• FarCry New Dawn
• Devil May Cry 5
• Borderlands 3
• Destiny 2 & Shadowkeep
• Ghost Recon Breakpoint
• The Outer Worlds

DX12 Games

• Ashes of the Singularity: Escalation
• Tom Clancy’s The Division
• Hitman
• Deus Ex: Mankind Divided
• Gears of War 4
• Civilization VI
• Sniper Elite 4
• Forza 7
• Total War: Warhammer II
• Warhammer: Vermintide 2
• Shadow of the Tomb Raider
• Hitman 2
• Battlefield V
• Metro Exodus
• Tom Clancy’s The Division 2
• Anno 1800
• F1 2019
• Control
• Gears 5
• Borderlands 3
• Call of Duty Modern Warfare

Vulkan Games

• DOOM
• Strange Brigade
• World War Z
• Wolfenstein: Youngblood
• Star Control Origins
• Red Dead Redemption 2

AMD Adrenalin Control Center Settings

All AMD settings are set so as to be apples-to-apples when compared to NVIDIA’s control panel settings – all optimizations are off, Vsync is forced off, Texture filtering is set to High, and Tessellation uses application settings.

We pick the above Custom Global Performance settings for our benchmarking which has to be precisely repeatable, so Surface Format and other optimizations are disabled. As above, Anisotropic Filtering is disabled by default but we always use 16X for all game benchmarks.

NVIDIA Control Panel settings

Here are the NVIDIA Control Panel settings that match AMD’s settings.

We did not adjust either card’s Power or Temperature limits since it does not benefit low-to-mid-power cards.

Let’s check the performance of our more than 60 game benchmark suite, and then head for our conclusion.

Performance Summary Charts

Here are the summary charts of more than 60 games and 3 synthetic tests. The highest settings are always chosen. The benches were run at 1920×1080 and 2560×1440. All results show average framerates in bold and the minimums are next to them in italics, and higher is always better. Games benched with OCAT also give average framerates but the minimums are expressed by the 99th percentile frametime in ms where lower numbers are better. “X” means the benchmark was not run.

The first two columns shows the RTX 2060 performance results with 2019 GeForce drivers in the first column compared with 2020 drivers in the second. The RX RX Vega 56’s results are compared in the last two columns; three shows Adrenalin 2020 driver performance and column four shows 2019 results. “Wins” between the RTX 2060 and the RX Vega 56 are based on averages and the higher performing card’s framerates are given in yellow text for 2019 performance results and in gold text for 2020 results unless there is a tie in which case both sets of results will be colored.

There is a bug with Control where the game refuses to launch in DX12 for the RX Vega 56 and we did not count the results. Also the benchmarks changed between 2019 and 2020 for Hitman 2 and for Destiny 2 so driver performance across the two years should not be compared for these games.

The RTX 2060 is still our fastest card for the majority of our games although the Vega performance bug has been fixed for Forza 7 and the RX Vega 56 has narrowed the gap for several of the titles that we tested both last year and this year. Last year, the RX Vega 56 won 23 benches and tied one to the RTX 2060’s 52 wins. This January, the RX 2060 won 64 benches and tied 3 while the RX Vega 56 took 34 individuals wins.

Let’s head for our conclusion.

The Conclusion

Both AMD and NVIDIA have made progress with their drivers but as in last January’s evaluation, the RTX 2060 FE 6GB is still overall faster than the RX Vega 56 8GB although the Red Devil has made up performance ground in some titles. If overclocking is taken into consideration, the RTX 2060 will pull away further.

The RX Vega series has been discontinued in favor of Navi Radeons although RX Vega 56es can still be found around for around $265 in the USA. The RTX 2060 offers entry-level GeForce ray tracing capabilities and can be found starting at $299.

Stay tuned, there is a lot coming from us at BTR. We have two video card reviews scheduled for early next week that you do not want to miss. And we have a T-FORCE LIQUID CARDEA M.2 SSD on deck followed by a ray traced game review and more VRSS with Boneworks.

Happy Gaming!

This is a follow-up to our June VR evaluation where we saw the RTX 2070 move ahead of the GTX 1080 and solidly beat the liquid-cooled RX Vega 64 in VR performance. We now present another eleven-game VR performance face-off between the Navi Red Devil RX 5700 versus the Turing RX 2060 Founders Edition (FE) measuring frametimes and unconstrained performance using FCAT VR.

Since we posted our original review over two and one-half years ago, we have benchmarked VR games for our follow-up reviews over the past thirty months. We have also compared FCAT VR with our own video benchmarks using a camera to capture images directly from the Rift’s lens. For BTR’s VR testing methodology, please refer to this evaluation.

We currently benchmark eleven Oculus Rift VR games at maxed-out settings using the Red Devil RX 5700 versus the RTX 2060 Founders Edition (FE). BTR’s testing platform is an Intel Core i7-8700K at 4.8GHz, an EVGA Z370 FTW motherboard and 16 GB of T-Force XTREEM DDR4 at 3866MHz on Windows 10 64-bit Home Edition. Here are the eleven VR games that we benchmark:

• Batman VR
• Chronos
• DiRT: Rally
• Elite Dangerous
• Fallout 4
• Hellblade: Sennua’s Sacrifice
• Obduction
• Project CARS 2
• Robinson: The Journey
• Skyrim
• The Vanishing of Ethan Carter

It is important to be aware of VR performance since poorly delivered frames will actually make a VR experience quite unpleasant even leading to VR sickness. It is also very important to understand how to accurately benchmark VR as explained here. But before we benchmark our eleven VR games, take a look at our Test Configuration.

Test Configuration – Hardware

• Intel Core i7-8700K (HyperThreading and Turbo boost is on to 4.8 GHz for all cores; Coffee Lake DX11 CPU graphics).
• EVGA Z370 FTW motherboard (Intel Z370 chipset, latest BIOS, PCIe 3.0/3.1 specification, CrossFire/SLI 8x+8x), supplied by EVGA
• T-Force XTREEM 16GB DDR4 (2×8 GB, dual channel at 3866MHz), supplied by Team Group
• RTX 2060 6 GB Founders Edition, stock FE clocks, on loan from NVIDIA
• Red Devil RX 5700, 8GB at Red Devil clocks, on loan from PowerColor
• 2 x 480GB Team Group SSDs – 1 for Radeon and 1 for GeForce
• 1.92 TB San Disk enterprise class SSD
• 2 TB Micron 1100 enterprise class SSD
• Seasonic 850W Gold Focus power supply unit
• EVGA CLC 280mm CPU water cooler, supplied by EVGA
• EVGA Nu Audio stereo PCIe sound card, supplied by EVGA
• Edifier R1280T active desktop speakers
• EVGA DG-77, mid-tower case supplied by EVGA
• Monoprice Crystal Pro 4K
• Oculus Rift CV1

Test Configuration – Software

• Nvidia’s Game Ready 436.15 WHQL drivers.
• Adrenalin Software Edition 19.8.2
• Unconstrained framerate results show average frame rates in bold including dropped and synthetic frames shown on the chart next to the averages in a smaller italics font.
• Highest quality sound (stereo) used in all games.
• Windows 10 64-bit Home edition v1903.
• Latest DirectX
• All 11 VR games are patched to their latest versions at time of publication.
• Precision X1
• Wattman
• FCAT-VR Capture v0.9.3202.0 UAC
• FCAT-VR Beta 17

11 VR Game benchmark suite & 2 synthetic tests

Synthetic

• VRMark Cyan Room
• Unigine Superposition VR Benchmark

Oculus Rift VR Games

• Batman VR
• Chronos
• DiRT: Rally
• Elite Dangerous
• Fallout 4
• Hellblade: Senua’s Sacrifice
• Obduction
• Project CARS 2
• Robinson: The Journey
• Skyrim
• The Vanishing of Ethan Carter

The Unreal 4 engine is one of the two most popular engines for VR development, and five of our games use it. The Creation engine is used for two games, while the Ego, COBRA, Cryengine, and Madness engines are each represented by one game. All of the engines are identified in each of the following game’s descriptions.

Let’s individually look at our eleven VR games’ performance using FCAT-VR plus two synthetic benches. First up, Batman Arkham VR.

Batman Arkham VR

Batman Arkham VR immerses you into Batman’s world without involving fighting. It is a short game using the Unreal Engine with an emphasis on detective work and puzzle solving. Although it is an older title, it has good visuals.

Batman Arkham VR has very few adjustable settings so we benchmark at its full resolution and with 100% pixel density

Here are the performance results of our competing cards compared using the FCAT-VR generated chart.

The unconstrained framerate of the RTX 2060 is 163.8 FPS with no dropped frames, and it is slower than the RX 5700 at 178.7 FPS with 12 dropped frames. Neither of our cards have to resort to generating synthetic frames.

Chronos

Chronos is a Rift launch title with graphics options that are good for GPU testing. It is a 3rd-person view RPG which includes many puzzles as well as requiring the player to be good with combat.

We picked Epic which is the highest settings for each option.

Here is the FCAT-VR generated chart for Chronos.The RTX 2060 manages 79.8 FPS with 4 dropped frames while 39% of its frames are synthesized while the RX 5700 leads with unconstrained 92.1 FPS with 7 dropped frames and 10% synthetic frames. We would recommend lowering setting for each card.

Let’s check out DiRT: Rally next.

DiRT: Rally

DiRT: Rally is built on the Ego Engine. It features a built-in benchmark that is very repeatable. DiRT: Rally is a fun and demanding VR game that requires its players to learn its challenging road conditions. DiRT: Rally has many individual settings although we benchmarked using the stock “Ultra” preset.

Here is the DiRT: Rally benchmark run by our 2 test cards as charted by FCAT-VR.

The RTX 2060 manages 67.4 unconstrained frames per second with 41 dropped frames and 15% synthetic frames while the RX 5700 pulls ahead with 100.8 FPS, 16 frames are dropped, and none of its frames are synthetic.

Elite Dangerous

Elite Dangerous is a very popular space sim built using the COBRA engine. It is difficult to find a repeatable and demanding benchmark outside of the training missions.

A player will spend a lot of time piloting a space cruiser while completing a multitude of tasks. Elite Dangerous is also co-op and multiplayer, and you need to be connected to the Internet. It has a very dedicated following of players.

We picked the default Ultra setting and maxed out our FoV. Here are the frametimes.

The RTX 2060 delivers 69.0 unconstrained frames per second with 1 dropped frame edging out the RX 5700 with 67.8 FPS with no dropped frames. For each card, 50% of the frames were synthetically generated.

Let’s check out another demanding game, Fallout 4.

Fallout 4 VR

Fallout 4 uses the Creation Engine. It has been somewhat optimized for the Oculus Rift and the controls work decently with it although we prefer using the Vive. We benchmark at its highest settings and with TAA.

Here is the frametime plot for Fallout 4.

In Fallout 4, the RTX 2060 delivers 69.7 FPS with 104 dropped frames and 10% synthetic frames while the RX 5700 delivered 58.7 unconstrained FPS with 390 dropped frames and also generated 10% synthetic frames.

Next we benchmark Hellblade: Senua’s Sacrifice.

Hellblade: Senua’s Sacrifice

Hellblade: Senua’s Sacrifice is a very impressive game visually using the Unreal 4 engine. It is an very dark and disturbing game that is far more intense in VR than playing the pancake version. We benchmark at Very High settings and with TAA.

Here is the frametime plot for Hellblade: Senua’s Sacrifice.

In Hellblade: Senua’s Sacrifice, the RTX 2060 delivers 93.7 FPS with 29 dropped frames. The RX 5700 does better with 111.8 unconstrained FPS with 17 dropped frames. Neither card needs reprojection.

Next we will check out another very demanding VR game, Obduction.

Obduction

Obduction is an adventure video game developed by Cyan Worlds using the Unreal 4 engine. Obduction is considered the spiritual successor to Myst and Riven. There is an emphasis on puzzle solving which get progressively more difficult as the player progresses.

Obduction’s visuals are graphically intensive scenes that exceed the Rift’s usual minimum recommendation of a GTX 970. There are adjustable menu sliders that will allow the player to dial down the settings although we picked Epic. We picked the area below to benchmark. The game still does not seem to be fully-optimized and there are occasional stutters and warp misses.

Here is the frametime plot.

The RTX 2060 delivers 72.2 FPS with 1 dropped frames but with 27% synthetic frames generated. The RX 5700 delivers 70.8 FPS with 6 dropped frames and 31% of the frames are synthesized.

Next we will check out another demanding VR game, Project CARS 2.

Project CARS 2

One has to experience Project CARS 2 as a player to appreciate it, and there is absolutely no way to convey the incredible sense of immersion that comes from playing it in VR using a wheel and pedals. It uses the in-house Madness engine and its physics implementation is outstanding.

Project CARS 2 offers many performance options and settings, and we prefer playing with SMAA Ultra which seems to offer a bit of temporal anti-aliasing.

We picked the SMAA Ultra pre-set and we used maximum settings except for Motion Blur which looked best to us on Low. The other setting were set at their highest.

Project CARS 2 makes good use of the Oculus Asynchronous Space Warp (ASW) feature so that a card delivering less than 90 FPS will intelligently reproject every other frame and will still be playable without causing VR sickness. We would also recommend lowering grass and reflections to maximize framerate delivery if necessary.

Here are the results of our FCAT-VR benching.

The RTX 2060 delivers 59.3 unconstrained FPS, no dropped frames and 50% synthetic frames generated, while the RX 5700 delivers 64.5 FPS with 3 dropped frames and 45% of its frames are synthesized.

Let’s benchmark Robinson: The Journey.

Robinson: The Journey

Robinson: The Journey is first person adventure/puzzle game developed by Crytek using the Cryengine. Just like with Crytek’s PC games, the visuals are among the very best to be found in any VR game.

For us, Crytek’s Robinson: The Journey was a blast to play in VR with a start to finish in about 6 hours. It’s probably Crytek’s best story and you even get a pet baby T-Rex.

We benchmarked at the highest settings as below but used 1.0X as the resolution setting.

Here is the frametime plot of our competing two video cards:

The RTX 2060 delivers 141.9 FPS with 3 dropped frames. The RX 5700 delivers 149.3 FPS with 1 dropped frame. No synthetic frames were needed to be generated by our cards.

Let’s benchmark Skyrim VR.

Skyrim VR

Skyrim VR is an older game that is not as demanding as many of the newer VR games so its performance is still very good on maxed-out settings using its Creation engine.

We benchmarked Skyrim VR using its highest settings, but we left the resolution at 100%.

The RTX 2060 delivers 114.8 unconstrained frames per second with 10 dropped frames, while the RX 5700 delivered 95.6 FPS with 9 dropped frames. No synthetic frames were needed to be generated by either card.

The Vanishing of Ethan Carter

The Vanishing of Ethan Carter is built on the Unreal 4 engine and it boasts amazing visuals. It is considered by some to be a “walking simulator”, but it is also an excellent detective game with great puzzles. Beware as its locomotion tends to make many players VR sick.

There are just a few in-game graphics options available, so we picked 100% resolution with TAA.

The RTX 2060 delivered 146.6 FPS with 8 dropped frames. The RX 5700 delivers 141.0 FPS with 7 dropped frames. No synthetic frames were needed to be generated.

If there is performance headroom, we recommend increasing the resolution multiplier as far as a player’s card can handle without the need to generate synthetic frames. The Vanishing of Ethan Carter is the last of our eleven tested games, so let’s look at the summary charts.

Unconstrained FPS, dropped frames, and synthetic frames

In this chart, we summarize the overall Unconstrained Framerates summing up our eleven test games. This time we also show the RX Vega 64, the GTX 1080, and the RTX 2070 on recent drivers for comparison. The Unconstrained Framerate is given by bold text, followed by the dropped frames (d), and the percentage of synthetic frames (s) in italics.

Let’s check out our conclusion.

Conclusion

The RX 5700 and the RTX 2060 are in approximately the same class when it comes to VR performance. The RX 5700 delivers higher unconstrained frames in six games while the RTX 2060 is faster in five games out of the eleven VR games we benchmarked.

Although the RX Vega 64 liquid-cooled edition is in the same class as the GTX 1080 and the RTX 2070, it falls far behind both of these cards in VR performance and it is also beaten by the RX 5700 in all but two games. We can conclude that Navi is better suited for VR than Vega.

We will continue our VR benchmarking as we transition from benchmarking with the Oculus Rift to benchmarking with the Vive Pro. The Pro is more demanding of video cards with better visuals and a higher resolution than the Rift.

Our upcoming review will introduce Vive Pro benchmarking in addition to using the Oculus Rift as we continue our “VR Wars” series by comparing the RTX 2060 SUPER FE and the RTX 2070 SUPER FE versus the 50th Anniversary Edition of the RX 5700 XT. Stay tuned!

Happy VR gaming!

This overclocking showdown is the follow-up to the RTX 2080 SUPER Founders Edition (FE) review where we are now focusing on its maximum overclocked performance versus the maximum overclocked original RTX 2080 FE. We have also matched the memory and core clocks on both cards to see what performance changes the upgrade of the RTX 2080 SUPER FE edition brings with its increased cache size and with two more SM’s enabled over the original RTX 2080 FE.

The RTX 2080 SUPER was released with about 4-8% more performance over the original RTX 2080 FE to differentiate itself more from the RTX 2070 SUPER which encroaches on the original’s performance territory. Besides the architectural upgrade, the RTX 2080 SUPER boasts a 15MHz faster core and 15.6Mbps GDDR6 which is clocked 750MHz higher than the original RTX 2080.

Besides matching their clocks, we have optimized and maxed-out our RTX 2080 SUPER FE and RTX 2080 FE overclocks with all performance options set to their highest limits to get the maximum performance from each card by using a carefully selected 7-game benchmark suite which uses the most accurate and reliable benchmarks that have been each run many more times than usual.

We use the latest WHQL public release Game Ready 431.60 drivers so our results may differ from the 431.56 press drivers used for the RTX 2080 SUPER review. In addition, the cards’ fan speeds have been set to at least 80% to eliminate thermal variability.

The RTX 2080 SUPER FE Overclocking

The RTX 2080 SUPER FE steadily averages 1935-1950MHz Boost under full gaming load after warm-up in a cool room with the card’s Power and Temperature limits maximized. We achieved a completely stable 120MHz offset to the core and a +500MHz offset to the GDDR6. We found that +600MHz was also possible, but the performance stopped increasing, and at +650MHz it began artifacting.

Our final overclock of the RTX 2080 SUPER FE gave us an offset of +120MHz to the core (2055MHz Boost) and a +500MHz offset to its faster memory (8250MHz).

The RTX 2080 Founders Edition Overclocking

The RTX 2080 FE averages 1905-1920MHz at stock settings in a cool room with its Power and Temperature limits maximized. Our final manual overclock of the original RTX 2080 FE achieves a +140MHz offset to the core (2055MHz Boost) and a +590MHz offset to the memory (7620MHz). This comparison should give a rough idea of how well these two cards each scale with overclocking and how their performance compares.

Matching the RTX 2080 FE’s Clocks to the RTX 2080 SUPER’s Clocks

To match the original RTX 2080 memory clocks, the RTX 2080 SUPER had its memory clocks lowered from 7750MHz to 7000MHz which we were able to do with Precision 1X but not with Afterburner. And to match the core clocks, the original RTX 2080 had +15MHz added to its core speed so that both cards held steady clocks, each card only fluctuating between 1920MHz and 1935MHz as the ambient temperatures were similar.

Our fan profile was set to allow both cards to ramp up from 80% to 85% for all of these tests so that neither card reached 70C even when fully overclocked so as to remove throttling caused by temperature variability.

Interestingly, both cards each reached 2055MHz boost when fully manually overclocked, so it is likely that most of the performance differences may come from the memory speed. The overclocked RTX 2080 SUPER can reach 8250MHz using faster GDDR6 while the original RTX 2080 can only handle 7590MHz.

The upgraded changes made from the original RTX 2080 to the RTX 2080 SUPER are (1) a GDDR6 memory speed upgrade to 15.6Gbps, (2) an additional speed increase to its core, (3) increased cache size, and with (4) two more SM’s enabled.

We feel that since both RTX 2080s have identical cooling and similar overall characteristics, lowering the memory clocks -750MHz on the SUPER to match the original RTX 2080 and increasing the core clock +15MHz of the original card to match the RTX 2080 SUPER may give us an idea of the performance changes from the increased cache size and additional SMs.

Let’s check the performance results of 7 games with our games at stock, matched, and overclocked speeds using the Founders Editions of the RTX 2080 SUPER versus the original RTX 2080 FE after the Testing configuration.

Testing Platform

We test 7 games and 4 synthetic benchmarks at 3840×2160 and at 2560×1440. Our testing platform is a recent install of Windows 10 64-bit Home Edition v1903, and we are using an i7-8700K which turbos all 6 cores to 4.8GHz, an EVGA Z370 FTW motherboard, and 16GB of XTREEM DDR4 at 3866MHz. The games, settings, and hardware are identical except for the cards being compared.

Before we run our benchmarks, let’s check out the test configuration.

Test Configuration – Hardware

• Intel Core i7-8700K (HyperThreading and Turbo boost are on to 4.8GHz for all cores; Coffee Lake DX11 CPU graphics).
• EVGA Z370 FTW motherboard (Intel Z370 chipset, latest BIOS, PCIe 3.0/3.1 specification, CrossFire/SLI 8x+8x), supplied by EVGA
• T-FORCE XTREEM 16GB DDR4 (2x8GB, dual channel at 3866MHz), supplied by Team Group
• RTX 2080 8GB Founders Edition, at stock FE clocks, matched to the RTX 2080 SUPER clocks, and overclocked, on loan from NVIDIA
• RTX 2060 SUPER 8GB Founders Edition, at stock FE clocks, matched to RTX 2080 clocks, and overclocked, on loan from NVIDIA.
• 480 GB Team Group SSD
• 1.92 TB San Disk enterprise class SSD
• 2 TB Micron 1100 SSD
• Seasonic 850W Gold Focus power supply unit
• EVGA CLC 280mm CPU water cooler, supplied by EVGA
• EVGA NuAudio soundcard, supplied by EVGA
• Edifier 1280T powered speakers
• EVGA DG-77, mid-tower case supplied by EVGA
• Monoprice Crystal Pro 4K

Test Configuration – Software

• GeForce 431.60 WHQL drivers
• VSync is forced off.
• AA enabled as noted in games; all in-game settings are specified
• Gaming results show average frame rates in bold including minimum frame rates shown on the chart next to the averages in a smaller italics font, and higher is better. Civilization VI uses frametimes measurements and lower is better.
• Highest quality sound (stereo) used in all games.
• Windows 10 64-bit Home edition latest version (v1903) and fully updated.
• Latest DirectX
• All games are updated to their latest versions at time of publication.
• MSI Afterburner, latest beta
• Precision X1 latest version
• Fraps, latest version

7 PC Game benchmark suite & 4 synthetic tests

Synthetic

• Firestrike – Basic & Extreme
• Time Spy DX12
• Superposition
• Heaven 4.0

DX11 Games

• For Honor
• Ghost Recon Wildlands
• Prey
• Total War: Warhammer II

DX12 Games

• Civilization VI
• Metro Exodus
• Metro Exodus – DLSS+Ultra RTX

Vulkan Game

• Strange Brigade

NVIDIA Control Panel settings

Here are the NVIDIA Control Panel settings that match AMD’s settings.

We used the latest beta of Afterburner to set both GeForce’s highest Power and Temperature targets. In addition we added .1mV for overclocking. By setting the Power Limits and Temperature limits to maximum, they do not throttle, but they can reach and maintain their individual maximum clocks. This is particularly beneficial for almost all higher power cards.

Let’s head to the performance results of our overclocking and matched-clocks showdown between the RTX 2080 FE and the RTX 2080 SUPER FE.

Performance summary charts

Here are the performance results of 7 games and 4 synthetic tests comparing the stock, matched, and overclocked RTX 2080 FE 8GB versus the stock, matched and overclocked RTX 2080 SUPER 8GB Founders Edition. The highest settings are always chosen and the settings are listed on the charts. The benches were run at 3840×2160 and at 2560×1440.

Gaming results show average framerates in bold text, and higher is better. Minimum framerates are next to the averages in italics and in a slightly smaller font. Civilization 6 results are given as frametimes in ms where lower numbers are better.

First up, the Overclocked results:

We see that both cards each gain performance from overclocking and the RTX 2080 SUPER pulls further away from the original RTX 2080 by virtue of its faster memory although they both achieve the same core overclock (2050MHz).

Now let’s check out the matched results where the clocks are equalized between both cards. The RTX 2080 SUPER has its memory clocks reduced by -750MHz to 7000MHz to match the original RTX 2080’s memory speeds, and the original RTX 2080 is overclocked by +15MHz to match the SUPER’s core speed.

Interestingly, the RTX 2080 SUPER tends to give slightly less performance when it is running at the same clocks as the RTX 2080. Of course, we only tested two samples with a dozen benchmarks, and all of the results are generally within 3% which could be attributed to benching ‘noise’ or margin of error.

There is one notable exception with the RTX 2080 SUPER running Strange Brigade. Lowering the SUPER’s memory clocks drastically dropped the minimum frame rates although we ran the benchmark many times with the same results. One may perhaps speculate that the RTX 2080 SUPER needs a bit more core speed and higher memory clocks to take full advantage of its extra two SMs and the larger cache.

Let’s check out our conclusion.

Conclusion

This has been quite an interesting exploration for us in evaluating the stock, clocked-matched, and overclocked RTX 2080 SUPER FE versus the stock, clock-matched, and overclocked original RTX 2080 FE. The RTX 2080 SUPER FE overall is a bit faster than the original RTX 2080 at stock. But overclocked, the new RTX 2080 SUPER pulls away even further away thanks to its significantly faster memory and architectural upgrade.

A gamer looking for the best value in the $700 range should pick either an original RTX 2080 or a RTX 2080 SUPER with confidence. Both cards overclock equally well on their cores and in our case, both reached 2050MHz, although the overclocked RTX 2080 SUPER card pulls further away from the overclocked original RTX 2080.

We had originally intended to get back to VR comparing the RX 5700 XT with the new SUPER cards, but our Anniversary Edition became defective and we sent it back to AMD for an exchange. We also received a Vive Pro from HTC Vive and can’t wait to bring you a mega-VR review. However, while waiting for our RX 5700 XT to return from RMA, we will bring you two Team Group reviews: a Vulcan 500GB SSD review and a T-FORCE DDR4 gaming review starting next week.

Stay tuned. In the meantime, if you have any comments or questions, feel free to post them in the comments section below.

Happy Gaming!

As a regular BabelTechReviews’ feature, this driver performance analysis will showcase the performance of the EVGA GTX 1060 SC and the RTX 2070 Founders Edition (FE) with 42 PC games using the latest GeForce Game Ready 419.67 Driver which released earlier this week. We will compare these drivers versus 418.91 for the GTX 1060 and 417.71 for the RTX 2070.

We document the performance changes of the current Game Ready 419.67 Driver on Windows 10 at 1920×1080 and at 2560×1440 resolutions using the latest games including The Division 2, Devil May Cry 5, and Metro Exodus.

Our testing platform is a recent install of Windows 10 64-bit Home Edition, and we are using a Core i7-8700K which turbos all 6 cores to 4.7 GHz, an EVGA Z370 FTW motherboard, and 16GB of HyperX DDR4 3333MHz. The games tested, settings, and hardware are identical except for the drivers being compared.

Test Configuration – Hardware

• Intel Core i7-8700K (HyperThreading and Turbo boost are on to 4.7 GHz for all cores; Coffee Lake DX11 CPU graphics).
• EVGA Z370 FTW motherboard (Intel Z370 chipset, latest BIOS, PCIe 3.0/3.1 specification, CrossFire/SLI 8x+8x), supplied by EVGA
• HyperX 16GB DDR4 (2×8 GB, dual channel at 3333 MHz), supplied by HyperX
• EVGA GTX 1060 6GB SC, stock SC clocks, on loan from EVGA
• RTX 2070 Founders Edition 8GB, at Founders Edition clocks, on loan from NVIDIA
• 480 GB Team Group SSD
• 1.92 TB San Disk enterprise class SSD
• 2 TB Micron 1100 SSD
• Seasonic 850W Gold Focus power supply unit
• EVGA CLC 280mm CPU water cooler, supplied by EVGA
• EVGA Nu Audio stereo PCIe sound card, on loan from EVGA
• Edifier R1280T active desktop speakers
• Grado SR60e headphones
• EVGA DG-77, mid-tower case supplied by EVGA
• Monoprice Crystal Pro 4K

Test Configuration – Software

• Nvidia GeForce 418.91 WHQL drivers are used for the GTX 1060 SC and 417.71 drivers used for the RTX 2070 FE and are compared with the latest 419.67 Game Ready Driver.
• VSync is forced off.
• AA enabled as noted in games; all in-game settings are specified
• Gaming results show average frame rates in bold including minimum frame rates shown on the chart next to the averages in a smaller italics font.
• Highest quality sound (stereo) used in all games.
• Windows 10 64-bit Home edition. All DX11 titles were run under DX11 render paths. DX12 titles are generally run under the DX12 render path unless performance is lower than with DX11. Three games use the Vulkan API.
• Latest DirectX
• All 42 games are patched to their latest versions at time of publication.
• MSI Afterburner, latest beta.
• OCAT, latest version
• Fraps, latest version

42 PC Game benchmark suite & 3 synthetic tests

Synthetic

• Firestrike – Basic & Extreme
• Time Spy DX12
• Superposition

DX11 Games

• Grand Theft Auto V
• The Witcher 3
• Fallout 4
• Rainbow Six Siege
• Battlefield 1
• For Honor
• Ghost Recon Wildlands
• Mass Effect: Andromeda
• Prey
• Hellblade: Senua’s Sacrifice
• Project CARS 2
• Middle Earth: Shadow of War
• Total War: Warhammer II
• Destiny 2
• Star Wars: Battlefront II
• Monster Hunter: World
• Kingdom Come: Deliverance
• Final Fantasy XV
• Far Cry 5
• The Crew 2
• F1 2018
• Assassin’s Creed: Odyssey
• Call of Duty: Black Ops 4
• Hitman 2
• Just Cause 4
• Resident Evil 2
• Devil May Cry 5

DX12 Games

• Tom Clancy’s The Division
• Ashes of the Singularity: Escalation
• Hitman
• Deus Ex Mankind Divided
• Gears of War 4
• Civilization VI
• Sniper Elite 4
• Forza 7
• Shadow of the Tomb Raider
• Battlefield V
• Metro Exodus
• Tom Clancy’s The Division 2

Vulkan Games

• DOOM
• Wolfenstein: The New Colossus
• Strange Brigade

NVIDIA Control Panel settings

Here are the NVIDIA Control Panel settings.

We used MSI’s Afterburner to set the RTX 2070 FE’s and the GTX 1060 SC’s highest Power and Temperature targets. By setting the Power Limits and Temperature limits to maximum, they can maintain their maximum clocks without throttling.

Game Ready GeForce 419.67 WHQL Drivers

This latest GeForce Game Ready 419.67 driver was released primarily for Battlefield V: Firestorm, Anthem, Shadow of the Tomb Raider, and Sekiro: Shadows Die Twice. It is supposed to provide the best experience for RTX GeForce gamers and it is mandatory if you play Anthem with DLSS. It also contains a lot of bug fixes found in earlier drivers, and we also found it introduces some bugs of its own.

The download links for the latest GeForce 419.67 drivers can be found starting here. The release notes can also be found here. The included highlights below are from NVIDIA’s website.

Game Ready Drivers provide the best possible gaming experience for all major new releases, including Virtual Reality games. Prior to a new title launching, our driver team is working up until the last minute to ensure every performance tweak and bug fix is included for the best gameplay on day-1.

Game Ready
Provides the optimal gaming experience for Battlefield V: Firestorm, Anthem, Shadow of the Tomb Raider, and Sekiro: Shadows Die Twice

Gaming Technology
Includes support for G-SYNC compatible surround mode and two new G-SYNC compatible monitors.

Let’s head to the charts to compare the driver changes from the GeForce 417.71 drivers for the RTX 2070 and the GeForce 418.91 drivers using the GTX 1060 SC to the latest 419.67 Game Ready Drivers.

As a returning feature for BabelTechReviews, this driver performance analysis will showcase the performance of 40 PC games using the latest GeForce Game Ready 418.11 Driver which released yesterday. We will compare these drivers versus the RTX 2060 Founders Edition 417.54 launch drivers.

We will document the performance changes of the current GeForce 418.81 driver from the 417.54 drivers by testing 40 games at 1920×1080 and at 2560×1440 plus 4 synthetic benchmarks including 3DMark’s Port Royal DLSS. In addition, Strange Brigade is benchmarked on the Vulkan and DX12 APIs, and Total War: Warhammer II is benchmarked on DX11 and on DX12. We now use the new Miami built-in benchmark of Hitman 2 which recently got patched into the game and our own custom time demo of Resident Evil 2.

Our testing platform is a recent install of Windows 10 64-bit Home Edition, and we are using a Core i7-8700K which turbos all 6 cores to 4.7 GHz, an EVGA Z370 FTW motherboard, and 16GB of HyperX DDR4 3333MHz. The games tested, settings, and hardware are identical except for the drivers being compared.

Test Configuration – Hardware

• Intel Core i7-8700K (HyperThreading and Turbo boost are on to 4.7 GHz for all cores; Coffee Lake DX11 CPU graphics).
• EVGA Z370 FTW motherboard (Intel Z370 chipset, latest BIOS, PCIe 3.0/3.1 specification, CrossFire/SLI 8x+8x), supplied by EVGA
• HyperX 16GB DDR4 (2×8 GB, dual channel at 3333 MHz), supplied by HyperX
• RTX 2060 6GB Founders Edition, stock Founders Edition clocks, on loan from NVIDIA
• 480 GB Team Group SSD
• 1.92 TB San Disk enterprise class SSD
• 2 TB Micron 1100 SSD
• Seasonic 850W Gold Focus power supply unit
• EVGA CLC 280mm CPU water cooler, supplied by EVGA
• Onboard Realtek Audio
• Genius SP-D150 speakers, supplied by Genius
• EVGA DG-77, mid-tower case supplied by EVGA
• Monoprice Crystal Pro 4K

Test Configuration – Software

• Nvidia GeForce 417.54 and 418.81 WHQL drivers are used for the RTX 2060.
• VSync is forced off.
• AA enabled as noted in games; all in-game settings are specified
• Gaming results show average frame rates in bold including minimum frame rates shown on the chart next to the averages in a smaller italics font.
• Highest quality sound (stereo) used in all games.
• Windows 10 64-bit Home edition. All DX11 titles were run under DX11 render paths. DX12 titles are generally run under the DX12 render path unless performance is lower than with DX11. Three games use the Vulkan API.
• Latest DirectX
• All 40 games are patched to their latest versions at time of publication.
• MSI Afterburner, latest beta.
• OCAT, latest version
• Fraps, latest version

40 PC Game benchmark suite & 4 synthetic tests

Synthetic

• Firestrike – Basic & Extreme
• Time Spy DX12
• Port Royal – Default and DLSS (1920×1080)
• Superposition

DX11 Games

• Grand Theft Auto V
• The Witcher 3
• Fallout 4
• Rainbow Six Siege
• Battlefield 1
• Resident Evil 7
• For Honor
• Ghost Recon Wildlands
• Mass Effect: Andromeda
• Prey
• Hellblade: Senua’s Sacrifice
• Project CARS 2
• Middle Earth: Shadow of War
• Total War: Warhammer II
• Destiny 2
• Star Wars: Battlefront II
• Kingdom Come: Deliverance
• Final Fantasy XV
• Far Cry 5
• The Crew 2
• Assassin’s Creed: Odyssey
• Call of Duty: Black Ops 4
• Hitman 2
• Resident Evil 2

DX12 Games

• Tom Clancy’s The Division
• Ashes of the Singularity: Escalation
• Hitman
• Rise of the Tomb Raider
• Deus Ex Mankind Divided
• Gears of War 4
• Civilization VI
• Sniper Elite 4
• Forza 7
• Total War: Warhammer II
• Shadow of the Tomb Raider
• Battlefield V
• Just Cause 4
• Strange Brigade

Vulkan Games

• DOOM
• Wolfenstein: The New Colossus
• Strange Brigade

NVIDIA Control Panel settings

Here are the NVIDIA Control Panel settings.

We used MSI’s Afterburner to set the RTX 2060’s highest Power and Temperature targets. By setting the Power Limits and Temperature limits to maximum, it can maintain its maximum clocks without throttling.

Game Ready GeForce 418.81 WHQL Drivers

This latest GeForce Game Ready 418.81 driver introduces public support for the new Port Royal DLSS Feature test. The download links for the latest GeForce 418.81 drivers can be found starting here. The release notes can also be found here. The included highlights below are from NVIDIA’s website.

Game Ready Drivers provide the best possible gaming experience for all major new releases, including Virtual Reality games. Prior to a new title launching, our driver team is working up until the last minute to ensure every performance tweak and bug fix is included for the best gameplay on day-1.

Gaming Technology
Includes support for new GeForce RTX laptops, and delivers the optimum performance in Futuremark’s 3DMark Port Royal benchmark, which has just added a new feature test for NVIDIA Deep Learning Super Sampling (DLSS).

Let’s head to the charts to compare the driver changes from the GeForce 417.54 drivers which were the RTX 2060 launch drivers to the latest GeForce 418.81 drivers.