This editor returned from a month-long working vacation in Honolulu last week and was able to review a PC Gamerz (PCGz) Hawaii ‘Blue Elixir‘ prebuild while there. We met the owner, Devin Wolery, last November and were so impressed with his company, we featured it in our article about the state of PC Gaming in Hawaii.

This time, we wanted hands on experience to underdstand what makes PC Gamerz prebuilt PCs different from other boutique and big box builders so that in a few short years their business increased tenfold to become one of the largest independent PC outlets in Hawaii and has attracted a fanbase of loyal customers. Since PC Gamerz is now expanding to the Mainland and internationally, it’s a good time to see if they are competitive enough to compete outside of Hawaii.

We arranged with Devin for a Blue Elixir Core i7-12700KF/DDR4/RTX 3080 FTW prebuilt and spent nearly a month playing games, benchmarking, and testing video cards with it including a review of the Hellhound RX 6650 XT against the RTX 3060 and RTX 3060 Ti. Is it worth its $3549 asking price and how does its performance compare with a no-expenses-spared i9-12900KF/DDR5/RTX 3080 Founders Edition (FE) PC?

Why a Prebuilt PC?

This editor’s first and last prebuilt desktop PC was in Hawaii in 1996. We were disappointed with what we got for what we paid and within two months, we learned how to build our own. It was a overclocked Celeron 300A PC with a ATi Rage Fury 32MB video card and it blew away the prebuild for less money. Learning PC building was a major time investment, and we realize that we are the exception, not the rule.

Last week, a friend of ours emailed us to ask us what RAM and NVMe SSD upgrade would breathe life into his aging i7-4790K PC until he could afford a new build. We gave him suggestions and he ordered the parts. Well, the upgrade turned out to be more complex than he liked as he had to update his motherboard’s BIOS to recognize NVMe and he didn’t know anything about SATA lanes and disk management.

PC Gamerz ‘Blue Elixir‘ Features & Specifications

• 12th Gen Intel Core i7-12700KF
• ASUS TUF GAMING H670-PRO WIFI D4 motherboard
• Trident Z 32GB 3600Mhz DDR4 with ARGB
• Lian-Li Galahad 360 AIO Cooler
• CoolerMaster TD500 Mesh White case
• EVGA 850W PSU (Bronze)
• Crucial P2 1TB NVMe SSD (2400MBps/1900MBps Read/Write)
• EVGA RTX 3080 FTW3 12GB 1800 MHz Boost Clock/12GB 384 bit GDDR6X video card
• Windows 10 Pro with optimizations

We asked why Windows 10 is being used instead of Windows 11. PC Gamerz is currently evaluating Windows 11 and will offer it later on. We are going to compare game and creative benchmarks of the Blue Elixir on Win 10 versus BTR’s flagship on Win 11 for this review. In part two, we will directly compare 40 games and creative benches using the Blue Elixir on Win 10 versus a clean installation we made of Win 11 to see if there are any performance disadvantages or advantages by upgrading.

Additional PCGz Information

• Game Ready builds – Ready to Ship Prebuilt Gaming PCs by PCGz
• Built with the latest high quality components from brands like ASUS, EVGA, Lian Li, Phanteks, Fractal Design and More!
• We tune and optimize all aspects of your build including latest BIOS, Windows updates, registry and startup optimizations, RAM timing, game optimizations, RGB software and more!
• Extensive stress testing and benchmarks, paired with high performance game testing allows us to prepare a PC “Ready to Game”, right out of the box. We analyze the metrics of the processor, graphics cards, RAM, etc. to ensure no component goes without optimization. Once all testing is complete and each PC runs smoothly, without bottlenecks/stability issues, it is ready to be received by our customers.
• Supported with a comprehensive 2 year parts and labor warranty.
• Responsive support that makes customers feel prioritized and understood. Each PC comes with lifetime expert technical support from our tech team with over 30 years of combined experience. We offer remote and telephone support to resolve issues as quickly and efficiently as possible.

The Test Bed

BTR’s test bed consists of 35 games and 5 synthetic game benchmarks at 1920×1080 as well as SPEC, workstation, CPU, memory, and GPGPU benchmarks. Our latest games include Total War: Warhammer III, God of War, Ghostwire: Tokyo, Elden Ring, Dying Light 2 and CoD: Vanguard.

The platform we compare the PCGz Blue Elixir with is BTR’s flagship PC using Windows 11 Professional: Intel Core i9-12900KF, an ASUS ROG Maximus Apex Z690 motherboard with 32GB T-FORCE DELTA 6400MHz CL40 DDR5, an RTX 3080 Ti Founders Edition (FE) overclocked +35MHz, and a T-FORCE M200 4TB USB 3.2 Gen2x2 Type-C Portable SSD that we used as game storage for both PCs. Please see the test bench configuration page for all of the details.

First, let’s take a closer look at the PCGz Blue Elixir.

A Closer Look

The Blue Elixir can be shipped or picked up at the PCGz retail outlet at the Navy Mall in Salt Lake or at their Aiea warehouse. We used Lyft to transport the Blue Elixir PC and loaner ACER display to our Waikiki condo in the Cooler Master case box.

The Blue Elixir is packed inside the sturdy box, but if you order it shipped, we highly recommend buying shipping insurance. Currently, PCGz is offering a $50 coupon for shipping to the Mainland. You can expect to pay $125-$150 for shipping depending on where you live without the discount coupon.

Opening the case box, we see an accessory box.

All PCGz PCs come with accessories, a quick setup guide, warranty information, and how to contact support.

The powercord, extra video card power cables, and a Wi-Fi antenna are included for the Blue Elixir.

If you pick up the PC, a tech will go over what to expect and a quick tour of its features. Everyone is instructed to turn the box upside down for ease of sliding out the PC instead of trying to lift it out.

We set the Blue Elixir up outside and took pictures of it in natural light. It is a very good looking white case build.

Turing it on we see RGB lighting on the fans, video card, and AIO which may be synched. We like that it is not an overdone RGB unicorn rainbow puke build.

We noticed the 12700KF was running warm even with a Lian Li 360mm cooler and suspected that the wrong backplate was used as all ASUS motherboards have holes for either LGA 1200/1151 or LGA 1700. This was confirmed when we returned the PC. Temperatures running Cinebench approached 100C on Core 5 but did not interfere with our benchmarks.

The fans used in the case are very quiet and they are tuned for performance in the BIOS by PCGz. We found their custom settings to be better than using ASUS Turbo fan settings.

The I/O panel has plenty of connectors and we made full use of the 2000MBps bandwidth provided by Type C Gen 2×2 for our 4TB M200 SSD. The Wi-Fi antennas are decent but we preferred to use a ethernet connection as the condo provided a solid 200Mbps connection for fast downloading of games and programs.

We were amazed at the incredibly neat cable management even behind the MB panel. Our own build cables are not so neat.

Upon returning the PC, we were informed that a new value added feature is being implemented for the Blue Elixir and other pre-built PCs. A screen can be custom programmed for the end user to display graphics or PC stats. Above is the Blue Elixir in black featuring the new screen.

We enjoyed the month we spent with the Blue Elixir playing games and benching which made our Honolulu working vacation even better.

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

Test Configuration – Hardware

PC Gamerz Hawaii Blue Elixir

• Intel Core i7-12700KF (HyperThreading/Turbo boost On) (All listed Blue Elixir hardware except the portable SSD supplied by PC GamerZ Hawaii)
• ASUS TUF Gaming H670-PRO WIFI D4 (Intel H670 chipset, latest BIOS, PCIe 5.0/5.0/3.0/3.1/3.2 specification, CrossFire/SLI 8x+8x)
• G.SKILL Trident Z 16GB DDR4 (2x16GB, dual channel at 3600MHz)
• Crucial P2 1TB NVMe SSD PCIe 3.0 (2400MBps/1900MBps Read/Write) for C: drive
• The T-FORCE M200 4TB USB 3.2 Gen2x2 Type-C Portable SSD (supplied by Team Group for game storage)
• EVGA 850B5, 850W Bronze PSU
• ACER (LC27G75TQSNXZA) 27? 1920×1080/165Hz monitor
• Lian-Li Galahad 360 AIO Cooler
• CoolerMaster TD500 Mesh White

BTR’s Flagship PC

• Intel Core i9-12900KF (HyperThreading and Turbo boost at stock settings).
• ASUS ROG Maximus Z690 Apex LGA 1700 motherboard (Intel Z690 chipset, latest BIOS, PCIe 5.0, DDR5)
• T-FORCE DELTA RGB PC5-51200 6400MHz DDR5 CL40 2x16GB kit, supplied by TeamGroup
• GeForce RTX 3080 Founders Edition 10GB (overclocked +35MHz), supplied by NVIDIA
• T-FORCE CARDEA A440 Pro Special Series 2TB M.2 NVMe PCIe 4.0 SSD, supplied by TeamGroup
• T-FORCE M200 4TB Portable Gen 2 x2 USB 3.2 Type C SSD, supplied by TeamGroup
• Super Flower LedEx, 1200W Platinum 80+ power supply unit
• MSI MEG CoreLiquid S360 – AIO CPU Liquid Cooler – 2.4″ IPS Display – 360mm Radiator – Triple 120mm Silent Gale P12 Fans
• Corsair 5000D ATX mid-tower (plus 1 x 140mm Noctua fan; 2 x 120mm Noctua fans)
• ASUS ROG Swift 360Hz PG259QNR 24.5” HDR Gaming Monitor, 1080P Full HD, Fast IPS, 1ms, G-SYNC, supplied by ASUS

Test Configuration – Software

• GeForce 512.77
• High Quality, prefer maximum performance, single display, set in the NVIDIA control panel; Vsync off.
• Optimizations are off, Vsync is forced off, Texture filtering is set to High Quality, and Power management prefer maximum performance
• 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 (1% lows/99 percentiles) shown on the chart next to the averages in a smaller italics font where higher is better.
• Windows 11 Pro edition for BTR’s PC and Windows 10 64-bit Pro edition for the Blue Elixir; latest updates. 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

Games

Vulkan

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

DX12

• God of War
• Ghostwire: Tokyo
• Elden Ring
• Dying Light 2
• Call of Duty: Vanguard
• Guardians of the Galaxy
• Far Cry 6
• Chernobylite
• Resident Evil Village
• Metro Exodus Enhanced Edition
• Godfall
• DiRT 5
• Assassin’s Creed: Valhalla
• Cyberpunk 2077
• Watch Dogs: Legion
• Horizon Zero Dawn
• Death Stranding
• F1 2021
• Borderlands 3
• Tom Clancy’s The Division 2
• Civilization VI – Gathering Storm Expansion
• Battlefield V
• Shadow of the Tomb Raider

DX11

• Total War: Warhammer III
• Days Gone
• Crysis Remastered
• Destiny 2 Shadowkeep
• Total War: Three Kingdoms
• Overwatch
• Assetto Corsa: Competizione

Synthetic

• TimeSpy (DX12)
• 3DMark FireStrike & Extreme
• Superposition
• VRMark Blue Room
• Cinebench
• GeekBench
• OctaneBench
• AIDA64 CPU, cache & memory, and GPGPU benchmarks
• Blender 3.01 benchmark
• Sandra 2021 CPU Benchmarks
• SPECviewperf 2020
• SPEC Workstation

NVIDIA Control Panel settings

Here are the NVIDIA Control Panel settings.

Next we look at overclocked performance.

Overclocking, temperatures and noise

Since the PC GamerZ PC ran rather warm, we did not attempt to overclock any hardware (memory or video card), not to mention that the H670 motherboard does not allow CPU overclocking. We think it is a mistake to pair a K or KF CPU with a H670 motherboard. A more logical pairing with be a H670 motherboard and a i7-12700 or i9-12900 (non-K) CPU which would also save money all round.

The other logical choice is to offer a slightly more expensive Z690 motherboard with 12x00K or KF CPUs for overclocking. BTR’s flagship i9-12900KF is able to overclock to 5.3GHz on Cores 0 and 1 and 5.1GHz on the rest of the P-cores but we left it stock for our comparison with the Blue Elixir. However, the RTX 3080 FE is only able to overclock +35MHz on its core while the EVGA RTX 3080 FTW is factory overclocked +80MHz (45MHz higher than the overclocked FE). So the FTW will definitely perform better than the FE by virtue of its higher clocks and 12GB DDR6 compared with the FE’s 10GB.

Let’s head to the performance charts

Performance Summary Charts

Here are the performance results of 35 games and 5 synthetic tests comparing BTR’s flagship with PCGz’ Blue Elixir. Click on each chart to open in a pop-up for best viewing.

All gaming results show average framerates in bold text, and higher is better. Minimum framerates (1% lows/99-percentiles) are next to the averages in italics and in a slightly smaller font. We picked the highest settings as shown on the charts. Wins are show by yellow text.

The PCGz Blue Elite by virtue of its faster EVGA RTX 3080 FTW trades blows with BTR’s flagship PC in 1080P gaming. Generally, there are very little performance differences. BTR’s PC excels when a faster CPU/DDR5 makes a positive performance difference. We would expect even less of an advantage in CPU-heavy games at 1440P or at 4K when the faster PCGz video card would probably pull ahead. Score a win or at least a tie for the PCGz i7-10700K/DDR4 prebuilt PC at less cost than BTR’s i9-12900KF/DDR5 build.

Let’s look at non-gaming applications next to see if the PC GamerZ Hawaii PC is also a solid performer in creative/workstation/pro tasks starting with Blender benchmarks.

Blender 3.01 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.

We benchmarked all three Open Data Blender.org benchmarks which measures both CPU and GPU performance by measuring samples per second by render production files.

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

The 12900KF spanks the 12700KF using Blender’s CPU rendering benchmarks, but the 12700KF/RTX 3080 FTW combination is faster than the 12900KF/RTX 3080 FE equipped PC in the GPU benchmarks.

Next, we move on to AIDA64 CPU, Cache & Memory, and GPGPU benchmarks.

AIDA64 v6.70

AIDA64 is an important industry tool for benchmarkers. Its GPGPU benchmarks measure performance and give scores to compare against other popular video cards while it’s CPU benchmarks compare relative performance of processors.

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/FPU Benchmark Results

CPU results are summarized below in two charts for comparison.

GPGPU Benchmark Summary

Here is the comparison summarized between the BTR’s 12900KF/DDR5 flagship PC and the Blue Elixir 12700KF/DDR4 PC.

Cache & Memory Benchmarks

Here is the summary chart.

For workstation, creative, and professional applications using the CPU, the 12900KF/DDR5 combination performs well ahead of the PCGz Blue Elite’s 12700KF/DDR4 PC. So let’s look at Sandra 2021 next.

SiSoft Sandra 2021

To see where the CPU, GPU, and motherboard performance results differ, there is no better tool than SiSoft’s Sandra 2021. 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. Sandra 2021 is the latest version, and we are using the full engineer suite courtesy of SiSoft. Sandra 2021 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.

We ran the latest version of Sandra’s intensive Processor benchmarks and summarize the overall results below.

In Sandra’s synthetic CPU benchmarks, the i9-12900KF is much stronger than the i7-12700KF.

Cinebench

Cinebench is based on MAXON’s professional 3D content creation suite, Cinema 4D. This latest R23 version of Cinebench can test up to 64 processor threads accurately and automatically. It is an excellent tool to compare CPU/memory performance and higher is better.

Cinebench’s Multi-Core benchmark will stress a CPU reasonably well over its 10-minute run and will show any weaknesses in CPU cooling. This is the test where we discovered that the Blue Elixir’s 12700K hit nearly 100C on Core 5 and lead us to conclude that the wrong LGA 1151 backplate was used by the PCGz builders instead of LGA 1700.

Here is the summary chart.

The 12900KF/DDR5 combination is the faster PC in Cinebench. However, the 12700KF/DDR4 PC has nothing to be ashamed of. Now we benchmark using GeekBench which measures CPU and GPU performance.

GeekBench

GeekBench is an excellent CPU/GPU benchmarking program which runs a series of tests and times how long the processor takes to complete its tasks.

Single Core Performance

First single core performance starting with the 12700KF/DDR4 Blue Elixir.

Now single core performance with the 12900KF/DDR5 PC.

The 12900KF has a stronger single core performance. Next we check multi-core.

Multi Core Performance

First CPU multi core performance starting with the 12700KF/DDR5 Blue Elixir.

Now CPU multi-core performance with BTR’s 12900KF.

Again, the i9-12900KF is stronger than the 12700KF in Multi-core performance benching.

Next we test the two PCs using OpenCL, CUDA, and Vulcan using GeekBench graphics-heavy benchmarks.

Open CL, CUDA, and Vulcan

First, we test the RTX 3080 FTW and the i7-12700KF using OpenCL, CUDA, and Vulcan

Finally, we test BTR’s RTX 3080 FE and the i9-12900KF using OpenCL, CUDA, and Vulcan

The summary charts below show the comparative performance scores.

In Geekbench, the stronger i9-12900KF/DDR5 PC pulls ahead of the i7-12700KF/DDR4 in every CPU test while the 12700KF/DDR5 Blue Elixir paired with the faster RTX 3080 FTW beats the i9-12900KF paired with the slower RTX 3080 FE in GPU compute except for the Vulcan benchmark.

Lets check out Octanebench, another GPU-heavy test

Octanebench

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

Here is the summary chart:

Again, the combination of the 12700KF with the RTX 3080 FTW is faster than the 12900KF with the RTX 3080 FE. Next up, SPECworkstation.

SPECworkstation3 (3.0.4) Benchmarks

All the SPECworkstation3 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 to vendors of computer-related products and/or services. The most comprehensive workstation benchmark is SPECworkstation3. 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.

SPECworkstation benchmarks are very demanding and all benchmarks were tested in an official run.

Here are the SPECworkstation Raw Scores which give the details.

We see the same thing repeated in all of the synthetic tests – the i9-12900KF/RTX 3080 FE is generally faster in CPU-heavy benchmarks while the i7-10700KF/RTX 3080 FTW is generally faster in GPU-heavy benching. In addition, storage benchmarks favor the faster Gen 4×4 NVMe SSD in BTR’s flagship.

Now, let’s look at a GPU-heavy SPEC benching suite, SPECviewperf 2020.

SPECviewperf 2020 GPU Benches

The SPEC Graphics Performance Characterization Group (SPECgpc) released a 2020 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 1900×1060 display resolution.

Here are SPECviewperf 2020 benchmarks summarized in the chart below.

Again the PC GamerZ PC is faster in eight of 9 individual benchmarks because of the faster RTX 3080 FTW.

The decision to buy a new CPU, memory, and video card should be based on the workflow and requirements of each user as well as their budget. Time is money depending on how these apps are used. However, the target demographic for the PC GamerZ Blue Elixir is primarily 1080P and 1440P gaming for gamers, and in this area, the Blue Elixir excels and it does not fare badly in the workstation and creative benchmarks.

Let’s head to our conclusion.

Final Thoughts

Our conclusion depends partly on pricing and partly on convenience. Let’s look at today’s pricing of the individual components on Amazon and on Newegg.

• 12th Gen Intel Core i7-12700KF – $363.99 (The 12700 is $342.99 which is a more logical choice for a H670 motherboard.)
• ASUS TUF GAMING H670-PRO WIFI D4 motherboard – $229.99
• Trident Z 32GB 3600Mhz DDR4 with ARGB – $174.99
• Lian-Li Galahad 360 AIO Cooler – $248.99
• CoolerMaster TD500 Mesh White case – $123.99
• EVGA 850W PSU (Bronze) – $99.99
• Crucial P2 1TB NVMe SSD (2400MBps/1900MBps Read/Write) – $84.99
• EVGA RTX 3080 FTW3 12GB 1800 MHz Boost Clock/12GB 384 bit GDDR6X video card – $1199.99
• Windows 10 Pro with optimizations – $129.00
• Video screen – You’d have to source, custom build, and probably program it yourself.

If you want to save $150, PCGz’ Blue Elixir offers a EVGA RTX 3080 XC3 12GB (as pictured below) which comes without RGB and is slightly slower (-25MHz) than the FTW but still faster than BTR’s overclocked RTX 3080 Founders Edition.

If you buy the individual parts and source them yourself, before tax and any shipping, it comes to $2655.92 – without any video screen. It also means that when you build it yourself, you get all the joy and headaches that come with building a PC. For BTR’s flagship build, it meant RMA’ing G.SKILL DDR5 6000 which took 10 days, and also replacing a MSI 360R cooler that was inadequate for cooling the 12900K. It also took many hours of work to optimize the build for stability in benchmarking. Any problems or issues that arise in the future are mine and will need to be researched and addressed.

For gamers that don’t have the time or the skill set to build a PC from parts, the PCGz Blue Elixir is a ready made $3500 top PC that you unbox and plug in – and it just works. BTR spent nearly $4000 on its flagship PC and it is basically about as fast in gaming as the Blue Elixir at 1080P – where the 12900KF CPU makes the most difference over a 12700K. Of course, we use it for more than gaming.

PC Gamerz high end Blue Elixir compares favoribly in pricing with the big box and boutique builders with the difference being that they offer a higher level of custom service with a completely tested PC that they will support – parts and labor – for two years plus lifetime tech support.

For example, Alienware offers a similar PC at a similar price to the Blue Elixir. It offers a slightly faster 12900K CPU, but a slower RTX 3080 10GB (instead of 12GB). It offers 64GB (that a gamer doesn’t need) of painfully slow DDR5 (4400MHz), and a very small 120mm AIO that is likely insufficient to properly cool the CPU in a stylish if cramped case that will be hard to upgrade. Included support/warranty lasts for only 1 year which isn’t that great having experienced Dell’s service for a gaming laptop. I would pick PCGz if I didn’t build PCs.

One thing we love about the Blue Elixir – besides its outstanding cable management – is that they chose a decent 850W PSU that will probably power the next generation of video cards – a simple remove and replace will allow for an upgrade unlike many big box PC builders that require the PSU to be upgraded also. Although we can nitpick ‘Bronze’, EVGA sells decent PSUs that deliver the wattage they specify. We spent a ridiculous amount of money on a 1200W Platinum PSU.

PCGz also gives a 1TB NVMe SSD which is generally sufficient for most gamers, and it is very easy to install a second NVMe SSD in the top slot without removing the video card if a gamer decides to expand storage. In addition, if Rocket Lake offers a solid upgrade over Alder Lake’s 12700 this year, it will be easy to drop in the new CPU with a BIOS flash as the Cooler Master case is large and easy to work in. PCGz has left room for expansion and upgrades with minimal fuss. All of the parts they chose are first class.

Let’s sum it up:

The Pros

• The PCGz Blue Elite 12700KF/DDR4/EVGA RTX 3080 12GB PC is a solid PC that competes well in gaming with Intel’s top DDR5 PC
• The case is large for expansion or upgrades and the fans are quiet
• The PSU is of sufficiently high capacity to upgrade video cards
• The cable management is exquisite
• Each PC is individually tested and benchmarked before being shipped out
• PCGz optimizes the BIOS and Win 10 for gamers
• Two years Parts & Labor Warranty and lifetime tech support
• New video screen adds extra value

Cons

• Pricing is high
• QC let the wrong CPU backplate slip by
• Mismatch of KF CPU with H670 MB. Either a non-K CPU should be used with H670 or a Z690 MB should be offered for OC potential with K CPUs

The Verdict:

If a gamer is looking for a very well made, solid and handsome PC that will just work perfectly for gamers with great support, the PC Gamerz Hawaii Blue Elixir should be considered. It is a solidly-built good-looking RGB desktop that trades blows with a top Intel DDR5 desktop in gaming. PCGz support is awesome and they justify their repeat customers and good reviews.

We will not be surprised to see PCGz become popular outside of Hawaii in the months to come.

Stay tuned for Part 2. We were somewhat surprised to see that PC GamerZ was still using Windows 10 so we decided to compare its performance with Win 11. We will post this review next week and then follow it up with a SSD review and then a VR review featuring the Hellhound RX 6650 XT versus the RX 6700 XT and versus the RTX 3060 Ti by next weekend.

We hope to follow up with PC GamerZ Hawaii the next time we visit the Islands, but until then we will say aloha and a hui ho (until we meet again).

Aloha and Happy Gaming!

BTR received the RTX 3050 XC Black Edition 8GB video card from EVGA and we have tested its VR performance by benchmarking nine virtual reality games using the Vive Pro. Although it’s a gaming card, we have added synthetic gaming and workstation benches. We will see if the RTX 3050 is a better entry level card for VR and an improvement over the GTX 1660 Super.

The RTX 3050 XC Black is a $249.99 MSRP EVGA card, but because of supply/demand issues, all suggested pricing is meaningless as only a very lucky few gamers will get them at or close to MSRP/SEP. It is out of stock at most retailers and can mostly be found on eBay starting around $450. The RTX 3050 is generally priced lower than the GTX 1660 Super which sells for around $500 currently.

There is hope in sight as the end of the COVID-19 pandemic may be in view, Etherium is transitioning to Proof of Stake from Proof of Work, and we may see lower prices and better availability before end of this year. Plus it is possible to buy video cards at close to retail if buyers exercise patience by signing up for EVGA’s queue, watch for Best Buy stock to come in, and participate in Newegg’s Shuffle.

Specifications

The RTX 3050 is a solid step up over the GTX 1650 which it replaces. The RTX 3050 is NVIDIA’s 1080P entry level gaming card. The RTX 3050 has also been equipped with Tensor and RT cores for ray tracing and DLSS. It’s also an upgrade over the GTX 1660, 1660 Super, and 1660 Ti which do not have these specialized cores. DLSS is now available on over 145 games and applications and it is also available in some VR games. We will specifically test No Man’s Sky to see if the RTX 3050’s DLSS VR performance is better than the GTX 1660 Super which cannot use DLSS.

We benchmark using Windows 11 Pro Edition with an Intel Core i9-12900KF at 5.3GHz/5.1 GHz and 32GB of T-FORCE Delta 6400MHz DDR5 on a ASUS ROG Maximum Apex Z690 motherboard. All games and benchmarks use the latest versions and the most recent drivers.

Let’s first look at the EVGA RTX 3050 XC Black before we check out the test configuration.

A Close-up of the EVGA RTX 3050 XC Black Edition

The 8GB XC Black Edition is EVGA’s entry level 2-slot dual-fan RTX 3050 which is rather small at 4.33 inches high by 7.94 inches long.

The back of the card is a bare PCB.

The RTX 3050 Black uses a single 8-pin connector and a 550W PSU is recommended as minimum.

The IO panel has four connectors. Three DisplayPort 1.4a connectors are included, and the HDMI port 2.1 allows for 4K/120Hz over a single HDMI cable and variable refresh rate (VRR) is supported.

As a small two-slot card, the EVGA RTX 3050 XC Black will fit into most cases and it looks good installed inside our Corsair 5000D.

Let’s check out our test configuration.

Test Configuration

Test Configuration – Hardware

• Intel Core i9-12900KF (Overclocked to 5.3GHz, Cores 1-2; 5.1GHz, Cores 3-8)
• ASUS ROG Maximus Apex motherboard (Intel Z690 chipset, latest BIOS with Resizable BAR enabled, PCIe 5.0/4.0/3.1/3.1 – USB 4.0 Type-C specification)
• T-FORCE DELTA 2x16GB DDR5 6400MHz CL40, supplied by TeamGroup
• GeForce RTX 3050 XC Black 8GB, supplied by EVGA
• ASUS GTX 1660 OC Super 6GB, supplied by ASUS
• T-FORCE CARDEA A440 1TB M.2 NVMe PCIe 4.0 SSD, supplied by TeamGroup as primary storage for games and benchmarks
• T-FORCE CARDEA Ceramic C440 2TB PCIe Gen 4 x4 NVMe SSD, C: drive
• 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

Test Configuration – Software

• GeForce 511.65 drivers
• Windows 11 Professional edition; latest updates/build
• Latest DirectX
• All benchmarking programs are updated to their latest versions
• All 9 VR games are patched to their latest versions at time of publication.
• FCAT-VR Capture v0.9.3202.0 UAC
• FCAT-VR Beta 17

VR Games

• Elite Dangerous
• Hellblade: Senua’s Sacrifice
• Myst 2021
• No Man’s Sky
• Obduction
• Skyrim
• The Walking Dead; Saints & Sinners
• The Vanishing of Ethan Carter
• Zombieland

Synthetic

• TimeSpy (DX12)
• 3DMark FireStrike – Ultra & Extreme
• VRMark – Orange & Cyan Benchmarks
• OpenVR Benchmark
• Superposition
• Blender 2.93.1 benchmark
• SPECworkstation3
• SPECviewperf 2020
• OctaneBench

NVIDIA Control Panel settings

Here are the NVIDIA Control Panel settings.

Let’s look at the RTX 3050 and GTX 1660 Super VR gaming using the Vive Pro and FCAT VR.

Next up, The Walking Dead: Saints & Sinners.

The Walking Dead: Saints & Sinners

The Walking Dead: Saints & Sinner is a first person survival horror adventure RPG with a strong emphasis on crafting. Its visuals using the Unreal 4 engine are outstanding and it makes good use of physics for interactions.

We benchmarked Saints and Sinners using its lowest preset and we left the Pixel Density at 100%.

Here is the frametime chart.

Here are the details as reported by FCAT-VR.

The GTX 1660 Super produced 154.16 unconstrained FPS with no dropped or synthesized frames.

The RTX 3050 gave 165.65 unconstrained FPS with no dropped or synthesized frames.

Playing The Walking Dead: Saints & Sinners is similar for both of our cards on the Low Preset although the RTX 3050 has a little extra performance headroom. Unfortunately, neither card can manage to deliver a steady 90 FPS on the Medium Preset, so we would suggest that a gamer experiment with the individual setting to give the best balance of performance to visuals.

Last up, Zombieland.

Zombieland

Zombieland VR: Headshot Fever is a fun arcade style light gun wave shooter that adds split-second racing mechanics while featuring just one mechanic – headshots that slow down time.

Here is the frametime chart using the Medium Preset.

Here are the details as reported by FCAT-VR.

The GTX 1660 Super produced 257.52 unconstrained FPS.

The RTX 3050 managed 252.52 unconstrained FPS. Neither card could deliver a steady minimum of at least 90 FPS using the High Preset because it boosts the resolution from 100% to 140%. We would again suggest that the player experiment with raising individual settings and adjusting the SteamVR resolution to deliver the best balance of performance to visuals

We see that overall the RTX 3050 is a stronger VR card than the GTX 1660 Super and it also has the advantage of being able to use DLSS.

We received an EVGA X20 MMO Mouse from EVGA a week ago together with the previously announced EVGA X20 Wireless mouse. We have been using the wireless X20 for more than a week alongside the EVGA X15 MMO mouse. Normally, many gamers use a typical Logitech mouse or not care about the name brand at all. Personally, I cannot be without a decent mouse that’s responsive, light, and has enough buttons that instantly lets me control my character in a game.

We present this review to give our impressions of these two premium customizable RGB mice. The just released today, EVGA X15 MMO mouse is launching at the incredible price of $39.99 for the launch window (normally $79.99) and the X20 Wireless mouse is $59.99 (normally $119.99). The EVGA X15 MMO mouse features 12 physical buttons, Light Strike LK Optical switches for fast response times, and fully customizable RGB. The X20 is very similar and features a premium wireless design that is light and responsive with fully customizable RGB.

Let’s take a deeper look at the two newest mice from EVGA.

Here are the EVGA X15 MMO mouse and EVGA X20 Wireless mouse features and specifications as presented by EVGA:

EVGA X20 GAMING Mouse Specifications

Sensor System
Main Sensor: PIXART 3335 Optical Sensor
Secondary Sensor: Lift Off Distance Detection Sensor
Third Sensor: Lift Off Distance Detection Sensor
Max DPI: 16,000
Tracking Speed: 400 ips
Acceleration: 40 g

DesignMain Switch Type: Omron – 60 Million Click Lifecycle
Number of Buttons: 10
LED Color: RGB Lighting 3-zone
Weight: 123?g
2 M Braided cable
Feet Type: Low Friction PTFE
Height: 44.52 mm
Length: 122.71 mm
Width: 70.27 mm
Max Number of Profiles: 5

Battery Life: 60 hrs (No lighting), 24 hrs (Default lighting)

Micro ProcessorFull Speed USB Max Polling/Report Rate: 1K Hz
32-bit Arm Cortex-M4 core (Mouse) / Arm Cortex-M0 core (Dongle)
USB data format: 16 bits/axis

EVGA X15 MMO Gaming Mouse Specifications:

Sensor System
Main Sensor: PIXART 3389 Optical Sensor
Max DPI: 16,000
Tracking Speed: 400 ips
Acceleration: 50 g

Design
Main Switch Type: LK Optical – 70 Million Click Lifecycle
Number of Buttons: up to 20
LED Color: RGB Lighting 3-zone
Weight: 110 g without cable
2 M Braided cable
Feet Type: Low Friction PTFE
Height: 42.7 mm
Length: 123.5 mm
Width: 71.7 mm
Max Number of Profiles: 5

Micro ProcessorHigh-Speed USB Max Polling/Report Rate: 8K Hz
32-bit Arm Cortex-M33 core
USB data format: 16 bits/axis

Mouse latency is one of the most important factors for gaming. Both mice feature some incredible tech and in our testing have been superb. I have used both for Destiny 2, Apex Legends, and Warzone and they both have been amazing weapons to use.

The features and specifications look great, but how do they handle while playing games? Besides playing fast-paced first-person shooters (FPS), we were particularly interested in the X20’s three modes of connecting via Bluetooth, USB, or wireless with a USB dongle. All were incredibly responsive and the wireless offering has a 1MS response time. The X15 impressed us most with its plethora of buttons to customize.

I have a significant amount of time in Destiny 2 using custom keybinds that have become second nature, but I did try to implement a ton of my various characters’ abilities and mappings to these buttons. I still prefer them on the KB but this was a very nice feature to have in the heat of the moment, and at times I can see the appeal.

Of course, mouse reviews are highly subjective and your own experience may vary. Please read on for our experiences with these two mice over the past week.

First, the unboxing.

Unboxing & Functionality

The EVGA X15 and X20 unboxing

The X20 wireless mouse comes in a small box that advertises dual wireless connectivity, USB connectivity, EVGA Unleash for RGB, but sadly no Qi wireless charging but instead charging with the included braided USB cable. At a MSRP of $119, I would have expected wireless everything including charging but it is not a deal-breaker and thankfully the mouse charges lighting fast. After the mouse died, we were at 50% in less than 20 minutes. On a full charge, the mouse lasts a little over 24 hours and 60 hours with RGB disabled.

As for the X15, you can see it boasts up to 20 custom buttons, 3 zone RGB control, and only 110G of weight.

The back of the box, below, goes into much more detail for both Mice.

The features of the X20 Wireless mouse are on the box. It features a 16000-dpi optical sensor with a 1000Hz polling rate, extra switches rated to last for 60 million clicks, and a 32-bit ARM Cortex-M4 Processor.

The truly impressive part of the EVGA X20 mouse is its triple sensors – a world first! The mouse has dual LOD sensors and a Pixart 3335 optical sensor. Most gaming mice traditionally use a LOD to detect lift-off distance when a mouse is picked up and moved. The X15 (and the X17 we are not reviewing at this time) uses a 3-Dimension Array to detect the shortest and most accurate detection we have seen. According to EVGA, this array is able to detect within a minimum of 0.4 mm and a max of 3mm from the mouse to the surface.

This is fully customizable to your liking depending on your playstyle in the EVGA Unleash RGB software. We found this simple to use and the program can remember different surfaces as well.

Below is pictured the basic equipment and the user guide.

The Teflon mouse feet are well aligned and they offer a nearly frictionless experience on a variety of surfaces. EVGA states that they use the highest-grade premium PTFE, Polytetrafluoroethylene, and in our testing on multiple surfaces, we can confirm the experience is great and nearly frictionless.

Common to Both Mice & Differences

Both mice are nearly identical except that one is wireless and the other must be connected by its attached USB 2.0 cable.

The X15 wireless mouse can use either Bluetooth for devices that support it or 2.4GHz by using a USB adapter that plugs into the PC. Pairing only takes a few seconds. In addition, although both mice support customizable RGB with logos and mouse wheels that light up, the EVGA X15 wireless mouse has extra RGB lighting for its many buttons, while both feature the same 3 zones for RGB lighting. The lighting is bright and fully customizable in the EVGA software suite but we wish it was able to sync with other hardware brands much more easily.

Both mice show off their RGB lighting in a darkened room best, but even throughout the day the effects are a nice addition, and they both pair perfectly with the EVGA Z12 keyboard.

Both mice come with braided USB cables, but only the Wireless X20 comes with a 2.4GHz wireless adapter that plugs into a USB port on your PC. Sadly, we do not see a way to easily replace the X20’s internal battery but it may be possible if you feel comfortable opening it up. However, for those looking to do it on the fly, you are out of luck.

Both mice are elegant and a nice addition to our office. They each fit my hand perfectly and grip well. The X20 wireless mouse is light and fun to use and has been an incredible asset during my FPS play sessions. The X15 MMO mouse is an MMO player’s dream and the buttons feel crisp, clean, and easy to use. The X15 is highly responsive, and on the fly DPI and profile conversion is intuitive and simple for the EVGA mice lineup.

The X20 features a massive “Sniper” button that is prominently featured on the side of the mouse. This intended feature is to allow you to drop your DPI – say from 800 normally to 400 while sniping to be more precise. I did not use this feature. I need muscle memory, and having to remember to click a “sniper button” while trying to not get my head blown off in-game is just not going to happen.

Thankfully, this is also fully customizable in the Unleash software suite and its bulkiness actually lent itself to being used more often. It allows for a “vertical” push so I could simply lift my thumb up a little more to melee enemies.

The Software

The EVGA RGB Unleash software allows for customization of either mouse. It needs some ease of use for syncing between other hardware brands but overall it is easy to use to customize your experience. By clicking on Device, we were able to customize the mice and were able to even sync them and their color patterns – just not with other third party hardware unless we choose a static color or array.

Both mice have similar options although some are specific to each mouse. The X15 in particular has a wide array of buttons, and with up to 5 profiles, the possibilities are wide open. The buttons feel very ergonomic and it encouraged us to use them fully during our play sessions.

All of the performance options can be changed to suit the style of each gamer including the LOD, RGB, DPI, and so on. Scenario profiles and macros can be individually customized to suit each gamer’s style and preference.

Color and lighting offer a high degree of customization and personalization and the logo, mouse wheel, and front lighting/button lighting for the X15 and X20 can be changed at will.

The lighting software is excellent, and if you have other EVGA hardware, the centralized experience is easy to navigate unlike Razer’s synapse software with many different menus. Still, the software needs some love and some more modern features such as game syncing, or game-specific mapping.

Let’s head for our game-playing experiences and the conclusion.

The EVGA Mice in Action & the Conclusion

Both of these EVGA mice have a premium feel to them, and both worked surprisingly well during our testing. As a competitive gamer, a strong mouse is key to the experience, and in the MMO world making your character fly with ease is paramount. The X15 is a no-brainer, a must buy, and BTR’s Editors Choice for the MMO gamer. We loved our experience with it and its robust features ensured we had precise control with some nice RGB to boot.

The X20 is a tricky one. At its half-off introductory price, I do not hesitate to recommend it. The X20 is a precision tool for any gamer looking to blast headshots wirelessly but its software holds it back slightly. If you are embedded using other hardware with RGB it may be hard to leave the ecosystem but EVGA is making a strong case for this mouse. It is very premium and reminds me of the old G9. If you are looking to build a PC and want a product that will help you play well without worry – go with EVGA now especially at its intro price level.

A gamer would be hard-pressed to choose between these two. Competition is heavy and there are many fantastic alternatives. But BabeltechReviews can wholeheartedly recommend the EVGA X15 MMO mouse and the EVGA X20 Wireless Mouse. It really comes down to the games you play.

At an MSRP of $119.99 the X20 is still an incredible value and a no-brainer right now at the intro price of $59.99. EVGA’s X15 MMO mouse is a superior option to much of its competition at an amazing price of $79.99 and its only $39.99 during the intro period. If you play MMOs this is an instant buy and we recommend it outright against the competition.

We give each of these mice a big recommendation and they each share BTR’s “Editor’s Choice Award”.

Happy Gaming!

EVGA is known for its great GPU offerings but it has steadily been increasing its product line for the peripheral market, offering high-quality products for a reasonable price. The core PC experience centers around the keyboard, and if it has no soul or feels cheap, it is instantly noticeable. Keyboards are a staple to any PC, and for us gamers, they are a part of the KB+M arsenal we need.

EVGA’s latest offering, releasing to the public on September 1, and for EVGA insiders on August 31, is the EVGA Z12 Gaming Keyboard. The Z12 keyboard offers an incredible value at an introductory $24.99 price at launch (normally $49.99 after the launch window) with full RGB backlighting, 5 programmable Macro Keys, Dedicated Media Keys, and full IP32 Water resistance.

EVGA may have a hit with the Z12’s pricing and by nailing the core needs of a gamer, but it is not without its shortcomings.

Let’s take a closer look at the Z12.

Unboxing the EVGA Z12

The back of the box boasts of some of the features for the Z12, including spill resistance, media buttons, and the fast response time of the keys.

The keyboard comes with a typical manual. There is not much to see here, and it mostly points out how to connect the board via USB, how to control RGB using the typical FN key and arrow keys, and how to use the media buttons.

The Z12 Showcase

The EVGA Z12 is a large keyboard, coming in at 5.96 in / 151.5 mm wide and 18.62 in / 473 mm long. It has been slightly extended to accommodate the Macro keys to the left of the keyboard. Here are its full specifications:

Overall, the Z12 feels very premium. The keys are nice to use and the font chosen isn’t some weird hybrid. The design is clean and we enjoyed using the EVGA Membrane switches.

Aesthetics and Design

The Z12 features 5 Macro keys that are fully customizable, including an EVGA “E” Game Mode key which simply disables a setlist of keys during gaming if you would prefer.

Macro keys are a common staple of some keyboards but I would much prefer a less cramped keyboard with bigger keys that are more important for gaming ala the ASUS ROG Strix Scope that we recently reviewed. I have never been a fan of these extra Macro keys and struggle to type comfortably as I tend to rest my hand slightly to the left.

In the heat of the moment working remotely, I often accidentally press a Macro key instead of shift or ctrl as intended leading to an awkward look down and see where to press moment.

Over time, one can adjust to this, and while some gamers rely on these Macros, I would much prefer they not exist for the genres of games I play. The same goes for the game mode key – it is a nice feature, but I rarely use it and often forget it exists until I am deep in a gaming session. I would also have preferred more volume control for the media controllers than just mute.

Brands tend to struggle with how much branding should be plastered over a product – EVGA does a decent job with a nice-looking logo at the top of the keyboard, although some may find it excessive. I like it and prefer the clean look of the rest of the board.

A keyboard lives and dies with its key choice – if something doesn’t feel good to type with, the keyboard can turn ugly and become a frustration. Luckily, I really enjoyed the membrane switches of the Z12. I need a board that is sturdy, reliable, and can withstand abuse all day at work. I love mechanical switches, but customers I deal with on a daily basis do not want to hear the click-clacks in the background, so the balance here is perfect.

The EVGA Z12’s keys are responsive, light, easy-to-use, and fun to type with. Besides the occasional struggle with the Macro keys, it is easy to recommend for any remote worker and gamer alike. The keys are also compatible with Cherry MX keys so a switch would be possible if you prefer that feel.

RGB Everything

The RGB trend is here to stay and I am here for it. I like a nice backlit keyboard and having full control of the colors is a must for me. Rainbow diarrhea is fun, but I need the ability to sync and manage my various products to fit the theme of my choosing. RGB can be often be hit or miss. Thankfully, the Z12 is sleek and elegant, and it features subtle but fantastic RGB lighting.

The EVGA Z12 Gaming Keyboard supports full key control using its “Unleash RGB” software, as well as 5-zone control for the Z12.

The software is decent and lets you control RGB easily for your EVGA peripherals – but unfortunately, it lacks the 3^rd party compatibility I would like. I do not like it when my RAM, CPU cooler, PC, keyboard, and mice are all spewing random rainbows and not syncing with each other. This is a major issue I have with RGB everything and one I struggle with here.

My T-Force RAM from TeamGroup, my MSI Magcore Liquid 240 R CPU cooler, my Asus Steel Legend Motherboard, my Razer Basilisk Mouse, Razer Firefly, and Lighstrips for my room all sync with each other using Razer Synapse and Alexa. With a couple of clicks, they all change, in sync with one another, all from different brands I like.

EVGA’s keyboard and mice we are reviewing today do not sync with other peripherals – so I will have to limit my choices to a solid color or a wave of sorts if I want my KB+M to be in sync with the rest of my setup.

Features

The Z12 lacks the audio passthrough of the EVGA Z20 or even a USB port that most higher-end keyboards feature – but it’s not a dealbreaker, and I would say that any gamer would love the Z12. The Z12 is crisp and responsive and easy to customize to your liking. However, I need a wrist rest as working all day with any keyboard can become quite tiresome.

The Z12 also does not feature the touted ToF, “Time-of-Flight,” sensor of the higher-end Z20. This is a sensor that notices when you are near the keyboard or have left to wake or sleep your system. A nice addition, but I don’t miss not having it.

My previous board was the Razer Cynosa V2, a $59.99 offering with similar features: RGB control, media keys, membrane keys, a nice media control section, and no Macro keys. When comparing the two, the Z12 is an easy buy. It feels premium, whereas the Razer is more expensive but feels distinctly plastic. However, I instantly missed the ability to sync with Razer Synapse and with the rest of my setup. If this is important to you, it is something to consider.

Final Thoughts

Overall, I’ve come to enjoy the Z12 for its presentation. The Z12 is an affordable option for any gamer looking for a sleek and reliable keyboard. The Z12 is a great buy that features sleek and subtle RGB with comfortable-to-use and responsive membrane keys. With a high degree of RGB, 5 macro, and key customization, the Z12 is fun to use with a consistent and smooth typing experience.

My main issue is my ecosystem – I am deep in the Razer Synapse world that allows easy RGB control for a wide variety of brands. If EVGA improved the software or its compatibility with other brands, it would be fantastic.

The Z12 is available to EVGA insiders today and to the public on September 1, for the half-off introductory price of $24.99. An insane value and an easy Editor’s Choice for gamers who are looking for an affordable gaming keyboard. Even at the standard price of $49.99 we wholeheartedly recommend the EVGA Z12 gaming keyboard.

The EVGA RTX 3070 FtW3 Ultra takes on the Reference RX 6800 in 35 Games, GPGPU & SPEC Workstation Benchmarks

EVGA sent BTR a RTX 3070 FTW3 Ultra ($609) and we have been evaluating it for the past couple of weeks. We also received a reference RX 6800 ($569) from AMD within the same timeframe. Since these cards have become direct competitors because of being priced within 7%, they each deserve a full review.

AMD sent us a reference RX 6800 the week after it launched. From our VR benchmarks, the RTX 3070 FE (Founders Edition $499) and the reference RX 6800 ($569) are in a similar class with the Radeon delivering higher unconstrained framerates which may support AMD’s justification for pricing it $70 higher than the FE. We use GPGPU, workstation, and SPEC benchmarks plus 35 games, directly comparing the reference RX 6800 with the $40 more expensive RTX 3070 FTW3 Ultra to see what the EVGA card at $609 brings over the RTX 3070 FE at $499.

BTR’s Test Bed

BTR’s test bed consists of 35 games and 3 synthetic game benchmarks at 1920×1080, 2560×1440, and at 3840×2160 as well as SPEC and GPGPU benchmarks. Our latest games include Cyberpunk 2077, Watch Dogs: Legions, Call of Duty Black Ops: Cold War, and Assassin’s Creed: Valhalla. We use 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.

Let’s split this review into two parts – First, let’s take a closer look at the EVGA RTX 3070 FTW3 Ultra on the next page.

The EVGA RTX 3070 FTW3 Ultra – Specifications & Features, Unboxing, Overclocking & noise

Specifications & Features

The RTX 3070 is not based on the GA102 chip like the RTX 3080 and the RTX 3090, but rather it uses a separate smaller GA104 GPU chip. The RTX 3070 FE uses 64 SMs, 5888 CUDA cores, 184 3rd Generation Tensor and 46 RT cores, along with 184 Texture Units and 96 ROPs. The Boost Clock is 1730MHz, and 8192MB of GDDR6 at 7000MHz on a 256-bit memory bus provide 448GB/s bandwidth, all within a 220W total GPU power envelope.

EVGA’s site lists the RTX 3070 FTW3 Ultra specifications and features and they are very impressive as they build upon the reference version with a 1815MHz Boost Clock and the ability to overclock further manually than the RTX 3070 Founders Editions because of selected GPUs and higher user accessible voltage.

Surprisingly, there is no mention of a dual-BIOS which is a great feature for overclockers. Although the OC BIOS mainly sports a more aggressive fan profile without an idle fan stop, we used the OC BIOS for our benchmarking. In addition, multiple sensors across the board monitor VRM/memory temps in Precision X1.

Unboxing

Unlike with NVIDIA’s Founders Editions, the EVGA RTX 3070 FTW3 Ultra comes in a traditional style box that advertises Ampere architecture, DLSS, and ray tracing.

The back of the box lists the key features in English, French, and Russian putting special emphasis on the Ampere architecture being second generation ray tracing cores and third generation tensor cores.

The box also states it is “VR Ready” but we did not see specifications, PSU requirements, nor information as to what is inside the box.

Inside the box and beneath the card are an installation guide and warranty information, plus a metal EVGA badge.

A completely redesigned shroud creates a very cool industrial look for the EVGA RTX 3070 FTW3 Ultra to provides a premium and solid heavy feel to it. It is a moderately heavy 2.75-slot card with three cooling fans. According to EVGA, triple HDB fans iCX3 technology offer higher performance cooling resulting in less acoustic noise. These iCX3 fans feature a second generation HDB active motor for silent 0dB mode, asyncronous fan mode, and a special upraised “E” pattern on the blades with the goal of a very quiet card.

It’s height is 5.38 in (136.75mm) and length is 11.81 in (300mm).

Turning the card over, we see a similar unique design with small hexagonal cutouts in the backplate. Some may not like the red line on the backplate as white, gray, or black may have been a more neutral choice fitting in with the overall design better. This card is designed to keep the GPU cool including by using a short PCB, and inside the card from its edges we see mostly all heatsink fins.

The plastic panel has RGB lighting with 60 LEDs that may be customized extensively using EVGA’s Precision X1 software. Two 8-pin PCIe connectors are required to supply external power to the card since most FTW cards are good overclockers. The FTW3 is rated to draw 220W at stock but may pull 50W or more when overclocked, and EVGA recommends a decent 650W PSU to power it.

There is very large surface area for cooling so the heat is readily transferred to the fin stack and the triple fans exhaust some heat out of the back and through the backplate cutouts, but especially from the top of the card into the case’s airflow.

The IO panel has a large air vent and four connectors. The connectors are similar to the Founders Edition of the RTX 3070. Three DisplayPort 1.4 connectors are included, and the HDMI port has been upgraded from 2.0 to 2.1 allowing for 4K/120Hz over a single HDMI cable.

In our opinion, the EVGA RTX 3070 FTW3 Ultra is a good-looking card with a unique industrial style and it looks good in any case. The FTW3 sports a Dual BIOS which is always a solid advantage for overclockers looking to push the card’s limits. And multiple sensors across the board for monitoring VRM/memory temps in Precision X1 giving enthusiasts more fine tuning controls.

The logo strip has 60 LEDs for a customizable RGB experience that can be synched to other EVGA component lighting. Unfortunately, we didn’t have a lot of luck photographing the customizable lighting, and it locked up leaving us with the logo in red and no way to adjust it even after multiple installs of Precision X1 and the drivers. EVGA’s tech support closed for much of the holidays so we will update this review with a short video if we can get the RGB working properly again. It looked very nice when it was working and there are a multitude of ways to customize the lighting, the colors, and the patterns.

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

Next, let’s take a closer look at the reference RX 6800.

The Reference RX 6800 – Big Navi 2 Specifications & Features, Unboxing, Overclocking & Noise

We received a reference RX 6800 from AMD after the launch together with a reference RX 6800 XT which we reviewed versus the Red Devil RX 6800 XT and versus the RTX 3080. The RX 6800 has the same architecture as the RX 6900 XT and the RX 6800 XT, but it has less compute units and stream processors and the game clocks are lower with an advantage of a lower power draw. However, all three cards sport 16GB DDR6 compared to the 10GB DDR6 of the RTX 3070 and RTX 3080 (DDR6X). The following chart shows the differences and similarities of the Big Navi 2 Radeon 6000 series.

The Big Navi 2 Radeon 6000 family

The RTX 6900 XT released at $999 to compete with the $1499 RTX 3090 while the RTX 6800 XT at $649 competes with the RTX 3080 at $699. Both of these cards are slower than their GeForce competitors but the Radeon 6800 competes with the $499 RTX 3070 and is priced higher at $579 as it is expected to be faster at rasterized games.

Just like NVIDIA, AMD has their own thriving ecosystem for gamers with many unique new features for the Radeon 6000 series. AMD touts the “open” system while NVIDIA’s tends to be more proprietary, each attracting fans who may prefer one gaming ecosystem over the other. Interestingly, AMD has adopted USB Type C for VR and displays while NVIDIA has dropped it from Ampere cards after using it for Turing.

Let’s unbox the reference RX 6800.

Unboxing

The reference RX 6800 box shows just a partial image of the card with “Radeon RX 6800” text. That’s it on the front.
The back of the box doesn’t advertise features nor give any useful information except to direct the buyer to www.amd.com/RadeonUserManual for a quickstart guide and warranty information.

Opening the well-padded box, we see the card inside (above). There are no bundled games and the packing is barebones compared with AMD’s partners’ cards such as PowerColor’s bundle.The reference RX 6800 is a large tri-fan card in a two slot design which is quite handsome using AMD’s reference design. The 6800 is very much a twin to the RX 6800 XT in looks and in size.

Above is the reference RX 6800 XT backplate. It has a very clean design top and bottom.

It uses two 1×8-pin PCIe connections and the heatsink can be seen along both edges.

Just like the other Big Navi 2 cards, the IO panel sports a Type-C USB connector for display and VR as well as two DisplayPort 1.4 ports plus a HDMI 2.1 port that can handle 4K/120Hz over a single cable.

We didn’t spend very much time overclocking the RX 6800. It has rather limited performance headroom with less than a 10% overclock as AMD evidently preferred giving everyone the highest performance they could get out of all of their new cards and Wattman is rather locked down regarding clocks and voltage. But the reference design is probably AMD’s best design to date, and the card runs very cool and quiet even under load.

Overclocking, temperatures and noise

The reference RX 6800 is a fair overclocker. All of our performance and overclocked testing are performed in a closed Phanteks Eclipse P400 ATX mid-tower case. Inside, the RX 6800 is a quiet card even when overclocked and we never needed to increase its fan speeds manually or change the stock fan profile.

Here are the default Wattman settings above using Heaven to load the GPU to above 98%.

Fan speeds are low and the RX 6800 boosts easily to 2257MHz at stock. Next we overclocked it.

The reference RX 6800 when overclocked between a minimum frequency of 2350Mhz and a maximum of 2450MHz, the Radeon Boost generally stays above 2400MHz with the memory at 2140MHz which provides better than a 4% performance boost for Heaven 4.0. We were unable to hit 2500MHz and memory is locked down to +7% overclock. It may be possible to fine tune this overclock further by trial-and-error undervolting/overclocking, but we found that it was necessary to increase the Power Limit to even hit 2450Mhz.

Even overclocked, the reference RX 6800 stays cool below 75C, and the fans are never irritating when they become audible under full load.

Let’s check out its performance compared with the EVGA RTX 3070 FTW3 Ultra and the RTX 3`070 FE after we look over our test configuration 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 Reference version 16GB, stock settings, on loan from AMD
• EVGA RTX 3070 FTW3 Ultra 8GB, at factory clocks and further manually overclocked, on loan from EVGA
• RTX 3070 Founders Edition 8GB, stock, on loan from NVIDIA
• Radeon RX 6800 XT Reference version 16GB, stock settings, on loan from AMD
• Radeon RX 5700 XT 8GB Anniversary Edition, stock AE clocks.
• RTX 3090 Founders Edition 24GB, stock clocks, on loan from NVIDIA
• RTX 3080 Founders Edition 10GB, 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 460.89 for the RTX 3070s, 456.96 for 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.
• Adrenalin 2020 Edition 20.12.1/2 drivers used for the RX 6800 and 6800 XT reference. 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
• Precision X1, latest beta

Games

Vulkan

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

DX12

• Cyberpunk 2077
• 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
• Gears 5
• Anno 1800
• Tom Clancy’s The Division 2
• Metro Exodus
• Civilization VI – Gathering Storm Expansion
• Battlefield V
• Shadow of the Tomb Raider
• Project CARS 2
• Forza 7

DX11

• Crysis Remastered
• Mech Warrior 5: Mercenaries
• Star Wars: Jedi Fallen Order
• The Outer Worlds
• Destiny 2 Shadowkeep
• Borderlands 3
• Total War: Three Kingdoms
• Far Cry New Dawn
• Assetto Corsa: Competitione
• 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.

We have recently upgraded our Core i7-8700K/Coffee Lake platform to i9-10900K/Comet Lake, and have been testing it using an ASRock Z490 Steel Legend motherboard. Since we were unable to gain any performance from overclocking, we asked EVGA for a review sample of the Z490 FTW motherboard.

The last EVGA motherboard that BTR evaluated was the Z370 FTW which allowed our i7-8700K to reach 4.8 GHz on all cores for regular use and for benching. We are going to review the $329 EVGA Z490 FTW motherboard using a RTX 2080 Ti by comparing it with the midrange ASRock Z490 Steel Legend motherboard with our i9-10900K at stock, and then we will overclock it.

It has been a long road to a solid performance increase from a stable overclock that required us to not only change the motherboard, but we also had to make several other upgrades. Core i9-10900 “K” CPUs are multiplier unlocked and can be overclocked beyond their stock speeds with hiqh-quality cooling. At stock, although all 10 cores are rated to run at 3.7 GHz, individual cores each have the potential to reach much higher speeds by selectively using Intel’s Boost and Velocity Boost which means there is very little room for increasing performance by manually overclocking. We started out using a EVGA CLC 280mm CPU cooler in a Focus G case, and later we upgraded to a Phanteks P400 case to house a DeepCool 360mm AIO cooler.

Comet Lake is Intel’s latest tenth generation platform and it brings some new features over the Coffee Lake platform although their IPC for gaming are identical So far, we have found that the biggest advantage to the Comet Lake flagship i9 CPU over Coffee Lake’s flagship i7 processor are the extra four cores of the i9. A Core i9 Comet Lake gamer using 10 cores plus HyperThreading doesn’t have to consider background processes while gaming, and the extra four cores over Coffee Lake’s CPU may be helpful for better handling upcoming multi-threaded game ports from the next generation of console games. In addition, we are also looking for a higher all core clockspeed well-above the 4.8GHz day-to-day overclock of our i7-8700K.

EVGA’s Z490 Motherboards

EVGA has introduced another Z490 motherboard in addition to the FTW that we are reviewing – the $549 Z490 DARK – their top board. The Z490 FTW is currently available from EVGA for $329. The primary difference, besides several added features between the two motherboards, are a 18 phase power delivery system for the DARK while the FTW is 14 phase. The ASRock Z490 Steel Legend uses a 11 phase power design.

The EVGA Z490 FTW is a step down from the DARK. Besides using a 18 phase power delivery instead of 14, the DARK has two extra native SATA ports, an extra USB header, and overclocking support for DDR4 to 4600MHz instead of 4400MHz. The DARK also has an upgraded Ethernet, an additional PCIe slot, and the PCB is ten layers instead of six. These are not really major downgrades for most enthusiasts although an extreme overclocker would definitely pick the DARK over the FTW. A motherboard may make a difference to achieving a higher CPU overclock. The FTW Z490 MB was able to overclock our i9-10900K while our ASRock Z490 Steel Legend couldn’t.

EVGA Z490 boards include cable cutouts to make cable management easier. These boards also feature metal-reinforced PCIe slots for supporting heavy video cards, 2-Way SLI/CrossFire support, multiple RGB headers, M.2 slots, Intel Gigabit NICs, switchable dual-BIOSes, and Realtek’s upgraded 7.1 Channel 1220 audio. Here are the EVGA Z490 FTW motherboard’s features from their website.

DESIGN DETAILS

• • Supported CPUs – Intel® Socket 1200, 10th Generation Intel® Core i9/i7/i5
  • Socket Type – Intel® Socket LGA1200
  • PCH – Intel® Z490
  • DIMM QTY – 4 DIMM Dual-Channel
  • Memory Type – DDR4 4400MHz+
  • Memory Capacity – 128GB
  • 6.0Gb/s Ports/Controller – 4/Intel® Z490 PCH
  • RAID Support – RAID 0, 1, 5, and 10
  • SATA 6.0Gb/s Ports/Controller – 2 / ASMedia ASM1061
  • USB 2.0 Ports/Controller – 5 Ports (4 from internal headers / 1 from Update Port for flashing BIOS)/ Intel® Z490 USB Hub
  • USB 3.2 Gen1 Ports/Controller – 4 (2 from internal header / Intel Z490 PCH
  • USB 3.2 Gen2 Ports/Controller – 4x USB3.2 Gen2 Type-A/1x USB3.2 Gen2x2 Type-C/1x USB3.2 Gen2 Type-C (internal header)
  • Network Speed – 10/100/1000
  • Network Ports/Controller – Intel® i219V PHY
  • WiFi/BT – Intel® AX201 WiFi 6/BT 5.1 module, preinstalled in the • M.2 Key-E 32mm Slot
  • Audio – 7.1 Channel Realtek + EVGA NU Audio
  • Audio Controller – Realtek ALC1220 + SV3H615
  • Display Output – DP 1.2 / HDMI 1.4
  • PCIe Slot Arrangement – 2×16, 1×1
  • PCIe x16 Mechanical Slots – 2
  • PCIe x16 Mechanical Arrangement – 1×16/8, 1×8 PCIe x1 Mechanical Slots – 1
  • M.2 Key-M – 2x 110mm (Up to 32Gbps)
  • M.2 Key-E – 1x 32mm (Vertical)
  • Fan Headers – 7x 4-Pin (2x CPU PWM, 5x PWM/DC)
  • BIOS Type – Latest UEFI BIOS with mouse/keyboard control, OC Robot, In-BIOS Stress
  • Software – EVGA ELEET X1 Tuning Utility

KEY SPECS

• Supports Intel® Core 10th Generation Processor Family for LGA1200 socket
• 150% Increased Gold Content
• Intel® Z490 Chipset
• NVIDIA® SLI® Ready Enthusiast Layout
• 4 DIMM Dual-Channel up to 128GB 4400MHz+
• PCI Express® 3.0 Ready
• 5 USB 2.0 Ports (4 from internal headers/1 from Update Port)
• 4 USB 3.2 Gen1 Ports (2 rear panel, 2 from 1 internal header)
• 5 USB 3.2 Gen2 Ports (4 Type-A rear panel, 1 type-C Header)
• 1 USB 3.2 Gen2x2 Port (1 Type-C rear panel)
• 1 PS/2 (Mouse+Keyboard support)
• Intel® Optane Support
• 6 SATA 6.0Gb/s (4 on Intel® Z490 PCH/2 from ASMedia ASM1061)
• 2 M.2 Key-M 110mm up to 32Gbps
• 1 M.2 Key-E 32mm
• 1 DisplayPort 1.2 / 1 HDMI 1.4
• 1 Intel® i219V Gigabit NIC (10/100/1000)
• 1 Intel® WiFi 6/ BT 5.1 module, preinstalled

DIMENSIONS

• Width: 9.6in – 244mm
• Length: 12in – 305mm
• Form Factor: ATX Form Factor

ACCESSORIES

• EVGA Quick Installation Guide
• Rear Case I/O Panel
• 2 SATA 6G Data Cables
• 2x M.2 Thermal Pad
• 2x Antenna for WiFi
• Case Badge
• USB Flash Drive
• Contains Driver and Manual

###

The specs and the design look great. An important consideration for any enthusiast are the warranty and product support should anything go wrong.

Warranty & Support

EVGA’s Z490 motherboards come with a 3 year warranty, and registration is recommended. A further warranty extension
is available upon registration within 30 days of purchase. For more details please visit: www.evga.com/warranty/motherboards/.

In addition, EVGA provides outstanding support including a 24/7 telephone hotline staffed by helpful professionals. We have had multiple opportunities to use their hotline over the past decade, and we have always received high-quality and fast service – even on the weekends. EVGA is a rare company that will allow you as a seller to transfer the remaining portion of your warranty to the second-hand buyer, and they can even cross ship RMA products so you can get back up and running quickly!

The EVGA Z490 FTW is a good-looking motherboard with excellent specifications, so let’s unbox it for a closer look before we install, test, overclock, and benchmark it.

The EVGA Z490 FTW motherboard

Unboxing and Installation

The EVGA Z490 FTW motherboard is a good-looking industrial design board using silver components to contrast with a black PCB without using any stenciled designs on the board. There are typical component placements, and some of the traces are visible. The FTW includes a silver shroud over the back panel with vents that extend to the VRM heatsinks. There are 4 RAM slots supporting up to 128GB DDR4, and 2 PCIe slots that are full X16 slots that will support mGPU in an 8x+8x configuration. If you need a high bandwidth (HB) SLI bridge for two GeForce cards, make sure to get the bridge with the 2-slot spacing.

There are headers on the board for RGB strips. The memory slots as well as the two primary x16 PCIe slots (1×16/8, 1×8 in dual graphics mGPU mode) are reinforced in order to support heavy graphics cards. We do not like the placement of the right angle connectors at the bottom of the board as the spacing is too tight for many cases including the Focus G and the Phanteks P400. You may have to remove the PSU – or as with the P400, you may have to remove the motherboard just to replace or install a new fan. The only vertical socket is for the ATX power connector and we would have instead preferred that it be right angle.

There is also a third PCIe slot at the bottom using the bandwidth from the chipset. There are two Key M M.2 slots for PCIe storage – 2x 110mm (Up to 32Gbps) – and one Key E M.2 slot with an Intel AX201 WiFi 6/BT 5.1 module preinstalled. Storage options include six SATA ports (RAID 0, 1, 5, and 10 support).

For convenience, there are seven 4-pin PWM fan headers (2x CPU PWM, 5x PWM/DC), and four are close to the CPU socket with the others at the bottom of the board. For audio, the FTW uses 7.1 Channel Realtek + EVGA NU Audio, and for networking, the FTW uses a single Intel I219-V PHY controller.

The EVGA Z490 FTW motherboard arrives in a rather plain box with the necessary components for a bare-bones installation.Bucking the modern trend, EVGA still supplies a separate rear case IO panel but we are used to the extra step. Besides the four SATA cables, IO rear panel cover, EVGA badge, mini-flash drive instead of a CD, quick-start guide and Wi-Fi antennas, the bundle is sparse. We really miss having a printed manual, but settled for the .pdf which is still rather basic for 163 pages. However, an experienced builder may not need it. The Z490 FTW is very similar to other EVGA motherboards, and anyone who has used one before will be familiar with its layout.

Here is the back of the FTW motherboard.

On the rear panel of the EVGA Z490 FTW, there is a Clear CMOS button, HDMI 1.4 and DisplayPort 2.0 video outputs, four USB 3.2 ports (red), the USB 3.1 ports (blue), a USB Type-C port, the Intel network port, the audio jacks, and the two Wi-Fi antennas screw right into their corresponding bases.

Originally we started with an ASRock Z490 Steel Legend which is a mid-range board. We pre-installed the memory and the CPU together with the IO backplate and CPU cooler standoffs and securely fastened them using 9 screws to the standoffs in the Focus G. We used 2x16GB of T-FORCE DARK Z DDR4 at stock speeds of 3600 MHz using the automatic XMP profile 1 in the BIOS. We installed the AIO cooler using Arctic Silver 5 and attached the rest of the cables and wiring.

The ASRock Z490 Steel Legend motherboard was OK for running our i9-10900K at stock speeds but it was unable to attain any performance-gaining overclock with a EVGA CLC 280mm CPU cooler in a Focus G case, so we switched it out for the EVGA Z490 motherboard. We started our build by moving the hardware from our ASRock motherboard to the EVGA motherboard, and then installed it into the same Focus G case.

After the FTW build was completed, we got a decent overclock, but the temperatures were too high for the 280mm cooler. So we upgraded to a Phanteks case to house a DeepCool 360mm AIO cooler.

It was now a very easy transplant installation since we had learned to pre-install the HD audio and fan plugs into the bottom connectors before we installed the motherboard.

Well, it was soon time to turn it on. And it started right up and went to the BIOS screen. We picked Advanced and ignored the other three choices for now: Default, OC Robot, and Gamer Mode which sets a conservative overclock.

BIOS

Upon reaching the BIOS by pressing delete after powering on, you are presented with an overclocking-friendly screen. You can navigate the BIOS with a keyboard or mouse, and although it is fully-featured, it is still somewhat minimalist. It’s an advantage as the options are easy to understand and they give full control over a multitude of overclocking options. Usually, screenshots can be saved to a USB flash drive, but it never worked reliably for us so we captured photos of a few of the BIOS options and overclocking-related settings that we used.

Overclocking is simple. You switch from Automatic (3.7 GHz with multiple automatic stages of Boost) to Manual and are presented with options to set the multiplier up or down. Setting the overall CPU Multiplier setting sets the individual cores to the same frequency or they may be adjusted individually as shown above.

We settled a multiplier of 51 for 5.1 GHz on all cores. We also made sure the AVX Ratio offset was set to zero because we wanted all 10 cores to turbo to 5.1 GHz without any exception for demanding software or even for ‘power viruses’ like Prime95.

The BIOS offers options to set the BCLK Frequency but we only overclocked the core. Individual cores may have their HyperThreading enabled or disabled for those who want the ultimate in overclocking options. We enabled HT on all cores at 5.1GHz for overall better performance and benchmark repeatability although we were able to get a higher overclock on some cores by selectively disabling it .

Unlike with many motherboards that allow excessive voltage to flow to a highly overclocked CPU, the FTW is a rare board that did not allow excessive spikes even with the automatic setting at 5.1 GHz. However, we also found that setting the voltage manually to 1.335V in Vcore was enough to stabilize the CPU at 5.1 GHz although it still occasionally spiked. Disabling or enabling Vdroop made no difference to overall stability that we could notice. In the Advanced options, the user is given many options including to further fine-tune the CPU.

All of the other regular options are in the BIOS, and flashing is made easy by having a dual BIOS. We tried v1.04 but found that we got a better overclock with v1.03.

Overclocking, Voltage, and Temperatures

An i9-10900K overclocked all-core to 4.9GHz or higher requires additional voltage overstock, and temperatures rise dramatically as the voltage is increased to maintain stability for higher overclocks. The ASRock motherboard could not hold a stable overclock without throttling. Although 5.1GHz could be applied, throttling by what we suspect is uneven power delivery caused the CPU cores to throttle significantly, and the overclocked performance was lower than stock performance even though temperatures were not excessive.

Our favorite new EVGA tool is the OC Robot. Watching it closely as it ran through the frequencies told us exactly where to start with our manual overclock and what approximate voltages worked, as well as letting us know where the CPU got into thermal trouble. Here is 4.9GHz.Only 1.247V are needed for all-cores at 4900MHz. Watch what happens when the OC Robot tests 5.0GHz.

It now takes 1.3V to stabilize an all core 5000MHz OC with the temperatures peaking at 85C. So the Robot continues.

The OC Robot settled on 5.1GHz since 5.2GHz raised the voltage and temperatures dramatically. OC Robot’s 5.1GHz OC is exactly the same overclock that we were able to finally achieve after spending hours manually trying setting after setting. The only difference is that we were able to stabilize 5100MHz at 1.335V instead of 1.355V, and thus we got lower temperatures. Overall, we would give the OC Robot an A- for being a time-saver and for being reasonably accurate. The OC Robot’s result is a great place for an experienced overclocker to begin manual testing and fine-tuning.

When we manually overclocked further to 5.2 GHz, the voltage requirements exceeded 1.4V and temperatures spiked into the 90sC under a full gaming load and near 100C under full load synthetic benches. These benches are often called power viruses and may include OCCT, the Blender Benchmark, and Prime95.

We were unable to use our EVGA 280mm CLC even with a modified aggressive fan profile at 5.2GHz, so we switched out the AIO to a 360mm DeepCool AIO and a larger Phanteks case. Although the temperatures dropped by 5-8C or better, thermal throttling still took place, and the 5.2GHz overclock resulted in overall lower performance than at 5.1GHz. We also tried 5.3GHz with a 2AVX offset, but we couldn’t get the i9 completely stable. So we settled on a maximum stable 10900K overclock with all 10 cores at 5.1GHz and set about benching and comparing performance with stock settings.

Driver Installation

The included EVGA motherboard drivers and programs are no longer on CD but on a flash drive, and they work quickly and painlessly to install all needed drivers. The latest drivers can always be found on EVGA’s website which are what we used before we discovered the tiny flash drive at the bottom of a plastic bag.

Let’s look at our test configuration before we do any benching

Test Configuration

Test Configuration – Hardware

• Intel Core i9-10900K (HyperThreading/Turbo boost On; stock and overclocked. Rocket Lake DX11 CPU graphics)
• EVGA Z490 FTW motherboard (Intel Z490 chipset, v1.3 and v1.4 BIOSes, PCIe 3.0/3.1/3.2 specification, CrossFire/SLI 8x+8x), supplied by EVGA
• ASRock Z490 Steel Legend motherboard (Intel Z490 chipset, latest BIOS, PCIe 3.0/3.1/3.2 specification, CrossFire/SLI 8x+8x)
• DEEPCOOL Castle 360EX AIO 360mm liquid CPU cooler
• EVGA 280mm CLC AIO liquid CPU cooler, supplied by EVGA
• T-FORCE DARK Z 32GB DDR4 (2x16GB, dual channel at 3600MHz), supplied by Team Group
• RTX 2080 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
• 2TB Micron 1100 SATA III SSD
• 1TB Team Group GX2 SATA III SSD
• 500GB T-FORCE Vulcan SSD, supplied by Team Group
• ANTEC HCG1000 Extreme, 1000W gold power supply unit
• BenQ EW3270U 32 Inch 4K HDR FreeSync Monitor
• SAMSUNG LC27G75TQSNXZA 27″ 2560 x 1440 240Hz 1ms G-Sync Monitor
• Fractal Design Focus G mid-tower PC case
• Phanteks Eclipse P400 ATX mid-tower

Test Configuration – Software

• Nvidia’s GeForce 451.67 WHQL drivers. High Quality, prefer maximum performance, single display
• VSync is off in the control panel
• AA enabled as noted in games; all in-game settings are specified with 16xAF always applied
• Highest quality sound (stereo) used in all games
• Windows 10 64-bit Pro edition; latest updates v2004
• Latest DirectX
• MSI’s Afterburner, latest version.
• CPUZ
• HWiNFO
• Intel XTU

Game-related

• Grand Theft Auto V
• Civilization Vi
• Anno 1800
• 3DMark – Time Spy & Physics
• Superposition
• VRMark Cyan

Synthetic

• Sandra 2020
• AIDA64
• PCMark 8
• PCMark 10
• RealBench
• Cinebench
• NovoBench
• Blender Benchmark
• Wprime

We used MSI’s Afterburner to set the RTX 2080 Ti’s power and temp limits to their maximums.

Let’s head to our performance charts.

Synthetic Benches

SiSoft Sandra 2020

To see where the CPU 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 consummate information & diagnostic utility in a single 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. It is highly recommended! Sandra 2020 R8 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.

We ran Sandra’s intensive benchmarks and charted the results summarizing our CPU-related benchmark testing. As with all of the following charts, the performance results of the ASRock Steel Legend motherboard at stock CPU speeds are first compared with the stock performance results of the EVGA FTW motherboard at stock and then with an all-core 5.1GHz i9 overclock.

Generally Z490 motherboards using the same CPU will give very similar results. Interestingly, the power management efficiency of the ASRock motherboard is higher than the EVGA board, and the FTW’s efficiency decreases as the power demands go up from overclocking.

A faster CPU has better results when it plays a major part in the calculations. When the tests are GPU oriented, the graphics card does most of the work. In the Sandra CPU benchmarks, overclocking the 10900K to 5.1GHz brings higher performance over stock.

We next feature AIDA64.

AIDA64 v6.00

As the successor to Everest, AIDA64 is an important industry tool for benchmarkers. Its memory bandwidth benchmarks (Memory Read, Memory Write, and Memory Copy) measure the maximum available memory data transfer bandwidth and its custom CPU benchmarks measure performance and give scores to compare against other popular CPUs.

AIDA64’s benchmark code methods are written in Assembly language, and they are extremely optimized for every popular AMD, Intel and VIA processor core variants 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.

The AIDA64 Memory Latency benchmark measures the typical delay beginning from when the CPU reads data from the system memory. Memory latency time means the time is accurately measured from the issuing of the read command until the data arrives to the integer registers of the CPU. It also tests Memory Read, Write, and Copy speeds besides Cache.

CPU Queen is an integer benchmark that focuses on the branch prediction capabilities of the CPU. It finds the solutions for the classic “Queens problem” on a 10 x 10 chessboard. CPU PhotoWorxx performs common tasks used during photo processing which stresses the SIMD integer arithmetic execution units of the CPU and also the memory subsystem. ZLib is a compression benchmark, while AES focus on Advanced Encryption Standard data encryption. SHA3 benchmarks use a standard hashing algorithm.

Here is the summary chart of the multiple AIDA64 memory benchmarks.As the CPU is overclocked, the bandwidth increases and the benchmark scores also scale favorably with overclocking. The performance results are similar between the ASRock and the EVGA motherboards when the CPU is at the same stock clocks, but the FTW’s 5.1GHz overclock delivers consistently higher performance.

Let’s look at PCMark 8 next to see if its benchmarks can reflect CPU speed increases.

PCMark 8

PCMark 8 has a great Creative test which uses real world timed benchmarks including web browsing, video group chat, photo, batch, and video editing, music and video tests, and even mainstream gaming. Since the PCMark 8 Storage Test does not test the CPU, we only used the Creative benchmark suite.

Here are the ASRock Steel Legend results with the i9 at stock settings – 9137.

Overclocking the i9-10900K to 5.1GHz in the ASRock motherboard brought no performance increase and delivered the same score.

Next are the stock i9 results in the FTW motherboard – 9715.

It’s a somewhat shocking comparison that may lead one to suspect that some of the demanding and prolonged PCMark 8 benchmarks may cause the ASRock motherboard to also throttle the i9 clocks even at stock settings.

Here are the results of the i9-10900K on the FTW motherboard now overclocked to 5.1GHz – 9967

In contrast to using the ASRock motherboard, overclocking the i9 to 5.1GHz in the FTW motherboard brought increased performance as evidenced by the improved score. Here is the summary chart.

We may perhaps infer from the summary chart that increasing the CPU speed to 5.1GHz with an all core overclock may help increase overall PCMark 8 performance as reflected by the results.

PCMark 10 is next.

PCMark 10

The PCMark 10 benching suite is the follow-up to PCMark 8 and it also uses real world timed benchmarks which include web browsing, video group chat, photo, batch, and video editing, music and video tests, and even mainstream gaming. The PCMark 10 test offers two primary tests and we chose the extended version.

First up is the i9–10900K at stock in the ASRock Steel Legend motherboard – 9848.

Next we overclock the i9–10900K to 5.1GHz in the ASRock Steel Legend motherboard – and it loses performance again – 9636.

Overclocking the i9 to 5.1GHz in the ASRock motherboard resulted in lower performance than at stock settings. So let’s look at the same i9 at stock and also overclocked in the FTW motherboard.

Next are the stock 10900K results in the EVGA Z490 FTW board – 10079.

Now lets see the results after we overclock the i9 to 5.1GHz in the FTW motherboard –10568.

Overclocking the i9-10900K CPU from stock to 5.1 GHz in the FTW motherboard makes for a decent performance improvement judging by the increased scores. Here is the summary chart.

We suspect that the ASRock motherboard may not throttle the CPU at stock speeds as it did with PCMark 8’s much longer tests, but overclocking it to 5.1GHz is a waste of time in a vain attempt to gain performance. We found the same issue with our other benchmarks and cannot recommend the Steel Legend for overclocking the i9-10900K based on our experiences with it.

Let’s look at our next synthetic test, RealBench.

RealBench v2.56

RealBench is a benchmarking utility by ASUS Republic of Gamers which benchmarks image editing, encoding, OpenCL, and Heavy Multitasking. Afterward, it gives individual results and an overall score for easy comparison off or online. Some of these tests are affected by CPU and memory speeds.

Here is the summary chart.

The stock results are close between the motherboards, but the performance and the scores generally increase with higher CPU clocks.

Next we benchmark using Cinebench.

Cinebench

CINEBENCH is based on MAXON’s professional 3D content creation suite, CINEMA 4D. This latest R20.0 version of CINEBENCH can test up to 64 processor threads accurately and automatically. It is an excellent tool to compare both CPU/memory and graphics OGL performance. We focus on the CPU whose results are given is cb, and higher is always better.

Here is the summary chart.

The ASRock motherboard scores higher but when we overclocked it to 5.1GHz, the score dropped to 5738! In contrast, the same overclock brought a small performance increases using the FTW motherboard. Next up, Novabench.

Novabench

Novabench is a very fast benching utility that spits out a 4 test results and an overall system score. We will focus on the CPU score and bandwidth speeds.

Overclocked, the i9-10900K gains performance overall. Here is the summary chart.

On to Wprime and number crunching.

WPrime v2.10

WPrime is a multi-threaded benchmark which may show the differences in IPC or clockspeeds between CPUs. Here are the tests using 10 threads, and we choose to calculate 1024 million digits and 32 million digits showing multiple runs.

Here is the summary chart:

An overclocked CPU calculates faster than a stock CPU and in all cases of calculation, overclocked is faster than at stock frequencies enabling the CPU to crunch numbers a little faster.

Let’s take a look at Blender.

Blender 2.83

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.

We have seen Blender performance increase with faster CPU speeds, so we decided to try several specific Blender 2.83 benchmarks which can measure CPU performance by timing how long it takes to render production files.

For the following chart, lower is better as the benchmark renders a scene multiple times and gives the results in minutes and seconds. This time, we tested the overclocked ASRock motherboard as well as multiple speeds of the 10900K on the FTW motherboard.

From the chart, the Blender benchmark performance is highest with the i9-10900K overclocked to 5.1GHz in the EVGA Z490 FTW motherboard. Overclocking beyond 5.1GHz delivers less performance.

Next, we move on to game-related benchmarks and games

Game Related Benching

Let’s look at quasi game-related benchmarks starting with 3DMark next.

Fire Strike Physics & Time Spy

Fire Strike Physics depends less on the GPU and more on the CPU which can benefit from increased CPU speeds while Time Spy is more dependent on the video card.

Overclocking again brings performance increases. Let’s look at VR next.

Virtual Reality (VR)

Superposition

Superposition is benchmarked at 720P low settings and we see no real difference in the scores between motherboards other than a small boost by overclocking the i9.

VRMark – Cyan Room

There is virtually no performance difference between motherboards or by CPU overclocking with VR. It is mostly dependent on the graphics card.

Although a VR experience depends on the GPU, a wireless adapter may make a faster CPU vital to an untethered VR experience. We are going to test our Vive Pro’s Wireless Adapter and will benchmark and compare its VR performance between using an i7-8700K at 4.8GHz and an i9-10900K at 5.1GHz later this month.

Let’s look at three PC games that may scale by increasing CPU speeds.

Gaming Performance Benchmarks

Below is the summary chart of three games that use accurate built-in benchmarks which also appear to be sensitive to scaling by increasing CPU clocks.

The highest settings are used, and the benches were run at 1920×1080 using a stock RTX 2080 Ti Founders Edition. Civilization 6’s built-in graphics benchmark uses frametimes but its AI benchmark measures the turn time in seconds – in both cases, lower is better. Anno 1800 is benched using OCAT and so the averages and the minimums are also expressed in frametimes where lower is better. The Grand Theft Auto V average/minimum results are given in FPS where higher is better.

Although the benchmarking margin of error may cloud the results, there is a trend showing that faster CPU clocks will increase framerates and improve frametimes for certain CPU-dependent games. We are going to follow up this review with an expanded review focusing on gaming as we compare a i7-8700K at 4.8GHz to the i9-10900K at 5.1GHz.

All Summary Charts

Here are all of the Summary charts.

If you are a primarily a gamer, you will get much higher framerates from overclocking your video card than from overclocking your CPU. If you game at 2560×1440, 3440×1440, or especially at 4K, you won’t notice any framerate increase from CPU scaling. However, for other tasks that primarily involve the CPU, overclocking may provide more dramatic results.

Let’s head for our conclusion.

Conclusion

There is no doubt that overclocking an i9-10900K from its stock frequencies of 3.7 GHz – even with Turbo and Velocity Turbo boost – to all-core 5.1GHz provides more performance. But any increased performance from overclocking is dependent on proper voltage delivery and stability from the motherboard plus well-controlled thermals.

We are impressed that the EVGA Z490 FTW motherboard was able to stably take our i9-10900K to 5.1 GHz after we failed to improve CPU performance by any manual overclock with our ASRock Steel Legend motherboard. Of course, there is a large price difference between the FTW and the Steel Legend, but it throttled excessively even at relatively cool temperatures; something the FTW motherboard never did.

As gamers, we see that CPU scaling is evident for the three highlighted games that are not GPU-bound with a small framerate increase when we increase our CPU’s core clocks to 5.1GHz from default. However, since there are issues with running a CPU at high frequency and with higher temperatures which will shorten a CPU’s life, each overclocker needs to balance the negatives against increased performance.

An extreme overclocker who wants every last bit of performance from his CPU will no doubt consider delidding their CPU to attempt to achieve an all core 5.3GHz overclock with watercooling. But a even dedicated watercooling may be insufficient on EVGA’s top motherboard to hit this overclock unless he wins the silicon lottery with a golden CPU.

Intel has pushed the i9-10900K right to its edge to maintain its gaming performance crown. Others may choose to moderate their overclocks, and for us, an all-core 5.1GHz OC provides a good balance for our general benchmarking tests and for 24/7 use with a reasonable voltage and with good temperatures under a full gaming load.

If you are upgrading to an i9-10900K, the EVGA Z490 FTW is an excellent choice as a fully-featured motherboard that will provide clean power and great features for overclockers. Intel did not leave a lot room for overclocking, but the FTW will help you maximize what is available.

EVGA Z490 FTW – Pros and Cons

Pros

• The EVGA Z490 FTW is a good-looking conservative industrial design motherboard at a competitive price
• The EVGA Z490 FTW is a fully-featured motherboard with premium components, and it has proved itself to be solid in providing stability for our i9-10900K at 5.1 GHz
• EVGA gives great 24/7 telephone support and 3 years of warranty service if you need help with your motherboard or if you need to RMA. You can also transfer your warranty to the next buyer and set up advanced RMAs for shorter down times
• In multi-tasking, encoding and almost every other task that we tested including gaming, an all core 5.1GHz overclocked Core i7-10900K is faster than at stock, and the FTW has the right overclocking tools to stabilize it
• Plenty of fan headers are available and cable management is easy using the provided FTW motherboard cutouts
• Dual BIOSes are very useful, and together with an external CMOS reset are very helpful for testing extreme overclocks
• The FTW BIOS continues to improve over earlier editions, and it has become less complex, easier to navigate, and more intuitive. Most overclocking tools are in the BIOS.
• The OC Robot is EVGA’s new star overclocking tool for inexperienced as well as well-seasoned overclockers

Cons

• We would have liked more room at the bottom of the board for right angle connectors without having to remove the PSU.
• The external Wi-Fi antennas are a funky solution and they get in the way of other connections.

We feel that EVGA has delivered another good overclocking motherboard in the form of the Z490 FTW. If you are an overclocker, you can choose this board with confidence. The EVGA Z490 FTW motherboard has become BTR’s flagship motherboard and we feel that it deserves BTR’s Editor’s Choice Award. After all, we are satisfied with our i9-10900K overclock and we will continue to use the FTW for benchmarking.

We have not finished benching the 10900K. We are going to continue by focusing on gaming as we compare our i7-8700K at 4.8GHz with our i9-10900K at stock and overclocked to see if it is a really worthwhile upgrade.

First we want take a close look at 240Hz gaming with our new Samsung 27″ 2560×1440 display to see if we can tell the difference from a really fast refresh rate compared to gaming at 120Hz or 100Hz. We certainly can tell the difference from playing our 32″ 4K BenQ FreeSync 2 display at 60Hz to a more fluid experience at 120Hz or even 100Hz on our ACER Predator 34″ G-SYNC 3440×1440 display.

Stay tuned!

Happy Gaming!

This overclocking showdown is the follow-up to the ASUS Dual GTX 1660 SUPER EVO review where we are now focusing on its maximum overclocked performance versus the maximum overclocked EVGA GTX 1660 Ti XC. The GTX 1660 SUPER was recently released at the same $229 with significantly more performance than the original vanilla GTX 1660. We found the new GTX 1660 SUPER performs within about 3% of the nearly fifty dollars more expensive Ti but we want to see if manually overclocking each card changes anything.

The ASUS Dual GTX 1660 SUPER EVO Overclocking

We reached a final stable overclock close to 2100MHz by adding +165MHz to the ASUS DUAL GTX 1660 SUPER EVO OC core’s offset. The GDDR6 is also highly overclockable as we were able to add +850MHz. Full overclocking details are here.

The EVGA GTX 1660 Ti XC Overclocking

We added 175MHz to the core to boost the EVGA GTX 1660 Ti XC’s clocks above 2050MHz. We also added 800MHz to the GDDR6 memory overclock and the full details of our overclocking experience is here.

Let’s check the performance results of 46 games with our games at factory-clocked and overclocked speeds using the EVGA GTX 1660 SUPER DUAL EVO OC and the ASUS GTX 1660 TI XC.

Testing Platform

We test 46 games and 3 synthetic benchmarks at 2560×1440 and at 1920×1080. 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 full 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
• ASUS DUAL GTX 1660 SUPER EVO OC, on loan from ASUS
• EVGA GTX 1660 Ti XC Gaming 6GB, stock GTX 1660 Ti clocks, on loan from EVGA
• 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
• Seasonic 850W Gold Focus power supply unit
• EVGA CLC 280mm CPU water cooler, supplied by EVGA
• 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 440.77 drivers used for the EVGA GTX 1660 Ti XC. . Press launch drivers 441.77 are used for the ASUS DUAL GTX 1660 SUPER EVO OC. See NVIDIA Control Panel image below.
• 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. Four games use the Vulkan API.
• Latest DirectX
• All 46 games are patched to their latest versions at time of posting.
• WattMan used to set Radeon cooling and power options.
• Afterburner used for GeForce settings and primarily for overclocking
• ASUS GPU Tweak II
• OCAT, latest version
• Fraps, latest version
• FrameView, latest beta
• Unigine Heaven 4.0 benchmark

46 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
• Overwatch
• For Honor
• Mass Effect: Andromeda
• ARK: Survival Evolved
• 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
• 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

• Civilization VI
• Sniper Elite 4
• Forza 7
• Total War: Warhammer II
• Total War: 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

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. By setting the Power Limits and Temperature limits to maximum, they do not throttle, but they can reach and maintain their individual maximum clocks.

Let’s head to the performance results of our overclocking showdown.

Performance summary charts

Here are the performance results of 46 games and 3 synthetic tests comparing the factory clocked EVGA and overclocked GTX 1660 Ti versus the ASUS stock and overclocked RTX 16600 SUPER. The highest settings are always chosen and the settings are listed on the charts. The benches were run at 1920×1080 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. Games benched with OCAT results are given as frametimes in ms where lower numbers are better.

The EVGA factory-clocked GTX 1660 Ti XC is in the first column and its overclocked results are in the second column compared with the overclocked ASUS GTX 1660 SUPER Dual EVO OC in the third column, followed by the SUPER’s factory-clocked results in the fourth column. Performance “wins” between the overclocked cards are given in yellow text and in gold text for the factory clocked results. Ties are expressed by having both sets of results given in colored text.

We see that both cards each gain performance from overclocking and at first glance the GTX 1660 Ti wins most of them. However, looking more closely, most of the results are usually up to a couple of percentage points apart and there is almost no noticeable practical performance difference between the cards playing most games.

Let’s check out our conclusion.

Conclusion

This has been quite an interesting exploration for us in evaluating the factory and manually overclocked ASUS GTX 1660 SUPER versus the factory-clocked and manually overclocked EVGA GTX 1660 Ti XC. The GTX 1660 SUPER overall is very slightly slower than the GTX 1660 Ti at factory settings, and it even manages to close the performance gap a little bit when each card is overclocked manually to its limits.

A gamer looking for the best value in the $230-$280 range should probably pick a GTX 1660 SUPER over a Ti. Both cards overclock equally well on their cores and memory from our samples as they remain in an almost identical performance class. NVIDIA has almost made the GTX 1660 Ti redundant by releasing the less expensive GTX 1660 SUPER. It’s very good news for gamers but a bit confusing without doing some research. To answer our original question regarding value between the two cards, the GTX 1660 Ti is generally not worth its current price premium over the GTX 1660 SUPER.

We are going to next look at these same two cards by measuring their VR performance using a Vive Pro with FCAT-VR early next week. After that, we will explore VR performance headroom using a RTX 2080 Super and a RTX 2080 Ti.

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

Happy Gaming!

The GTX 1660 is the sixth GeForce GPU based on NVIDIA’s Turing architecture. Just as with the GTX 1660 Ti, there is no Founders Edition (FE) so it is represented in this review by the factory overclocked EVGA GTX 1660 XC. The Turing 1660s are successors to the Pascal GTX 1060 and they are not equipped with RT nor Tensor cores which allows the GTX 1660 to launch at $229. It is aimed directly at the RX 590. Since the RTX 1660 Ti’s launch, RX 590s can now be found below $230 with a 3-game bundle.

The GTX 1660 is NVIDIA’s very latest mainstream non-RTX Turing card and it will be available globally today starting at its $229 pricing and up for factory-overclocked cards. The mildly overclocked EVGA GTX 1660 XC is priced at $229 with a $10 mail-in rebate plus a Grip game and EVGA skin bundle.

Like the GTX 1060, the GTX 1660 and 1660 Ti each have 6GB of vRAM. In the EVGA line-up, they look almost identical.

We will highlight the GTX 1660’s differences between the GTX 1060, the RTX 2060, the GTX 1660 Ti, and afterward we will focus on its performance. We will benchmark it versus the $259 Red Devil RX 590 with an expanded 41-game benching suite which now includes Devil May Cry 5 and Metro Exodus to see how capable it is at 1920×1080 and at 2560×1440. We also want to see how the GTX 1660 performs compared with a Galaxy GTX 970 EXOC.

The GTX 1660 and the GTX 1660 Ti are both based on their own Turing TU106 GPU without tensor nor RT cores so they are less complex and less expensive than the RTX 2060. The GeForce GTX 1660s feature the Turing shader core which allows them to exceed Pascal performance in newer games with more complex shaders. TU116 includes support for Concurrent Floating Point and Integer Operations, a Unified Cache Architecture with larger L1 cache, and Adaptive Shading. NVIDIA claims that the GTX 1660 is 68% faster overall than the GTX 970 so we will check it out.

The TDP of GTX 1660 and Ti are each only 120 watts which makes them an easy upgrade from a GTX 970 or GTX 960 as long as the same PSU has an 8-pin power connector. The GTX 1660 Ti features 1536 CUDA Cores and a minimum GPU Boost clock of 1770 MHz as shown by NVIDIA’s chart below.

The GTX 1660 has 1,408 CUDA cores, down from the GTX 1660 Ti’s 1536. While the GTX 1660 Ti uses 6GB of GDDR6, the GTX 1660 uses slower 6GB of GDDR5 memory with a 192-bit memory bus, for a combined memory bandwidth of 192GB/sec which is down considerably from the Ti’s 288.1GB/s. Base and boost clocks are 1530MHz and 1785MHz, respectively which are just a notch up above the Ti’s clocks. Here are the GTX 1660 specifications from NVIDIA’s charts.

The EVGA GTX 1660 XC Gaming edition is factory overclocked with a 45MHz offset over the stock core and its specifications are as below. EVGA offers some of the best warranty and RMA support anywhere. A three year warranty is offered with a further extension possible upon registration.

BTR received a GTX 1660 review sample on an extended loan from EVGA on Monday, and we have put it through its paces. We test all ten of our cards with recent drivers on a clean installation of Windows 10 64-bit Home edition, using a Core i7-8700K with all six cores overclocked to 4.7 GHz by the BIOS, and 16GB of Kingston’s 3333MHz DDR4.

First, let’s unbox the EVGA GTX 1660 XC Gaming.

UNBOXING

The EVGA GTX 1660 XC Gaming

The EVGA GTX 1660 XC Gaming comes in an average sized box that advertises its features.

The features are detailed on the back of the box in multiple languages.

Here are the minimum system requirements.

The EVGA GTX 1660 is a single-fan compact card but it is relatively thick.

Turning it around, we see the heatsink extends to the other edge.Here is the end.

The connectors consist of one DisplayPort, a HDMI connector, and a DVI connector. The new NVIDIA Type-C/VirtualLink connector is an option that EVGA decided to do without although some partner versions may offer it.

The PCB is raw and there is no backplate as befits an entry-level overclocked card.

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

Test Configuration – Hardware

• Intel Core i7-8700K (HyperThreading and Turbo boost is on to 4.7GHz 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 (2x8GB, dual channel at 3333 MHz), supplied by HyperX
• EVGA GTX 1660 XC Gaming 6GB, stock GTX 1660 Ti clocks, on loan from EVGA
• EVGA GTX 1660 Ti XC Black 6GB, at factory clocks, on loan from EVGA
• RTX 2060 6GB Founders Edition, stock RTX 2060 clocks, on loan from NVIDIA
• RTX 2070 Founders Edition 8GB, stock RTX 2070 FE clocks, on loan from NVIDIA
• GTX 1070 Ti 8GB Founders Edition, stock FE clocks, on loan from NVIDIA
• GTX 1070 8GB Founders Edition, stock FE clocks, on loan from NVIDIA
• EVGA GTX 1060 SC 6GB, factory SC clocks, on loan from EVGA
• Red Devil RX Vega 56 8GB, at factory overclocked settings, on loan from PowerColor
• Red Devil RX 590 8GB, at factory overclocked settings, on loan from PowerColor
• Gigabyte RX 480 G1 8GB at factory overclocked settings.
• 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
• Seasonic 850W Gold Focus power supply unit
• EVGA CLC 280mm CPU water cooler, supplied by EVGA
• 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

• Nvidia’s GeForce 419.35 drivers used for the GTX 1660 and for the GTX 970. The 419.17 drivers are used for the GTX 1660 Ti, and for the GTX 1070. 418.81 is used for the RTX 2060 FE, 417.71 WHQL drivers are used for the GTX 2070 FE, 418.91 for the GTX 1060, and 417.54 is used for the GTX 1070 Ti. See NVIDIA Control Panel image below.
• AMD Adrenalin Software 19.3.1 used for the RX 480 and the RX 590. The 19.2.3 drivers used for the RX Vega 56. 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.
• 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 39 games are patched to their latest versions at time of publication.
• WattMan used to set Radeon cooling and power options.
• Precision X1 used for all GeForce settings and for overclocking the GTX 1660 Ti.
• OCAT, latest version
• Fraps, latest version
• Unigine Heaven 4.0 benchmark

41 PC Game benchmark suite & 4 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
• Total Wars: Warhammer II
• Middle Earth: Shadow of War
• Destiny 2
• Star Wars: Battlefront II
• Monster Hunter: World
• Kingdom Come: Deliverance
• Final Fantasy XV
• Far Cry 5
• The Crew 2
• 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
• Rise of the Tomb Raider
• Deus Ex Mankind Divided
• Gears of War 4
• Civilization VI
• Sniper Elite 4
• Forza 7
• Shadow of the Tomb Raider
• Battlefield V
• Metro Exodus

Vulkan Games

• DOOM
• Wolfenstein: The New Colossus
• Strange Brigade

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 use Wattman to set the Radeons’ power, temperature and fan settings to their maximums.

NVIDIA Control Panel settings

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

We used the latest beta of Precision X1 to set all GeForces’ highest Power and Temperature targets and for our overclock of the GTX 1660.

By setting the Power Limits and Temperature limits to maximum for each card, they do not throttle, but they can each reach and maintain their individual maximum clocks. This is particularly beneficial for high power cards.

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

Overclocking, temperatures & noise

The XC Edition of the EVGA GTX 1660 is a low-power and quiet card even when overclocked. We could not hear the card over the many fans of our PC even when it ramps up, unlike with the Red Devil RX 590. We will spend much more time manually overclocking in our follow-up overclocking showdown.

Here is Heaven 4.0 running at stock clocks which allowed the Boost to average around 1920MHz. We notice that the card does not benefit – nor does it lose performance – by setting the Temperature target to Maximum as the Power Target is already maxed-out in Precision X1.

We ran Precision X1’s scan several times which indicated that +109MHz to 115MHz could be safely added to the core clocks. Temperatures mostly remained in the 60s C under this overclock.

With the preliminary overclock offset set at 109MHz to the core, the clocks mostly boosted above 1965MHz. We think that we can do better with a manual overclock and that the Precision X1 results are just a starting point. We did not try a GDDR5 memory overclock but will save it for the upcoming overclocking showdown versus the RX 590.

Let’s check the EVGA GTX 1660 XC’s performance compared with nine other cards using 41 games and then head for our conclusion.

This overclocking showdown is the follow-up to BTR’s launch review of the EVGA GTX 1660 Ti XC Black versus the PowerColor Red Devils RX 590 and RX Vega 56, and versus the GTX 1070. Today we have optimized and maxed-out all individual overclocks with all performance options set to their highest limits to get the most performance from each card.

At stock, the GTX 1660 Ti won nearly all of the 40 games we tested over the Red Devil RX 590, and it also beat the GTX 1070, but it fell short of the premium Red Devil RX Vega 56’s performance. This time, we will overclock all four cards manually each as far as they will go to see where they stand in relation to each other and to see if anything changes.

The GTX 1660 Ti OC

We found our own final stable manual overclock was much higher than Precision X1’s +114MHz scan recommendation. We added 175MHz to the core to boost the clocks close to 2050MHz. We also added 800MHz to the GDDR6 memory overclock, and although it could probably go higher, we found that our core overclock began to destabilize possibly due to the comparatively limited power delivery capabilities of an EVGA entry-level Black GTX 1660 Ti.

Our core overclock of the GTX 1660 Ti seems to be in line with what is expected from Turing GPUs.

The Red Devil RX Vega 56 OC

The premium PowerColor Red Devil RX Vega 56 uses the same two 8-pin PCIe power connectors as the reference RX Vega 56. However, we saw the original reference RX Vega 56 throttle its clocks regularly under load as its power delivery was apparently insufficient for overclocking vs. the GTX 1070 Ti. In contrast, the Red Devil RX Vega 56 features a 12-phase power delivery system plus a huge heatsink and three fans. The Red Devil’s clocks flatline at maximum boost unlike the reference RX Vega 56 which tends to throttle under load.

For our own gaming, we undervolt our RX Vega 56 and take a slight performance hit for big power savings, but for overclocking maximum-performance-damn-the-power-consumption, it requires brute force. We set the power limit and voltages up as high as they can go and set the clocks right to the edge of instability while using a somewhat loud and aggressive fan profile.

Our PowerColor Red Devil RX Vega 56 uses Hynix memory which doesn’t overclock particularly well in this case so we found an offset of +150MHz is its ceiling for stable performance paired with a stock-clocked core. However, for overall rock solid stability and overall maximum performance, we settled on a 4.0% overclock for an average boost above 1625MHz with its memory clocks overclocked +135MHz to 935MHz.

The GTX 1070 OC

Overclocking the Founders Edition of the GTX 1070

We devoted a separate evaluation to overclocking the GTX 1080 Founders Edition which also applies to the GTX 1070. Our Founders Edition’s final stable offset was +140 MHz to the core which settled in around 2012MHz with GPU Boost, and we added 520MHz to achieve a 4519MHz final stable memory clock.

We did not need to adjust the fan profile, but left it on automatic. The fan never becomes obtrusive as we left its profile at stock, and the GPU remained relatively cool and did not throttle under overclocking.

The Red Devil RX 590 OC

The $279 Red Devil of the RX 590 8GB is clocked up from the reference 1545MHz to its maximum boost speeds of 1576MHz. The details of our original overclocking may be found here. We found that as long as the the Power and Temperature limits are maximized, it will not throttle even with the Silent BIOS profile.

We settled on a 2.5% overclock or +40MHz to the core for a 1615MHz boost, with memory clocks overclocked +175MHz to 2175MHz. We found that higher memory clocks gained significant performance over a slightly higher core overclock. Adjusting the voltage – undervolting or undervolting – made no practical difference, and we achieved stability in all of our 40 tested games again using overclocking brute force.

Testing Platform

We test 40 games and 3 synthetic benchmarks at 1920×1080 and at 2560×1440. Our platform is a recent installation of Windows 10 64-bit Home Edition, and we are using an i7-8700K which turbos all 6 cores to 4.7GHz, an EVGA Z370 FTW motherboard, and 16GB of HyperX DDR4 3333MHz. The games, settings, and hardware are identical except for the cards being compared.

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

Test Configuration – Hardware

• Intel Core i7-8700K (HyperThreading and Turbo boost is on to 4.7GHz 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 (2x8GB, dual channel at 3333 MHz), supplied by HyperX
• EVGA GTX 1660 Ti XC Black 6GB, stock GTX 1660 Ti clocks, on loan from EVGA
• GTX 1070 8GB Founders Edition, stock FE clocks, on loan from NVIDIA
• Red Devil RX Vega 56 8GB, at factory overclocked settings, on loan from PowerColor
• Red Devil RX 590 8GB, at 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
• Seasonic 850W Gold Focus power supply unit
• EVGA CLC 280mm CPU water cooler, supplied by EVGA
• 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

• Nvidia’s GeForce 418.91 press drivers and 419.17 which have the same functionality and performance. See NVIDIA Control Panel image below.
• AMD Adrenalin Software 19.2.3 drivers. 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.
• 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.
• WattMan used to set Radeon cooling and power options.
• Precision X1 used for all GeForce settings and for overclocking.
• OCAT, latest version
• Fraps, latest version
• Unigine Heaven 4.0 benchmark

40 PC Game benchmark suite & 4 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
• Total Wars: Warhammer II
• Middle Earth: Shadow of War
• Destiny 2
• Star Wars: Battlefront II
• Monster Hunter: World
• Kingdom Come: Deliverance
• Final Fantasy XV
• Far Cry 5
• The Crew 2
• Assassin’s Creed: Odyssey
• Call of Duty: Black Ops 4
• Hitman 2
• Just Cause 4
• 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
• Shadow of the Tomb Raider
• Battlefield V
• Metro Exodus

Vulkan Games

• DOOM
• Wolfenstein: The New Colossus
• Strange Brigade

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 use Wattman to set the Radeons’ power, temperature and fan settings to their maximums.

NVIDIA Control Panel settings

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

We used the latest beta of Precision X1 to set all GeForces’ highest Power and Temperature targets and for our overclocks.

By setting the Power Limits and Temperature limits to maximum for each card, they do not throttle, but they can each reach and maintain their individual maximum clocks. This is particularly beneficial for high power cards.

Let’s check out the performance of our 4 target cards to conclude how they stand each manually overclocked in relation to each other.

Performance summary charts

Here are the performance results of 40 games and 3 synthetic tests comparing the stock and overclocked GTX 1660 Ti XC Black versus the Red Devils RX Vega 56 and RX 590 and versus the GTX 1070 Founders Edition. The highest settings are always chosen and the settings are listed on the charts. The benches were run at 1920×1080 and at 2560×1440.

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. A few games benched with OCAT show average framerates but the .1 minimums are expressed by frametimes in ms where lower numbers are better. All overclocked numbers are represented by yellow text while stock values are in white text.

The first four columns are devoted to the Red Devil RX 590 versus the GTX 1660 Ti. The next two columns are devoted to the GTX 1070 and the final two columns to the Red Devil RX Vega 56. “OC” refers to the overclocked cards’ performance and the overclocked results are shown by yellow text. As always, open each chart in a separate tab for the best viewing.

We see three of our cards each gain decent performance from manual overclocking while the Red Devil RX 590 gains a bit less as it doesn’t appear to have a lot of headroom. The Red Devil is a premium RX Vega 56 that is in a higher performance and price range than the other three cards and its performance should not be considered representative of a reference RX Vega 56, stock or overclocked.

The stock or overclocked GTX 1660 Ti Black wins more benchmarks than it loses against the GTX 1070 FE and it beats the Red Devil RX 590 in 39 of the 40 games we benchmarked. Nothing has changed significantly from overclocking although a few of the individual results have shifted their positions due to overclocking variability.

Let’s check out our conclusion.

Conclusion

This has been a fun and interesting overclocking exploration evaluating the manually overclocked EVGA GTX 1660 Ti XC Black versus the overclocked Red Devils RX 590 and RX Vega 56 and versus the overclocked GTX 1070 FE. All four cards appear to scale well with each of their respective overclocks although their overall ranking has not changed from the stock values reported in our original GTX 1660 Ti launch review.

The EVGA GTX 1660 Ti XC Ti Black is the quietest of the four cards. It is quieter than the GTX 1070 FE and significantly quieter than either Red Devil RX Vega 56 or RX 590 especially when all of the cards are overclocked and under full load. The Radeons also tend to use a lot of power when overclocked especially if undervolting isn’t also used.

The overall pricing hasn’t changed from ten days ago when the GTX 1660 Ti launched. The EVGA GTX 1660 Ti Black sits at $279 with the entry-level cards although it distinguishes itself by bundling a racing game, Grip. The Red Devil RX 590 is at $259 generally with a 3-game bundle, the GTX 1070 is still priced pretty high at $359, and the aftermarket RX Vega 56 cards are still in the $349 and up range price in the USA market. There are a lot of good choices for mainstream gamers and now may be an excellent time to upgrade from an older generation card.

Later this week, we will continue our Driver Performance Analyses series with a new post as a GeForce driver has dropped this morning. Let us know in the comments below which two GeForce cards you would like to see benchmarked.

We will also continue our VR performance benching series next week.

Happy Gaming!

The GeForce GTX 1660 Ti is the fifth GPU based on NVIDIA’s Turing architecture. There is no Founders Edition (FE) so it is represented by BTR’s review of the EVGA GTX 1660 Ti XC Black. It is the true successor to the GTX 1060 and it does not come with RT or Tensor cores which allows it to launch at $279, and its performance is supposed to approximate a $329 GTX 1070. It is $20 more expensive than the $259 Red Devil RX 590 which comes with a 3-game bundle from AMD, and so it is expected to outperform it.

The GTX 1660 Ti is NVIDIA’s new mainstream non-RTX Turing card and it will be available globally today starting at its $279 pricing and up for factory-overclocked cards. Like the GTX 1060 it replaces, the GTX 1660 Ti has 6GB of vRAM, but it is now equipped with GDDR6.

We will highlight the differences between the GTX 1060, the RTX 2060, and the GTX 1660 Ti, and then we will focus on its performance by benchmarking it versus the Red Devil RX 590 and the GTX 1070 with an expanded 40-game benching suite including with Resident Evil 2 and Metro Exodus to see how capable it is.

We will pay particular attention to the GTX 1660 Ti’s performance versus its main and lower-priced competitor, the factory overclocked PowerColor Red Devil RX 590 at $259, and also versus the GTX 1070 at $329. We will also compare performance by using a total of eight video cards in BTR’s Big Picture, and will compare how far the GTX 1660 Ti has progressed versus the GTX 1060 SC at 1920×1080 and at 2560×1440.

The GTX 1660 Ti is based on its own Turing TU106 GPU but it has been modified to not have tensor nor RT cores so it is less complex and less expensive than the RTX 2060.

The GeForce GTX 1660 Ti sports the Turing shader core which allows it to excel over Pascal in newer games with more complex shaders.

NVIDIA claims the GTX 1660 Ti is 1.5 times faster than the GTX 1060 and 3 times faster than the GTX 960. Here is a comparison of the GTX 1060 vs. the GTX 1660 Ti.

The GTX 1660 Ti features 1536 CUDA Cores and a minimum GPU Boost clock of 1770 MHz while the TDP is a modest 120 Watts and external power is provided by a single 8-pin power connector. Here are the GTX 1660 Ti specifications from NVIDIA’s chart.

The Black line-up is EVGA’s entry-level, non RGB video cards and they are stock-clocked with their specifications as below. EVGA offers some of the best warranty and RMA support anywhere. A three year warranty is offered with a further extension possible upon registration.

BTR received a GTX 1660 Ti review sample on an extended loan from EVGA last week, and we have put it through its paces. We test all of our eight cards with recent drivers on a clean installation of Windows 10 64-bit Home edition, using a Core i7-8700K with all six cores overclocked to 4.7 GHz by the BIOS, and 16GB of Kingston’s 3333MHz DDR4.

First, let’s unbox the EVGA GTX 1660 Ti XC Black.

UNBOXING

The EVGA GTX 1660 Ti XC Black

The EVGA GTX 1660 Ti XC Black comes in an average sized box that advertises its features.

The features are detailed on the back of the box in multiple languages.

Here are the GTX 1660 Ti minimum system requirements.

The EVGA GTX 1660 Ti is a compact card similar in outline to the EVGA GTX 1060 SC.

However, it is much thicker.

Turning it around, we see the heatsink extends to the other edge.

The other side.

Here is the end.

The connectors consist of one DisplayPort, a HDMI connector, and a DVI connector. The new NVIDIA Type-C/VirtualLink connector is an option that EVGA decided to do without although some partner versions may offer it.

The PCB is raw and there is no backplate as befits an entry-level card.

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

Test Configuration – Hardware

• Intel Core i7-8700K (HyperThreading and Turbo boost is on to 4.7GHz 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 (2x8GB, dual channel at 3333 MHz), supplied by HyperX
• EVGA GTX 1660 Ti XC Black 6GB, stock GTX 1660 Ti clocks, on loan from EVGA
• RTX 2060 6GB Founders Edition, stock RTX 2060 clocks, on loan from NVIDIA
• RTX 2070 Founders Edition 8GB, stock RTX 2070 FE clocks, on loan from NVIDIA
• GTX 1070 Ti 8GB Founders Edition, stock FE clocks, on loan from NVIDIA
• GTX 1070 8GB Founders Edition, stock FE clocks, on loan from NVIDIA
• EVGA GTX 1060 SC 6GB, factory SC clocks, on loan from EVGA
• Red Devil RX Vega 56 8GB, at factory overclocked settings, on loan from PowerColor
• Red Devil RX 590 8GB, at 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
• Seasonic 850W Gold Focus power supply unit
• EVGA CLC 280mm CPU water cooler, supplied by EVGA
• 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

• Nvidia’s GeForce 418.91 drivers are used for the GTX 1660 Ti, the GTX 1070, and for the GTX 1060. 418.81 is used for the RTX 2060 FE, 417.71 WHQL drivers are used for the GTX 207 FE, and 417.54 is used for the GTX 1070 Ti. See NVIDIA Control Panel image below.
• AMD Adrenalin Software 19.2.2 drivers were used for the RX 590, and 18.12.3 for the RX Vega 56. 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.
• 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 39 games are patched to their latest versions at time of publication.
• WattMan used to set Radeon cooling and power options.
• Precision X1 used for all GeForce settings and for overclocking the GTX 1660 Ti.
• OCAT, latest version
• Fraps, latest version
• Unigine Heaven 4.0 benchmark

40 PC Game benchmark suite & 4 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
• Total Wars: Warhammer II
• Middle Earth: Shadow of War
• Destiny 2
• Star Wars: Battlefront II
• Monster Hunter: World
• Kingdom Come: Deliverance
• Final Fantasy XV
• Far Cry 5
• The Crew 2
• Assassin’s Creed: Odyssey
• Call of Duty: Black Ops 4
• Hitman 2
• Just Cause 4
• 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
• Shadow of the Tomb Raider
• Battlefield V
• Metro Exodus

Vulkan Games

• DOOM
• Wolfenstein: The New Colossus
• Strange Brigade

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 use Wattman to set the Radeons’ power, temperature and fan settings to their maximums.

NVIDIA Control Panel settings

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

We used the latest beta of Precision X1 to set all GeForces’ highest Power and Temperature targets and for our overclock of the GTX 1660 Ti.

By setting the Power Limits and Temperature limits to maximum for each card, they do not throttle, but they can each reach and maintain their individual maximum clocks. This is particularly beneficial for high power cards.

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

Overclocking, temperatures & noise

The Black Edition of the EVGA GTX 1660 Ti is a quiet card even when overclocked. We could not hear the card over the many fans of our PC even when it ramps up, unlike with the Red Devil RX 590. We will spend more time manually overclocking in our follow-up overclocking showdown versus the overclocked RX 590 and versus the overclocked GTX 1070.

Here is Heaven 4.0 running at stock clocks which allowed the Boost to average around 1875MHz.

We ran Precision X1’s scan which told us that +114MHz could be added to the core clocks.

We set our own manual overclock according to Precision X1’s recommendation adding 114MHz to the core to boost the clocks mostly above 1980MHz with peak Boost around 1995MHz. We did not try a GDDR6 memory overclock but will save it for the upcoming overclocking showdown.

This is only a preliminary overclock picked by the Precision X1 automatic scan and it seems a bit low. Our GTX 1060 Ti’s manual maxed-out overclocked performance will be explored later this weekend when we pit it against the Red Devil RX 590 and against the GTX 1070 FE in an overclocking showdown.

Let’s check the GTX 1660 Ti’s performance compared with seven other cards using 40 games and then head for our conclusion.

EVGA released the high-end $249 Nu Audio PCIe sound card aimed at PC gamers who are also stereo music aficionados with better than average speakers and headphones. Critiquing audio hardware is a difficult yet important task and a challenge for any reviewer. Properly describing how music “sounds” on audio equipment is not about giving bitrate, synthetic tests, specifications, latency pings, or other measurements. High-quality audio components are created to reproduce music as accurately as possible in a way that EVGA calls “lifelike audio”.

Primary Comparison Method

Unboxing, Specs, & Software

There is some exciting news for audiophiles and gamers from our friends at EVGA which is timed with CES 2019. Here is their press release.

January 8^th, 2019 – For nearly 20 years, EVGA has built the most powerful graphics cards to play your games at the highest settings, powerful motherboards to run your system at optimal settings, and reliably efficient power supplies to power your system. Now, EVGA extends its enthusiast tradition by partnering with Audio Note (UK), Ltd. to provide the most immersive audio and lifelike gaming experienced on a PC with EVGA NU Audio Card.

Engineered by Audio Note

Audio Note (UK), Ltd. has been in the high-end audio business for over 30 years, making a name for itself by producing a wide variety of analog and digital devices. A core component of Audio Note’s philosophy is to research, design, and build its own components – often custom-made for the specific application – without financial limitations to create the finest audio products available. With this in mind, EVGA partnered with Audio Note (UK), Ltd. to select audiophile-grade digital and analog components and carefully craft the NU Audio card.

With the EVGA NU Audio Card you can experience:

• True Audio – Hear audio as it is
• Lifelike Gaming – Improved and precise enemy detection, and increased immersion.
• Studio and Audiophile Grade – Hear all the subtle sounds, emotional dialog, and powerful explosions to keep you on the edge of your seat.
• Ambient RGB Lighting – Set the RGB lighting to match your mood, or use Audio Reactive Lighting options to let your audio control the effect.

A Completely Different Gaming and Multimedia Experience

High-quality audio is built on a simple premise: everything must be solid from start to finish. One weak link in the chain can turn a symphony into a cacophony. The NU Audio Card is designed to maximize the performance in every major component section:

• Analog – Detail. Clarity. Reality.
  • The analog section is responsible for returning audio without distortion and fine-tuning the sound to ensure it remains as faithful to the source as possible.
• Digital – From Source to Playback
  • The digital side of the EVGA NU Audio Card is responsible for processing your source audio at the preferred format or converting the audio to and from analog, without affecting the internal makeup of the source audio.
• Power – Good Clean Power
  • Powered by Specialized Audio Note (UK), Ltd. audio-grade capacitors and audio-grade resistors, your audio remains noise- and distortion-free as it passes through to your choice of speakers or headphones.

NU Audio Software

From simple volume controls with a sliding headphone amplifier, to creating quick custom EQ profiles, the Nu Audio software avoids the clutter of features that you never use. Customizable audio reactive RBG lighting lets your music choose how your card looks from moment-to-moment.

Available January 16, 2019

Learn more about the EVGA NU Audio Card.

Happy Gaming!