• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Has Radeon VII caught up with RTX 2080?

Joined
Aug 6, 2017
Messages
4,025 (5.93/day)
Location
Poland
System Name skurwiel szatan
Processor i7 5775c @4.3GHz/1.385v/EDRAM @2GHz
Motherboard Z97X Gaming 5
Cooling Noctua D15S
Memory Crucial Ballistix Tactical LP 1600 CL8 @2133 CL8
Video Card(s) MSI GTX 1080 Ti Gaming X Trio 2GHz
Storage SU900 128 (OS)/850 PRO 256+256+ 512,860 EVO 500,XPG SX950U 480 (games)/4TB of external HDDs (3+1)
Display(s) Acer XB241YU+Dell S2716DG dual monitor setup
Case Full tower
Audio Device(s) Mad Catz FREQ wireless
Power Supply Superflower Leadex Gold 850W
Mouse G403 wireless + Steelseries DeX + Roccat rest
Keyboard Razer Deathstalker
Software Windows 10
Benchmark Scores A LOT
Three months ago we found the Radeon VII to be 7% slower at 1440p and today it’s 8% slower despite a few new titles added to the mix. World War Z has been a good addition for AMD, but beyond that titles such as Rage 2, Anthem, DiRT Rally 2, The Division 2, Apex Legends, Generation Zero, and in particular Ace Combat 7: Skies Unknown, have not been so positive.



additional test suite from pcgh.de,21 games across 3 resolutions,updated May 2019

RVII 9% slower than 2080

 

HTC

Joined
Apr 1, 2008
Messages
3,597 (0.88/day)
Location
Portugal
System Name HTC's System
Processor Ryzen 5 2600X
Motherboard Asrock Taichi X370
Cooling NH-C14, with the AM4 mounting kit
Memory G.Skill Kit 16GB DDR4 F4 - 3200 C16D - 16 GTZB
Video Card(s) Sapphire Nitro+ Radeon RX 480 OC 4 GB
Storage 1 Samsung NVMe 960 EVO 250 GB + 1 3.5" Seagate IronWolf Pro 6TB 7200RPM 256MB SATA III
Display(s) LG 27UD58
Case Fractal Design Define R6 USB-C
Audio Device(s) Onboard
Power Supply Corsair TX 850M 80+ Gold
Mouse Razer Deathadder Elite
Software Ubuntu 19.04 LTS
Stupid question: the comparison's author knows that, by making any change to the game titles in the comparison VS the original benchmark essentially makes the comparison worthless, right?

If he wanted to make this comparison properly, he should have used ONLY the games that were present in the original benchmark: only then one could see the improvements to the various games through driver updates, from both camps.
 
Joined
Feb 2, 2015
Messages
2,709 (1.70/day)
Location
On The Highway To Hell \m/
Short answer: Yes.

It's even slightly faster than the RTX 2080 in some instances. Though slightly slower in others. On average though, they're probably fairly equal in performance. Until, like I said before, you turn RTX on and the Radeon VII smokes it. :pimp:
 
Joined
Aug 6, 2017
Messages
4,025 (5.93/day)
Location
Poland
System Name skurwiel szatan
Processor i7 5775c @4.3GHz/1.385v/EDRAM @2GHz
Motherboard Z97X Gaming 5
Cooling Noctua D15S
Memory Crucial Ballistix Tactical LP 1600 CL8 @2133 CL8
Video Card(s) MSI GTX 1080 Ti Gaming X Trio 2GHz
Storage SU900 128 (OS)/850 PRO 256+256+ 512,860 EVO 500,XPG SX950U 480 (games)/4TB of external HDDs (3+1)
Display(s) Acer XB241YU+Dell S2716DG dual monitor setup
Case Full tower
Audio Device(s) Mad Catz FREQ wireless
Power Supply Superflower Leadex Gold 850W
Mouse G403 wireless + Steelseries DeX + Roccat rest
Keyboard Razer Deathstalker
Software Windows 10
Benchmark Scores A LOT
Joined
Sep 27, 2012
Messages
24 (0.01/day)
Location
Croatia
On average RTX2080 is 10% faster @4K, they are not equal. In fact Radeon is only noticeably faster (10+%) in 1 title out of 38, while RTX got 18 of those.
Basically R7 is even worse at 4K, which goes against popular narrative that massive BW and large memory buffer will help it beat the RTX2080.
 
Joined
Feb 3, 2017
Messages
1,367 (1.59/day)
Processor i7-6700k
Motherboard asus z170i pro gaming
Cooling ekwb custom loop for cpu/gpu running on d5 and 480 rad
Memory 2*16gb ddr4-2400
Video Card(s) msi geforce gtx 1080 ti aero
Storage 250gb 950 pro, 2*500gb samsung 850 evo
Display(s) asus pg279q, eizo ev2736w
Case thermaltake core p5
Power Supply seasonic platinum 660
Mouse logitech g700
Keyboard corsair k60
If he wanted to make this comparison properly, he should have used ONLY the games that were present in the original benchmark: only then one could see the improvements to the various games through driver updates, from both camps.
I agree. So, final 1440p graphs from these two reviews:

And games both charts have in common with changes in percentages:
  • Prey: 0 (14)
  • Battlefield V: -2 (13 > 11)
  • Strange Brigade: -15 (10 > -5)
  • Shadow of Tomb Raider: -5 (3 > -2)
  • Resident Evil 2: -4 (1 > -3)
  • Hitman 2: -6 (-2 > -8)
  • Sniper Elite 4: -1 (-2 > -3)
  • Forza Horizon 4: +4 (-7 > -3)
  • Warhammer Vermintide II: -2 (-8 > -10)
  • Kingdom Come Deliverance: -1 (-8 > -9)
  • For Honor: 0 (-8)
  • Deus Ex: Mankind Divided: 0 (-8)
  • Quake Champions: 0 (-9)
  • Just Cause 4: -6 (-9 -15)
  • Ghost Recon Wildlands: +1 (-10 > -9)
  • Star Wars Battlefront II: 0 (-10)
  • Middle-earth Shadow of War: -3 (-10 > -13)
  • Project Cars 2: -2 (-11 > -13)
  • F1 2018: 0 (-12)
  • Wolfenstein II The New Colossus: +3 (-12 > -9)
  • Apex Legends: -8 (-12 > -20)
  • Assassin's Creed Odyssey: +7 (-13 > -6)
  • Witcher 3 Wild Hunt: +1 (-13 > -12)
  • Rainbow Six Siege: -1 (-14 > -15)
  • Total War Warhammer II: 0 (-15)
  • Monster Hunter World: 0 (-16)
  • World of Tanks HD Client: -1 (-16 > -17)
  • Fortnite: -1 (-20 > -21)
  • Warframe: +13 (-23 > -10)
If I wrote something down wrong, please correct me.

Out of these 29 games:
- RTX2080 improved its standing in 15, in 5 significantly (by 5% or more)
- Radeon VII improved its standing in 6, in 2 significantly (by 5% or more)
- Standings in remaining 8 games stayed the same.

Edit:
Average difference in old test: -7.8%
Average difference in new test: -8.8%
 
Last edited:
Joined
Apr 14, 2019
Messages
25 (0.40/day)
System Name Violet
Processor AMD Ryzen 1600x
Motherboard MSI X370 Gaming Carbon PRO
Cooling Be quiet! Dark Rock Pro 4
Memory G.Skill Flare x 16GB 3333MHZ 14-14-14-26 1T
Video Card(s) Sapphire RX Vega 56 Nitro+ 1722MHz 1140mv 945MHz mem
Storage Western Digital WD Black SN750 1TB
Display(s) Philips 349X7 3440x1440
Case NZXT H440 New Edition Window Purple
Audio Device(s) Bose Companion 50
Power Supply Corsair RM850x
Mouse Razer NAGA TRINITY
Keyboard Corsair K95 RGB
Software Windows 10 64bit
Nvidia is a (little) faster and use a (little) less power.
AMD has better drivers and better/more features and is smoother in gameplay.
 
Joined
Feb 3, 2017
Messages
1,367 (1.59/day)
Processor i7-6700k
Motherboard asus z170i pro gaming
Cooling ekwb custom loop for cpu/gpu running on d5 and 480 rad
Memory 2*16gb ddr4-2400
Video Card(s) msi geforce gtx 1080 ti aero
Storage 250gb 950 pro, 2*500gb samsung 850 evo
Display(s) asus pg279q, eizo ev2736w
Case thermaltake core p5
Power Supply seasonic platinum 660
Mouse logitech g700
Keyboard corsair k60
Nvidia is a (little) faster and use a (little) less power.
AMD has better drivers and better/more features and is smoother in gameplay.
225W vs 300W is 33%, one third.
Turing is on feature parity with Vega and has some extra bits mainly in RT and Tensor cores.
Better drivers and smoother in gameplay are subjective.
 
Joined
Jan 8, 2014
Messages
199 (0.10/day)
System Name maipc
Processor 4790k @ 4Ghz / 1.05v
Motherboard Asus vii hero
Cooling Noctua NH-D14
Memory 16GB (2x8GB) Hyperx fury 1866 / CL10
Video Card(s) MSI GTX1060 Gaming X 6GB
Storage 1TB 860EVO + 1TB 7200.12 + 2TB WD Green + 4TB WD Red
Display(s) Samsung 2243BW 22'' 1680x1050 @ 77Hz
Case Cooler Master Centurion 6
Power Supply Seasonic SSP-450 RT (80+ Gold OEM PSU)
Mouse Razer DeathAdder 3.5G
Keyboard CM Storm Quickfire Rapid-i
Software Win 7 64bit
Stupid question: the comparison's author knows that, by making any change to the game titles in the comparison VS the original benchmark essentially makes the comparison worthless, right?

If he wanted to make this comparison properly, he should have used ONLY the games that were present in the original benchmark: only then one could see the improvements to the various games through driver updates, from both camps.
Just because it's not a 100% apples to apples comparison does not mean it's worthless. You can get useful information out of it. What I got from it was that the Radeon 7 has not gotten any faster compared to the RTX 2080 over the past couple of months.

Also, testing the latest games that did not exist before makes sense because these new games will get played by Radeon 7 and RTX 2080 users. So knowing how both cards perfom in new games is useful information.
 

HTC

Joined
Apr 1, 2008
Messages
3,597 (0.88/day)
Location
Portugal
System Name HTC's System
Processor Ryzen 5 2600X
Motherboard Asrock Taichi X370
Cooling NH-C14, with the AM4 mounting kit
Memory G.Skill Kit 16GB DDR4 F4 - 3200 C16D - 16 GTZB
Video Card(s) Sapphire Nitro+ Radeon RX 480 OC 4 GB
Storage 1 Samsung NVMe 960 EVO 250 GB + 1 3.5" Seagate IronWolf Pro 6TB 7200RPM 256MB SATA III
Display(s) LG 27UD58
Case Fractal Design Define R6 USB-C
Audio Device(s) Onboard
Power Supply Corsair TX 850M 80+ Gold
Mouse Razer Deathadder Elite
Software Ubuntu 19.04 LTS
Just because it's not a 100% apples to apples comparison does not mean it's worthless. You can get useful information out of it. What I got from it was that the Radeon 7 has not gotten any faster compared to the RTX 2080 over the past couple of months.
The review itself isn't worthless: it's the comparison that's worthless because the conditions (games) were changed between them. Also, and i forgot to mention, any other changes other then drivers used (OS patches, for example) between the comparison also makes it worthless.
 
Joined
Mar 18, 2008
Messages
4,282 (1.04/day)
System Name Virtual Reality / Bioinformatics
Processor Undead CPU
Motherboard Undead TUF X99
Cooling Noctua NH-D15
Memory GSkill 128GB DDR4-3000
Video Card(s) EVGA 2080Ti
Storage Samsung 960 Pro 1TB + 860 EVO 2TB + WD Black 5TB
Display(s) Acer K272HUL, HTC Vive
Case Fractal Design R5
Audio Device(s) BOSE 2.0
Power Supply Seasonic 850watt
Mouse Logitech Master MX
Keyboard Corsair K70 Cherry MX Blue
Software Windows 10 Professional/Linux Mint
I agree. So, final 1440p graphs from these two reviews:

And games both charts have in common with changes in percentages:
  • Prey: 0 (14)
  • Battlefield V: -2 (13 > 11)
  • Strange Brigade: -15 (10 > -5)
  • Shadow of Tomb Raider: -5 (3 > -2)
  • Resident Evil 2: -4 (1 > -3)
  • Hitman 2: -6 (-2 > -8)
  • Sniper Elite 4: -1 (-2 > -3)
  • Forza Horizon 4: +4 (-7 > -3)
  • Warhammer Vermintide II: -2 (-8 > -10)
  • Kingdom Come Deliverance: -1 (-8 > -9)
  • For Honor: 0 (-8)
  • Deus Ex: Mankind Divided: 0 (-8)
  • Quake Champions: 0 (-9)
  • Just Cause 4: -6 (-9 -15)
  • Ghost Recon Wildlands: +1 (-10 > -9)
  • Star Wars Battlefront II: 0 (-10)
  • Middle-earth Shadow of War: -3 (-10 > -13)
  • Project Cars 2: -2 (-11 > -13)
  • F1 2018: 0 (-12)
  • Wolfenstein II The New Colossus: +3 (-12 > -9)
  • Apex Legends: -8 (-12 > -20)
  • Assassin's Creed Odyssey: +7 (-13 > -6)
  • Witcher 3 Wild Hunt: +1 (-13 > -12)
  • Rainbow Six Siege: -1 (-14 > -15)
  • Total War Warhammer II: 0 (-15)
  • Monster Hunter World: 0 (-16)
  • World of Tanks HD Client: -1 (-16 > -17)
  • Fortnite: -1 (-20 > -21)
  • Warframe: +13 (-23 > -10)
If I wrote something down wrong, please correct me.

Out of these 29 games:
- RTX2080 improved its standing in 15, in 5 significantly (by 5% or more)
- Radeon VII improved its standing in 6, in 2 significantly (by 5% or more)
- Standings in remaining 8 games stayed the same.

Edit:
Average difference in old test: -7.8%
Average difference in new test: -8.8%
So FineWine RTX??
 
Joined
Oct 1, 2006
Messages
4,321 (0.93/day)
Location
Hong Kong
Processor Ryzen Threadripper 1950X
Motherboard X399 Fatal1ty Professional Gaming
Cooling Custom loop water, 3x 420 Rad
Memory Team T-FORCE XCALIBUR RGB 4000 @3466 CL14
Video Card(s) Zotac RTX 2080 Ti
Storage Samsung 960 Evo 512GB 3-way Raid 0
Display(s) Samsung C27HG70
Case Thermaltake Level 20 XT
Audio Device(s) Xonar Essence STX II
Power Supply FSP Aurum PT 1200W
Software Windows 10 Pro 64-bit
So FineWine RTX??
Unlike Pascal / Maxwell, Turing supports pretty much every new API feature that Vega 20 does and some more.
So, not surprising. On DX12 / Vulkan Turing benefit just as much if not more so than GCN.
 
Last edited:
Joined
Aug 6, 2017
Messages
4,025 (5.93/day)
Location
Poland
System Name skurwiel szatan
Processor i7 5775c @4.3GHz/1.385v/EDRAM @2GHz
Motherboard Z97X Gaming 5
Cooling Noctua D15S
Memory Crucial Ballistix Tactical LP 1600 CL8 @2133 CL8
Video Card(s) MSI GTX 1080 Ti Gaming X Trio 2GHz
Storage SU900 128 (OS)/850 PRO 256+256+ 512,860 EVO 500,XPG SX950U 480 (games)/4TB of external HDDs (3+1)
Display(s) Acer XB241YU+Dell S2716DG dual monitor setup
Case Full tower
Audio Device(s) Mad Catz FREQ wireless
Power Supply Superflower Leadex Gold 850W
Mouse G403 wireless + Steelseries DeX + Roccat rest
Keyboard Razer Deathstalker
Software Windows 10
Benchmark Scores A LOT
225W vs 300W is 33%, one third.
Turing is on feature parity with Vega and has some extra bits mainly in RT and Tensor cores.
Better drivers and smoother in gameplay are subjective.
turing can do int32+fp32 simultaneously while vega can do int16+fp16

as for the "better drivers" it's always an unsubstantiated claim from people who got nothing left to say,amd are much more hit or miss with theirs but it doesn't get reported as much as any pefromance variances on nvidia,as an example look at V64 vs GTX 1080



throughout the testing,4 amd drivers out of 10 reduced the previously achieved performance increases by 5-10%.
meanwhile nvidia achieved +20% and then +10% increases,and that's on a showcase game for amd,with vulkan,async and rapid packed math.you'd think amd would pull away given 32% advantage at launch,but they're actually 7% apart now.

 
Last edited:
Joined
Feb 3, 2017
Messages
1,367 (1.59/day)
Processor i7-6700k
Motherboard asus z170i pro gaming
Cooling ekwb custom loop for cpu/gpu running on d5 and 480 rad
Memory 2*16gb ddr4-2400
Video Card(s) msi geforce gtx 1080 ti aero
Storage 250gb 950 pro, 2*500gb samsung 850 evo
Display(s) asus pg279q, eizo ev2736w
Case thermaltake core p5
Power Supply seasonic platinum 660
Mouse logitech g700
Keyboard corsair k60
So FineWine RTX??
Not really. For GPUs on a new architecture it is common to have some kinks ironed out early-ish in the life cycle. RTX2080 vs Radeon VII is a bit unique in this regard as Radeon VII is on the mature Vega architecture this time while Turing was about 6 months old at the time of first benchmark roundup.
 
Joined
Aug 6, 2017
Messages
4,025 (5.93/day)
Location
Poland
System Name skurwiel szatan
Processor i7 5775c @4.3GHz/1.385v/EDRAM @2GHz
Motherboard Z97X Gaming 5
Cooling Noctua D15S
Memory Crucial Ballistix Tactical LP 1600 CL8 @2133 CL8
Video Card(s) MSI GTX 1080 Ti Gaming X Trio 2GHz
Storage SU900 128 (OS)/850 PRO 256+256+ 512,860 EVO 500,XPG SX950U 480 (games)/4TB of external HDDs (3+1)
Display(s) Acer XB241YU+Dell S2716DG dual monitor setup
Case Full tower
Audio Device(s) Mad Catz FREQ wireless
Power Supply Superflower Leadex Gold 850W
Mouse G403 wireless + Steelseries DeX + Roccat rest
Keyboard Razer Deathstalker
Software Windows 10
Benchmark Scores A LOT
In fact I suspect turing to get incremental increases over r7 with time.It beats vega on feature level,it even gets more performance increase from async compute than vega.
Gotta find the link but I think I saw this on computerbase.de

Yup,rage 2 test,turing gets 8% with async on,vega gets 4%,pascal gets 3%.
 
Joined
Apr 30, 2012
Messages
3,250 (1.25/day)
I agree. So, final 1440p graphs from these two reviews:

And games both charts have in common with changes in percentages:
  • Prey: 0 (14)
  • Battlefield V: -2 (13 > 11)
  • Strange Brigade: -15 (10 > -5)
  • Shadow of Tomb Raider: -5 (3 > -2)
  • Resident Evil 2: -4 (1 > -3)
  • Hitman 2: -6 (-2 > -8)
  • Sniper Elite 4: -1 (-2 > -3)
  • Forza Horizon 4: +4 (-7 > -3)
  • Warhammer Vermintide II: -2 (-8 > -10)
  • Kingdom Come Deliverance: -1 (-8 > -9)
  • For Honor: 0 (-8)
  • Deus Ex: Mankind Divided: 0 (-8)
  • Quake Champions: 0 (-9)
  • Just Cause 4: -6 (-9 -15)
  • Ghost Recon Wildlands: +1 (-10 > -9)
  • Star Wars Battlefront II: 0 (-10)
  • Middle-earth Shadow of War: -3 (-10 > -13)
  • Project Cars 2: -2 (-11 > -13)
  • F1 2018: 0 (-12)
  • Wolfenstein II The New Colossus: +3 (-12 > -9)
  • Apex Legends: -8 (-12 > -20)
  • Assassin's Creed Odyssey: +7 (-13 > -6)
  • Witcher 3 Wild Hunt: +1 (-13 > -12)
  • Rainbow Six Siege: -1 (-14 > -15)
  • Total War Warhammer II: 0 (-15)
  • Monster Hunter World: 0 (-16)
  • World of Tanks HD Client: -1 (-16 > -17)
  • Fortnite: -1 (-20 > -21)
  • Warframe: +13 (-23 > -10)
If I wrote something down wrong, please correct me.

Out of these 29 games:
- RTX2080 improved its standing in 15, in 5 significantly (by 5% or more)
- Radeon VII improved its standing in 6, in 2 significantly (by 5% or more)
- Standings in remaining 8 games stayed the same.

Edit:
Average difference in old test: -7.8%
Average difference in new test: -8.8%
Correct me if im wrong but on the first test he used a Reference "OC" RTX 2080 the second test hes using a Gigabyte Auros Xtreme - Difference is 90mhz and capability to sustain the higher clock due to cooling.

Hes using the same Radeon VII but different 2080s
 
Joined
Aug 6, 2017
Messages
4,025 (5.93/day)
Location
Poland
System Name skurwiel szatan
Processor i7 5775c @4.3GHz/1.385v/EDRAM @2GHz
Motherboard Z97X Gaming 5
Cooling Noctua D15S
Memory Crucial Ballistix Tactical LP 1600 CL8 @2133 CL8
Video Card(s) MSI GTX 1080 Ti Gaming X Trio 2GHz
Storage SU900 128 (OS)/850 PRO 256+256+ 512,860 EVO 500,XPG SX950U 480 (games)/4TB of external HDDs (3+1)
Display(s) Acer XB241YU+Dell S2716DG dual monitor setup
Case Full tower
Audio Device(s) Mad Catz FREQ wireless
Power Supply Superflower Leadex Gold 850W
Mouse G403 wireless + Steelseries DeX + Roccat rest
Keyboard Razer Deathstalker
Software Windows 10
Benchmark Scores A LOT
Correct me if im wrong but on the first test he used a Reference "OC" RTX 2080 the second test hes using a Gigabyte Auros Xtreme - Difference is 90mhz and capability to sustain the higher clock due to cooling.

Hes using the same Radeon VII but different 2080s
Look at londiste's number breakdown
 
Joined
Feb 3, 2017
Messages
1,367 (1.59/day)
Processor i7-6700k
Motherboard asus z170i pro gaming
Cooling ekwb custom loop for cpu/gpu running on d5 and 480 rad
Memory 2*16gb ddr4-2400
Video Card(s) msi geforce gtx 1080 ti aero
Storage 250gb 950 pro, 2*500gb samsung 850 evo
Display(s) asus pg279q, eizo ev2736w
Case thermaltake core p5
Power Supply seasonic platinum 660
Mouse logitech g700
Keyboard corsair k60
Arent those numbers under the assumption its the same 2080 being used on the two test ?
On Feb 11: He test with RTX 2080 Founders Edition
On May 22: He test with RTX 2080 Auros Xtreme
Good catch.
Typical HardwareUnboxed. I am surprised they even mention Aorus being used in the newer article, I have given up on trying to find what cards they use. No idea what they used in the older one though.
This makes comparison much more meaningless than before.

Edit:
From reviews, that Aorus Extreme does not realise its potential for higher clocks, it should be comfortably within 1-2% of the FE. It is not the point though, it should be methodologically verified if there is a difference and how large it is.
 
Last edited:
Joined
Apr 12, 2013
Messages
938 (0.42/day)
Location
Europe
Processor Ryzen 1200
Motherboard X370 Prime A
Cooling Wraith Prism
Memory 16gb @ 3000
Video Card(s) RX580 8gb Nitro+
Case Kolink Aviator
Power Supply Seasonic 620w
Software Win 10 x64
So Radeon VII is 8,8% slower then RTX2080 while power draw difference is 215W vs 300W, a quick check on the lowest price on newegg, Radeon VII is 680$ and RTX2080 is 730$
 
Joined
Aug 23, 2018
Messages
59 (0.20/day)
System Name Primary Systems
Processor AMD Ryzen 7 2700X @ PBO | Intel Core i7-7700K @ 5.0+ GHz
Motherboard Asus ROG Strix B450-I Gaming | Asus ROG Maximus IX Apex
Cooling Noctua NH-U9S (Push-Pull Fans) | Noctua NH-D15
Memory Team Dark Pro DDR4-3200 CL14 @ 3600 CL14 | G.Skill Trident Z DDR4-3600 CL15 @ 4266 CL17
Video Card(s) Sapphire Nitro+ Radeon RX Vega 64 @ -200mV, 1050 HBM | EVGA GeForce GTX 1070 FTW Gaming @ +120 MHz
Storage Samsung 950 Pro (512GB) /Samsung 850 Evo (2TB) | Samsung 950 Pro (512GB)/Samsung 850 Evo (1TB)
Display(s) Samsung QLED UHD (4K) 120Hz FreeSync TV | Various Monitors
Case None - Cases are inconvenient for frequent parts swapping + a lot of systems = a lot of space
Audio Device(s) HDMI and/or DisplayPort and/or On-board Audio
Power Supply Seasonic Focus+ Gold 1000 | Corsair HX850
Mouse Cheap Mice
Keyboard Cheap Keyboards
Software Windows 10 Pro and/or Various Linux Distros
Benchmark Scores Better than stock
Steve usually does a good job benchmarking hardware and drawing fair comparisons, however, I take issue with this particular test.

He used an RTX 2080 with one of the highest factory overclocks and a current price that is $110 (USD), about 16%, more than current prices for Radeon VII. In terms of value, that makes the Radeon VII the winner. The Gigabyte card performs 8% better, but costs 16% more, and that's factoring in a stupid $30 mail-in rebate, otherwise it's $140, about 21%, more expensive.

I really don't care which GPU comes out ahead, but I'm going to guess that a more balanced comparison with an RTX 2080 that is priced at or below its $700 (USD) suggested price and/or without such a high factory overclock will perform very similarly to the Radeon VII.
 
Last edited:
Joined
Feb 3, 2017
Messages
1,367 (1.59/day)
Processor i7-6700k
Motherboard asus z170i pro gaming
Cooling ekwb custom loop for cpu/gpu running on d5 and 480 rad
Memory 2*16gb ddr4-2400
Video Card(s) msi geforce gtx 1080 ti aero
Storage 250gb 950 pro, 2*500gb samsung 850 evo
Display(s) asus pg279q, eizo ev2736w
Case thermaltake core p5
Power Supply seasonic platinum 660
Mouse logitech g700
Keyboard corsair k60
RTX 2080 factory overclocks are a tricky thing, they may not be quite as good as you would expect. Looking at TPU reviews (sadly, that Aorus Extreme has not been through here):
NVIDIA GeForce RTX 2080 Founders Edition: 1800MHz | 1897/1905MHz
Gigabyte GeForce RTX 2080 Gaming OC: 1815MHz (1%) | 1885/1935MHz (-1%/2%)
Palit GeForce RTX 2080 Gaming Pro OC: 1815MHz (1%) | 1920/1935MHz (1%/2%)
Palit GeForce RTX 2080 Super Jetstream: 1860MHz (3%) | 1924/1920MHz (1%/1%)
ASUS GeForce RTX 2080 STRIX OC: 1860MHz (3%) | 1931/1980MHz (2%/4%)
MSI GeForce RTX 2080 Gaming X Trio: 1860MHz (3%) | 1969/1965MHz (4%/3%)
Zotac GeForce RTX 2080 AMP Extreme: 1905MHz (6%) | 1993/1995MHz (5%/5%)

What is even more important - keep in mind that these cards are primarily power-limited:
Zotac is 280W - https://www.zotac.com/hk/product/graphics_card/zotac-gaming-geforce-rtx-2080-amp-extreme#spec
MSI is 260W - https://www.msi.com/Graphics-card/GeForce-RTX-2080-GAMING-X-TRIO/Specification

Gigabyte does not have the TDP spec for Aorus Extreme and reviews do not agree on what the limit is set to. Some clearly have the limit at the FE-comparable 225W while others seem to have it around 260W.
 
Last edited:
Joined
Oct 1, 2006
Messages
4,321 (0.93/day)
Location
Hong Kong
Processor Ryzen Threadripper 1950X
Motherboard X399 Fatal1ty Professional Gaming
Cooling Custom loop water, 3x 420 Rad
Memory Team T-FORCE XCALIBUR RGB 4000 @3466 CL14
Video Card(s) Zotac RTX 2080 Ti
Storage Samsung 960 Evo 512GB 3-way Raid 0
Display(s) Samsung C27HG70
Case Thermaltake Level 20 XT
Audio Device(s) Xonar Essence STX II
Power Supply FSP Aurum PT 1200W
Software Windows 10 Pro 64-bit
A bit sloppy on Steve's side.
For science he really should have used the same cards as round one, since the point was to test software improvements between the two cards.
Even if he wants to do a more "practical" comparison (since most people do no buy the FE version), he should have used something like an MSI Ventus that usually sits around the MSRP of the 2080.
Not saying it makes enough of a difference to make the Radeon VII "win" but still...
 
Last edited:
Joined
Dec 22, 2011
Messages
2,813 (1.03/day)
System Name Zimmer Frame Rates
Processor Intel i7 920 @ Stock speeds baby
Motherboard EVGA X58 3X SLI
Cooling True 120
Memory Corsair Vengeance 12GB
Video Card(s) Palit GTX 980 Ti Super JetStream
Storage Of course
Display(s) Crossover 27Q 27" 2560x1440
Case Antec 1200
Audio Device(s) Don't be silly
Power Supply XFX 650W Core
Mouse Razer Deathadder Chroma
Keyboard Logitech UltraX
Software Windows 10
Benchmark Scores Epic
The results seem to be inline with W1zzards recent test system rebench:


At best it's good to see the 1080 Ti now has solid competition, ignoring the node shrinked elephant in the room.
 
Top