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

[Tomshardware] AMD vs Nvidia: Whose Driver Updates Improve Performance More?

Joined
Aug 6, 2017
Messages
1,697 (4.53/day)
Likes
932
Location
Poland
System Name skurwiel szatan
Processor i7 5775c
Motherboard Z97X Gaming 5
Cooling Noctua D15S
Memory Crucial Ballistix Tactical LP DDR3L 2133MHz 9-9-9-27-1T 1.512v.
Video Card(s) GTX 1080 SuperJetstream overcucked to ~2152MHz
Storage Samsung 850 Pro 256(x2)+512GB / 3TB+1TB HDDs
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
#1
Interesting

https://www.tomshardware.com/reviews/amd-nvidia-driver-updates-performance-tested,5707.html

As expected, new video drivers increase frame rates in most titles, but the differences are often subtle. AMD averages a 2.3% speed-up across our benchmark suite, while Nvidia averages 4.3%. In some games, you will see noticeable improvements, with frame rate boosts as high as 15%. But in others, the change is so modest that you might not notice it. It's unlikely that a driver update will make a game that was formerly sluggish or unplayable suddenly run well on the same hardware.

Which graphics card comes out ahead in this experiment? Whether you count the overall improvement from 2016 to 2018 or tally the total number of average frame rate wins, Nvidia's GeForce GTX 1060 6GB bags the greatest number of victories. That's why AMD followed up its Radeon RX 480 with the RX 580 less than a year later.

It's really interesting to see that in five dx12 games they tested (BF1,ROTR,Hitman,Division,AOTS) Nvidia wins in 3, one is a tie (0,3% difference) and RX480 is really only able to outperform 1060 by 4% in Hitman. Generally 1060 and 480 are the same performance regardless if you use dx11 or dx12.

AMD lost performance in three games (-3,5% avg) while nvidia lost 0,2% in Wiildlands (margin of error).
 
Last edited:
Joined
Feb 3, 2017
Messages
425 (0.76/day)
Likes
151
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
#2
Performance updates for games are game-specific optimizations in drivers. within a few percent is definitely equal. Especially as there is no clear pattern for performance improvement for a specific game. Sometimes it is done pre-release, sometimes after, sometimes long after - especially for games that change or have a long life. Most of the games they tested have gotten performance improvements both from patches as well as drivers over time.

There are games in the list that one or both manufacturers heavily target on due to high profile and/or being a very common benchmark. RotR, Hitman, AotS, GTA5, Division.
 
Last edited:

FordGT90Concept

"I go fast!1!11!1!"
Joined
Oct 13, 2008
Messages
22,317 (6.21/day)
Likes
11,396
Location
IA, USA
System Name BY-2015
Processor Intel Core i7-6700K (4 x 4.00 GHz) w/ HT and Turbo on
Motherboard MSI Z170A GAMING M7
Cooling Scythe Kotetsu
Memory 2 x Kingston HyperX DDR4-2133 8 GiB
Video Card(s) PowerColor PCS+ 390 8 GiB DVI + HDMI
Storage Crucial MX300 275 GB, Seagate 6 TB 7200 RPM
Display(s) Samsung SyncMaster T240 24" LCD (1920x1200 HDMI) + Samsung SyncMaster 906BW 19" LCD (1440x900 DVI)
Case Coolermaster HAF 932 w/ USB 3.0 5.25" bay
Audio Device(s) Realtek Onboard, Micca OriGen+
Power Supply Enermax Platimax 850w
Mouse SteelSeries Sensei RAW
Keyboard Tesoro Excalibur
Software Windows 10 Pro 64-bit
Benchmark Scores Faster than the tortoise; slower than the hare.
#3
If you read the article, AMD focused on raising minimum framerate in the drivers and they succeeded. That cost some frames per second but reduced or eliminated stutter that was present in 2016 drivers.
 
Joined
Aug 6, 2017
Messages
1,697 (4.53/day)
Likes
932
Location
Poland
System Name skurwiel szatan
Processor i7 5775c
Motherboard Z97X Gaming 5
Cooling Noctua D15S
Memory Crucial Ballistix Tactical LP DDR3L 2133MHz 9-9-9-27-1T 1.512v.
Video Card(s) GTX 1080 SuperJetstream overcucked to ~2152MHz
Storage Samsung 850 Pro 256(x2)+512GB / 3TB+1TB HDDs
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
#4
If you read the article, AMD focused on raising minimum framerate in the drivers and they succeeded. That cost some frames per second but reduced or eliminated stutter that was present in 2016 drivers.
https://www.tomshardware.com/reviews/amd-nvidia-driver-updates-performance-tested,5707-10.html

look at min fps and frametimes.

min fps went down and the latest drivers (blue line) still have frametime variance problem while nvidia's fps (both avg and min) went up and the latest drivers (blue) show fewer/smaller frametime spikes than amd.


it's especially eye-opening in case of BF1 dx12, countless times I've heard people on TPU claim that amd performs better than nvidia both in dx12 and frostbite.
 
Last edited:

FordGT90Concept

"I go fast!1!11!1!"
Joined
Oct 13, 2008
Messages
22,317 (6.21/day)
Likes
11,396
Location
IA, USA
System Name BY-2015
Processor Intel Core i7-6700K (4 x 4.00 GHz) w/ HT and Turbo on
Motherboard MSI Z170A GAMING M7
Cooling Scythe Kotetsu
Memory 2 x Kingston HyperX DDR4-2133 8 GiB
Video Card(s) PowerColor PCS+ 390 8 GiB DVI + HDMI
Storage Crucial MX300 275 GB, Seagate 6 TB 7200 RPM
Display(s) Samsung SyncMaster T240 24" LCD (1920x1200 HDMI) + Samsung SyncMaster 906BW 19" LCD (1440x900 DVI)
Case Coolermaster HAF 932 w/ USB 3.0 5.25" bay
Audio Device(s) Realtek Onboard, Micca OriGen+
Power Supply Enermax Platimax 850w
Mouse SteelSeries Sensei RAW
Keyboard Tesoro Excalibur
Software Windows 10 Pro 64-bit
Benchmark Scores Faster than the tortoise; slower than the hare.
#5

Look at the blue line. Less average FPS but much, much better 99.9 percentile = smoother.


Division doesn't surprise me how both fell. There's so many graphics issues (like water and lighting) that both AMD and NVIDIA have in that game and Ubisoft never bothered to fix. They likely tweaked broke shaders in their subsequent updates that reduced performance.
 
Last edited:
Joined
Aug 6, 2017
Messages
1,697 (4.53/day)
Likes
932
Location
Poland
System Name skurwiel szatan
Processor i7 5775c
Motherboard Z97X Gaming 5
Cooling Noctua D15S
Memory Crucial Ballistix Tactical LP DDR3L 2133MHz 9-9-9-27-1T 1.512v.
Video Card(s) GTX 1080 SuperJetstream overcucked to ~2152MHz
Storage Samsung 850 Pro 256(x2)+512GB / 3TB+1TB HDDs
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
#6

Look at the blue line. Less average FPS but much, much better 99th percentile = smoother.
1060 got an even bigger improvement in 99th (mid 20s to mid 60s) than 480 (mid 20s to mid 50s), by 10 fps.
 
Joined
Feb 3, 2017
Messages
425 (0.76/day)
Likes
151
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
#7
These improvements for min framerate and stutter are very game-dependent.
 

FordGT90Concept

"I go fast!1!11!1!"
Joined
Oct 13, 2008
Messages
22,317 (6.21/day)
Likes
11,396
Location
IA, USA
System Name BY-2015
Processor Intel Core i7-6700K (4 x 4.00 GHz) w/ HT and Turbo on
Motherboard MSI Z170A GAMING M7
Cooling Scythe Kotetsu
Memory 2 x Kingston HyperX DDR4-2133 8 GiB
Video Card(s) PowerColor PCS+ 390 8 GiB DVI + HDMI
Storage Crucial MX300 275 GB, Seagate 6 TB 7200 RPM
Display(s) Samsung SyncMaster T240 24" LCD (1920x1200 HDMI) + Samsung SyncMaster 906BW 19" LCD (1440x900 DVI)
Case Coolermaster HAF 932 w/ USB 3.0 5.25" bay
Audio Device(s) Realtek Onboard, Micca OriGen+
Power Supply Enermax Platimax 850w
Mouse SteelSeries Sensei RAW
Keyboard Tesoro Excalibur
Software Windows 10 Pro 64-bit
Benchmark Scores Faster than the tortoise; slower than the hare.
#8
Except BF1 isn't the only game AMD traded frames for 99.9 percentile. They did it in Ashes of the Singularity as well:
 
Joined
Aug 6, 2017
Messages
1,697 (4.53/day)
Likes
932
Location
Poland
System Name skurwiel szatan
Processor i7 5775c
Motherboard Z97X Gaming 5
Cooling Noctua D15S
Memory Crucial Ballistix Tactical LP DDR3L 2133MHz 9-9-9-27-1T 1.512v.
Video Card(s) GTX 1080 SuperJetstream overcucked to ~2152MHz
Storage Samsung 850 Pro 256(x2)+512GB / 3TB+1TB HDDs
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
#9
Except BF1 isn't the only game AMD traded frames for 99.9 percentile. They did it in Ashes of the Singularity as well:
There are as many (or more) cases of where 99th got worse than those of where they improved.











Looks like 17.7 were better overall for rx480. Would really like to see rx580 over time to see if it's consistent on both or nerfing the old one.


I've seen results like that before and I thought it was just an error in methodology, but now I don't know. Look at 390 and Fury, they lose performance (look at 1080p, 1440 has different settings between two tests)

July 2017

https://www.purepc.pl/karty_graficz...eforce_gtx_1060_9_gbps_msi_gaming_x?page=0,13

July 2016

https://www.purepc.pl/karty_graficz..._radeon_rx_480_test_kart_graficznych?page=0,8
 
Last edited:

FordGT90Concept

"I go fast!1!11!1!"
Joined
Oct 13, 2008
Messages
22,317 (6.21/day)
Likes
11,396
Location
IA, USA
System Name BY-2015
Processor Intel Core i7-6700K (4 x 4.00 GHz) w/ HT and Turbo on
Motherboard MSI Z170A GAMING M7
Cooling Scythe Kotetsu
Memory 2 x Kingston HyperX DDR4-2133 8 GiB
Video Card(s) PowerColor PCS+ 390 8 GiB DVI + HDMI
Storage Crucial MX300 275 GB, Seagate 6 TB 7200 RPM
Display(s) Samsung SyncMaster T240 24" LCD (1920x1200 HDMI) + Samsung SyncMaster 906BW 19" LCD (1440x900 DVI)
Case Coolermaster HAF 932 w/ USB 3.0 5.25" bay
Audio Device(s) Realtek Onboard, Micca OriGen+
Power Supply Enermax Platimax 850w
Mouse SteelSeries Sensei RAW
Keyboard Tesoro Excalibur
Software Windows 10 Pro 64-bit
Benchmark Scores Faster than the tortoise; slower than the hare.
#10
Hitman and Metro: Last Light Redux looks like margin of error. AMD nor developer did anything to change performance in those games between the tested driver versions.

Witcher 3 clearly did get driver love between 16.6.2 and 17.7.2. What happens at 99th percentile looks like margin of error to me.

As I already said, Division is a mess for everyone. AMD hasn't touched it driver wise...ever? Looking back at release notes, there were problems with crossfire and intermittent stuttering on RX 480. There was also mention of texture corruption in DX12. I can't find any references to performance changes. The newest driver could be the worst performing simply because of the fixes for the problems. It's hard to say but NVIDIA has had a similar experience with The Division. There's still graphics issues that no one is bothering to fix. Hell, Steam overlay still crashes the game in DX12.
 
Last edited:
Joined
Jun 14, 2010
Messages
364 (0.12/day)
Likes
187
Location
City 217
Processor AMD Phenom II X4 925
Motherboard Asus M4A78LT-M
Cooling Ice Hammer IH-4***
Memory 2x4GB DDR3 Corsair
Video Card(s) Sapphire HD7870 2GB
Storage 500GB SATAII Samsung | 500GB SATAII Seagate
Display(s) 23" LG 23EA63V-P
Case Thermaltake V3 Black Edition
Audio Device(s) VIA VT1708S
Power Supply Corsair TX650W
Software Windows 10 x64
#11
Why do they compare 1060 to 480 and not 580?
 
Joined
Feb 3, 2017
Messages
425 (0.76/day)
Likes
151
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
#12
Why do they compare 1060 to 480 and not 580?
Both are released at the same time, June/July 2016.
RX580 is just an overclocked RX480 anyway.
 
Joined
Aug 6, 2017
Messages
1,697 (4.53/day)
Likes
932
Location
Poland
System Name skurwiel szatan
Processor i7 5775c
Motherboard Z97X Gaming 5
Cooling Noctua D15S
Memory Crucial Ballistix Tactical LP DDR3L 2133MHz 9-9-9-27-1T 1.512v.
Video Card(s) GTX 1080 SuperJetstream overcucked to ~2152MHz
Storage Samsung 850 Pro 256(x2)+512GB / 3TB+1TB HDDs
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
#13
Why do they compare 1060 to 480 and not 580?

Which graphics card comes out ahead in this experiment? Whether you count the overall improvement from 2016 to 2018 or tally the total number of average frame rate wins, Nvidia's GeForce GTX 1060 6GB bags the greatest number of victories. That's why AMD followed up its Radeon RX 480 with the RX 580 less than a year later.
 

qubit

Overclocked quantum bit
Joined
Dec 6, 2007
Messages
15,356 (3.93/day)
Likes
8,964
Location
Quantum Well UK
System Name Quantumville™
Processor Intel Core i7-2700K at stock (hits 5 gees+ easily)
Motherboard Asus P8Z68-V PRO/GEN3
Cooling Noctua NH-D14
Memory 16GB (4 x 4GB Corsair Vengeance DDR3 PC3-12800 C9 1600MHz)
Video Card(s) Zotac GTX 1080 AMP! Extreme Edition
Storage Samsung 850 Pro 256GB | WD Green 4TB
Display(s) BenQ XL2720Z | Asus VG278HE (both 27", 144Hz, 3D Vision 2, 1080p)
Case Cooler Master HAF 922
Audio Device(s) Creative Sound Blaster X-Fi Fatal1ty PCIe
Power Supply Corsair HX 850W v1
Software Windows 10 Pro 64-bit
#14
The reliability angle seems to be at odds with what I've seen on TPU over the years. Basically, NVIDIA had less niggles.
 
Joined
Feb 3, 2017
Messages
425 (0.76/day)
Likes
151
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
#16
Wasn't 368.64 a beta driver? Press drivers usually are. I suspect they used the closest public driver.
 
Joined
Feb 3, 2017
Messages
425 (0.76/day)
Likes
151
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
#18
Barring some major bugs I would not think driver version here or there matters much.

Edit:
Actually, there are 2 WHQL drivers between 368.64 and 368.95 - 368.69 and 368.81 but neither of these officially supports GTX 1060.
368.95 is the first driver release where GTX 1060 is supported.
 
Joined
Mar 18, 2015
Messages
632 (0.51/day)
Likes
394
#20
Generally 1060 and 480 are the **same performance** regardless if you use dx11 or dx12.
Well not really .... This is only true when you ignore the the substantil overclocking ability of nVidia versus AMD. Because AMD very agressively clocks their cards *in the box*, the overclocking ability of their cards is almost always in single digits while the nVidia cards are in double digits, sometimes over 30%. There are various factors which may warrant one particular card to be a better choice for someone rather than another. While every factor may not matter to every person, each factor should be evaluated.

1. Out of the Box performance - So let's compare two cards from the same (MSI) manufacturer and model line (Gaming X). From above link:

As a result, the card is 4% faster than the RX 480 reference and 6-7% slower than the [reference] GeForce GTX 980, GTX 1060, and Radeon R9 Fury, which all have roughly the same performance at 1080p.


2. AIB Cards - From the above, we see that the MSI RX 480 is 7% faster overall in TPUs 16 game test suite. From Below, the MSI 1060 Gaming X is 3% faster than the reference 1060 ... so we can can conclude that at the time of testing the MSI 1060 was 10% faster than the MSI 480 in the 16 game test suite

3. Overclocking - We see there that the MSI 480 overclocks 8.6% and the MSI 1060 overclocks 15.1%.. So when the 1060 (10% performance advantage) is overclocked, the relative difference would be:

https://www.techpowerup.com/reviews/MSI/RX_480_Gaming_X/26.html
https://www.techpowerup.com/reviews/MSI/GTX_1060_Gaming_X/27.html

110% x (115.1 / 108.6) = 116.6% of the 480s speed or 16.6 % faster

4. Initial Driver improvements - AMDs driver improvements have improved the performance of the 480 since originally tested. As we can see from the link here, TPU tested the results from the latest driver improvements and found an increase if 2.1% at 1080 p average across 21 games:

https://www.techpowerup.com/reviews/AMD/Radeon_Crimson_ReLive_Drivers/6.html

Unfortunately, we had no info at the time on what improvements have resulted from newer nVidia drivers but suffice to say, those improvements did not erase that 10% gap outta the box (16.6% if both overclocked).

What the THG data shows is that the gap has widened, but the reality was... the 480 was never really competitive in the 1st place when both cards were overclocked.
 
Joined
Dec 21, 2005
Messages
400 (0.09/day)
Likes
90
Location
Texas, U.S. of A.
System Name Eric's Battlestation
Processor Core i7 6700k (soon to be OC)
Motherboard GIGABYTE G1 Gaming GA-Z170X-Gaming 7
Cooling Fractal Design Celsius S24
Memory Corsair Vengeance LPX 16GB DDR4 3200
Video Card(s) SAPPHIRE RX Vega 56 (64 BIOS Flash-mod)
Storage Samsung 850 EVO 1TB, Crucial MX500 1TB
Display(s) Viewsonic XG2401 24"
Case Fractal Design Define R5
Power Supply EVGA G2-XR 80 PLUS GOLD 750W
Mouse Logitech G603
Keyboard Logitech G810
Software Microsoft Windows 10 Home
#21
Doesn't seem like fine wine is the case anymore. Most games get one patch after release and that's it. Only Blockbusters like TW3 get any love. Oh if I could only get that lock solid 60fps in TW Warhammer 2...
 
Joined
Feb 14, 2012
Messages
772 (0.33/day)
Likes
516
Location
Romania
System Name No name
Processor i5 8600k@4800Mhz 1.22v
Motherboard Gigabyte Z370 UD3P
Cooling Noctua U12P
Memory 16GB 3000Mhz
Video Card(s) Evga 980Ti SC + Accelero Xtreme III
Storage Kingston AV400 256GB, WD30PURX
Display(s) Asus MG279Q
Case Regnum RG4F Frosty White
Audio Device(s) XFI Music
Power Supply Seasonic G650 650W
Mouse Cougar 600M
Keyboard Asus Strix MX Brown
Software Windows 10 Pro x64
Benchmark Scores 5948 3dmark TimeSpy
#22
Oh if I could only get that lock solid 60fps in TW Warhammer 2...
I gave up that quest on my rig, turning off MSI AB was the best choice, the game is perfectly playable at 40-50 fps.
 
Joined
Sep 17, 2014
Messages
5,475 (3.83/day)
Likes
4,430
Location
Duiven, Netherlands
Processor i7 8700k 4.8Ghz @ 1.31v
Motherboard AsRock Fatal1ty K6 Z370
Cooling beQuiet! Dark Rock Pro 3
Memory 16GB Corsair Vengeance LPX 3200/C16
Video Card(s) MSI GTX 1080 Gaming X @ 2100/5500
Storage Samsung 850 EVO 1TB + Samsung 830 256GB + Crucial BX100 250GB + Toshiba 1TB HDD
Display(s) Eizo Foris FG2421
Case Fractal Design Define C TG
Power Supply EVGA G2 750w
Mouse Logitech G502 Protheus Spectrum
Keyboard Sharkoon MK80 (Brown)
Software W10 x64
#23
Doesn't seem like fine wine is the case anymore. Most games get one patch after release and that's it. Only Blockbusters like TW3 get any love. Oh if I could only get that lock solid 60fps in TW Warhammer 2...
I barely get any drops below 60 in TW WH2... and when I do its GPU load, not CPU. What API are you using?

As for fine wine, that was just a last resort AMD fan response with no data or proof behind it. Just some twisted benchmark numbers of some specific games that ran well on AMD hardware. Fury X already proved fine wine was complete BS - that card lost so much performance and still does with every new game that gets released because it is a card with bad balance. We know by now that Polaris after the initial patch love ran well and that was somehow regarded as 'fine wine' as well... but in fact its just AMD catching up with software to their hardware release. You know, regular support like everyone does. Prior to that, with 7970 and the like, it wasn't fine wine, it was just the fact they kept rebranding the same stuff so obviously it gets more driver love.

As for the Tom's review... look at the immense careface <- We're talking about a few % across several years of driver releases. Get real. There's nothing really substantial going on. The outliers with large perf boosts were just badly optimized or supported from release, what's new here?
 
Last edited:
Joined
Aug 6, 2017
Messages
1,697 (4.53/day)
Likes
932
Location
Poland
System Name skurwiel szatan
Processor i7 5775c
Motherboard Z97X Gaming 5
Cooling Noctua D15S
Memory Crucial Ballistix Tactical LP DDR3L 2133MHz 9-9-9-27-1T 1.512v.
Video Card(s) GTX 1080 SuperJetstream overcucked to ~2152MHz
Storage Samsung 850 Pro 256(x2)+512GB / 3TB+1TB HDDs
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
#24
I barely get any drops below 60 in TW WH2... and when I do its GPU load, not CPU. What API are you using?

As for fine wine, that was just a last resort AMD fan response with no data or proof behind it. Just some twisted benchmark numbers of some specific games that ran well on AMD hardware. Fury X already proved fine wine was complete BS - that card lost so much performance and still does with every new game that gets released because it is a card with bad balance. We know by now that Polaris after the initial patch love ran well and that was somehow regarded as 'fine wine' as well... but in fact its just AMD catching up with software to their hardware release. You know, regular support like everyone does. Prior to that, with 7970 and the like, it wasn't fine wine, it was just the fact they kept rebranding the same stuff so obviously it gets more driver love.

As for the Tom's review... look at the immense careface <- We're talking about a few % across several years of driver releases. Get real. There's nothing really substantial going on. The outliers with large perf boosts were just badly optimized or supported from release, what's new here?
I expect same fate for Vega.If it can't go past gtx 1080 in doom vulkan, bf1 dx12 and hitman, it never will.I think only way to see V64 gain with age is rapid packed math, and even then it's 10% advantage like we see in FC5."Finewine" was mostly due to incremental improvements that amd made in dx11, but if you have polaris/vega cards go head to head with pascal in those that support dx12,vulkan,async and all that stuff, that's kinda giving amd cards all they've got already.
 
Joined
Dec 21, 2005
Messages
400 (0.09/day)
Likes
90
Location
Texas, U.S. of A.
System Name Eric's Battlestation
Processor Core i7 6700k (soon to be OC)
Motherboard GIGABYTE G1 Gaming GA-Z170X-Gaming 7
Cooling Fractal Design Celsius S24
Memory Corsair Vengeance LPX 16GB DDR4 3200
Video Card(s) SAPPHIRE RX Vega 56 (64 BIOS Flash-mod)
Storage Samsung 850 EVO 1TB, Crucial MX500 1TB
Display(s) Viewsonic XG2401 24"
Case Fractal Design Define R5
Power Supply EVGA G2-XR 80 PLUS GOLD 750W
Mouse Logitech G603
Keyboard Logitech G810
Software Microsoft Windows 10 Home
#25
I barely get any drops below 60 in TW WH2... and when I do its GPU load, not CPU. What API are you using?

As for fine wine, that was just a last resort AMD fan response with no data or proof behind it. Just some twisted benchmark numbers of some specific games that ran well on AMD hardware. Fury X already proved fine wine was complete BS - that card lost so much performance and still does with every new game that gets released because it is a card with bad balance. We know by now that Polaris after the initial patch love ran well and that was somehow regarded as 'fine wine' as well... but in fact its just AMD catching up with software to their hardware release. You know, regular support like everyone does. Prior to that, with 7970 and the like, it wasn't fine wine, it was just the fact they kept rebranding the same stuff so obviously it gets more driver love.

As for the Tom's review... look at the immense careface <- We're talking about a few % across several years of driver releases. Get real. There's nothing really substantial going on. The outliers with large perf boosts were just badly optimized or supported from release, what's new here?
I use DX12, since I swapped my 970 for the Vega 56. The battles actually run very smooth unless I'm zerging with the Skaven. The drops I encounter are on the campaign maps. I notice when I'm nearby the Vortex the FPS frequently dips into the 40s. Mortal Empires campaign runs similar for me regardless of where I'm present, but that doesn't bother me as much considering the scope of the map.
 
Top