Overclock.net › Forums › Industry News › Hardware News › [PCper] Frame Rating Dissected: Full Details on Capture-based Graphics Performance Testing
New Posts  All Forums:Forum Nav:

[PCper] Frame Rating Dissected: Full Details on Capture-based Graphics Performance Testing - Page 20

post #191 of 420
I'll have to watch it again at home where I can download the video, but it's only Sleeping Dogs that has obvious differences that I could see. So the question is, how much does the extra lighting source affect latency? I don't know, but I'd guess not much. I'm more interested in finding out whether those lights are missing on all Nvidia cards though, not any impact on frame rates. They affect the visual look of the game pretty significantly (irrespective of whatever they do to performance).

Edit: But I really doubt PCPer was so stupid as to have used different settings for the different cards (plus TR, and other sites using FRAPS, have found similar results), so it seems like a mnaufacturer difference, but that's pretty common. Wasn't there a similar issue recently where AMD had a driver issue that affected quality? Can't remember the game, but I vaguely remember the discussion.

I do hope PCPer or one of the other sites looks into it though.
Edited by Forceman - 3/28/13 at 5:12pm
post #192 of 420
Quote:
Originally Posted by zGunBLADEz View Post

How you can show the smoothness when one is showing more stuff on screen than the other which is my point...

I mean if i turn shadows from very high to high on my rig difference in performance is night and day..
Do i notice a visual difference? is very hard to tell if you are not really looking for it..

biggrin.gif

Im saying amd/nvidia dont have problems with multiple gpus setups? no...
It's not a question of whether NVidia is better than AMD, it's pointing out a problem that needs to be fixed. Whatever problems Nvidia has with image quality, it does not excuse the issues with runt/dropped frames in crossfire.
1
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 930 @ 3.6 Gigabyte x58a ud3r Gigabyte GV-R6870C-1GD Mushkin Redline 
Hard DriveOptical DriveMonitorKeyboard
2x 1TB Spinpoint samsung dvd burner Samsung p2370 + Mitsubishi Diamond Pro 930SB WASD keyboards v1 semi custom w/ cherry browns 
PowerCaseMouse
Antec CP-850 Antec P183 CM Storm Spawn 
  hide details  
Reply
1
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 930 @ 3.6 Gigabyte x58a ud3r Gigabyte GV-R6870C-1GD Mushkin Redline 
Hard DriveOptical DriveMonitorKeyboard
2x 1TB Spinpoint samsung dvd burner Samsung p2370 + Mitsubishi Diamond Pro 930SB WASD keyboards v1 semi custom w/ cherry browns 
PowerCaseMouse
Antec CP-850 Antec P183 CM Storm Spawn 
  hide details  
Reply
post #193 of 420
Quote:
Originally Posted by TranquilTempest View Post

It's not a question of whether NVidia is better than AMD, it's pointing out a problem that needs to be fixed. Whatever problems Nvidia has with image quality, it does not excuse the issues with runt/dropped frames in crossfire.

I agreed with you to a certain point.. Also go all open like pcper and say that crossfire is a placebo that actually is worst than single card is making the big claim...

Also nvidia suffer from this as well... So lets not throw the hot potato to amd alone XD
post #194 of 420
Another thing i like to point out..
No matter how hard i try configure my mouse wireless and my gaming mouse wired....
It feels alot smoother with the 360 gamepad than the mouse itself...
post #195 of 420
Quote:
Originally Posted by TranquilTempest View Post

It's not a question of whether NVidia is better than AMD, it's pointing out a problem that needs to be fixed. Whatever problems Nvidia has with image quality, it does not excuse the issues with runt/dropped frames in crossfire.

but if true, it does put into question just how valid any comparison is.

You are right thought, it's still an issue that AMD needs to fix and obviously we would need a lot more info before we point any fingers at Nvidia for skimping on IQ.
Bubbles
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel i5 3570k @4.7 GA-z77x-ud5h Crossfire 7950's / Heatkiller GPU-X³ 79X0 16gb Samsung 30nm  
Hard DriveHard DriveCoolingCooling
Samsung F3 OCZ Vertex 3 XSPC Raystorm EK Coolstream 360 + XSPC RX240 
OSMonitorKeyboardCase
Windows 8 Pro Crossover Q27 CM Quick Fire Rapid (Reds) NZXT Switch 810 
Mouse
Roccat Savu  
  hide details  
Reply
Bubbles
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel i5 3570k @4.7 GA-z77x-ud5h Crossfire 7950's / Heatkiller GPU-X³ 79X0 16gb Samsung 30nm  
Hard DriveHard DriveCoolingCooling
Samsung F3 OCZ Vertex 3 XSPC Raystorm EK Coolstream 360 + XSPC RX240 
OSMonitorKeyboardCase
Windows 8 Pro Crossover Q27 CM Quick Fire Rapid (Reds) NZXT Switch 810 
Mouse
Roccat Savu  
  hide details  
Reply
post #196 of 420
Quote:
Originally Posted by specopsFI View Post

I think PCper is a bit harsh saying adding another Radeon adds nothing to the user experience, but the point is that the traditional meter of CF scaling ("90% increase in frame rate, booyah!") is quite ignorant or in some cases self-deceptive.

There is another point worth making: the separation of benchmarking and gaming. Everyone should take note here: choosing a GPU solution for gaming can go wrong if it's based on pure benchmarking preferences. Benchmarking is all about fps numbers and this new method makes it obvious that a good gaming experience is a way more complex objective.

That happens in general, what do you think I've been saying this entire time in regards to CPU performance in gaming? There is a difference in FPS, etc but the differences comes down to be nearly unnoticeable in terms of user experience.
Quote:
Originally Posted by reqq View Post

No vishera. Some more here, seems its get better with more powerfuls GPUs.

Because games are GPU limited, so any issues with the CPU (whether it's high latency, latency going up and down quickly like that or lower performance in that scenario, etc) are masked more and more.
Quote:
Originally Posted by Kane2207 View Post

I know this is going back and forward and people have started to introduce metrics like "user experience" (never going to be measurable as it's totally subjective rolleyes.gif ) and "frame rates and frame times don't tell the whole story!" (not what I've seen people bang on and on and on about around here but hey..).

That's because those are what matters, 20fps or 120fps you're still going to be going by the experience on your own PC/a friends PC/a demo PC in the end. It's why I think that reviewers should also record a test and upload the raw video.
Quote:
Originally Posted by Kane2207 View Post

The difference between 1:1 scaling and 1:5 isn't subjective at all if all you get is 21 pixel high frames, it's measurable and quantifiable

Nice work on actually dodging answering the question there though huh?

So all of the people with CFX setups are lying/misinformed when they see that it is smoother?
Is there an issue? Yes. Does it need to be fixed? Yes.
Quote:
Originally Posted by Kane2207 View Post

This is exactly what I don't get, why are people defending it based on metric such as 'user experience' when AMD have gone on record and confirmed that there is an issue?

kookoo.gif

Because when you play a game, the experience is what you're experiencing, (Shock horror) you can go on about the metrics and maths but when it comes down to it, your brain is different to mine...30fps may be smooth to you, but maybe 20fps is to me, there's no one standard for smoothness that applies to every single person due to the nature of how we perceive it...
Quote:
Originally Posted by TranquilTempest View Post

The point isn't that there are graphical differences, Gunblade's accusation was that PCPer changed settings between runs, which I've seen no evidence of. http://www.overclock.net/t/1375702/pcper-frame-rating-dissected-full-details-on-capture-based-graphics-performance-testing/160#post_19624287

Actually, that is all that matters...Either PCPer has lowered settings on nVidia or nVidia has decided to use IQ lowering optimizations in their drivers...Not the first time for either nVidia or AMD to do that, either.
Quote:
Originally Posted by Forceman View Post

Not the first time there have been image differences between the companies, so I wouldn't jump to the Nvidia cheating conclusion anymore than the AMD cheating assumption about runt frames. I'd like to see some comparison videos from TR or Anand tech though. It's also unclear how those differences would affect latency as opposed to just performance.

This. We're way too early into it to really pull any conclusions out of this so far.
Quote:
Originally Posted by Sm0keydaBear View Post

I mean from my experience in way back in the day, Nvidia was known to kinda skimp on some graphical qualities in turn for better FPS and frame time deliveries whereas ATi was rendering full scenes with full graphical details vs. frame rate delivery. So all in all, it's like Nvidia = skimp a little on the edges for a better experience, ATi = skimp a little on frame rates for graphics quality. Again, this is my perspective from the pasts of both companies, and kind of what the graphs are saying regardless.

AMD/ATi has done the same in the past, iirc there was a big stink over some camo tent thing not being rendered back in 2006 or so.
Quote:
Originally Posted by Forceman View Post

Yes, Sleeping Dogs is obvious with the missing light sources. I wonder if that's consistent on all the cards - does the 680/7970 video show the same thing, for example, or if that's a dynamic game quirk.

There needs to be a retest and video for sure.
    
CPUMotherboardGraphicsRAM
Intel Core i5 3570k @ 4.5Ghz ASRock Z77 Pro3 Powercolor Radeon HD7950 3GB @ 1150/1350 4x4GB G.Skill Ares 2000Mhz CL9 
Hard DriveHard DriveHard DriveHard Drive
Samsung 840 250GB Western Digital Black 1TB WD1002FAEX Seagate Barracuda 3TB ST3000DM001 Samsung Spinpoint EcoGreen 2TB 
Optical DriveCoolingCoolingCooling
Pioneer DVR-220LBKS Noctua NH-D14 Scythe Gentle Typhoon 1850rpm Corsair AF140 Quiet Edition 
CoolingOSMonitorMonitor
Arcitc Cooling Acclero Twin Turbo II Arch Linux x86-64, amdgpu BenQ G2220HD BenQ G2020HD 
KeyboardPowerCaseMouse
Ducky Shine III Year of the Snake, Cherry Blue Silverstone Strider Plus 600w CoolerMaster CM690 II Black and White SteelSeries Sensei Professional 
Mouse PadAudioOther
Artisan Hien Mid Japan Black Large ASUS Xonar DX NZXT Sentry Mesh 30w Fan Controller 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Intel Core i5 3570k @ 4.5Ghz ASRock Z77 Pro3 Powercolor Radeon HD7950 3GB @ 1150/1350 4x4GB G.Skill Ares 2000Mhz CL9 
Hard DriveHard DriveHard DriveHard Drive
Samsung 840 250GB Western Digital Black 1TB WD1002FAEX Seagate Barracuda 3TB ST3000DM001 Samsung Spinpoint EcoGreen 2TB 
Optical DriveCoolingCoolingCooling
Pioneer DVR-220LBKS Noctua NH-D14 Scythe Gentle Typhoon 1850rpm Corsair AF140 Quiet Edition 
CoolingOSMonitorMonitor
Arcitc Cooling Acclero Twin Turbo II Arch Linux x86-64, amdgpu BenQ G2220HD BenQ G2020HD 
KeyboardPowerCaseMouse
Ducky Shine III Year of the Snake, Cherry Blue Silverstone Strider Plus 600w CoolerMaster CM690 II Black and White SteelSeries Sensei Professional 
Mouse PadAudioOther
Artisan Hien Mid Japan Black Large ASUS Xonar DX NZXT Sentry Mesh 30w Fan Controller 
  hide details  
Reply
post #197 of 420
Yeah, sorry but I don't at all buy into this accusation that CF 7970's don't look any better than a single card. I'm right in the middle of doing an exhaustive 7970 vs Titan comparison using both single and dual cards and I can say for certain that there is a good deal of benefit to the second 7970 in all the games I used for the test. I'm not sure what the goal here is from PCPer but their claims on that point at least are pure BS in my opinion...
post #198 of 420
I'm grateful that PcPer crew did a wonderful job creating the sophisticated charts that enable our current understanding of the situation. I most certainly trust them more than Guru3d, not to skew results in Nvidia's favour. Yet, I find it curious that they did not test Gtx680's v-sync capabilities on the Battlefield 3 1440p test. I end up assuming that they hesitated to highlight the obvious: 7970CF can play constant animation at 60 fps on 2560x1440p setting, while gtx680 would not.
The Machine
(14 items)
 
Nexus 7 2013
(11 items)
 
 
CPUMotherboardGraphicsRAM
A10 6800K Asus F2A85-V MSI 6870 Hawx, VTX3D 5770, AMD HD6950(RIP), Sap... G.skill Ripjaws PC12800 6-8-6-24 
Hard DriveOptical DriveOSMonitor
Seagate 7200.5 1TB NEC 3540 Dvd-Rom Windows 7 x32 Ultimate Samsung P2350 23" 1080p 
PowerCaseMouseAudio
Seasonic s12-600w CoolerMaster Centurion 5 Logitech G600 Auzen X-Fi Raider 
CPUMotherboardGraphicsRAM
Quad Krait 300 at 1.5Ghz Qualcomm APQ8064-1AA SOC Adreno 320 at 400mhz 2GB DDR3L-1600 
Hard DriveOSMonitorKeyboard
32GB Internal NAND Android 5.0 7" 1920X1200 103% sRGB & 572 cd/m2 LTPS IPS Microsoft Wedge Mobile Keyboard 
PowerAudio
3950mAh/15.01mAh Battery Stereo Speakers 
  hide details  
Reply
The Machine
(14 items)
 
Nexus 7 2013
(11 items)
 
 
CPUMotherboardGraphicsRAM
A10 6800K Asus F2A85-V MSI 6870 Hawx, VTX3D 5770, AMD HD6950(RIP), Sap... G.skill Ripjaws PC12800 6-8-6-24 
Hard DriveOptical DriveOSMonitor
Seagate 7200.5 1TB NEC 3540 Dvd-Rom Windows 7 x32 Ultimate Samsung P2350 23" 1080p 
PowerCaseMouseAudio
Seasonic s12-600w CoolerMaster Centurion 5 Logitech G600 Auzen X-Fi Raider 
CPUMotherboardGraphicsRAM
Quad Krait 300 at 1.5Ghz Qualcomm APQ8064-1AA SOC Adreno 320 at 400mhz 2GB DDR3L-1600 
Hard DriveOSMonitorKeyboard
32GB Internal NAND Android 5.0 7" 1920X1200 103% sRGB & 572 cd/m2 LTPS IPS Microsoft Wedge Mobile Keyboard 
PowerAudio
3950mAh/15.01mAh Battery Stereo Speakers 
  hide details  
Reply
post #199 of 420
Quote:
Originally Posted by Majin SSJ Eric View Post

Yeah, sorry but I don't at all buy into this accusation that CF 7970's don't look any better than a single card. I'm right in the middle of doing an exhaustive 7970 vs Titan comparison using both single and dual cards and I can say for certain that there is a good deal of benefit to the second 7970 in all the games I used for the test. I'm not sure what the goal here is from PCPer but their claims on that point at least are pure BS in my opinion...

I think the important thing to remember is that it is their conclusion (that the second card adds little) that is the point of debate - pretty much all the sites have identified the runt frame problem, but only PCPer makes the claim that the second card is of little benefit. TR's charts show the second card being more effective, although still not as effective as the pure FPS number would show, but they didn't draw the same conclusion as PCPer. More study and debate needs to occur on exactly what the runt frames mean for performance and user experience - I think it is too early to draw any simple conclusions like "crossfire is pointless".

Now that we have these tools, I'm hoping someone will go back and compare the results between the earlier AMD drivers and the newer ones that are said to fix the latency issue in the DX9 games and see how the data changes.

Edit: The more I think about it, the more I wonder about their conclusion. The game engine thinks the runt frames are real frames, and so it is going to continue to move the simulation along when they are produced, which means the animations will move, the interactions will occur, and the viewpoint will change with the game and mouse movement - so what you end up seeing on screen would be the same thing you'd see as if they were displayed, just maybe less fluid. So it isn't exactly like the frames are wasted, they aren't from a game engine perspective, just that they don't really convey much useful visual information since they aren't displayed for long. Of course, it could be the other way and the short frame could completely screw up the simulation clock and cause all sorts of issues, including more runt frames down the line, but I guess that's why we need more testing.

Someone needs to modify the tools so they can just capture the colored bars, instead of the whole frame, and then release them to the general public. Then maybe we could get some user testing without the expense of the dedicated capture rig. Wonder if that is feasible?
Edited by Forceman - 3/28/13 at 8:15pm
post #200 of 420
Quote:
Originally Posted by zGunBLADEz View Post

How you can show the smoothness when one is showing more stuff on screen than the other which is my point...

I mean if i turn shadows from very high to high on my rig difference in performance is night and day..
Do i notice a visual difference? is very hard to tell if you are not really looking for it..

biggrin.gif

Im saying amd/nvidia dont have problems with multiple gpus setups? no...

I downloaded the 1080p videos and watched them pretty carefully, and I think the differences are in contrast and vibrancy, not in quality. The shadows in Far Cry 3 are darker on the Nvidia card, but they aren't any softer on one or the other, they look pretty much the same. There are also serious compression issues with the video, which I think makes any comparison pretty pointless - turning makes the screen a complete blurry mess, even on the full size videos (not the ones from YouTube). The lighting problem in Sleeping Dogs is an issue, but I think the others are pretty much the same.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Hardware News
Overclock.net › Forums › Industry News › Hardware News › [PCper] Frame Rating Dissected: Full Details on Capture-based Graphics Performance Testing