Overclock.net › Forums › Industry News › Hardware News › [PCPer] Frame Rating Comparison - HD 7950 CrossFire vs GTX 660 Ti SLI (full review)
New Posts  All Forums:Forum Nav:

[PCPer] Frame Rating Comparison - HD 7950 CrossFire vs GTX 660 Ti SLI (full review) - Page 13  

post #121 of 542
Quote:
Originally Posted by Forceman View Post

So 20 FPS for 1 card and 80 FPS for 2 cards doesn't sound strange?

I was wondering what you were talking about, so I looked into it. If anything this article shows is that FRAPS is not reliable.

FRAPS


Frame Rating Results

Quote:
The frame rates for the HD 7950s in CrossFire differ greatly once again from the FRAPS reported metrics to our new Frame Rating results.

Was this what you were referring to?
2010rig
(14 items)
 
Galaxy S3
(8 items)
 
 
CPUMotherboardGraphicsRAM
X5660 @ 4.5  ASUS P6X58D-E 980TI? 12GB OCZ Platinum - 7-7-7-21 
Hard DriveCoolingOSMonitor
1 80GB SSD x25m - 3TB F3 + F4 NH-D14 Windows 7 Ultimate LG 47LH55 
KeyboardPowerCaseMouse
Natural Wireless Keyboard Corsair 750HX CM 690 II Advanced MX 518 
CPUGraphicsRAMHard Drive
Snapdragon S4 Dual core 1500mhz Adreno 225 Samsung 2GB 16GB Onboard Flash 
OSMonitorPowerCase
Android 4.4.2 - CM11 4.8" AMOLED 1280x720 2100 mAh battery Otterbox Defender 
  hide details  
2010rig
(14 items)
 
Galaxy S3
(8 items)
 
 
CPUMotherboardGraphicsRAM
X5660 @ 4.5  ASUS P6X58D-E 980TI? 12GB OCZ Platinum - 7-7-7-21 
Hard DriveCoolingOSMonitor
1 80GB SSD x25m - 3TB F3 + F4 NH-D14 Windows 7 Ultimate LG 47LH55 
KeyboardPowerCaseMouse
Natural Wireless Keyboard Corsair 750HX CM 690 II Advanced MX 518 
CPUGraphicsRAMHard Drive
Snapdragon S4 Dual core 1500mhz Adreno 225 Samsung 2GB 16GB Onboard Flash 
OSMonitorPowerCase
Android 4.4.2 - CM11 4.8" AMOLED 1280x720 2100 mAh battery Otterbox Defender 
  hide details  
post #122 of 542
PCPer should do one test with tri-SLI. I remember reading a long time ago, an article of [H], that talked about AMD 6900-series CF microstutter, and how the frame rates became much smoother after adding in the third card. Or maybe it was Anandtech. Can't remember, but I'll see if I can find it anyway...

Edit : I was wrong. Was neither. Was Tom's Hardware...

http://www.tomshardware.com/reviews/radeon-geforce-stutter-crossfire,2995-6.html

Obviously not to FCAT standards, but it shows that adding a third card could really smooth things out.
Edited by Creator - 4/2/13 at 2:02pm
CUDA Workstation
(12 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 5960X Asus X99-E WS GTX Titan Tri-SLI Crucial Ballistix Sport 
Hard DriveHard DriveCoolingOS
Samsung XP941 Sandisk Ultra II 960GB Water Windows 8.1 
MonitorPowerCaseAudio
Asus ROG Swift EVGA Supernova 1300 G2 NZXT Switch 810 Creative SB ZXR 
  hide details  
CUDA Workstation
(12 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 5960X Asus X99-E WS GTX Titan Tri-SLI Crucial Ballistix Sport 
Hard DriveHard DriveCoolingOS
Samsung XP941 Sandisk Ultra II 960GB Water Windows 8.1 
MonitorPowerCaseAudio
Asus ROG Swift EVGA Supernova 1300 G2 NZXT Switch 810 Creative SB ZXR 
  hide details  
post #123 of 542
@2010rig: Specifically the top chart, tht shows a FRAPS FPS of around 80 for Crossfire (and 40 for SLI) but only 20 for both single cards. If that is what FRAPS is reporting, why has no one else seen those results? I realize the FCAT part (and hence the lower graph) is new, but if FRAPS is reporting 80 FPS that should have been noticed by now. Why did PCPer not address that obviously impossible Crossfire scaling?
post #124 of 542
Quote:
Originally Posted by Forceman View Post

@2010rig: Specifically the top chart, tht shows a FRAPS FPS of around 80 for Crossfire (and 40 for SLI) but only 20 for both single cards. If that is what FRAPS is reporting, why has no one else seen those results? I realize the FCAT part (and hence the lower graph) is new, but if FRAPS is reporting 80 FPS that should have been noticed by now. Why did PCPer not address that obviously impossible Crossfire scaling?

Yeah, they didn't really address it, all they said was this.
Quote:
The frame rates for the HD 7950s in CrossFire differ greatly once again from the FRAPS reported metrics to our new Frame Rating results.

If you want to talk about weird...



Crossfire getting less FPS than a single card.

Quote:
Observed frame rates on the HD 7950s in CrossFire are much lower again with some runts and dropped frames but the SLI results are consistent in both FRAPS and Frame Rating capture testing.

They really need to go into more details.
2010rig
(14 items)
 
Galaxy S3
(8 items)
 
 
CPUMotherboardGraphicsRAM
X5660 @ 4.5  ASUS P6X58D-E 980TI? 12GB OCZ Platinum - 7-7-7-21 
Hard DriveCoolingOSMonitor
1 80GB SSD x25m - 3TB F3 + F4 NH-D14 Windows 7 Ultimate LG 47LH55 
KeyboardPowerCaseMouse
Natural Wireless Keyboard Corsair 750HX CM 690 II Advanced MX 518 
CPUGraphicsRAMHard Drive
Snapdragon S4 Dual core 1500mhz Adreno 225 Samsung 2GB 16GB Onboard Flash 
OSMonitorPowerCase
Android 4.4.2 - CM11 4.8" AMOLED 1280x720 2100 mAh battery Otterbox Defender 
  hide details  
2010rig
(14 items)
 
Galaxy S3
(8 items)
 
 
CPUMotherboardGraphicsRAM
X5660 @ 4.5  ASUS P6X58D-E 980TI? 12GB OCZ Platinum - 7-7-7-21 
Hard DriveCoolingOSMonitor
1 80GB SSD x25m - 3TB F3 + F4 NH-D14 Windows 7 Ultimate LG 47LH55 
KeyboardPowerCaseMouse
Natural Wireless Keyboard Corsair 750HX CM 690 II Advanced MX 518 
CPUGraphicsRAMHard Drive
Snapdragon S4 Dual core 1500mhz Adreno 225 Samsung 2GB 16GB Onboard Flash 
OSMonitorPowerCase
Android 4.4.2 - CM11 4.8" AMOLED 1280x720 2100 mAh battery Otterbox Defender 
  hide details  
post #125 of 542
What is "Observed FPS"? Is that number of frames we actually see on the monitor? If so how come CF is as fast as single card in most cases? Something is not right there.
post #126 of 542
Quote:
Originally Posted by Forceman View Post

Now posted:

http://www.pcper.com/reviews/Graphics-Cards/Frame-Rating-GeForce-GTX-660-Ti-and-Radeon-HD-7950

Something weird with those Far Cry 3 1440p results. Fraps showing Crossfire to be 4x the performance of a single card?

Is this the review?
post #127 of 542
Quote:
Originally Posted by Ha-Nocri View Post

What is "Observed FPS"? Is that number of frames we actually see on the monitor? If so how come CF is as fast as single card in most cases? Something is not right there.

Full explanation here:
http://www.pcper.com/reviews/Graphics-Cards/Frame-Rating-Dissected-Full-Details-Capture-based-Graphics-Performance-Test-2
2010rig
(14 items)
 
Galaxy S3
(8 items)
 
 
CPUMotherboardGraphicsRAM
X5660 @ 4.5  ASUS P6X58D-E 980TI? 12GB OCZ Platinum - 7-7-7-21 
Hard DriveCoolingOSMonitor
1 80GB SSD x25m - 3TB F3 + F4 NH-D14 Windows 7 Ultimate LG 47LH55 
KeyboardPowerCaseMouse
Natural Wireless Keyboard Corsair 750HX CM 690 II Advanced MX 518 
CPUGraphicsRAMHard Drive
Snapdragon S4 Dual core 1500mhz Adreno 225 Samsung 2GB 16GB Onboard Flash 
OSMonitorPowerCase
Android 4.4.2 - CM11 4.8" AMOLED 1280x720 2100 mAh battery Otterbox Defender 
  hide details  
2010rig
(14 items)
 
Galaxy S3
(8 items)
 
 
CPUMotherboardGraphicsRAM
X5660 @ 4.5  ASUS P6X58D-E 980TI? 12GB OCZ Platinum - 7-7-7-21 
Hard DriveCoolingOSMonitor
1 80GB SSD x25m - 3TB F3 + F4 NH-D14 Windows 7 Ultimate LG 47LH55 
KeyboardPowerCaseMouse
Natural Wireless Keyboard Corsair 750HX CM 690 II Advanced MX 518 
CPUGraphicsRAMHard Drive
Snapdragon S4 Dual core 1500mhz Adreno 225 Samsung 2GB 16GB Onboard Flash 
OSMonitorPowerCase
Android 4.4.2 - CM11 4.8" AMOLED 1280x720 2100 mAh battery Otterbox Defender 
  hide details  
post #128 of 542

Yes, it is actually what we see, so CF doesn't work in most games according to them, which I find very suspicious at least
post #129 of 542
Quote:
Originally Posted by Ha-Nocri View Post

What is "Observed FPS"? Is that number of frames we actually see on the monitor? If so how come CF is as fast as single card in most cases? Something is not right there.

I think he's taking the average frame rate in cf and subtracting instances of runt/dropped frames to get a number that hopefully represents what we actually see on screen. trouble is that stutter, like beauty, is in the eye of the beholder. that might mean that observed fps will vary from person to person.

edit: ty 2010rig.
My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
i7 4770K @ 4.7ghz Gigabyte Z87X-UD4H EVGA GTX 980TI ACX 2.0  8GB G.Skill Trident X 2400 mhz 
Hard DriveHard DriveHard DriveCooling
Samsung 830 SSD 128GB 3TB Toshiba  3TB Toshiba Custom Water 
OSMonitorPowerCase
Window 7 Ultimate 64 bit HP 23xi Corsair HX750i Phanteks Enthoo Luxe 
Mouse
Logitech G600 
  hide details  
My System
(15 items)
 
  
CPUMotherboardGraphicsRAM
i7 4770K @ 4.7ghz Gigabyte Z87X-UD4H EVGA GTX 980TI ACX 2.0  8GB G.Skill Trident X 2400 mhz 
Hard DriveHard DriveHard DriveCooling
Samsung 830 SSD 128GB 3TB Toshiba  3TB Toshiba Custom Water 
OSMonitorPowerCase
Window 7 Ultimate 64 bit HP 23xi Corsair HX750i Phanteks Enthoo Luxe 
Mouse
Logitech G600 
  hide details  
post #130 of 542
Quote:
Originally Posted by PureBlackFire View Post

I think he's taking the average frame rate in cf and subtracting instances of runt/dropped frames to get a number that hopefully represents what we actually see on screen. trouble is that stutter, like beauty, is in the eye of the beholder. that might mean that observed fps will vary from person to person.

edit: ty 2010rig.

Not sure what are 'runt' frames.

but dropped frames are called like that b/c they aren't actually being displayed? So they do not exist even if our eye could catch them. Right? So "Observed FPS" is what is actually output to the monitor?

I am just trying to understand what they are measuring here b/c I know ppl who have CF are saying that they see noticeable improvements in games compared to sigle card.
Edited by Ha-Nocri - 4/2/13 at 2:35pm
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Hardware News
This thread is locked  
Overclock.net › Forums › Industry News › Hardware News › [PCPer] Frame Rating Comparison - HD 7950 CrossFire vs GTX 660 Ti SLI (full review)