Overclock.net › Forums › Industry News › Hardware News › [PCPer] Frame Rating: GTX 660 vs HD 7870, plus HD 7790, HD 7850, GTX 650 Ti BOOST
New Posts  All Forums:Forum Nav:

[PCPer] Frame Rating: GTX 660 vs HD 7870, plus HD 7790, HD 7850, GTX 650 Ti BOOST - Page 11

post #101 of 182
Quote:
Originally Posted by th3illusiveman View Post

Really? when did they say they would increase their frametime variations? I think what they said was an option for some added latency (maybe to Nvidias level since AMD cards do have lower latency currently) or the current level of frametime variation which doesn't seem to bother most people.

I meant current levels by saying worse, I wasn't comparing to any current situation just that the frame times with that option are worse than with the other option.

One of the alternatives gives worse frame times but better latency and the other is the opposite.

How much each alternative improves latency/frame times remains to be seen of course.
Quote:
Originally Posted by EastCoast View Post

From what I've observed most are. tongue.gif

Their findings are clearly visible on the captured footage, the only reason someone would deny the raw data is if they are in denial. The conclusion that you draw from that is another thing.
 
Benching
(17 items)
 
 
CPUMotherboardGraphicsRAM
[i7 5960X @ 4.8GHz] [Rampage V Extreme] [Titan 1400MHz (1500MHz bench)] [Various] 
Hard DriveCoolingCoolingCooling
[250GB 840EVO +2x SpinpointF3 1TB RAID0] [LD PC-V2 SS Phase Change] [XSPC X2O 750 pump/res] [Monsta 360 full copper + EK XT 360 + XT 240] 
MonitorPowerCaseAudio
[Crossover 27Q LED-P 1440p+ASUS 1200p+LG 1080p] [Corsair AX1200] [Dimastech Easy v3.0] [Sennheiser HD558s] 
CPUCPUMotherboardGraphics
FX 8320, FX 8350, Phenom II x2 555BE i7 3930K, i7 860, i7 4770K, 68x Celeron D CVF, commando, 2x RIVE, Z87X-OC Asus 4870x2, Sapphire 4870 
GraphicsGraphicsGraphicsGraphics
2x 5870, 5850, 5830, 5770 2x 3870x2, 3870 GTX Titan, GTX 480, GTX 590 GTX 285, GTX 260, 4x 9800GT, 8800GTX 
RAMHard DriveCoolingCooling
4x4GB vengeance, 2x4GB predatorX, 2x1GB OCZ DDR2 Intel X25-M 80GB LD PC-V2 SS Phase Change OCN Marksman 
CoolingCoolingOSPower
2x old tek slims (GPU) Various watercooling stuff win7, winxp AX1200 
Case
test bench / cardboard box 
  hide details  
Reply
 
Benching
(17 items)
 
 
CPUMotherboardGraphicsRAM
[i7 5960X @ 4.8GHz] [Rampage V Extreme] [Titan 1400MHz (1500MHz bench)] [Various] 
Hard DriveCoolingCoolingCooling
[250GB 840EVO +2x SpinpointF3 1TB RAID0] [LD PC-V2 SS Phase Change] [XSPC X2O 750 pump/res] [Monsta 360 full copper + EK XT 360 + XT 240] 
MonitorPowerCaseAudio
[Crossover 27Q LED-P 1440p+ASUS 1200p+LG 1080p] [Corsair AX1200] [Dimastech Easy v3.0] [Sennheiser HD558s] 
CPUCPUMotherboardGraphics
FX 8320, FX 8350, Phenom II x2 555BE i7 3930K, i7 860, i7 4770K, 68x Celeron D CVF, commando, 2x RIVE, Z87X-OC Asus 4870x2, Sapphire 4870 
GraphicsGraphicsGraphicsGraphics
2x 5870, 5850, 5830, 5770 2x 3870x2, 3870 GTX Titan, GTX 480, GTX 590 GTX 285, GTX 260, 4x 9800GT, 8800GTX 
RAMHard DriveCoolingCooling
4x4GB vengeance, 2x4GB predatorX, 2x1GB OCZ DDR2 Intel X25-M 80GB LD PC-V2 SS Phase Change OCN Marksman 
CoolingCoolingOSPower
2x old tek slims (GPU) Various watercooling stuff win7, winxp AX1200 
Case
test bench / cardboard box 
  hide details  
Reply
post #102 of 182
Quote:
Originally Posted by Alatar View Post

Their findings are clearly visible on the captured footage, the only reason someone would deny the raw data is if they are in denial. The conclusion that you draw from that is another thing.
There various reasons why they are ignoring it. Some already discussed but it's not about denial of anything. For one, if someone isn't experiencing it won't take much stock in it no matter how often you reference capture video. Others don't take the idea of one competitor using their tools to benchmark another very well no matter how well intended you feel it is. While others are aware that AMD is coming out with their own 7990 so new CF drivers are already on the way. Furthermore, they are already committed to memory management changes and other improvements for their drivers. So, beating a dead horse to them isn't ideal...because after a while it begins to stink.

I'm sure there are others, but it isn't necessarily a result of denial.
Edited by EastCoast - 4/7/13 at 11:14am
post #103 of 182
AMD just need to sync their GPU's in CF me think. IT is purely driver issue so I am not sure why it takes them so long. But ofc, I don't know what is involved in making drivers tongue.gif

AMD obviously have raw power, just need to articulate it through drivers. I don't think we will see worse scaling then we see with nVidia.
post #104 of 182
Quote:
Originally Posted by Ha-Nocri View Post

AMD just need to sync their GPU's in CF me think. IT is purely driver issue so I am not sure why it takes them so long. But ofc, I don't know what is involved in making drivers tongue.gif

AMD obviously have raw power, just need to articulate it through drivers. I don't think we will see worse scaling then we see with nVidia.

trouble is, with kepler nvidia added some hardware to deal with this,

"Improved Frame rate Metering



Kepler introduces hardware based frame rate metering, a technology that helps to minimize stuttering. In SLI mode, two GPUs share the workload by operating on successive frames; one GPU works on the current frame while the other GPU works on the next frame. But because the workload of each frame is different, the two GPUs will complete their frames at different times. Sending the frames to the monitor at varying intervals can result in perceived stuttering.

The GeForce GTX 690 features a metering mechanism (similar to a traffic meter for a freeway entrance) to regulate the flow of frames. By monitoring and smoothing out any disparities in how frames are issued to the monitor, frame rates feel smoother and more consistent."

AMD is going to have a pure software solution for a while,

another reason to test Fermi vs Kepler on SLI, some GTX 580 SLI vs GTX 660 SLI with this method would be great.
Current PC
(11 items)
 
  
CPUMotherboardGraphicsRAM
i5 3GHz OEM Pegatron h61 HD 5850 @ 925MHz 8GB 
Hard DriveCoolingOSMonitor
Samsung 1tb C2D e6300 Intel HSF Win10 HP 19" 
KeyboardMouseAudio
generic AT with PS/2 adapter wheel mouse optical 1.1 alc 662 
  hide details  
Reply
Current PC
(11 items)
 
  
CPUMotherboardGraphicsRAM
i5 3GHz OEM Pegatron h61 HD 5850 @ 925MHz 8GB 
Hard DriveCoolingOSMonitor
Samsung 1tb C2D e6300 Intel HSF Win10 HP 19" 
KeyboardMouseAudio
generic AT with PS/2 adapter wheel mouse optical 1.1 alc 662 
  hide details  
Reply
post #105 of 182
Quote:
Originally Posted by HMBR View Post

trouble is, with kepler nvidia added some hardware to deal with this,

"Improved Frame rate Metering



Kepler introduces hardware based frame rate metering, a technology that helps to minimize stuttering. In SLI mode, two GPUs share the workload by operating on successive frames; one GPU works on the current frame while the other GPU works on the next frame. But because the workload of each frame is different, the two GPUs will complete their frames at different times. Sending the frames to the monitor at varying intervals can result in perceived stuttering.

The GeForce GTX 690 features a metering mechanism (similar to a traffic meter for a freeway entrance) to regulate the flow of frames. By monitoring and smoothing out any disparities in how frames are issued to the monitor, frame rates feel smoother and more consistent."

AMD is going to have a pure software solution for a while,

Interesting, didn't know that. REP for you. That probably causes some input lag, but it is very worth it as we see.
Quote:
another reason to test Fermi vs Kepler on SLI, some GTX 580 SLI vs GTX 660 SLI with this method would be great.
Agree. Would be interesting to see if SLI was a waste of money like CF was all these years.
post #106 of 182
it's hilarious that people call CF a "waste of money"
there is a VERY noticeable improvement going from single to CF, despite stutter
zen dream
(13 items)
 
the old nag
(13 items)
 
 
CPUMotherboardGraphicsRAM
Zen 1700 Asrock B350 Pro4 MSI RX 480 4GB Crucial Ballistix DDR4 2400 2x4GB 
Hard DriveCoolingMonitorKeyboard
*ntel 600p 256GB NVMe Thermalright TRUE Spirit 140 Direct Dell SE2717HX Filco Magestouch MX Brown Tenkeyless 
PowerCaseMouseMouse Pad
SeaSonic M12II 620W Antec 300 Logitech G400 Glorious PC Gaming Large 
Audio
Sennheiser HD598 
CPUMotherboardRAMHard Drive
Phenom II x4 955 Asus M3N78 PRO 2GB Corsair XMS2-8500 DDR2 5-5-5-15 (2) OCZ 128GB SSD 
Hard DriveHard DriveHard DriveHard Drive
OCZ 30 GB SSD Samsung 2TB WD Black 2TB  WD Green 2TB 
OSMonitorPowerCase
Windows 10 x64 Dell P2414H Corsair CX430 v2 Fractal Design Define R5 
Other
Linksys EA6350 AC1200+  
  hide details  
Reply
zen dream
(13 items)
 
the old nag
(13 items)
 
 
CPUMotherboardGraphicsRAM
Zen 1700 Asrock B350 Pro4 MSI RX 480 4GB Crucial Ballistix DDR4 2400 2x4GB 
Hard DriveCoolingMonitorKeyboard
*ntel 600p 256GB NVMe Thermalright TRUE Spirit 140 Direct Dell SE2717HX Filco Magestouch MX Brown Tenkeyless 
PowerCaseMouseMouse Pad
SeaSonic M12II 620W Antec 300 Logitech G400 Glorious PC Gaming Large 
Audio
Sennheiser HD598 
CPUMotherboardRAMHard Drive
Phenom II x4 955 Asus M3N78 PRO 2GB Corsair XMS2-8500 DDR2 5-5-5-15 (2) OCZ 128GB SSD 
Hard DriveHard DriveHard DriveHard Drive
OCZ 30 GB SSD Samsung 2TB WD Black 2TB  WD Green 2TB 
OSMonitorPowerCase
Windows 10 x64 Dell P2414H Corsair CX430 v2 Fractal Design Define R5 
Other
Linksys EA6350 AC1200+  
  hide details  
Reply
post #107 of 182
So you're getting worse framerate with SLI 660 than a single card or am I getting the graph wrong?
built in Feb 08
(15 items)
 
built in Nov 16
(15 items)
 
 
CPUMotherboardGraphicsRAM
QX9650 Extreme Asus P5E Rampage Formula MSI GTX660 TwinFrozr 4gb Corsair Dominator 8500 
Hard DriveOptical DriveCoolingOS
Crucial M550/WD Black  LG-sata Zalman CNPS9700 LED Windows 8.1 Pro x64 
MonitorKeyboardPowerCase
Dell P2414H G510 Enermax Infiniti 720W Coolermaster Stacker 831 
MouseMouse PadAudio
Razer Diamondback 3G Razer Mantis control Samson Meteor mic 
CPUMotherboardGraphicsRAM
i7 6700k Asus Maximus VIII Ranger Zotac 1070 Amp Edition Corsair Vengeance LPX  
Hard DriveHard DriveCoolingOS
Samsung 960 Evo  Crucial M550 Cryorig H5 Uni Windows 10 Pro x64  
MonitorKeyboardPowerCase
Dell 24" CM Storm QuickFire TK Enermax Revolution Xt II 80 Plus Gold 750W Fractal Design Define C 
MouseMouse PadAudio
Razer Diamondback 5G Razer Speed / Control Onboard ESS Sabre 
  hide details  
Reply
built in Feb 08
(15 items)
 
built in Nov 16
(15 items)
 
 
CPUMotherboardGraphicsRAM
QX9650 Extreme Asus P5E Rampage Formula MSI GTX660 TwinFrozr 4gb Corsair Dominator 8500 
Hard DriveOptical DriveCoolingOS
Crucial M550/WD Black  LG-sata Zalman CNPS9700 LED Windows 8.1 Pro x64 
MonitorKeyboardPowerCase
Dell P2414H G510 Enermax Infiniti 720W Coolermaster Stacker 831 
MouseMouse PadAudio
Razer Diamondback 3G Razer Mantis control Samson Meteor mic 
CPUMotherboardGraphicsRAM
i7 6700k Asus Maximus VIII Ranger Zotac 1070 Amp Edition Corsair Vengeance LPX  
Hard DriveHard DriveCoolingOS
Samsung 960 Evo  Crucial M550 Cryorig H5 Uni Windows 10 Pro x64  
MonitorKeyboardPowerCase
Dell 24" CM Storm QuickFire TK Enermax Revolution Xt II 80 Plus Gold 750W Fractal Design Define C 
MouseMouse PadAudio
Razer Diamondback 5G Razer Speed / Control Onboard ESS Sabre 
  hide details  
Reply
post #108 of 182
Quote:
Originally Posted by .Sup View Post

So you're getting worse framerate with SLI 660 than a single card or am I getting the graph wrong?

you are probably looking @frame times. This is the right graph:
post #109 of 182
Quote:
Originally Posted by Ha-Nocri View Post

Interesting, didn't know that. REP for you. That probably causes some input lag, but it is very worth it as we see.
Agree. Would be interesting to see if SLI was a waste of money like CF was all these years.

I've owned both setups and neither was a waste of money. The conclusions PCPer are drawing based on their data is clearly not fully fleshed out. Anyone whos experienced these things first hand will tell you as much.

Raw data is great, it clearly points to a problem for AMD. That however does not mean that their conclusion (CFX not being worth it) is correct.
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 #110 of 182
Quote:
Originally Posted by CaptainChaos View Post

I've owned both setups and neither was a waste of money. The conclusions PCPer are drawing based on their data is clearly not fully fleshed out. Anyone whos experienced these things first hand will tell you as much.

Raw data is great, it clearly points to a problem for AMD. That however does not mean that their conclusion (CFX not being worth it) is correct.

This is how I see it. Ppl were playing with FRAPS turned on and they were telling themselves 'it runs great'. Some of them saw some stuttering and were saying they had smoother game-play with single card(probably b/c they lowered video settings to match what FRAPS was showing), but wasn't sure what is it all about. But these reviews show observed FPS. We are talking about exact numbers here, not some subjective feeling users have. If this is all lie by PCper, they have ruined their reputation for eternity, nothing would be able to fix this lie, so I don't want to believe that. Soon other reviewers will use the same method, AnandTech among others, and we will know for sure.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Hardware News
Overclock.net › Forums › Industry News › Hardware News › [PCPer] Frame Rating: GTX 660 vs HD 7870, plus HD 7790, HD 7850, GTX 650 Ti BOOST