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 16

post #151 of 420
Quote:
Originally Posted by zGunBLADEz View Post

Then how so you guys explain the few games that were fixed and there after tested again. Which give same performance?

I mean if I were trying to cheat the system I would do it on single cards instead on dual gpus. Most people opt out for 1card.

AMD is not trying to cheat the system. They are just trying to fix a problem by causing another. They are trying to come to a happy medium as they release newer and newer drivers.
Super P's rig
(21 items)
 
  
CPUMotherboardGraphicsRAM
i7-6800K @ 4.4GHZ 1.37v / 3.6GHZ Cache 1.2v ASUS X99-A II Zotac GTX 1070FE @ 2088/9216 Corsair Vengeance DDR4 3000 
Hard DriveHard DriveHard DriveOptical Drive
MyDigitalSSD BPX NVMe Samsung 850 EVO Seagate Momentus XT 500 GB External DVDRW 
CoolingCoolingCoolingOS
EK-XLC Predator 240 GTX 1070 XSPC Full Cover Blade Swiftech 240mm Radiator Windows 10 
MonitorKeyboardPowerCase
Samsung 40" 4K - UN40KU6290 G710+ EVGA SuperNOVA 850G2 Fractal Design Define S 
MouseMouse PadAudioAudio
G700s Vipamz Extended XXXL Asus U7 M-Audio AV40 
Audio
Sennheiser HD 439 
  hide details  
Reply
Super P's rig
(21 items)
 
  
CPUMotherboardGraphicsRAM
i7-6800K @ 4.4GHZ 1.37v / 3.6GHZ Cache 1.2v ASUS X99-A II Zotac GTX 1070FE @ 2088/9216 Corsair Vengeance DDR4 3000 
Hard DriveHard DriveHard DriveOptical Drive
MyDigitalSSD BPX NVMe Samsung 850 EVO Seagate Momentus XT 500 GB External DVDRW 
CoolingCoolingCoolingOS
EK-XLC Predator 240 GTX 1070 XSPC Full Cover Blade Swiftech 240mm Radiator Windows 10 
MonitorKeyboardPowerCase
Samsung 40" 4K - UN40KU6290 G710+ EVGA SuperNOVA 850G2 Fractal Design Define S 
MouseMouse PadAudioAudio
G700s Vipamz Extended XXXL Asus U7 M-Audio AV40 
Audio
Sennheiser HD 439 
  hide details  
Reply
post #152 of 420
Quote:
Originally Posted by zGunBLADEz View Post

Everything in this industry is subjective..
I prefer the other users experiences over a any of this so called journalists...

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?
Meh
(11 items)
 
  
CPUMotherboardGraphicsRAM
i7 2600k @4.7GHz Asus P8Z68-V GEN3 GTX Titan 16GB Mushkin Redline Ridgeback 
Hard DriveHard DriveCoolingOS
Crucial M4 128GB  2x 500GB WD Green RAID1 + WD 2TB Green  Noctua Nh-DH 14 Win 8 Pro x64 
MonitorPowerCase
Shimian Achieva 2560x1440p IPS Corsair AX850 Lian-Li PC-A05N 
  hide details  
Reply
Meh
(11 items)
 
  
CPUMotherboardGraphicsRAM
i7 2600k @4.7GHz Asus P8Z68-V GEN3 GTX Titan 16GB Mushkin Redline Ridgeback 
Hard DriveHard DriveCoolingOS
Crucial M4 128GB  2x 500GB WD Green RAID1 + WD 2TB Green  Noctua Nh-DH 14 Win 8 Pro x64 
MonitorPowerCase
Shimian Achieva 2560x1440p IPS Corsair AX850 Lian-Li PC-A05N 
  hide details  
Reply
post #153 of 420
I don't know why people are getting so defensive about this. No one is accusing AMD of "cheating" or trying to purposefully inflate the numbers. There are obviously some issues with AMD's driver execution (which they've admitted) and there is probably a solution they can implement in software to mitigate the problem. It is good for AMD users that they have recognized the problem and are working to fix it - AMD users should be rejoicing, not denying.

As for the performance issue itself, spending time rendering a frame that is never displayed (or only displayed for a portion of a frame) is a wase of resources, and eliminating those should help overall performance. Instead of rendering two 20ms frames and a 5 ms one, it might be better to just render two 16ms frames instead.
post #154 of 420
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?

Im not dodging no question, to be honest with you. I dont know how pcper get to that conclusion anyway..
I can tell you one thing tho, i dont feel no placebo on cfx thats for sure if is scaling to 90% or not.....
post #155 of 420
Quote:
Originally Posted by Forceman View Post

I don't know why people are getting so defensive about this. No one is accusing AMD of "cheating" or trying to purposefully inflate the numbers. There are obviously some issues with AMD's driver execution (which they've admitted) and there is probably a solution they can implement in software to mitigate the problem. It is good for AMD users that they have recognized the problem and are working to fix it - AMD users should be rejoicing, not denying.

As for the performance issue itself, spending time rendering a frame that is never displayed (or only displayed for a portion of a frame) is a wase of resources, and eliminating those should help overall performance. Instead of rendering two 20ms frames and a 5 ms one, it might be better to just render two 16ms frames instead.

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
Meh
(11 items)
 
  
CPUMotherboardGraphicsRAM
i7 2600k @4.7GHz Asus P8Z68-V GEN3 GTX Titan 16GB Mushkin Redline Ridgeback 
Hard DriveHard DriveCoolingOS
Crucial M4 128GB  2x 500GB WD Green RAID1 + WD 2TB Green  Noctua Nh-DH 14 Win 8 Pro x64 
MonitorPowerCase
Shimian Achieva 2560x1440p IPS Corsair AX850 Lian-Li PC-A05N 
  hide details  
Reply
Meh
(11 items)
 
  
CPUMotherboardGraphicsRAM
i7 2600k @4.7GHz Asus P8Z68-V GEN3 GTX Titan 16GB Mushkin Redline Ridgeback 
Hard DriveHard DriveCoolingOS
Crucial M4 128GB  2x 500GB WD Green RAID1 + WD 2TB Green  Noctua Nh-DH 14 Win 8 Pro x64 
MonitorPowerCase
Shimian Achieva 2560x1440p IPS Corsair AX850 Lian-Li PC-A05N 
  hide details  
Reply
post #156 of 420
Quote:
Originally Posted by Kane2207 View Post

What I'm really struggling with is this: What use is a 21 (or less) pixel frame to anybody?

My question is how many of those are there? Sky discards few he says. Where is the threshold that separates a runt from a good frame? If there is no definition each reviewer can make its own assumption -not exactly scientific. Or are those runt frames actually genuine? After reading articles on how both AMD and nvidia agree that fraps collects data too early in the pipeline I am puzzled that FCAT does something similar. Or if they are actually genuine, do they have value after all? Is the experience better,worse or the same when short frames are involved?
Mastodon Ryzen
(12 items)
 
HP Z220
(8 items)
 
 
CPUMotherboardGraphicsRAM
R7 1800X Asus Crosshair VI Hero Asus R9 Nano Gskill TridentZ 
Hard DriveOptical DriveCoolingOS
Pny SSD 240GB Crucial MX100 CM Nepton 280L Win 10 
MonitorPowerCaseMouse
Acer Predator XG270HU Freesync XFX 750W Pro HAF XM Logitech G502 
CPUMotherboardGraphicsCooling
i7 3770 HP Quadro K2000 HP 
OSPowerCaseMouse
Win 7  HP 400W HP CMT RAT 7 
  hide details  
Reply
Mastodon Ryzen
(12 items)
 
HP Z220
(8 items)
 
 
CPUMotherboardGraphicsRAM
R7 1800X Asus Crosshair VI Hero Asus R9 Nano Gskill TridentZ 
Hard DriveOptical DriveCoolingOS
Pny SSD 240GB Crucial MX100 CM Nepton 280L Win 10 
MonitorPowerCaseMouse
Acer Predator XG270HU Freesync XFX 750W Pro HAF XM Logitech G502 
CPUMotherboardGraphicsCooling
i7 3770 HP Quadro K2000 HP 
OSPowerCaseMouse
Win 7  HP 400W HP CMT RAT 7 
  hide details  
Reply
post #157 of 420
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

There is an issue.. Saying cfx is just a placebo its not the issue...

Micro stuttering is old as hell it is nothing new, been experiencing micro stutters since i started playing pc games back in the days when no dual cards setups even been thought to exist...
Edited by zGunBLADEz - 3/28/13 at 12:26pm
post #158 of 420
Quote:
Originally Posted by Kuivamaa View Post

My question is how many of those are there? Sky discards few he says. Where is the threshold that separates a runt from a good frame? If there is no definition each reviewer can make its own assumption -not exactly scientific. Or are those runt frames actually genuine? After reading articles on how both AMD and nvidia agree that fraps collects data too early in the pipeline I am puzzled that FCAT does something similar. Or if they are actually genuine, do they have value after all? Is the experience better,worse or the same when short frames are involved?

If you read the articles again, runt frame tolerance is configurable, They set the tolerance low at 21px and still found that tons were chucked.

I'm aware that SKY has had different results, and when they are published I'll read them too.


Current situation though: 4 sites, all found the same dud frames, AMD on record, am I missing something? lol
Meh
(11 items)
 
  
CPUMotherboardGraphicsRAM
i7 2600k @4.7GHz Asus P8Z68-V GEN3 GTX Titan 16GB Mushkin Redline Ridgeback 
Hard DriveHard DriveCoolingOS
Crucial M4 128GB  2x 500GB WD Green RAID1 + WD 2TB Green  Noctua Nh-DH 14 Win 8 Pro x64 
MonitorPowerCase
Shimian Achieva 2560x1440p IPS Corsair AX850 Lian-Li PC-A05N 
  hide details  
Reply
Meh
(11 items)
 
  
CPUMotherboardGraphicsRAM
i7 2600k @4.7GHz Asus P8Z68-V GEN3 GTX Titan 16GB Mushkin Redline Ridgeback 
Hard DriveHard DriveCoolingOS
Crucial M4 128GB  2x 500GB WD Green RAID1 + WD 2TB Green  Noctua Nh-DH 14 Win 8 Pro x64 
MonitorPowerCase
Shimian Achieva 2560x1440p IPS Corsair AX850 Lian-Li PC-A05N 
  hide details  
Reply
post #159 of 420
FCAT doesn't collect data where FRAPS does, it injects the color bar there but it measures the data at the end of the pipeline. So it should capture the entire pipeline.
Quote:
So what is FCAT? FCAT is essentially a collection of tools, but at its most fundamental level FCAT is a simple, yet ingenious method to measure frame latency at the end of the rendering pipeline.

Edited by Forceman - 3/28/13 at 12:34pm
post #160 of 420
Quote:
Originally Posted by Kane2207 View Post

If you read the articles again, runt frame tolerance is configurable, They set the tolerance low at 21px and still found that tons were chucked.

I'm aware that SKY has had different results, and when they are published I'll read them too.


Current situation though: 4 sites, all found the same dud frames, AMD on record, am I missing something? lol

They all used same hardware, it would be peculiar if their findings were different. What I am saying is this: why 21 and not 20 or 22? What is the effect of 22 over 21 that sets it apart?
Quote:
Originally Posted by Forceman View Post

FCAT doesn't collect data where FRAPS does, it injects the color bar there but it measures the data at the end of the pipeline. So it should capture the entire pipeline.

Yeah,my main concern is the possibility of getting stuff "compromised" the moment color bar is injected somehow. Although to be frank I do not have much doubt in the process itself,I have faith it is proper. My question is more towards the definition and interpretation of data.
Edited by Kuivamaa - 3/28/13 at 1:33pm
Mastodon Ryzen
(12 items)
 
HP Z220
(8 items)
 
 
CPUMotherboardGraphicsRAM
R7 1800X Asus Crosshair VI Hero Asus R9 Nano Gskill TridentZ 
Hard DriveOptical DriveCoolingOS
Pny SSD 240GB Crucial MX100 CM Nepton 280L Win 10 
MonitorPowerCaseMouse
Acer Predator XG270HU Freesync XFX 750W Pro HAF XM Logitech G502 
CPUMotherboardGraphicsCooling
i7 3770 HP Quadro K2000 HP 
OSPowerCaseMouse
Win 7  HP 400W HP CMT RAT 7 
  hide details  
Reply
Mastodon Ryzen
(12 items)
 
HP Z220
(8 items)
 
 
CPUMotherboardGraphicsRAM
R7 1800X Asus Crosshair VI Hero Asus R9 Nano Gskill TridentZ 
Hard DriveOptical DriveCoolingOS
Pny SSD 240GB Crucial MX100 CM Nepton 280L Win 10 
MonitorPowerCaseMouse
Acer Predator XG270HU Freesync XFX 750W Pro HAF XM Logitech G502 
CPUMotherboardGraphicsCooling
i7 3770 HP Quadro K2000 HP 
OSPowerCaseMouse
Win 7  HP 400W HP CMT RAT 7 
  hide details  
Reply
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