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 15

post #141 of 420
Guys, frame times and latency are not end all be all, just as fps isn't either. When consumers and reviewers need to argue and discern this much over one metric of reviewing, then clearly we havent hit the nail on the head with this stuff just yet. When you have a GPU producer giving out testing software, that should be another giveaway that we haven't gotten there yet.

It is clear many of you haven't bothered to even read the articles by anandtech or TR (which vary slightly in lean but hold a lot of good info) and are just talking out of your behind for whatever reason. I know, this is a rabbit in the road and is easy pickings to the vultures, problem is, this is way more complicated then most of us could even begin to explain in complete detail.

This isn't a denial of any problems because clearly there are, and some massive ones as well. The CFx section of the anandtech article explains these blob/dropped frame graphs and why they are misleading merely because of how and is approaching the issue.
     
CPUMotherboardGraphicsRAM
intel 3960x @4.7 ghz asrock extreme9 x79 3x msi 7970 lightnings g.skill ripjawx 1600 cl8 
Hard DriveOptical DriveCoolingMonitor
ssd: kingston hyperx 120gb hdd: seagate barracu... bd-rwh water for cpu benq 2420t 24" 120hz 
Case
mountain mods extended asension 
CPUMotherboardGraphicsRAM
3960x Asrock Extreme9 2x Sapphire 6950 unlocked G.skill CL8 12800 x4 (4gb) 
Hard DriveHard DriveOptical DriveCooling
Kingston hyperx  Seagate barracuda xt 7200/ sata3/ 64 mb cache LiteOn BD-R x12 Watercooling 
OSMonitorPowerCase
win 7 x64 asus 23" 1080p 1000W Thermaltake toughpower HAX 932 Advanced (modified for Radiators on ins... 
MouseAudio
Razer Deathhadder 3.5g sennheiser PC 330, Game blaster with Extreme9 
  hide details  
Reply
     
CPUMotherboardGraphicsRAM
intel 3960x @4.7 ghz asrock extreme9 x79 3x msi 7970 lightnings g.skill ripjawx 1600 cl8 
Hard DriveOptical DriveCoolingMonitor
ssd: kingston hyperx 120gb hdd: seagate barracu... bd-rwh water for cpu benq 2420t 24" 120hz 
Case
mountain mods extended asension 
CPUMotherboardGraphicsRAM
3960x Asrock Extreme9 2x Sapphire 6950 unlocked G.skill CL8 12800 x4 (4gb) 
Hard DriveHard DriveOptical DriveCooling
Kingston hyperx  Seagate barracuda xt 7200/ sata3/ 64 mb cache LiteOn BD-R x12 Watercooling 
OSMonitorPowerCase
win 7 x64 asus 23" 1080p 1000W Thermaltake toughpower HAX 932 Advanced (modified for Radiators on ins... 
MouseAudio
Razer Deathhadder 3.5g sennheiser PC 330, Game blaster with Extreme9 
  hide details  
Reply
post #142 of 420
Quote:
Originally Posted by Derp View Post

How disgusting..... +rep for posting this, thank you. Did your source also check vishera? I bet AMD's implementation of "fake" cores doesn't suffer from this.

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









post #143 of 420
Quote:
Originally Posted by jomama22 View Post

Guys, frame times and latency are not end all be all, just as fps isn't either. When consumers and reviewers need to argue and discern this much over one metric of reviewing, then clearly we havent hit the nail on the head with this stuff just yet. When you have a GPU producer giving out testing software, that should be another giveaway that we haven't gotten there yet.

It is clear many of you haven't bothered to even read the articles by anandtech or TR (which vary slightly in lean but hold a lot of good info) and are just talking out of your behind for whatever reason. I know, this is a rabbit in the road and is easy pickings to the vultures, problem is, this is way more complicated then most of us could even begin to explain in complete detail.

This isn't a denial of any problems because clearly there are, and some massive ones as well. The CFx section of the anandtech article explains these blob/dropped frame graphs and why they are misleading merely because of how and is approaching the issue.
It's also clear from your first paragraph that you don't understand that frame times and FPS is attempting to capture the same data. FPS is simply an average of the former once per second.

Look at it this way. FPS and Frame Times are measuring the same thing. Frame times measures it with millimeters while FPS measures it in meters. It's an inherently more accurate way of testing.

How the data is presented is where it can differ. The reason why frame times are being used to measure fluidity is because it's accurate enough to do so, while FPS isn't. You can still use frame time to determine things like average frame rate.
zfz.Professional
(16 items)
 
zfz.Prodigy
(9 items)
 
zfz.Bench
(15 items)
 
CPUMotherboardGraphicsRAM
2500K ASUS P8Z77-I Deluxe EVGA GTX 670 8GB Samsung 30nm 
Hard DriveCoolingOSCase
Samsung 830 128GB H60 Win 7 64 Bit BitFenix Spectre Pro 
Audio
NuForce uDAC 2 
CPUMotherboardGraphicsRAM
3570K Gigabyte Z77X-UP7 ASUS RoG Matrix 7970 8GB Samsung DDR3 1600 
Hard DriveCoolingOSMonitor
128GB Crucial M4 Prolimatech Megahalems Black Win 7 64bit BenQ XL2420T 
PowerCase
Corsair AX650 TripT Custom Bench 
  hide details  
Reply
zfz.Professional
(16 items)
 
zfz.Prodigy
(9 items)
 
zfz.Bench
(15 items)
 
CPUMotherboardGraphicsRAM
2500K ASUS P8Z77-I Deluxe EVGA GTX 670 8GB Samsung 30nm 
Hard DriveCoolingOSCase
Samsung 830 128GB H60 Win 7 64 Bit BitFenix Spectre Pro 
Audio
NuForce uDAC 2 
CPUMotherboardGraphicsRAM
3570K Gigabyte Z77X-UP7 ASUS RoG Matrix 7970 8GB Samsung DDR3 1600 
Hard DriveCoolingOSMonitor
128GB Crucial M4 Prolimatech Megahalems Black Win 7 64bit BenQ XL2420T 
PowerCase
Corsair AX650 TripT Custom Bench 
  hide details  
Reply
post #144 of 420
Quote:
Originally Posted by Michalius View Post


Look at it this way. FPS and Frame Times are measuring the same thing. Frame times measures it with millimeters while FPS measures it in meters. It's an inherently more accurate way of testing.

not really... this whole frame times vs rates really is a lot less confusing that its made out to be. really it is just basic first order statistics....mean & variance

mean = average... average FPS over a second or a femtosecond, same thing

variance = how much deviation from the average on average....


the interesting part and actual "innovation" (stretching it) is the measurement of frame times at different point in the pipeline, and how AMD and nvidia do things differently

carry on folks. smile.gif
yunojelly
(17 items)
 
  
CPUMotherboardGraphicsGraphics
2600K P8P67 EVO GTX Titan 560 Ti GE 
RAMHard DriveHard DriveHard Drive
16GB 1600/CL8 64GB C300 RAID0 128GB M4 256GB M4 
CoolingOSMonitorPower
NH-D14 Windows 8 Pro VP2770-LED 850AX 
Case
PCV1020B 
  hide details  
Reply
yunojelly
(17 items)
 
  
CPUMotherboardGraphicsGraphics
2600K P8P67 EVO GTX Titan 560 Ti GE 
RAMHard DriveHard DriveHard Drive
16GB 1600/CL8 64GB C300 RAID0 128GB M4 256GB M4 
CoolingOSMonitorPower
NH-D14 Windows 8 Pro VP2770-LED 850AX 
Case
PCV1020B 
  hide details  
Reply
post #145 of 420
No no no.

So frame times shows you the data for how long each frame took to render. FPS shows you the same thing, but it takes all of the frames rendered over the course of one second instead.

In a perfect world, if you have 120 frames consecutively drawn in 8.3ms, you would have 120FPS. 1000ms/8.3ms = 120 frames in one second.

How the data is put into different statistics is how you can find variance and mean.

The important thing to note is that FPS only polls data once a second, and then averages it. Over the course of that second, you could have frames being rendered at 16.7, some at 33.3, and some at 8.3. Lets say this averaged out to 60 frames per second. That isn't indicative of what actually transpired though.
Edited by Michalius - 3/28/13 at 11:18am
zfz.Professional
(16 items)
 
zfz.Prodigy
(9 items)
 
zfz.Bench
(15 items)
 
CPUMotherboardGraphicsRAM
2500K ASUS P8Z77-I Deluxe EVGA GTX 670 8GB Samsung 30nm 
Hard DriveCoolingOSCase
Samsung 830 128GB H60 Win 7 64 Bit BitFenix Spectre Pro 
Audio
NuForce uDAC 2 
CPUMotherboardGraphicsRAM
3570K Gigabyte Z77X-UP7 ASUS RoG Matrix 7970 8GB Samsung DDR3 1600 
Hard DriveCoolingOSMonitor
128GB Crucial M4 Prolimatech Megahalems Black Win 7 64bit BenQ XL2420T 
PowerCase
Corsair AX650 TripT Custom Bench 
  hide details  
Reply
zfz.Professional
(16 items)
 
zfz.Prodigy
(9 items)
 
zfz.Bench
(15 items)
 
CPUMotherboardGraphicsRAM
2500K ASUS P8Z77-I Deluxe EVGA GTX 670 8GB Samsung 30nm 
Hard DriveCoolingOSCase
Samsung 830 128GB H60 Win 7 64 Bit BitFenix Spectre Pro 
Audio
NuForce uDAC 2 
CPUMotherboardGraphicsRAM
3570K Gigabyte Z77X-UP7 ASUS RoG Matrix 7970 8GB Samsung DDR3 1600 
Hard DriveCoolingOSMonitor
128GB Crucial M4 Prolimatech Megahalems Black Win 7 64bit BenQ XL2420T 
PowerCase
Corsair AX650 TripT Custom Bench 
  hide details  
Reply
post #146 of 420
Quote:
Originally Posted by Kane2207 View Post

A second card is rendering the frames, you can run additional detail and higher AA, this much is obvious as AFR shares the work load.

The issue is the current research suggests: whilst you think you're going from 30fps to 60 fps, with all of the little frames less than 21 scan lines high that actually add nothing to the quality of motion or the 'smoothness', the scaling factor becomes more like going from 30fps to 40fps. This is because AMD prioritized latency above all else. If a frame is very slow to render, it could be discarded altogether, hence microstutter.

What will be interesting to see is when AMD fix this by applying some kind of frame metering, will they lose some of the advances in raw FPS seen in the latest drivers? I wouldn't be surprised if the normal metric of absolute FPS drops but you end up with a smoother feeling experience due to better frame pacing.

LoL.

Even better, go play some console games and come back to pc and play the same thing at higher res with AA cfx and everything.

Seriously guys the first ones that would be screaming if cfx were a placebo are those on higher res than 1080P or multi screen setup users.

Of course the 2nd card is been used.

Do you think a single 7970 can give you a smooth experience at 1440p+ with AA on modern games?

I play a lot console games and I been playing on pc for so long.. I know what's 30vs60 feels like.
Edited by zGunBLADEz - 3/28/13 at 11:54am
post #147 of 420
Quote:
Originally Posted by zGunBLADEz View Post

Quote:
Originally Posted by Kane2207 View Post

A second card is rendering the frames, you can run additional detail and higher AA, this much is obvious as AFR shares the work load.

The issue is the current research suggests: whilst you think you're going from 30fps to 60 fps, with all of the little frames less than 21 scan lines high that actually add nothing to the quality of motion or the 'smoothness', the scaling factor becomes more like going from 30fps to 40fps. This is because AMD prioritized latency above all else. If a frame is very slow to render, it could be discarded altogether, hence microstutter.

What will be interesting to see is when AMD fix this by applying some kind of frame metering, will they lose some of the advances in raw FPS seen in the latest drivers? I wouldn't be surprised if the normal metric of absolute FPS drops but you end up with a smoother feeling experience due to better frame pacing.

LoL.

Even better, go play some console games and come back to pc and play the same thing at higher res with AA cfx and everything.

Seriously guys the first ones that would be screaming if cfx were a placebo are those on higher res than 1080P or multi screen setup users.

Of course the 2nd card is been used.

I don't think anyone is saying the second card is not being used. They are saying that with a second card in cf, some frames that are being counted, are not complete frames and are useless.
Super P's rig
(20 items)
 
  
CPUMotherboardGraphicsRAM
5960x ASUS X99-A II Asus GTX 1080 Ti Corsair Vengeance DDR4 3000 
Hard DriveHard DriveHard DriveOptical Drive
MyDigitalSSD BPX NVMe Samsung 850 EVO Seagate Momentus XT 500 GB External DVDRW 
CoolingCoolingOSMonitor
EK-XLC Predator 240 Swiftech 240mm Radiator Windows 10 Samsung 40" 4K - UN40KU6290 
KeyboardPowerCaseMouse
G710+ EVGA SuperNOVA 850G2 Fractal Design Define S G700s 
Mouse PadAudioAudioAudio
Vipamz Extended XXXL Asus U7 M-Audio AV40 Sennheiser HD 439 
  hide details  
Reply
Super P's rig
(20 items)
 
  
CPUMotherboardGraphicsRAM
5960x ASUS X99-A II Asus GTX 1080 Ti Corsair Vengeance DDR4 3000 
Hard DriveHard DriveHard DriveOptical Drive
MyDigitalSSD BPX NVMe Samsung 850 EVO Seagate Momentus XT 500 GB External DVDRW 
CoolingCoolingOSMonitor
EK-XLC Predator 240 Swiftech 240mm Radiator Windows 10 Samsung 40" 4K - UN40KU6290 
KeyboardPowerCaseMouse
G710+ EVGA SuperNOVA 850G2 Fractal Design Define S G700s 
Mouse PadAudioAudioAudio
Vipamz Extended XXXL Asus U7 M-Audio AV40 Sennheiser HD 439 
  hide details  
Reply
post #148 of 420
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.
post #149 of 420
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..).

What I'm really struggling with is this: What use is a 21 (or less) pixel frame to anybody?
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 #150 of 420
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 )

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