Overclock.net › Forums › Industry News › Hardware News › [TR] Inside the second with Nvidia's frame capture tools
New Posts  All Forums:Forum Nav:

[TR] Inside the second with Nvidia's frame capture tools - Page 4

post #31 of 43
Quote:
Originally Posted by xlink View Post

I'm waiting for graphics cards reviewers to start looking at frame times and inverted frame times and to start doing ARIMA analysis(or similar) as part of their performance measure.


I do statistics for a living. Methods like that might be overkill in some cases, but they can shed new light on things if handled well.
If any graphics reviewer wants to take a look at things and really dive down deep, check out R
http://www.r-project.org/

Surely you realize that a graph of inverted frame time is a fps graph, right?
Overkill
(19 items)
 
  
CPUMotherboardGraphicsRAM
i7 5820k @ 4.625Ghz ASUS Rampage V Extreme GTX 1080 Ti 32GB G.Skill RIPJAWS 3200Mhz CL14 
Hard DriveHard DriveHard DriveHard Drive
Intel 750 400GB Samsung 850 Pro 512GB Samsung 850 EVO 500GB WD Black 4TB 7200RPM 
Hard DriveCoolingOSMonitor
Hitachi 2TB 7200RPM Corsair H80i GT Windows 10 Professional Acer Predator X34 3440x1440 100hz IPS 
KeyboardPowerCaseMouse
Razer BlackWidow Chroma Corsair RM1000i SilverStone RV02 Razer DeathAdder 3G 
AudioAudioAudio
TEAC HA-501 HRT Music Streamer II+ DAC Sennheiser HD650 
  hide details  
Reply
Overkill
(19 items)
 
  
CPUMotherboardGraphicsRAM
i7 5820k @ 4.625Ghz ASUS Rampage V Extreme GTX 1080 Ti 32GB G.Skill RIPJAWS 3200Mhz CL14 
Hard DriveHard DriveHard DriveHard Drive
Intel 750 400GB Samsung 850 Pro 512GB Samsung 850 EVO 500GB WD Black 4TB 7200RPM 
Hard DriveCoolingOSMonitor
Hitachi 2TB 7200RPM Corsair H80i GT Windows 10 Professional Acer Predator X34 3440x1440 100hz IPS 
KeyboardPowerCaseMouse
Razer BlackWidow Chroma Corsair RM1000i SilverStone RV02 Razer DeathAdder 3G 
AudioAudioAudio
TEAC HA-501 HRT Music Streamer II+ DAC Sennheiser HD650 
  hide details  
Reply
post #32 of 43
Quote:
Originally Posted by Forceman View Post

I know where it comes from, but it is a stupid, pejorative name. They aren't fake frames, which would imply they were made solely to inflate the framerate, they are real frames which just don't get shown on screen for very long. Calling them fake implies that AMD is doing it on purpose, and no objective observer is saying that.

This.

AMD was concentrating on getting the raw fps numbers up. Using previous methods, you couldn't detect runt frames, they would be just counted as frames.
Upstairs Rig
(11 items)
 
  
CPUMotherboardGraphicsRAM
4770k Asus Maximus VI Hero evga 1080 Ti with Hybrid mod Corsair Vengeance Pro 2133 mhz 
Hard DriveHard DriveCoolingOS
Samsung 850 EVO 500gb WD Caviar Black Corsair h100i GTX Windows 8.1 64bit 
MonitorPowerCase
xb280hk EVGA Supernova 1000 G2 Corsair Carbide Air 540 
  hide details  
Reply
Upstairs Rig
(11 items)
 
  
CPUMotherboardGraphicsRAM
4770k Asus Maximus VI Hero evga 1080 Ti with Hybrid mod Corsair Vengeance Pro 2133 mhz 
Hard DriveHard DriveCoolingOS
Samsung 850 EVO 500gb WD Caviar Black Corsair h100i GTX Windows 8.1 64bit 
MonitorPowerCase
xb280hk EVGA Supernova 1000 G2 Corsair Carbide Air 540 
  hide details  
Reply
post #33 of 43
To my understanding, "runt" frames are still full frames that need to be rendered and all- they just don't remain on screen long enough. Most likely a driver issue.
Mastodon Ryzen
(12 items)
 
HP Z220
(8 items)
 
 
CPUMotherboardGraphicsRAM
R7 1800X Asus Crosshair VI Hero Sapphire RX Vega 64 reference 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 Sapphire RX Vega 64 reference 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 #34 of 43
Quote:
Originally Posted by Kuivamaa View Post

To my understanding, "runt" frames are still full frames that need to be rendered and all- they just don't remain on screen long enough. Most likely a driver issue.

Pretty much. This breaks how fluid the movement appears so is perceived at stutter
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 #35 of 43
Quote:
Originally Posted by looniam View Post

i would suspect that is true but would it not artificially inflate the measured fps?

Yes, to a certain extent. However, so far in my testing, the number of dropped frames is typically a very small fraction of the overall total. For example, in a 60 second benchmark running at 60 FPS (3600 frames in total) there may be 30 or so dropped frames, or less than 1%. The impact of those frames on the average framerate count is negligible at best.

Even if you increase the number of dropped frames to 10% or 360, the actual change they impart upon average framerates is minimal and likely won't make all that much difference in the positioning of a product.

Hopefully people remember this fact since the real focus of frame time testing should always be to demonstrate fluidity. Focusing too much on these intangibles risks derailing the goal of frame time testing.
post #36 of 43
I don't think frame time testing is just about fluidity. I think that because it's a more accurate metric, it *can* be used to measure fluidity. However, switching to frame time testing for *all* aspects is the way to go. You can get a more accurate version of average FPS with 99th percentile, for example.

It needs to stop being viewed as the 'fluidity' standard. It's the same data as FPS, it's just looking at all of it rather than an average once a second. That's inherently more accurate.
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 #37 of 43
be interesting to see what quad 7970's really gets in Heaven now...
Main Rig
(13 items)
 
Secondary Rig
(15 items)
 
 
CPUMotherboardGraphicsGraphics
i7 3770k MSI Z77A-GD65 Gigabyte 780 OC Gigabyte 780 OC 
RAMHard DriveHard DriveOptical Drive
Ripjaw  Samsung F3  Samsung 830  Asus CD-Rom 
CoolingMonitorMonitorPower
Cooler master hyper 212+  Asus  27in 1440p Auria EQ276W 760 Watt Silencer 
Case
Haf 912 
CPUMotherboardGraphicsGraphics
i7 3770k ASRock Z77 Pro4-M 290 XFX DD R9 280X Gigabyte 
GraphicsGraphicsRAMHard Drive
R9 280X MSI Gaming R9 280X Power Cooler Crucial Ballistix  Samsung 830 series 
Hard DriveOptical DriveOSMonitor
WD Black Asus 24x Windows 7 Asus 24in  
MonitorPowerCase
27in 1440p Auria CM 750 600T 
  hide details  
Reply
Main Rig
(13 items)
 
Secondary Rig
(15 items)
 
 
CPUMotherboardGraphicsGraphics
i7 3770k MSI Z77A-GD65 Gigabyte 780 OC Gigabyte 780 OC 
RAMHard DriveHard DriveOptical Drive
Ripjaw  Samsung F3  Samsung 830  Asus CD-Rom 
CoolingMonitorMonitorPower
Cooler master hyper 212+  Asus  27in 1440p Auria EQ276W 760 Watt Silencer 
Case
Haf 912 
CPUMotherboardGraphicsGraphics
i7 3770k ASRock Z77 Pro4-M 290 XFX DD R9 280X Gigabyte 
GraphicsGraphicsRAMHard Drive
R9 280X MSI Gaming R9 280X Power Cooler Crucial Ballistix  Samsung 830 series 
Hard DriveOptical DriveOSMonitor
WD Black Asus 24x Windows 7 Asus 24in  
MonitorPowerCase
27in 1440p Auria CM 750 600T 
  hide details  
Reply
post #38 of 43
Quote:
Originally Posted by SKYMTL View Post

Yes, to a certain extent. However, so far in my testing, the number of dropped frames is typically a very small fraction of the overall total. For example, in a 60 second benchmark running at 60 FPS (3600 frames in total) there may be 30 or so dropped frames, or less than 1%. The impact of those frames on the average framerate count is negligible at best.

Even if you increase the number of dropped frames to 10% or 360, the actual change they impart upon average framerates is minimal and likely won't make all that much difference in the positioning of a product.

Hopefully people remember this fact since the real focus of frame time testing should always be to demonstrate fluidity. Focusing too much on these intangibles risks derailing the goal of frame time testing.
That's true in normal situations, but if you look at the CFX Eyefinity results, you see a TON of dropped frames. So you still need to verify.

http://www.pcper.com/image/view/21644?return=node%2F56813
Edited by TranquilTempest - 3/28/13 at 2:03pm
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 #39 of 43
Quote:
We do want to be careful to note that frame delivery as measured by FCAT is just one part of a larger picture. Truly fluid animation requires the regular delivery of frames whose contents are advancing at the same rate. What happens at the beginning of the pipeline needs to match what happens at the end. Relying on FCAT numbers alone will not tell that whole story; we'd just be measuring the effectiveness of frame metering techniques. We've come too far in the past couple of years in how we measure gaming performance to commit that error now.

Quoted for emphasis, just so people don't take this as gospel.
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 #40 of 43
Quote:
Originally Posted by CaptainChaos View Post

Quoted for emphasis, just so people don't take this as gospel.
Right. Fraps frame time + FCAT frame time = whole picture.
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
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Hardware News
Overclock.net › Forums › Industry News › Hardware News › [TR] Inside the second with Nvidia's frame capture tools