Results 1 to 7 of 7

Thread: Challenging FPS: Testing SLI And CrossFire Using Video Capture

  1. #1
    Senior Member
    Join Date
    Jun 2012
    Posts
    2,401
    Thanks
    87
    Thanked
    151 times in 145 posts
    • Willzzz's system
      • Motherboard:
      • Gigabyte
      • CPU:
      • 4670K
      • PSU:
      • FD Newton R3 600W
      • Case:
      • Corsair 350D

    Challenging FPS: Testing SLI And CrossFire Using Video Capture

    http://www.tomshardware.co.uk/graphi...iew-32658.html

    Wow, now this is a really interesting way of looking at "inside the second" frame times.
    Haven't finished reading yet but it is fascinating.

  2. #2
    Senior Member
    Join Date
    Jun 2012
    Posts
    2,401
    Thanks
    87
    Thanked
    151 times in 145 posts
    • Willzzz's system
      • Motherboard:
      • Gigabyte
      • CPU:
      • 4670K
      • PSU:
      • FD Newton R3 600W
      • Case:
      • Corsair 350D

    Re: Challenging FPS: Testing SLI And CrossFire Using Video Capture

    Not quite sure why there is no FRAPS measurement for the Nvidia cards, would be interesting to know how accurate FRAPS is on Nvidia.

    Also it's obviously not really a fair test, as a 660 Ti is about £40 more than a 7870. Not as "closely-priced" as Tom's claims.

  3. #3
    Not a good person scaryjim's Avatar
    Join Date
    Jan 2009
    Location
    Gateshead
    Posts
    15,196
    Thanks
    1,231
    Thanked
    2,291 times in 1,874 posts
    • scaryjim's system
      • Motherboard:
      • Dell Inspiron
      • CPU:
      • Core i5 8250U
      • Memory:
      • 2x 4GB DDR4 2666
      • Storage:
      • 128GB M.2 SSD + 1TB HDD
      • Graphics card(s):
      • Radeon R5 230
      • PSU:
      • Battery/Dell brick
      • Case:
      • Dell Inspiron 5570
      • Operating System:
      • Windows 10
      • Monitor(s):
      • 15" 1080p laptop panel

    Re: Challenging FPS: Testing SLI And CrossFire Using Video Capture

    Hmmm, nvidia designed tools used by nvidia for performance-tuning show nvidia graphics cards in good light!

    Does make you wonder what kind of special sauce goes into nvidia drivers and stuff to enable this kind of smoothing effect - it was very interesting to read the article where AMD basically said "we assumed everyone has this problem and there was nothing we could do about it", as this shows nvidia have known about it - and been doing something about it - for years. Also, got to wonder how many 22 line frames - or 24 line frames, or 26 line frames - the nividia pair render, and what Tom's will find when they start playing with the settings. This could still end up being negative spin from nvidia, and I suspect AMD will have something to say about being judged based on an oob tool from a direct competitor...

    That said, the biggest eye opener for me in all this was that the output on your screen can be made up of multiple frame slices. I think I always assumed that if you weren't producing enough frames (or too many and were using vsync) that whole frames either got repeated or dropped. It never really occurred to me that the monitor might draw parts of several different frames at once...

  4. #4
    Senior Member
    Join Date
    Jun 2012
    Posts
    2,401
    Thanks
    87
    Thanked
    151 times in 145 posts
    • Willzzz's system
      • Motherboard:
      • Gigabyte
      • CPU:
      • 4670K
      • PSU:
      • FD Newton R3 600W
      • Case:
      • Corsair 350D

    Re: Challenging FPS: Testing SLI And CrossFire Using Video Capture

    You'd think they could do FPS/average lines per frame

    Or indeed frame lines per second (FLPS).

    Ideally 60 FPS should equal 60,000 FLPS, I'd be really interested to know what the figure ACTUALLY was.

  5. #5
    Senior Member
    Join Date
    Jun 2012
    Posts
    2,401
    Thanks
    87
    Thanked
    151 times in 145 posts
    • Willzzz's system
      • Motherboard:
      • Gigabyte
      • CPU:
      • 4670K
      • PSU:
      • FD Newton R3 600W
      • Case:
      • Corsair 350D

  6. Received thanks from:

    CAT-THE-FIFTH (28-03-2013)

  7. #6
    Moosing about! CAT-THE-FIFTH's Avatar
    Join Date
    Aug 2006
    Location
    Not here
    Posts
    31,621
    Thanks
    3,757
    Thanked
    5,060 times in 3,912 posts
    • CAT-THE-FIFTH's system
      • Motherboard:
      • Less E-PEEN
      • CPU:
      • Massive E-PEEN
      • Memory:
      • RGB E-PEEN
      • Storage:
      • Not in any order
      • Graphics card(s):
      • EVEN BIGGER E-PEEN
      • PSU:
      • OVERSIZED
      • Case:
      • UNDERSIZED
      • Operating System:
      • DOS 6.22
      • Monitor(s):
      • NOT USUALLY ON....WHEN I POST
      • Internet:
      • FUNCTIONAL

    Re: Challenging FPS: Testing SLI And CrossFire Using Video Capture

    Quote Originally Posted by scaryjim View Post
    Hmmm, nvidia designed tools used by nvidia for performance-tuning show nvidia graphics cards in good light!

    Does make you wonder what kind of special sauce goes into nvidia drivers and stuff to enable this kind of smoothing effect - it was very interesting to read the article where AMD basically said "we assumed everyone has this problem and there was nothing we could do about it", as this shows nvidia have known about it - and been doing something about it - for years. Also, got to wonder how many 22 line frames - or 24 line frames, or 26 line frames - the nividia pair render, and what Tom's will find when they start playing with the settings. This could still end up being negative spin from nvidia, and I suspect AMD will have something to say about being judged based on an oob tool from a direct competitor...

    That said, the biggest eye opener for me in all this was that the output on your screen can be made up of multiple frame slices. I think I always assumed that if you weren't producing enough frames (or too many and were using vsync) that whole frames either got repeated or dropped. It never really occurred to me that the monitor might draw parts of several different frames at once...
    We now have two Tom's Hardware labs enabled with the hardware and software to run tests using FCAT, and we're already in the process of testing for a much more comprehensive Part 2.
    The thing is RadeonPro has been around for a while,so TBH I am surprised AMD had not looked at the stuttering problems earlier. I wonder if AMD will release their own tool at some point?? They did mention that future driver releases will be user configurable between better latency or higher framerates:

    http://www.anandtech.com/show/6857/a...-roadmap-fraps



    Interesting comparison of FRAPs and FCAT by TR,especially with the part about Skyrim.
    Last edited by CAT-THE-FIFTH; 28-03-2013 at 02:28 PM.

  8. #7
    Senior Member
    Join Date
    Jun 2012
    Posts
    2,401
    Thanks
    87
    Thanked
    151 times in 145 posts
    • Willzzz's system
      • Motherboard:
      • Gigabyte
      • CPU:
      • 4670K
      • PSU:
      • FD Newton R3 600W
      • Case:
      • Corsair 350D

    Re: Challenging FPS: Testing SLI And CrossFire Using Video Capture

    Skyrim seems to be a really odd game, the 7970 CF just doesn't work at all. About half the frames come at a similar speed to a single 7970 and the other half come so quickly, that they are barely displayed at all (runt frames). In many respects it looks as though the single 7970 is preferable.

    BF3 demonstrates similar problems, although not nearly as bad, but still around a quarter of frames are totally superfluous in a CF situation.

    In other games, Borderlands 2 for example, there appear to be precious little difference between Nvidia and AMD.
    Tom's found a few small issues in BL2 although with the much cheaper 7870 CF.

    It is quite clear that cards from both companies ought to perform very similarly, however in certain games there are driver issues that have very pronounced effects.

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •