Creative Communities of the World Forums

The peer to peer support community for media production professionals.

Activity Forums Creative Community Conversations CC-CS6-X render comparisons

  • Oliver Peters

    July 15, 2013 at 10:13 pm

    [Jason Van Patten] “Specifically the OP’s machine specs. He’s using a single 5870 card”

    That’s not exactly correct. I ran tests with BOTH an ATI 5870 AND an NVIDIA Quadro 4000. Only one card was installed at a time, but the PProCS6, PProCC and FCP X tests were run with each card.

    FWIW – I was mainly interested in comparing CS6 to CC. The FCP X comparison was thrown in out of curiosity, since I have also found (and posted here) that I’ve had better render times with Motion than with AE on various types of similar compositions.

    [Jason Van Patten] “His results are so good because he didn’t transcode.”

    I tested only with ProRes, to keep the test a constant. I simply don’t ever work natively with AVCHD, H.264, etc. if I can avoid it. So yes, that would affect performance.

    It’s also too much of a variable between Adobe and Apple, since each company has deemed to optimize their applications for different native codecs and wrappers. No way to control that. I would image that a test using only P2 AVC-Intra on a fast HP workstation and a fast NVIDIA card might yield results faster than X, too.

    Again, not the original point of the test.

    – Oliver

    Oliver Peters Post Production Services, LLC
    Orlando, FL
    http://www.oliverpeters.com

  • Ronny Courtens

    July 15, 2013 at 10:15 pm

    No Jason, Oliver has used different GPUs that are good for CUDA as well as for OpenCL acceleration.

    Your card is perfect for CUDA, not good for OpenCL. This is what makes your comparison debatable. The fact that these tests were done on an older MacPro only has an influence on the overall results, not on the comparison.

    And like I said: we do work natively in certain workflows and I see no difference between exporting from a ProRes timeline or a timeline with native footage. The only difference I see is that editing in a ProRes timeline is faster in FCPX.

  • Jason Van patten

    July 15, 2013 at 10:20 pm

    [Ronny Courtens] “No Jason, Oliver has used different GPUs that are good for CUDA as well as for OpenCL acceleration. Your card is perfect for CUDA, not good for OpenCL.”

    *head slap* Never mind.

    I’ll take “Missing the Point” for $600, Alex.

  • Morten

    July 16, 2013 at 8:19 am

    Would be interesting to see a similar test with H264 footage.

    – No Parking Production –

    2 x Finalcut Studio3, 2 x Prod. bundle CS6, 2 x MacPro, 2 x ioHD, Ethernet File Server w. X-Raid…. and FCPX on trial

  • Kevin Monahan

    July 16, 2013 at 9:17 pm

    [Oliver Peters] “Fast Corrector filter used for color correction in Premiere.”

    Hi Oliver,
    I was curious as to why you are using a real time effect in testing rendering speed? And what is emulation mode? Are you referring to Software Only processing?

    Cheers,
    Kevin

    Kevin Monahan
    Social Support Lead
    Adobe After Effects
    Adobe Premiere Pro
    Adobe Systems, Inc.
    Follow Me on Twitter!

  • Oliver Peters

    July 16, 2013 at 9:35 pm

    [Kevin Monahan] “I was curious as to why you are using a real time effect in testing rendering speed? And what is emulation mode? Are you referring to Software Only processing?”

    Emulation mode is MPE Software-Only.

    Since it’s an accelerated effect it gains the benefit of CUDA and/or OpenCL hardware acceleration, since this hardware acceleration affects scaling and color correction only for the most part. Yes, it’s “real-time”, but that also means it’s faster rendering, too.

    Also it’s an effect that *should* be similar processing to what Apple is doing with their built-in Color Board. As such it’s sort of an “apple-to-apples” comparison.

    In my opinion, ultimately nothing in Premiere Pro is truly real-time, since it requires rendering at some point, even upon export. As you can see from the render and export times, clearly there’s a difference based on software emulation, CUDA and/or OpenCL acceleration. At least that what it seems to me.

    Am I misunderstanding something?

    – Oliver

    Oliver Peters Post Production Services, LLC
    Orlando, FL
    http://www.oliverpeters.com

  • Dennis Radeke

    July 19, 2013 at 7:54 pm

    I’ve been meaning to write up a quick post on this for many a month and this thread prompted me to finish it up in my spare moments.

    Render Taxes

    Have a great weekend everybody.

    Dennis – Adobe guy

  • Jason Van patten

    July 20, 2013 at 9:04 pm

    [Dennis Radeke] ” I’ve been meaning to write up a quick post on this for many a month and this thread prompted me to finish it up in my spare moments.”

    It basically re-iterates what I wrote in my first reply in this thread. Just in, perhaps, an easier-to-swallow format. 😉

Page 4 of 4

We use anonymous cookies to give you the best experience we can.
Our Privacy policy | GDPR Policy