I didn't look back at the link, but if Premiere CS6 was running on a 2011 MacBook Pro, then it was also using OpenCL (as long as it was OSX Lion).
Results 11 to 20 of 36
-
Senior Member
- Join Date
- Jun 2007
- Location
- Austin, TX
- Posts
- 4,630
07-11-2012 07:16 AM
-
07-12-2012 02:33 AM
So this test was sponsored by Adobe, and some of the footage was provided by Nvidia? Hmmm...
As Ben mentioned the computer used was nicely optimized for Premiere with the mercury engine, but not so much for FCPX.
If there was a test sponsored by Apple with footage provided by ATI/AMD people would be up in arms about possible bias, so what would be the point....?
-
07-12-2012 03:52 AM
indentical performing machines with different graphics (at same clock speeds and ram maybe) could make this significant... the big render shootout? muhaha
Stefan Huber
Director Cinematographer Editor
www.stefanhuberfilms.at
vimeo.com/stefanhuberfilms
facebook.com/stefanhuberfilms
instagram.com/stefanhuberfilms
-
07-12-2012 04:42 AM
What's the point? Do you really think, footage that comes from ATI renders faster on a Mac?
And how can you "nicley optimize" a Mac at all? Itīs not a PC, there ain't much you can do on a Mac.
So what does the Test say? Premiere with a Cuda card performs better than FCPX with the same card, using Open GL.
No surprise, it's common knowledge that Cuda is faster than GL.
The point was not, this vs. that, the point was: what to buy when you need top performance.
Would FCPX have performed better with an ATI card?
I doubt it - but I reserve the right to be wrong.
Frankfrankglencairn.wordpress.com
http://twitter.com/FrankGlencairn
Real men edit their films in a hex editor.
-
07-12-2012 05:59 AM
No worries man.
However I also reserve the right to question a test when I see that the results promote 2 companies, both of whom have had a hand in the test (irrespective of how neglible the input was). Fair enough?
Hey, it could be that the world is full of happy coincidences.
For a guy who doesent like Macs you sure spend a lot of time talking about them!
-
Senior Member
- Join Date
- Jun 2007
- Location
- Austin, TX
- Posts
- 4,630
07-12-2012 06:26 AM
I don't think it matters in the end anyway. Isn't it pretty much common knowledge that if you need the absolute cutting-edge in raw processing speed, that Macs are not the route you should take? If I, personally, had the resources to build the (at this moment in time of course) fastest workstation possible, it would have to be a Windows machine.
I also don't think that a difference of a minute or two in render times will turn someone off from FCPX. I think there are plenty of other factors that are more important when it comes time for an editor to choose an NLE of choice.
-
07-12-2012 06:48 AM
Fair enough.
For the record: It's not that I don't like Macs (I grow up on them and never touched Windows till 2001) - I just don't buy them anymore.
Give me a Mac with the same hardware specs as my Workstation, but for a better (or even same) price and I'm all in.
Frank
Last edited by Postmaster; 07-12-2012 at 06:54 AM.
frankglencairn.wordpress.com
http://twitter.com/FrankGlencairn
Real men edit their films in a hex editor.
-
Member
- Join Date
- Apr 2006
- Posts
- 81
07-12-2012 06:50 AM
I see your point, but as an FCP X user myself, it's a bit deceiving to say it's about a time difference of one minute or so. It's about Premiere Pro being often 2 times or 3 times as fast (1 minute export versus 3 minute export... Not a 2 minute difference on a 60 minute program, for instance). Considering how CS6 seems like a pretty mature suite, and FCP X is still maturing (but interesting none the less), these kinds of tests are important for making decisions.
That being said, it seems this isn't a truly independent test in the sense that the system was optimized for Premiere Pro but not for FCP X.
I would like to see a comparison with FCP X and another graphics card, and see if that makes the difference (a lot) smaller or not.
-
Senior Member
- Join Date
- Jun 2007
- Location
- Austin, TX
- Posts
- 4,630
07-12-2012 07:10 AM
FCPX utilizes OpenCL for its GPU acceleration purposes. When Adobe was first looking into hardware-acceleration for the Mercury Engine, they originally went solely CUDA because OpenCL hadn't been ratified yet. Given CUDA's three-year head start, I wouldn't be surprised if it's better optimized, and thus would explain some difference.
-
07-12-2012 08:18 AM
I understand. I was the other way round; always preferred the low cost of PC's until a couple of years ago when I jumped on the mac bandwagon (to do with my perceived reliability issues with windows, and a number of laptops that died on me).
Horses for courses. Cheers.






