Render speed - OS comparison

Become a member of the CGSociety

Connect, Share, and Learn with our Large Growing CG Art Community. It's Free!

THREAD CLOSED
 
Thread Tools Search this Thread Display Modes
  11 November 2012
Or Vray for that matter. . . (sidenote: Furryball is GPU based. So curious how OSes affect it)

But 40 minutes seems a bit extreme compared to the commercial work I currently do. Average time for frames as 1080HD for a layer asset is about 10-20 minutes with motion blur.

Complex frames at 1080HD are typically sub-3 hours.

http://www.themill.com/work/norfolk...sibilities.aspx

Frames were brute force Unified mental ray on CentOS, typically less than an hour a frame. The longest frames were less than 3 hours a frame with longer motion blur, where the train whizzes by the robot for example near the end.

I want to say our render times marginally improved on CentOS compared to Fedora.
__________________
My opinions are always my own...and maybe a friend's, but never my employer's.
 
  11 November 2012
Given how linux drivers are for GPUs. . .that might be a fun test. . .of sanity.
__________________
My opinions are always my own...and maybe a friend's, but never my employer's.
 
  11 November 2012
the Linux Quadro drivers are fine.
 
  12 December 2012
I did a new linux vs windows 7 mental ray test rendering a 4k image on a modern 4.7ghz 6-core 3930k i7:

Cent OS linux:
57:53 = 3473 seconds

Windows 7
1:03:35 = 3815 seconds


so with this same high poly scene from my previous test, win7 appears to be 9.847% slower than linux


Compared to my previous linux vs XP64 4-core Phenom II machine tests that showed at 12% difference, I'm not sure if it's mainly win7 performing drastically different than XP64 or if the it has to do with how the OS's handl 4 vs 6 cores, or if it's an AMD vs Intel issue.

Last edited by sentry66 : 12 December 2012 at 08:19 PM.
 
  12 December 2012
it's the compiler. Realflow, Maxwell and other Maya plugins like Fracture FX, Houdini sims, etc are faster on OS X than on Windows because of the compiler. V-Ray is the only one that is slightly faster in Windows than in Linux/OS X because they are using the more expensive Intel C++ compiler on Windows and the usual GCC compiler on OS X/Linux. When they used MS' compiler, it was slightly slower in Windows as well:

http://polygonspixelsandpaint.tumbl...post/3353016788
 
  12 December 2012
so if everyone used the expensive intel compiler, would windows theoretically be the faster OS?

I just don't see if all compilers were equal, how a lightweight stripped down OS like linux could be slower than bogged down OS like windows.
 
  12 December 2012
Geforce GTX670 or Quadro 2000 ?

I want to build a new workstation( windows 7 -64 bit ) ,will be use for maya animation ,rendering ,effects.
I will starting with video card choose . My options are Geforce GTX670 or Quadro 2000 !
Can you tell me wich of this are better for my purpose and why ?
__________________
Genius bravo !
 
  12 December 2012
Originally Posted by graphicos: I want to build a new workstation( windows 7 -64 bit ) ,will be use for maya animation ,rendering ,effects.
I will starting with video card choose . My options are Geforce GTX670 or Quadro 2000 !
Can you tell me wich of this are better for my purpose and why ?


I would choose the Geforce. The quadros we're using (that are recommended by autodesk no less) has maaad lag in the hypershade. The ones using geforce are much snappier.
__________________
"No, the movement is good, it just has to travel faster from spot A to spot B in the same amount of time."
 
  12 December 2012
Originally Posted by sentry66: so if everyone used the expensive intel compiler, would windows theoretically be the faster OS?

I just don't see if all compilers were equal, how a lightweight stripped down OS like linux could be slower than bogged down OS like windows.


well, I don't know enough about compilers to say but the Intel compiler is available for all platforms so who knows what they all look like when on equal footing there.

I'd be more interested to see how OS X performs with the clang llvm compiler that is actually recommended for use with Xcode/Cocoa. GCC is deprecated in OS X and no energy is being spent improving it for OS X. It's been like that for a few years now.
 
  12 December 2012
Originally Posted by lostparanoia: I would choose the Geforce. The quadros we're using (that are recommended by autodesk no less) has maaad lag in the hypershade. The ones using geforce are much snappier.


Which Quadros are you using lostparanoia? And are they exhibiting the same performance issues in the new 'Node Editor'?
__________________
"Even the Christmas vacation will be darkened by New Zealand scripts…"
~ J.R.R. Tolkien, The Letters of J.R.R. Tolkien, Letter 34
 
  12 December 2012
Thread automatically closed

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.
__________________
CGTalk Policy/Legalities
Note that as CGTalk Members, you agree to the terms and conditions of using this website.
 
Thread Closed share thread



Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
CGSociety
Society of Digital Artists
www.cgsociety.org

Powered by vBulletin
Copyright ©2000 - 2006,
Jelsoft Enterprises Ltd.
Minimize Ads
Forum Jump
Miscellaneous

All times are GMT. The time now is 05:52 AM.


Powered by vBulletin
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.