Hello all,
I was hoping for some assistance with the following issue. Any help would be appreciated.
I have just received an evaluation version of the H10 Master Education from the good people at SFX and my first taks is to evaluate it's ability to render using RenderMan to our existing render farm consisting of 233 RMan nodes. We put this together on 2 CRAY CX1's (96 cores) and on 21 DELL XPS 720's (4 core each). The farm is RenderManProServer-14.3 and we normaly shoot the renders from Maya (we have RenderManStudio-2.0.1-maya2009) via Alfred.
The real question is: Can H10 Master Education, out of the box, use our existing RenderMan render farm via Alfred? It would seem a simple matter of putting down an Alfred Render Node and setting up that node to talk with the farm. Is there any documentation on this at all?
Of course, that isn't the end of the issue by a long shot but I felt it as good a place to start as any.
Any pointers, tutorials, documents etc. etc. whould be most appreciated. I, of course, will talk with our Pixar guys (we have a fairly good relationship) as well but I am not holding out hope that they work the Houdini angle as much as the Maya angle.
Thank you.
David Bennett
H10 Master Educational and Pixar's RenderManProServer-14.3
3761 1 0-
- Barad_Dur
- Member
- 46 posts
- Joined: Oct. 2006
- Offline
-
- JOEMI
- Member
- 128 posts
- Joined: July 2005
- Offline
There's 3 issues, you need to solve to setup farm for Houdini-alfred-prman job rendering without any additional tweaks of standard scripts
1. Alfred node produced distibuted rib-genration request, so your farm nodes requires batch license for using it.
2. Nodes have to run alfserver with good houdini environment, with all HFS, PATH and so on variables set. You may set them from alfserver's envkeys mechanics.
3. You need correctly setup paths in Alfred and Renderman nodes, all - temporary and resources. Also you may require to include in all takes (if you use takes) some parameters, which houdini batch tying to change during rib-generation.
1. Alfred node produced distibuted rib-genration request, so your farm nodes requires batch license for using it.
2. Nodes have to run alfserver with good houdini environment, with all HFS, PATH and so on variables set. You may set them from alfserver's envkeys mechanics.
3. You need correctly setup paths in Alfred and Renderman nodes, all - temporary and resources. Also you may require to include in all takes (if you use takes) some parameters, which houdini batch tying to change during rib-generation.
Anything that flies
-
- Quick Links

