AccuRender nXt

advanced rendering for AutoCAD

hola necesitaria trabajar mas rapido , tengo una maquina

amd x6  2.8 con 4 gb , mi pregunta es:

1: como tendria que hacer para poner 2 o 3 maquinas trabajando en red

2 : cual es es costo

3 : si en todas las maquinas tienen que estar instaladas las bliblotecas de texturas o solo en la principal,

4 : cual es  la mejor recomendacion de equipo y que  cantidad de memoria para la o las maquinas

5 : si la velocidad se directamente proporcional a la cantidad de equipos ...osea 2 equipos = mitad de tiempo

6 : si es como pienso ya que la traduccion no es del todo fiel , que una misma imagen la procesan varios equipos a la vez

7 : si es asi que pasa si algunas de las maquina se cuelga, si las otras siguen prosesando normalmente esa misma imagen

pd espero poder entendernos

Views: 64

Reply to This

Replies to This Discussion

There isn't a definitive answer to most of these-- you'll need to experiment.  You get a two-node farm for free, so you can try with that.  For more, you'll need to purchase a render farm license (I don't remember how much it is-- but not very much.)  In general, if you mix fast and slow machines on the network you won't get very good results.  The rule of thumb for performance (# of cores X clock speed) still holds-- but the farm does depend (also) on the speed of your network and how much data you're sending around.  You shouldn't need to have materials or other libraries installed on the nodes-- the package it sends around for rendering should be completely self-contained.  

 

************

 

The render farm generally works pretty well-- although I might suggest waiting for my next update (this weekend?) since I haven't updated the farm in a while.  The results should be consistent.  The farm should be pretty robust if machines are crashing, although that shouldn't be happening anyway.

 

RSS

Search

Translate

Latest Activity

Roy Hirshkowitz replied to Peter Milner's discussion Sun settings
"In this case, yes, it's subtle.  It's only readily apparent in the reflection of the window in the upper right quadrant of the sphere (as expected) where a small, bright, reflection of the solar disk can be seen.  I'm…"
Wednesday
Peter Milner replied to Peter Milner's discussion Sun settings
"Here is the same test with a glossy object. There is a very subtle difference in the reflection and the lighting. "
Wednesday
Rich Rosemann replied to Peter Milner's discussion Sun settings
"I'm very good at optical mis-conclusions. "
Apr 26
Roy Hirshkowitz replied to Peter Milner's discussion Sun settings
"Looks like there's some code to handle this in both the path tracer and hybrid engines.  A little unclear in the packet tracer."
Apr 26
Roy Hirshkowitz replied to Peter Milner's discussion Sun settings
"I'll have to look to see which engines respect that switch, I can't remember off-hand.  It controls whether glossy reflections get a contribution from the direct sun component.  The easiest way to see it would be to create a…"
Apr 26
Peter Milner replied to Peter Milner's discussion Sun settings
"If you flick between the two images full screen, there is actually no difference at all. Seeing one above the other makes them look slightly different, but it's just an optical illusion."
Apr 26
Rich Rosemann replied to Peter Milner's discussion Sun settings
"One image does seem to be a bit brighter.  Maybe it opens up the shadows a bit.  Or raises the ambient light. .....or something."
Apr 26
Peter Milner posted a discussion

Sun settings

Under the Sun settings, what does Sun Highlight do?I've tried it set to 0 and 1, but there doesn't seem to be any difference.See More
Apr 26

© 2016   Header image courtesy Peter Milner   Powered by

Badges  |  Report an Issue  |  Terms of Service