I want one of these…
http://news.com.com/2100-1006\_3-6119618.html?part=rss&tag=6119618&subj=news
I was reading a related article where one of the intel guys was talking about replacing graphics cards with real time raytracing on the cpu in about the same time span.
Holy crap…I cant wate. 8)
Wow. :shock: Microsoft Word will be smokin’ with 80 cores! I think we’d definitely need a new benchmark for V-Ray too, sorry Percy.
So, in DR, you are limited to 10 machines, so it is 10 x 80 buckets, so 800 buckets, hummmmmm… ok, just daydreamin’ out loud
With power like that, Vray (or any other physically accurate rendering engine for that matter) would have no need for all of the rendering options we have currently. It would just render everything in pure PPT with millions of samples - all materials would be designed for the perfect result and caustics and other volumetric phenomenon would just always be there and be perfect as a result. Whoa…
Hey no worries LOL. I’ve already suggested need a new benchmark that takes longer then 2 minutes to render. :shock:
Yes, percy, that maybe something like hard SSS with 300 subdivision on the glossy refractions, etc… 
Best regards,
nikki Candelero
Of course, to go along with this you’ll need Intel’s new terabit network chip I saw in an article in Information Week. Of course, that was also 5 years out…
5 years? I won’t be getting my hopes up! Anyone remember SPACE 1999??
Regards Peter.
unfortinantly vray only supports 256 cpus. And as it has been pointed out, there is no real point of using more then 10 machines, because of various slowdowns of the network etc.
However network rendering would be a cool thing…imagine 10 machines and you have as many cpus as orphanage has 
ha ha 