*Double post
TUrning off HT would typically reduce performance by 20-25%, so thats really not an option …
Does anyone know how v-ray handles 2x 12 cores (48 threads) or 2x 16 cores (64 threads) on a single machine? We may go that route.
2x12 cores is not a problem at all. It scales very close to 100%. Just make sure all the memory settings are optimal and reduce the bucket size. I found that a bucket size of around 10 (+\-) is optimal for 48 threads.
What is the rhyme/reason behind picking a specific bucket size or is it purely experimental on what works best per scene? Are small buckets to keep the L1-4 fetches to RAM consolidated so that all cores are working on similar data and each core’s cache isn’t requesting totally different data and bottlenecking the RAM pipe?
I don’t think it has anything to do with RAM caches, but more with CPU load balancing.
Best regards,
Vlado
reduced the bucket size from 64 to 12 on my 48 thread machine… adds ~50% rendertime… ?
Its just experimenting for the specific scene, but for 48 threads its generally around size 10. Vlado kan probably elaborate on why ![]()
All I can say is that by reducing the bucket size, things generally renders faster. If I use a low core \ high frequency machine of the same CPU-architecture as a bar for 100% theoretical performance, I get very close to optimal performance on a 48 thread machine
Anyone that wants to test a build with support for processor groups, please email me to vlado@chaosgroup.com with your 3ds Max version number so that I can get you download links.
Best regards,
Vlado
Vlado,
“64+ threads support – Corona Renderer now correctly works with more than 64 threads without any workarounds”
http://cgpress.org/archives/corona-renderer-1-3-for-3ds-max-released-future-developments-announced.html
I understand all the implications you mentioned earlier in this thread, and that its possible to get a build with 64+ threads enabled. But maybe its time to at least have a switch for this somewhere in the official build? ![]()
I did post above that anyone interested in a build can email me to vlado@chaosgroup.com
But maybe its time to at least have a switch for this somewhere in the official build?There is a switch for the next official build coming out in December and it may even be enabled by default, depending on what the beta testing shows.
Best regards,
Vlado
Very nice ![]()
and does the new version also adress the issues with Multiscatter on a machine with multi processor groups (more then 64 processors)?
Wouldn’t that be a question for the Multiscatter guys?
Best regards,
Vlado
nope, you , Vlado, already confirmed that this is a Vray problem, has to do with Embree library version… so any date on updating embree library?
We are working on it. We’ve done a lot of custom modifications to Embree, so updating it is a complicated process. I will be able to tell you more in about two weeks, if all goes well.
Best regards,
Vlado
great great great, really great ![]()
Hi Vlado,
Any news sofar?
We tested the new vray (35501) and stil much slower compared to rendering with vrayspawner
Well, this moves a lot slower than I would have liked, unfortunately
My goal is to have beta 2 released with Embree 2.12 and see how that goes.
Best regards,
Vlado