Vray rendering time with many more core(s)/cpu's

[Simontam]

I'm trying to speed up my rendering time dramatically. I have compared the render time on my i7 2600k for one model and it takes about 1min.

I put Vray up in the cloud on Amazon AWS instance using a very large machine (2 * Xeon E5-2670) basically 16 cores * 2 threads each. This should be substantially larger hardware and increase rendering time. Vray reports it is setting up 32 threads. However the total rendering time does not decrease. In fact it takes the same amount of time. Map generation doesn't seem to be utilizing all the cores at 100%.

What might be going on here? I would think I should have seen dramatically improved performance.

[Simontam]

Sorry I mean decrease rendering time.

[vlado]

Keep in mind that on the cloud, the software is running on a virtual machine; there are several layers of abstraction between the software and the hardware. There might even be other OS'es running on the same underlying hardware at the same time.

With that said, in our own tests on an actual 2x16 cores machine V-Ray seemed to perform more or less as expected (about 3 times faster than an i7), with the exception of some issues with V-Ray proxy objects, which we have already fixed in our internal builds.

Best regards,
Vlado