V-Ray CPU Comparison: Xeon Scalable vs Core i7 8700K, Core X, and Threadripper

Introduction

V-Ray, from Chaos Group, is a widely used rendering engine for creating realistic 3D graphics. Two versions are currently available:

• V-Ray Adv, which uses the CPU and scales with both core count and clock speed
• V-Ray RT, which originally used CUDA to run calculations on NVIDIA GPUs but has been expanded recently to run on CPU cores as well

Chaos Group also publishes a benchmark utility for V-Ray, which can test both CPU and GPU performance. This is a fantastic utility for comparing different processors and video cards, and we run it here at Puget Systems on every workstation we build.

Intel released a new generation of multi-socket Xeon processors this year, dubbed "Xeon Scalable". These can provide a tremendous number of CPU cores - especially when used in pairs. To see how well they handle rendering, we have run several of them through the V-Ray Benchmark here in our test lab. For comparison, some Intel and AMD single-CPU systems are also included in the charts below.

Test Setup

To see how these new CPUs perform in V-Ray, we selected a few of the Xeon Scalable chips to work with. Intel makes something like fifty different SKUs in this series, so there is no way to test them all - but we picked models with both a wide range of core count (4, 12, 18, 20, and 28 cores) and clock speeds (from 2.4 to 3.6GHz base clocks). We are also including the top Intel mainstream (i7 8700K) and enthusiast (i9 7980XE) single-socket processors as points of comparison, along with AMD's top Threadripper model (the 1950X).

All of those CPUs were run through the free V-Ray Benchmark utility, in CPU mode. If you would like more details about the full hardware configurations involved in these tests, click here to expand the following section.

Benchmark Results

Here are the results for the various CPUs we tested in V-Ray Benchmark 1.0.8:

Analysis

It is clear from the graph above that rendering in V-Ray scales very well with additional CPU cores - but that clock speed is important as well. The fastest result came from the dual 28-core Xeon Platinum 8180 chips, but in second place were the Gold 6154s... even though they have fewer cores than the 6148s. The higher clock speed on the 6154s allowed them to outpace the extra two cores per CPU on the 6148 models.

The only Xeons that did poorly were the pair of Gold 5122s, but that was to be expected: they are only quad-core chips, so together they had far fewer cores than some of the single-socket systems we tested. Speaking of those single-socket workstations, the Intel Core i9 7980XE turned in a very good performance. It packs 18 cores on a single CPU, with fairly high clock speeds as well. The big advantage it has, though, is the price: it is a hair under half the speed of the dual Xeon Platinum 8180 system, but it costs only 1/5th of a *single* 8180 processor! Going to dual Xeons, especially at the high end, gives amazing performance... with a very high price tag.

Conclusion

As discussed above, the new Xeon Scalable processors take the crown for top performance in V-Ray Adv rendering. If you look at the performance per dollar when buying a whole computer, though, Intel's Core i9 7980XE easily takes the lead (along with other models in the Core X series). Because of this split, we will continue to offer both single- and dual-CPU options for V-Ray workstations - so you can decide if you want the absolute best performance or the best value for your money. You can expect the new Xeon Scalable processors to show up as options in the dual CPU systems below, once we wrap up their qualification.

V-Ray Workstations and Render Nodes