Puget Systems print logo

https://www.pugetsystems.com

Read this article at https://www.pugetsystems.com/guides/1557
Article Thumbnail

Cinema 4D CPU Rendering Performance: Intel Xeon-W 3200 Series Processors

Written on August 21, 2019 by William George
Share:

Introduction

Intel recently launched a new batch of Xeon W processors - the workstation side of their Xeon family. The previous generation of these CPUs was based heavily on Intel's enthusiast Core X series, with up to 18 cores and using the same socket 2066. This time around, Intel used their server-focused Xeon Scalable line as the basis for the new Xeon W processors, leading to models with up to 28 cores and six memory channels (up from four on the last generation). We are testing these chips across a wide range of applications, but the focus of this article is on CPU-based rendering in Cinema 4D and how they compare up to other modern Intel and AMD processors.

We also took a look at performance in V-Ray Next in another article, for those who use that engine instead of the native C4D renderer.

Cinebench R20 Screenshot

Cinebench R20 in the process of testing on a Core i9 9980XE

Looking for a Cinema 4D Workstation?

Puget Systems offers a range of workstations that are tailor-made for your unique workflow. Our goal is to provide the most effective and reliable system possible so you can concentrate on your work and not worry about your computer.

Configure a System!

Test Hardware

Our focus with this round of testing is on the new Xeon W processors from Intel, of which we have three samples. Unfortunately we do not have the top-end model, the W-3275, but we should be able to estimate how well it will perform based on the behavior of the three we can test. For comparison, we have included recent results from a wide range of other Intel and AMD processors as well.

Intel Xeon W Test Platform
CPU Intel Xeon W-3265
Intel Xeon W-3245
Intel Xeon W-3225
CPU Cooler Noctua NH-U12S DX-3647
Motherboard Asus PRO WS C621-64L SAGE/10G
RAM 12x DDR4-2933 ECC Reg 16GB (196GB total)
Video Card NVIDIA GeForce RTX 2080 Ti 11GB
Hard Drive Samsung 970 Pro 1TB
Software Windows 10 Pro 64-bit (version 1903)
Cinebench R20
AMD Ryzen Test Platform
CPU AMD Ryzen 9 3900X
AMD Ryzen 7 3800X
AMD Ryzen 7 3700X
AMD Ryzen 7 2700X
CPU Cooler AMD Wraith PRISM
Motherboard Gigabyte X570 Aorus Ultra
RAM 4x DDR4-2666 16GB (64GB total)
Video Card NVIDIA GeForce RTX 2080 Ti 11GB
Hard Drive Samsung 960 Pro 1TB
Software Windows 10 Pro 64-bit (version 1903)
Cinebench R20
Intel Core Test Platform
CPU Intel Core i9 9900K
Intel Core i7 9700K
CPU Cooler Noctua NH-U12S
Motherboard Gigabyte Z390 Designare
RAM 4x DDR4-2666 16GB (64GB total)
Video Card NVIDIA GeForce RTX 2080 Ti 11GB
Hard Drive Samsung 960 Pro 1TB
Software Windows 10 Pro 64-bit (version 1903)
Cinebench R20

Benchmark Details

We used the latest version of Cinebench, from Maxon, for this comparison. The primary test it conducts looks at how quickly the CPU is able to render one full frame, and it benefits greatly from both core count and clock speed / per-core performance. Other operations in Cinema 4D are not nearly as well threaded, so do not look at the data here as a guide for the best 3D modeling or animation system.

Cinebench R20 was run three times on each of the new Xeon processors, and the fastest result was used for our chart below. The other CPUs had all been tested recently for our Ryzen 3rd Gen article, so we just brought those results over directly rather than re-run them.

Results

Here is a chart with the full results, color-coded based on CPU family. The new Intel Xeon W models are in purple, while Intel's various Core models are in blue, and then AMD Threadripper chips are shown in orange and AMD Ryzen is red.

Cinebench R20 CPU Comparison Chart with Intel Xeon W, Intel 9th Gen, Intel X-series, AMD Ryzen 3 and AMD Threadripper 2

Analysis

Intel's new Xeon W processors perform pretty closely to AMD's Threadripper models at the same core counts. We don't have the top-end 28-core model (W-3275) to test, but given its specs I would expect it to come in just shy of the 32-core TR 2990WX. Considering the much higher price tag on the Xeon chips, that isn't a great deal - unless you need the higher RAM capacity they offer. Getting ahold of a workstation-oriented motherboard for these Xeons is also a lot harder, and that is something which may hold us up from carrying them.

What strikes me as a bit odd, though, is that Intel's Core X chips outperform these new Xeons at similar core counts. This updated Xeon W line has more memory bandwidth - six channels instead of four - and at the 16-core size has higher base and turbo clock speeds as well... so I cannot figure out why the W-3245 is a full 10% behind the Core i9 9960X. AMD's Threadripper 2970WX beats them both in terms of price and performance, though, so I guess it is a moot point in this particular workload.

Conclusion

AMD's Threadripper Processors Beat Intel Core X and Xeon W on Both Price and Performance in Cinema 4D Rendering

The new Xeon W processors perform reasonably well when rendering in Cinema 4D, but you can get similar performance for a lower price with AMD's Threadripper line. Those Threadripper chips aren't as fast at single-threaded workloads though, like modeling and animation within Cinema 4D, so which CPU is best for you will depend on how much of your time is spent performing different tasks.

Looking for a Cinema 4D Workstation?

Puget Systems offers a range of workstations that are tailor-made for your unique workflow. Our goal is to provide the most effective and reliable system possible so you can concentrate on your work and not worry about your computer.

Configure a System!

Tags: Cinema4D, Cinebench, CPU, Rendering, Performance, AMD Threadripper 2nd Gen, AMD Ryzen 3rd Gen, AMD Ryzen 2nd Gen, Core X, Intel, Intel 9th Gen, Intel vs AMD, Intel X-series, AMD, Xeon W
Misha Engel
What strikes me as a bit odd, though, is that Intel's Core X chips
outperform these new Xeons at similar core counts. This updated Xeon W
line has more memory bandwidth - six channels instead of four - and at
the 16-core size has higher base and turbo clock speeds as well... so I
cannot figure out why the W-3245 is a full 10% behind the Core i9 9960X.

Cinebench runs from the cache memory so it might have something to do with the longer path to the cache memory for the W-3245(based on 28 core silicon) in comparisson to the i9-9960x(based on 18 core silicon). An other explenation could be the bios setting of the motherboard (some boards are standard overclocked where others stick to the specs given). Maybe you can try to run them at a fixed frequency of 3 GHz all-core and than see what happens (if the i9 is still faster, it is probably the cache latency of the W-3245).

Posted on 2019-08-22 22:30:24

That is a fascinating theory! Unfortunately, the Xeon W system has moved on from Labs to our product qualification department, so I cannot go back and test it again at a fixed clock speed. I do know that I didn't have any overclocking going on with any of the other testbeds, and while I didn't go over the BIOS options on the Xeon in detail I didn't see anything that looked out of place in my cursory examination (when I did things like turn the boot-up logo off... I hate those...). Perhaps I'll be able to revisit this again in the future :)

Posted on 2019-08-23 16:26:54
Misha Engel

You get amazing high scores compared to other users with the intel i9-99xxX series (and pretty low scores with the Ryzen 3000).

https://www.cgdirector.com/cinebench-r20-scores-updated-results/
https://www.guru3d.com/articles_pages/amd_ryzen_7_3800x_review,10.html

I have a dozen more links when needed.

How did you manage that?

Posted on 2019-08-24 21:55:36

There really isn't that big of a disagreement between the scores on those two sites and our results. I made a chart to illustrate this:

https://uploads.disquscdn.c...

Questionable scores are highlighted in red, and scores that largely agree are in green. Variance - the difference in percentage between the lowest and highest scores - is shown to the right.

The only CPUs where there is more than a 5% difference in scores are the Core i9 9920X and 9940X, and one low outlier on the i9 9900K. All three of the low scores on those CPUs are from the CGDirector site, even though they are much closer to our results on other Intel processors. I am not sure what is going on there, but since Guru3D agrees with us on the 9900K I am inclined to think that one at least is "correct" on our end.

Something to keep in mind, especially with Intel processors, is that recent months have seen multiple patches to Windows and BIOS updates that were aimed at mitigating Intel bugs related to the Spectre and Meltdown style vulnerabilities. These really aren't a big deal on workstation systems (or home PCs) anyway, but Microsoft, Intel, and OEMs have still been working to correct them... sometimes with impacts on CPU performance in certain workloads. Perhaps differences between patch versions led to some of the bigger variances you noticed? I'm not 100% certain, but that would be my first guess.

Posted on 2019-08-26 16:38:29