Techgage logo

AMD Ryzen Threadripper 2970WX & 2920X Workstation Performance

AMD Ryzen Threadripper Wraithripper Thumb

Date: January 29, 2019
Author(s): Rob Williams

With its Ryzen Threadripper series, AMD wants you to “create with heavy metal”, but we all have different needs, and thus need different models to choose from. Having released both the 16- and 32-core Threadrippers last summer, the company recently followed-up with two more: 2920X, and 2970WX. Let’s put them to the test.



Introduction

Right this minute, there’s some intense competition going on in the HEDT space. AMD’s Ryzen Threadripper is proving to be a viable contender in a market Intel has long dominated, causing Chipzilla to mull the ways it could strike back. While Intel still tops many performance charts today, Threadripper’s price-to-performance ratio can prove much stronger – depending on the workload.

I was genuinely surprised when AMD unveiled a 32-core processor last summer. It’s not that I didn’t think the company had the capability, I just couldn’t personally see it happening. It’s clear now that the company has become extremely aggressive. That’s why we see a 32-core option on its side, and an 18-core on the other (soon to be 28-core, when the Xeon W-3175X drops).

Even AMD’s first step down from the top of the Threadripper stack bests Intel’s current top core count chip for the desktop, with 24. Those 24 cores find themselves in the Threadripper 2970WX, while 12 others call the 2920X home. We’ve put these chips through the grinder, and have spit out some results across the next couple of pages to see how they measure up.

AMD Ryzen Threadripper in MSI X399 MEG CREATION

Here’s an overview of AMD’s current CPU offerings:

AMD’s Ryzen Processor Lineup
CoresClock (Turbo)L2+L3MemoryTDPPrice
Threadripper WX-series
2990WX32 (64T)3.0 GHz (4.2)16+64MBQuad250W$1799
2970WX24 (48T)3.0 GHz (4.2)12+64MBQuad250W$1299
Threadripper X-series
2950X16 (32T)3.5 GHz (4.4)8+32MBQuad180W$899
2920X12 (24T)3.5 GHz (4.3)6+32MBQuad180W$649
Ryzen 7
R7 2700X8 (16T)3.7 GHz (4.3)4+16MBDual105W$329
R7 27008 (16T)3.2 GHz (4.1)4+16MBDual95W$299
Ryzen 5
R5 2600X6 (12T)3.6 GHz (4.2)3+16MBDual95W$219
R5 26006 (12T)3.4 GHz (3.9)3+16MBDual65W$189
R5 1600X6 (12T)3.6 GHz (4.0)3+16MBDual95W$219
R5 16006 (12T)3.2 GHz (3.6)3+16MBDual65W$189
R5 1500X4 (8T)3.5 GHz (3.7)2+16MBDual65W$174
R5 14004 (8T)3.2 GHz (3.4)2+8MBDual65W$169
Ryzen 3
R3 1300X4 (4T)3.5 GHz (3.7)2+8MBDual65W$129
R3 12004 (4T)3.1 GHz (3.4)2+8MBDual65W$109
Ryzen w/ Radeon Vega Graphics
R5 2400G4 (8T)3.6 GHz (3.9)0.5+4MBDual65W$169
R3 2200G4 (4T)3.5 GHz (3.7)0.5+4MBDual65W$99

For most people looking to build a capable workstation for all-around use, the Threadripper 2950X is probably the best choice in this lineup, as it offers a lot of value for its price. 16-cores isn’t market-leading, but it’s hardly unimpressive, especially at $899 (but regularly less at Amazon).

The two biggest Threadrippers are best-suited for those with more specific workloads – namely, those workloads that will actually take proper advantage of as many cores that these chips offer. It’s not hard to find workloads that do scale amazingly on either of these chips, but their lower clock speed and less-than-ideal memory design (which increases latency) means they’re for specific users.

If you want a better primer, I’d encourage checking out our 2990WX and 2950X launch review, as well another performance look catering to Linux, an OS that complements such a many-core architecture better than Windows (so far).

This past fall, AMD released its Dynamic Local Mode feature. It helps manage intensive threads by making sure they’re being put on the cores that have direct access to the memory. We have a content piece coming up taking a look at Default vs. Dynamic Local Mode performance, and with the testing already done, we can say that it’s made almost no difference to our workloads (there’s one instance in particular where it helped, but not greatly).

Just ahead of CES, a tool called Coreprio by developer Bitsum was released, which similar to Dynamic Local Mode associates the appropriate affinity at and after application load. This tool will be tackled in the same article, and fortunately, I can say our experience with it has been more impressive than AMD’s own solution. That’s not to say AMD’s solution isn’t effective in some situations, but in our particular workloads, the setting didn’t change much.

Coreprio, meanwhile, made a considerable performance difference in some scenarios, and in addition, general reliability of that performance was increased. If you are buying a 2970WX or 2990WX, we can already encourage use of Coreprio – but definitely do your own before treating it as a de facto solution for your specific needs.

With that, we can move onto a look at the testing methodologies, and finally, the results. These CPUs came out a few months ago, and we’ve known everything we’ve needed to know about them for months before that, so, let’s get to it:

Test Methodology & Systems

Benchmarking a CPU may sound like a simple enough task, but in order to deliver accurate, repeatable results, and not to mention results that don’t favor one vendor over another, strict guidelines need to be adhered to. That in turn makes for rigorous, time-consuming testing, but we feel that the effort is worth it.

This page exists so that we can be open about how we test, and give those who care about testing procedures an opportunity to review our methodology before flaming us in the comments. Here, you can see a breakdown of all of our test machines, specifics about the tests themselves, and other general information that might be useful.

Let’s start with a look at the test platforms, for AMD’s TR4 (MSI’s MEG X399 Creation and Aorus’ X399 GAMING 7) and AM4 (ASUS’ CROSSHAIR VII HERO), along with Intel’s LGA2011-v3 (ASUS’ ROG STRIX X299-E GAMING), and LGA1151 (EVGA Z370 FTW and ASUS’ ROG STRIX Z390-E GAMING).

To prevent unexpected performance results, the “Multi-Core Enhancement” optimizations (effectively overclocking all cores to max turbo, instead of just two cores) offered by ASUS on its respective motherboards is disabled. There is, however, an exception. For Ryzen 2 testing, we have to use the DOCP profile on the motherboard to take proper advantage of the memory kit (it doesn’t appear to overclock anything).

All platforms were run with DDR4-3200 speeds, and 14-14-14 timings. Accomplishing that kind of thing used to be extremely difficult, but fortunately it’s a lot easier today, and helps us keep things as close to apples-to-apples as possible.

Here’s the full breakdown of the test rigs:

Techgage’s CPU Testing Platforms

AMD TR4 Test Platform #1
ProcessorAMD Ryzen Threadripper 2990WX (3.0GHz, 32C/64T)
AMD Ryzen Threadripper 2970WX (3.0 GHz, 24C/48T)
AMD Ryzen Threadripper 2920X (3.5 GHz, 12C/24T)
MotherboardMSI MEG X399 Creation
2990WX tested with BIOS 1.0EN43 (Jul 27, 2018)
2970WX and 2920X test with BIOS 1.20 (Nov 14, 2018)
MemoryG.SKILL TridentZ (F4-3200C14-8GTZ) 8GB x 4
Operates at DDR4-3200 14-14-14 (1.35V)
GraphicsNVIDIA TITAN Xp (12GB; GeForce 398.82)
StorageHost OS: WD Blue 3D NAND 1TB (SATA 6Gbps)
Power SupplyCooler Master Silent Pro Hybrid (1300W)
ChassisCooler Master MasterCase H500P Mesh
CoolingCooler Master Wraithripper Tower Cooler
Et ceteraWindows 10 Pro (Build 17134), Ubuntu 18.04 (4.15 kernel)
AMD TR4 Test Platform #2
ProcessorAMD Ryzen Threadripper 2950X (3.5GHz, 16C/32T)
AMD Ryzen Threadripper 1950X (3.4GHz, 16C/32T)
MotherboardGIGABYTE X399 Aorus Gaming 7
CPU tested with BIOS F10 (July 10, 2018)
MemoryG.SKILL Flare X (F4-3200C14-8GFX) 8GB x 4
Operates at DDR4-3200 14-14-14 (1.35V)
GraphicsNVIDIA TITAN Xp (12GB; GeForce 398.82)
StorageHost OS: WD Blue 3D NAND 1TB (SATA 6Gbps)
Power SupplyEnermax RevoBron 80+ Bronze (600W)
ChassisEnermax Equilence
CoolingEnermax Liqtech TR4 AIO (240mm)
Et ceteraWindows 10 Pro (Build 17134), Ubuntu 18.04 (4.15 kernel)
NVIDIA GeForce GTX 1080 Ti - SLI Configuration NVIDIA GeForce GTX 1080 Ti - SLI Configuration
As tested configuration: AMD Ryzen Threadripper 2990WX
NVIDIA GeForce GTX 1080 Ti - SLI Configuration
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
As tested configuration: AMD Ryzen Threadripper 2970WX
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
As tested configuration: AMD Ryzen Threadripper 2950X
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
As tested configuration: AMD Ryzen Threadripper 1950X
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
As tested configuration: AMD Ryzen Threadripper 2920X
Sony Ps4 Pro Angled View

It’s a matter of circumstance that these chips were tested across two different PCs. Whenever a suite overhaul arrives, we’ll probably whittle it down to just one machine. Wraithripper makes it easy to swap CPUs, and it’s more than capable of handling even the 32-core behemoth. Nonetheless, neither of these motherboards tried to apply an overclock when the DDR4-3200 speeds were set.


AMD AM4 Test Platform
ProcessorsAMD Ryzen 7 2700X (3.7GHz, 8C/16T)
MotherboardASUS Crosshair VII HERO Wi-Fi
CPU tested with BIOS 0804 (July 9, 2018)
MemoryG.SKILL Flare X (F4-3200C14-8GFX) 8GB x 4
Operates at DDR4-3200 14-14-14 (1.35V)
GraphicsNVIDIA TITAN Xp (12GB; GeForce 398.82)
StorageHost OS: WD Blue 3D NAND 1TB (SATA 6Gbps)
Power SupplyEVGA Bronze 600B1 (600W)
ChassisFractal Design Define C
CoolingNoctua NH-U12S SE-AM4 (1x120mm)
Et ceteraWindows 10 Pro (Build 17134), Ubuntu 18.04 (4.15 kernel)
NVIDIA GeForce GTX 1080 Ti - SLI Configuration NVIDIA GeForce GTX 1080 Ti - SLI Configuration
As tested configuration: AMD Ryzen 7 2700X
NVIDIA GeForce GTX 1080 Ti - SLI Configuration

Nothing special had to be done on the CROSSHAIR VII HERO to get up and running quickly. As covered above, we had to run with the DOCP Standard profile to get the memory to run at optimum settings. Again, in testing, we couldn’t see how that improves performance outside of applying the proper memory settings, so we believe the setting requires further input for an actual boost to performance (eg: overclocking).


Intel LGA2011-3 Test Platform
ProcessorsIntel Core i9-7980XE (2.6GHz, 18C/36T)
Intel Core i9-7960X (2.8GHz, 16C/32T)
Intel Core i9-7900X (3.3GHz, 10C/20T)
MotherboardASUS ROG STRIX X299-E GAMING
CPU tested with BIOS 1004 (Nov 14, 2017)
MemoryG.SKILL TridentZ (F4-3200C14-8GTZ) 8GB x 4
Operates at DDR4-3200 14-14-14 (1.35V)
GraphicsNVIDIA TITAN Xp (12GB; GeForce 398.82)
StorageHost OS: WD Blue 3D NAND 1TB (SATA 6Gbps)
Power SupplyCorsair Professional Series Gold AX1200 (1200W)
ChassisCorsair Carbide 600C
CoolingNZXT Kraken X62 AIO (280mm)
Et ceteraWindows 10 Pro (Build 17134), Ubuntu 18.04 (4.15 kernel)
CyberPowerPC AMD VR Gaming PC - Keyboard Switches CyberPowerPC AMD VR Gaming PC - Keyboard Switches
As tested configuration: Intel Core i9-7980XE
CyberPowerPC AMD VR Gaming PC - Keyboard Switches
CyberPowerPC AMD VR Gaming PC - Keyboard Switches CyberPowerPC AMD VR Gaming PC - Keyboard Switches
As tested configuration: Intel Core i9-7960X
CyberPowerPC AMD VR Gaming PC - Keyboard Switches
CyberPowerPC AMD VR Gaming PC - Keyboard Switches CyberPowerPC AMD VR Gaming PC - Keyboard Switches
As tested configuration: Intel Core i9-7900X
CyberPowerPC AMD VR Gaming PC - Keyboard Switches

This ASUS motherboard tries to trick us into accepting core boost optimizations after selecting the XMP profile. That’s disregarded, to keep things as apples-to-apples as possible.


Intel LGA1151 Test Platform #1
ProcessorsIntel Core i7-8700K (3.7GHz, 6C/12T)
MotherboardEVGA Z370 FTW
CPU tested with BIOS 1.08 (March 20, 2018)
MemoryG.SKILL Flare X (F4-3200C14-8GFX) 8GB x 4
Operates at DDR4-3200 14-14-14 (1.35V)
GraphicsNVIDIA TITAN Xp (12GB; GeForce 398.82)
StorageHost OS: WD Blue 3D NAND 1TB (SATA 6Gbps)
Power SupplyCorsair RM650X (1200W)
ChassisNZXT S340 Elite Mid-tower
CoolingCorsair Hydro H100i V2 AIO Liquid Cooler (240mm)
Et ceteraWindows 10 Pro (Build 17134), Ubuntu 18.04 (4.15 kernel)
Intel LGA1151 Test Platform #2
ProcessorsIntel Core i9-9900K (3.6GHz, 8C/16T)
MotherboardASUS ROG STRIX Z390-E Gaming
CPU tested with BIOS 0602 (October 19 , 2018)
MemoryG.SKILL TridentZ (F4-3200C14-8GTZ) 8GB x 4
Operates at DDR4-3200 14-14-14 (1.35V)
GraphicsNVIDIA TITAN Xp (12GB; GeForce 398.82)
StorageHost OS: WD Blue 3D NAND 1TB (SATA 6Gbps)
Power SupplyCorsair RM650X (1200W)
ChassisNZXT S340 Elite Mid-tower
CoolingCorsair Hydro H100i V2 AIO Liquid Cooler (240mm)
Et ceteraWindows 10 Pro (Build 17134), Ubuntu 18.04 (4.19.4 kernel)
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
As tested configuration: Intel Core i9-9900K
Sony Ps4 Pro Angled View
Intel Core i7-8700K CPU-Z & GPU-Z As-tested Screenshots Intel Core i7-8700K CPU-Z & GPU-Z As-tested Screenshots
As tested configuration: Intel Core i7-8700K
Intel Core i7-8700K CPU-Z & GPU-Z As-tested Screenshots

On the EVGA board, there’s nothing strange to report, because there’s no boosts built-in after setting the DRAM to 3200. The ASUS motherboard did offer to apply core optimizations for the 9900K, which we rejected.

Testing Considerations

For the bulk of our testing, we use Windows 10 build 17134 with full updates as the base. Our basic guidelines are:


Encoding Tests

Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Encoding: Adobe Premiere Pro (Media Encoder)
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Encoding: LameXP
Sony Ps4 Pro Angled View

Adobe Premiere Pro
LameXP
MAGIX Vegas

(You can click each name to go straight to that result.)


Rendering Tests

Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Rendering: Adobe Dimension CC
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Rendering: Arnold in Maya 2018
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Rendering: Blender
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Rendering: Cinebench
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Rendering: Cinema 4D
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Rendering: Corona in 3ds Max 2019
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Rendering: POV-Ray
Sony Ps4 Pro Angled View
Sony Ps4 Pro Angled View Sony Ps4 Pro Angled View
Rendering: Radeon ProRender in Maya 2018
Sony Ps4 Pro Angled View
dBpoweramp - Convert FLAC to MP3 dBpoweramp - Convert FLAC to MP3
Rendering: V-Ray Next in 3ds Max 2019
dBpoweramp - Convert FLAC to MP3
Intel Core i7-6700K (CPU-Z & GPU-Z) Intel Core i7-6700K (CPU-Z & GPU-Z)
Rendering: V-Ray Benchmark
Intel Core i7-6700K (CPU-Z & GPU-Z)

Adobe Dimension
Arnold (Maya 2018) (Also relevant to: 3ds Max, C4D, Houdini, Katana, Softimage)
Blender
Cinebench
Cinema 4D
Corona (3ds Max 2019) (Also relevant to: C4D)
KeyShot
POV-Ray
Radeon ProRender (Maya 2018) (Also relevant to: 3ds Max, Blender, C4D, SolidWorks)
V-Ray Next (3ds Max 2019) (Also relevant to: C4D, Houdini, Maya, Rhino, SketchUp)
V-Ray Benchmark
SiSoftware Sandra 2018

(You can click each name to go straight to that result.)


If you think there’s some information lacking on this page, or you simply want clarification on anything in particular, don’t hesitate to leave a comment.

Encoding: Premiere Pro, Vegas, & LameXP

We’re going to kick off this performance look with a handful of encoding tests. Encoding is one of those scenarios that can be extremely hit-or-miss when it comes to taking good advantage of big CPUs. Sometimes, applications will give the impression that they’re making proper use of the CPU, but we’ve found more than once that some applications actually just use the entire CPU very poorly.

Fortunately, the situation is getting a lot better over time. As an example, for most of its life, Adobe’s Lightroom didn’t use more than a few cores and threads. Today, the application can use most of whatever CPU you can hand it, although we’re not sure it uses it entirely effectively.

A couple of tests were scrapped for this article that were present in the 2990WX review, including DaVinci Resolve and Agisoft PhotoScan. PhotoScan didn’t give expected results, likely due to the fact that we didn’t detail the testing procedure well enough when we began running the test for that previous review. DaVinci was skipped over for the simple fact that our tests need to improve. HandBrake also got a cut because of issues with previous testing, but it’s likely to be re-added with the next suite overhaul.

In time, we plan to add DaVinci back in, as well as some others. If we can ever get reliable repeated testing out of Adobe Lightroom, it will hopefully be reintroduced at some point, as well.

Tests on this page include Adobe Premiere Pro and MAGIX Vegas Pro video encoding, as well as FLAC to MP3 conversions with LameXP.

Adobe Premiere Pro CC

Adobe Premiere Pro AVC CPU Encode Performance (AMD Ryzen Threadripper 2970WX and 2920X)
Adobe Premiere Pro HEVC CPU Encode Performance (AMD Ryzen Threadripper 2970WX and 2920X)

It’s useful that the very first benchmark in this article exhibits a performance regression we can talk about. On both the 2970WX and 2990WX, the unique memory design results in increased latency for some threads, and thus results in occasions where seemingly nonsensical performance can occur. With simple encodes of one file to a different format, there’s no performance regression at all – the top two AMD chips all perform great for both AVC and HEVC.

We mentioned earlier that we’ll be testing three different thread-handling modes on the 2990WX, and posting content around it very shortly after this article. As this is being written, the full suite hasn’t been run with the Coreprio tool, but rapid progress is being made. We can say now that the tool does in fact improve the regression here, though it’s not completely bullet-proof, as we’ll get into in that look (namely, performance is corrected 90% of the time.)

Looking beyond the 2970WX, the 2920X proves to be a strong value here. It’s a $649 chip that performs better than Intel’s $999 i9-9900X 10-core across all three tests.

Here’s what happens when we introduce CUDA GPU encoding into the mix:

Adobe Premiere Pro AVC CUDA CPU plus GPU Encode Performance (AMD Ryzen Threadripper 2970WX and 2920X)
Adobe Premiere Pro HEVC CUDA CPU plus GPU Encode Performance (AMD Ryzen Threadripper 2970WX and 2920X)

The performance regression keeps doing what it does best on the 2990WX and 2970WX, but once again, the straight-forward encodes deliver impressive performance. On the Threadrippers without the regression potential, the performance here is super-strong, and once the regressions are fixed, the top two AMD chips are going to show their strengths a lot better.

MAGIX Vegas

MAGIX Vegas Pro CPU Encode Performance (AMD Ryzen Threadripper 2970WX and 2920X)

In MAGIX’s popular Vegas video editing suite, scalability across the entire stack is really good. That helps the 24-core 2970WX jump ahead a wee bit of Intel’s 18-core Core i9-7980XE. The 32-core 2990WX uses its extra cores to keep separated a good distance from the others.

In the i9-7900X match-up again, AMD’s 12-core 2920X performed well, again pulling ahead. While our testing could include a much greater variety of CPUs, it’s interesting that the smallest chip in this particular lineup has six cores. It’s nice to see that the many-core ecosystem has finally exploded.

LameXP

LameXP MP3 Encode Performance (AMD Ryzen Threadripper 2970WX and 2920X)

Intel’s strong IPC seemed to help its chips top the performance chart here, thanks in part to the fact that LameXP can’t use more than 32 threads. The fact it can use that many at all is impressive (and useful). After that point, the I/O might become a bit of a bottleneck, but it’s a bottleneck we’d love to have the chance to stress test!

Rendering: Arnold, Blender, KeyShot, Radeon ProRender & V-Ray Next

There are few things we find quite as satisfying as rendering: seeing a bunch of assets thrown into a viewport that turn into a beautiful scene. Rendering also happens to be one of the best possible examples of what can take advantage of as much PC hardware as you can throw at it. This is true both for CPUs and GPUs.

On this page and next, we’re tackling many different renderers, because not all renderers behave the same way. That will be proven in a few cases. If you don’t see a renderer that applies to you, it could to some degree in the future, should you decide to make a move to a different design suite or renderer. An example: V-Ray supports more than just 3ds Max; it also supports Cinema 4D, Maya, Rhino, SketchUp, and recently announced, Houdini.

SolidAngle Arnold

SolidAngle Arnold (Maya 2018) CPU Render Performance (AMD Ryzen Threadripper 2970WX and 2920X)

With Arnold, the 2970WX performs great. It again keeps ahead of the 18-core Intel competition. The 2920X didn’t have the same kind of luck against its 10-core 7900X competition, despite coming ahead in all but one test on the previous encoding page. Since Arnold only uses the CPU for rendering, it’s nice to see that the software will scale really well . GPU support is coming, and you can bet your ass we’ll be doing some heterogeneous testing once it becomes available.

Blender

Blender Cycles CPU Render Performance (AMD Ryzen Threadripper 2970WX and 2920X)

Performance in Blender isn’t that different from Arnold. The 2970WX sits in second place, behind its 32-core bigger sibling, while the 2920X has redeemed itself in its battle against Intel’s 10-core 7900X. “Redeemed” might be a poor way to word it, though, since AMD has a definite cost advantage.

KeyShot

KeyShot CPU Render Performance (AMD Ryzen Threadripper 2970WX and 2920X)

You might recall that Coreprio has been mentioned a couple of times in this article. That’s the tool that helps fix performance regressions on the 2970WX and 2990WX, like the top two orange results here. This is one benchmark where Coreprio actually smooths out the regression. That fix decreases the 161s result on the 2990WX to 82s. This and other performance differences will be seen in a forthcoming article.

We left the regressed performance result here because out-of-the-box, this odd performance will strike some users, so it needs to be known about. Also, AMD’s own Dynamic Local Mode helped with performance in this same regression instance, but not nearly to the effect of Coreprio. So, people need to know about Coreprio in order to fix potential regressions.

Radeon ProRender

AMD Radeon ProRender (Maya 2018) CPU Render Performance (AMD Ryzen Threadripper 2970WX and 2920X)

As odd as it might be, AMD’s own many-core CPUs do not perform too well in its renderer. That may be largely fine since ProRender focuses on the GPU, but the software recently gained heterogeneous rendering capabilities, so it’d be great to see better scaling on the 24- and 32-core chips. Somehow, Intel’s 16-core chip rules this graph, sitting ahead of AMD’s core-equivalent 2950X.

The version of ProRender used for this round of testing is a version out-of-date, so it could be that the newest version out there will improve performance on those big chips. Since this performance testing wrapped-up, we switched our ProRender platform to 3ds Max, since it’s easier to deal with for benchmarking purposes.

Chaos Group V-Ray Next

Chaos Group V-Ray (3ds Max 2019) CPU Render Performance (AMD Ryzen Threadripper 2970WX and 2920X)

AMD’s 32-core 2990WX does battle well with Intel’s chip, but the latter company can boast that it could get the job done just as fast, and with fewer cores. The gains in V-Ray from top to bottom are not as great as it was with other renderers – such as Arnold, which showed the 2990WX to be over 3.5x quicker than the 8700K.

For future V-Ray testing, even in CPU reviews, we’ll introduce the GPU for the sake of showing the full picture, since many will take advantage of heterogeneous rendering if they’re willing to introduce the GPU into the mix. We don’t believe that Chaos Group considers its GPU renderer to be suitable for final production rendering, but we’re edging closer to that reality. For iterative rendering, combining the powers of the CPU and GPU is a no-brainer.

Chaos Group V-Ray Standalone Benchmark CPU Score (AMD Ryzen Threadripper 2970WX and 2920X)

With a real V-Ray 4.X in 3ds Max, the performance delta between the top and bottom wasn’t too great, but according to Chaos Group’s 3.5-based standalone benchmark, disagreements are made about that kind of scaling. It could be that in specific projects, scaling will in fact be impressive, but with the two projects we use in testing, no such gains were seen.

Rendering: Adobe Dimension, Cinebench, Cinema 4D, Corona, POV-Ray

We hope you’re not sick of rendering, because there’s a lot more of it on this page. On the previous page, many big-name renderers were tackled, but on this one, others are being gazed at, along with some synthetic tests that will allow you to test your own PC against ours.

Adobe Dimension

Adobe Dimension CC CPU Render Performance (AMD Ryzen Threadripper 2970WX and 2920X)

Intel rules the Dimension CC roost. Despite AMD’s core advantage, Intel’s top two chips of the 7th-gen Core X-series lead the pack. At launch, the 2990WX had to be tested without SMT, but the 24-core chip made through fine in this test, due to what could be a variety of reasons. However, there are still some Dimension projects we have that will not open on the top Threadrippers well. We’re hoping to secure better projects for the next full round of testing.

Cinema 4D & Cinebench

Cinebench R15 Multi-thread CPU Score (AMD Ryzen Threadripper 2970WX and 2920X)
Cinebench R15 Single-thread CPU Score (AMD Ryzen Threadripper 2970WX and 2920X)

Cinebench is a go-to benchmark for both AMD and Intel, and for the most part, it’s for good reason. It’s easy to feel like Cinebench is used by these companies only because it scales so well, but the fact of the matter is, Cinebench isn’t only a benchmark; it’s a benchmark for Maxon’s Cinema 4D design suite. For anyone who uses C4D, Cinebench performance scaling is directly relevant to them.

AMD’s 2990WX, out-of-the-box, surpasses the 5K mark, while the 24-core sibling sits under that – but still well ahead of the third-place i9-7980XE. The 2920X again performs quite a bit better than the 10-core Intel i9-7900X – but would the real Cinebench agree with such a statement?

Close, but not quite:

Cinema 4D CPU Render Performance (AMD Ryzen Threadripper 2970WX and 2920X)

Here, the 2920X performs close to the 7900X, while the 2990WX again graces the top of the charge. Sometimes, we think quick renders may not scale as well as the lengthier ones, but here we can see the 2990WX still set itself apart from the 2970WX. That said, the Intel competition isn’t lagging behind.

Corona Renderer

Corona (3ds Max 2019) CPU Render Performance (AMD Ryzen Threadripper 2970WX and 2920X)

With Corona, we’re seeing a result that mimics some others we’ve seen. Here, the 12-core fails to beat out the 7900X, but the 2970WX exhibits strong performance against the 7980XE. IPC clearly matters in Corona just as much as core count, so if you have a chip that excels with both, great performance will be seen. Take a look at that 7980XE performance, and imagine if it were boosted to 32 cores.

POV-Ray

POV-Ray Multi-thread CPU Score (AMD Ryzen Threadripper 2970WX and 2920X)
POV-Ray Single-thread CPU Score (AMD Ryzen Threadripper 2970WX and 2920X)

In Cinebench, the 24-core 2970WX managed to pull a good deal ahead of Intel’s 18-core 7980XE, but POV-Ray tightens the gap between the chips significantly. In both cases, the 2920WX was displaced by Intel’s 10-core. The single-thread results between CB and POV-Ray are also similar, with Intel reigning the top, and AMD underneath, propping those results up. We can’t wait for Zen 2.

System: SiSoftware Sandra 2018

While this article has no lack of synthetic benchmarks, SiSoftware’s Sandra makes it very easy to get reliable performance information on key metrics, such as arithmetic, multimedia, cryptography, and memory. Sandra is designed in such a way that it takes the best advantage of any architecture it’s given, so each CPU always has its best chance to shine.

That means a couple of things. This is definitely the “best” possible performance outlook for any chip, and doesn’t necessary correlate with real-world performance in other tests. It’s best used as a gauge of what’s possible, and to see where one architecture obviously differs from another.

Multimedia

SiSoftware Sandra Multimedia CPU Performance (AMD Ryzen Threadripper 2970WX and 2920X)

Sandra’s multi-media test effectively tests each CPU you it give it, because it’s aware of the instruction sets that could be engaged to boost performance. Ultimately, Intel dominates this test, which likely won’t be a surprise to those who’ve been watching the CPU market for a while. This isn’t going to correlate with all (or most) real-world use cases, but could represent the performance boost if Intel’s bells and whistles are all engaged.

Arithmetic

SiSoftware Sandra Arithmetic CPU Performance (AMD Ryzen Threadripper 2970WX and 2920X)

Math is hard, which is why more cores working on the same problem can be hugely beneficial. Whereas AMD struggled in the multimedia test, it excels in the arithmetic one, with strong gains seen by both top AMD chips over the top-ranking Intel ones. This is another case where AMD’s 2920X easily surpasses the performance of the 7900X.

Cryptography

SiSoftware Sandra Cryptography CPU Performance (AMD Ryzen Threadripper 2970WX and 2920X)

In a couple of generations, AMD improved its cryptography performance pretty significantly on both CPUs and GPUs. Interestingly, the 24-core chip managed to replace the 2990WX at the top for crypto bandwidth, but scaling normality resumes with hashing. This strong crypto performance trickles down to the 2920X, which also pulls ahead of its best competition (7900X) in hashing, and matching it in crypto bandwidth.

Memory Bandwidth

SiSoftware Sandra Memory Bandwidth (AMD Ryzen Threadripper 2970WX and 2920X)

If you need memory bandwidth, you need to go with an enthusiast platform that grants you the use of a quad-channel memory controller. At a certain point, you will reach diminishing returns on memory performance, so as long as you have fast memory and with as much of it as you can get, a quad-channel platform is going to suit you best.

Final Thoughts

We’re reaching the end of our look at AMD’s 2920X and 2970WX, so it’s time to establish some conclusions. Fortunately, that’s fairly easily done. In most cases, the performance offered by both chips speaks volumes in our charts, but it still only tells half of the story. AMD continues to deliver its top chips with aggressive pricing, but that pricing isn’t going to be enough for everyone.

Where IPC and single-threaded performance is concerned, Intel continues to reign. On the flip-side, AMD reigns where cores-per-dollar are concerned, and if your workload takes proper advantage of the chips, big performance gains over the Intel competition can be seen.

AMD Ryzen Threadripper in MSI X399 MEG CREATION

If AMD’s Threadripper series had improved single-threaded performance, and managed to avoid caveats like the performance regressions that can be run against in some software, the products would become more attractive on a grand scale. Things are improving, and Zen 2 is shaping up to bring on much-needed improvements – so let’s hope it all plays out.

On the regression front, Microsoft will eventually release an update which will integrate kernel updates to better accommodate chips like the 2970WX and 2990WX. I am under the assumption that we won’t see this fix land in the next major Windows 10 update (~April), but will instead see it in the fall update. Given the core functionality of Windows that such an update would implement, it doesn’t seem likely that the fix would come to us as a regular Patch Tuesday update – but I’d love to be wrong.

It’s been mentioned a couple of times throughout this article that further regression testing is coming, thanks especially to the recent introduction of the Coreprio tool. With it, an alternative to AMD’s own Dynamic Local Mode became available, so if that solution doesn’t fix a regression, Coreprio may.

The below is a quick preview of what’s in store for this Coreprio article. KeyShot featured a notable regression in our testing, where one project rendered fine, but another ran worse on AMD’s top-core processor than another with half the number of cores. In the chart below, you can see the same projects run through default settings, with Dynamic Local Mode, and also Coreprio:

KeyShot CPU Render Performance (AMD Ryzen Threadripper 2990WX Regression Testing)

As with most images, this one doesn’t tell the full story, as there are occasions when Coreprio even gets it wrong. One run may encounter the regression, while most others won’t. Across extensive testing, we’ve found Coreprio to be successful 90% of the time, but your success rate may depend on the app or even the way you launch the application.

You can’t expect speed-ups like the above across all workloads that seem to have a regression, and in fact, this is the most extreme example we’ve come across. Another good example would be with Adobe’s Premiere Pro. There, one of our YouTube encodes finishes the job in 7m 50s, whereas with Coreprio, that drops to 4m 40s, but that improvement hasn’t proven to have a 100% success rate.

To beat a dead horse that’s about rotted at this point, with anything workstation-related, it really pays to know your workload. We occasionally see instances where a smaller chip can greatly outperform a bigger one, completely depending on the workload. If you use one of the tools featured in the article, the results will be able to show you what you can expect. If your software isn’t here, a web search will hopefully help you grasp what you can expect.

Overall, AMD’s Ryzen Threadripper processors are powerhouses that boast a better core-to-dollar ratio over the competition. However, that assumes that the solution you’re running with is able to take proper advantage of the chip. Some workloads may favor IPC over cores, as some of the results in this article have shown. Fortunately, if you use a CPU-bound renderer, Threadripper will serve you extremely well. KeyShot was the outlier of our rendering testing, and as covered above, Coreprio can help smooth that out.

In many cases throughout the article, AMD’s 12-core 2920X outperformed Intel’s Core i9-7900X, despite being priced at $350 less. That doesn’t mean that Intel’s solution isn’t worth considering, though, since it has faster single-threaded performance, and higher IPC in general, and still manages to dominate in some scenarios. Remember what we keep saying about “knowing your workload”? The 2970WX likewise outperformed the i9-7980XE often, but not all of the time, despite its huge core advantage.

If you’re still left with questions of whether or not these Threadrippers (or any other chips) are for you, feel free to leave a comment.

Copyright © 2005-2019 Techgage Networks Inc. - All Rights Reserved.