The AMD Ryzen Threadripper 1950X and 1920X Review: CPUs on Steroids
by Ian Cutress on August 10, 2017 9:00 AM ESTCPU Office Tests
The office programs we use for benchmarking aren't specific programs per-se, but industry standard tests that hold weight with professionals. The goal of these tests is to use an array of software and techniques that a typical office user might encounter, such as video conferencing, document editing, architectural modeling, and so on and so forth.
All of our benchmark results can also be found in our benchmark engine, Bench.
Chromium Compile (v56)
Our new compilation test uses Windows 10 Pro, VS Community 2015.3 with the Win10 SDK to compile a nightly build of Chromium. We've fixed the test for a build in late March 2017, and we run a fresh full compile in our test. Compilation is the typical example given of a variable threaded workload - some of the compile and linking is linear, whereas other parts are multithreaded.
One of the interesting data points in our test is the Compile, and it is surprising to see the 1920X only just beat the Ryzen 7 chips. Because this test requires a lot of cross-core communication, the fewer cores per CCX there are, the worse the result. This is why the 1950X in SMT-off mode beats the 3 cores-per-CCX 1920X, along with lower latency memory support. We know that this test is not too keen on victim caches either, but it does seem that the 2MB per core ratio does well for the 1950X, and could explain the performance difference moving from 8 to 12 to 16 cores under the Zen microarchitecture.
PCMark8: link
Despite originally coming out in 2008/2009, Futuremark has maintained PCMark8 to remain relevant in 2017. On the scale of complicated tasks, PCMark focuses more on the low-to-mid range of professional workloads, making it a good indicator for what people consider 'office' work. We run the benchmark from the commandline in 'conventional' mode, meaning C++ over OpenCL, to remove the graphics card from the equation and focus purely on the CPU. PCMark8 offers Home, Work and Creative workloads, with some software tests shared and others unique to each benchmark set.
Strangely, PCMark 8's Creative test seems to be failing across the board. We're trying to narrow down the issue.
SYSmark 2014 SE: link
SYSmark is developed by Bapco, a consortium of industry CPU companies. The goal of SYSmark is to take stripped down versions of popular software, such as Photoshop and Onenote, and measure how long it takes to process certain tasks within that software. The end result is a score for each of the three segments (Office, Media, Data) as well as an overall score. Here a reference system (Core i3-6100, 4GB DDR3, 256GB SSD, Integrated HD 530 graphics) is used to provide a baseline score of 1000 in each test.
A note on context for these numbers. AMD left Bapco in the last two years, due to differences of opinion on how the benchmarking suites were chosen and AMD believed the tests are angled towards Intel processors and had optimizations to show bigger differences than what AMD felt was present. The following benchmarks are provided as data, but the conflict of opinion between the two companies on the validity of the benchmark is provided as context for the following numbers.
Facebook
Twitter
RSS
347 Comments
View All Comments
Lolimaster - Friday, August 11, 2017 - link
A single 1950X destroyed 80% of the intel xeon lineup.Lolimaster - Friday, August 11, 2017 - link
Any cpu after nehalem perform enough at single thread except for software optimized too much for certain brands, like dolphin and intel.Lolimaster - Friday, August 11, 2017 - link
Specially when every cpu right now autoclocks to 4Ghz on ST tasks. Single thread is just an obsolete metric when just the most basic of tasks will use it, tasks the last thing you will worry is speed, maybe curse about that piece of c*rap not using 80% of you cpu resources.ZeroPointEF - Thursday, August 10, 2017 - link
I would love to see more VM benchmarking on these types of CPUs. I would also love to see how a desktop performs on top of a Server 2016 hypervisor with multiple servers (Windows and Linux) running on top of the same hypervisor.ZeroPointEF - Thursday, August 10, 2017 - link
I should have made it clear that I loved the review. Ian's reviews are always great!I would just like to see these types of things in addition. It seems like we are getting to a point where we can have our own home lab and a desktop all on one machine on top of a hypervisor, but this idea may be my own strange dream.
smilingcrow - Thursday, August 10, 2017 - link
And others would like to know how it works at video editing or as a DAW etc.To add a whole bunch of demanding benchmarks just for HEDT systems is a hell of a lot of work for little return for a site whose main focus is the mainstream.
Try looking at more specialised reviews.
johnnycanadian - Thursday, August 10, 2017 - link
This, please! My TR purchase is hinging on the performance of multiple VMWare VMs all running full-out at least 18 hours per day.Ian, I'd love to see some of your compute-intensive multi-core benches running on a Linux host with Linux-based VMWare VMs (OpenCV analysis, anyone? Send me that 1950x and I'll happily run SIFT and SURF analysis all day long for you :-). I was delighted by the non-gaming benchmarks shown first in this review and hope to see more professional benches on Anand. Leave the gamerkids to Tom's or HardOCP (or at least limit gaming benchmarks to hardware that is built for it): Anandtech has always been more about folks who make their living on HPDC, and I have nothing but the highest respect for the technical staff at this publication.
I don't give a monkey's about RGB lighting, tempered glass cases, 4k gaming or GTAV FPS. How machines like Threadripper perform in a HPC environment is going to keep AMD in this market, and I sincerely hope they prove to be viable.
mapesdhs - Thursday, August 10, 2017 - link
Yes, I was pleased to see the non-gaming tests presented first, makes a change, and at least a subtle nod to the larger intended market for TR.Ian.
pm9819 - Friday, August 18, 2017 - link
Your going to spend a $1000 on cpu but have no clue how it handles the tasks you need it for, smh. As a VMWare customer they will tell you which cpu has been certified to handle a specific tasked. You don't need a random website to tell you that.nitin213 - Thursday, August 10, 2017 - link
Hi IanIt's a great review but i do have some suggestions on the test suite. The test suite for this CPU was not materially different from test suites of many of the other desktop CPUs done earlier. I think it would be great to see some tests which explicitly put to use the multi-threaded capabilities and the insane IOs of the system to test, e.g server hosting with how many users being able to login, virtual machines, more productivity test suites when put together with a multi-GPU setup (running adobe creator or similar) etc. I think a combination of your epyc test suite and your high-end GPU test suite would probably be best suited for this.
Also, for the gaming benchmark, it seemed you had 1080, 1060, rx580 and rx480 GPUs. Not sure if these were being bottlenecked by GPU with differences in framerates being semantic and not necessarily a show of PC strength. Also, Civ 6 AI test suite would a great addition as that really stresses the CPU.
i completely understand that there is only so much that can be done in a limited timeframe typically made available for these reviews but would be great to see these tests in future iterations and updates.