The AMD Ryzen Threadripper 1950X and 1920X Review: CPUs on Steroids
by Ian Cutress on August 10, 2017 9:00 AM ESTCPU Legacy Tests
Our legacy tests represent benchmarks that were once at the height of their time. Some of these are industry standard synthetics, and we have data going back over 10 years. All of the data here has been rerun on Windows 10, and we plan to go back several generations of components to see how performance has evolved.
All of our benchmark results can also be found in our benchmark engine, Bench.
3D Particle Movement v1
3DPM is a self-penned benchmark, taking basic 3D movement algorithms used in Brownian Motion simulations and testing them for speed. High floating point performance, MHz and IPC wins in the single thread version, whereas the multithread version has to handle the threads and loves more cores. This is the original version, written in the style of a typical non-computer science student coding up an algorithm for their theoretical problem, and comes without any non-obvious optimizations not already performed by the compiler, such as false sharing.
CineBench 11.5 and 10
Cinebench is a widely known benchmarking tool for measuring performance relative to MAXON's animation software Cinema 4D. Cinebench has been optimized over a decade and focuses on purely CPU horsepower, meaning if there is a discrepancy in pure throughput characteristics, Cinebench is likely to show that discrepancy. Arguably other software doesn't make use of all the tools available, so the real world relevance might purely be academic, but given our large database of data for Cinebench it seems difficult to ignore a small five minute test. We run the modern version 15 in this test, as well as the older 11.5 and 10 due to our back data.
x264 HD 3.0
Similarly, the x264 HD 3.0 package we use here is also kept for historic regressional data. The latest version is 5.0.1, and encodes a 1080p video clip into a high-quality x264 file. Version 3.0 only performs the same test on a 720p file, and in most circumstances the software performance hits its limit on high-end processors, but still works well for mainstream and low-end. Also, this version only takes a few minutes, whereas the latest can take over 90 minutes to run.
The 1950X: the first CPU to score higher on the 2nd pass of this test than it does on the first pass.
Facebook
Twitter
RSS
347 Comments
View All Comments
launchcodemexico - Thursday, August 10, 2017 - link
Why did you end all the gaming review sections with something like "Switching it to Game mode would have made better numbers..."? Why didn't you run the benchmarks in Gaming mode in the first place?Ian Cutress - Thursday, August 10, 2017 - link
Gaming mode is not default, and we run gaming mode alongside the default - there's two sets of values in each graming test.DanNeely - Thursday, August 10, 2017 - link
You might want to call that out more clearly in the text. I also missed that you have two sets of 1950X results; and probably wouldn't've figured out what the -G suffix meant without a hint.Ian Cutress - Thursday, August 10, 2017 - link
I mentioned it in the Game vs Creator mode page, but I'll propagate it through.lordken - Thursday, August 10, 2017 - link
read before you complain, it is stated at beginning of the review that -G is for game mode...DanNeely - Thursday, August 10, 2017 - link
Especially during the work day a lot of people just are doing quick glances at the most interesting parts. I'll end to end read it sometime tonight.mapesdhs - Thursday, August 10, 2017 - link
If people quick-glance, that's their problem for missing key info. :D When learning about something as new as this, I read everything. Otherwise, it's like the tech equivalent of crossing a road while gawping at a phone. :}Last time I read so much about a new CPU launch was Nehalem/X58.
Ian.
smilingcrow - Thursday, August 10, 2017 - link
It seemed really clear to me but for people who didn't read the long text on NUMA etc maybe not.The dangers of skimming!
mapesdhs - Friday, August 11, 2017 - link
Indeed. :D Reminds me of when a long time ebay seller told me that long item decriptions are pointless, because most bidders only read the first paragraph, often only the first sentence.Ian Cutress - Thursday, August 10, 2017 - link
The test suite is a global glove: rather than have 20 tests for each segment, it's a global band of 80 tests for every situation. Johan does different tests as his office is several hundred miles away from where I am (and we're thousands of miles away from any other reviewer).For the gaming benchmarks, there are big differences in 99th percentile frame rates and Time Under analysis. As games become more and more GPU bottlenecked for average frame rates, this is where the differentiation point is. It's a reason why we still test 1080p as well. With regards the AI test, I've asked the Civ team repeatedly to make the AI test accessible from the command line so I can rope it into my testing scripts easily (they already do it with the main GPU test). But like many other game studios, getting them to unlock a flag is a frustrating endeavor when they don't even respond to messages.