AMD Threadripper 2990WX 32-Core and 2950X 16-Core Review • Hacker News Board

Re: AMD Threadripper 2990WX 32-Core and 2950X 16-Core Review

#121

Post by floatboth » Mon, Aug 13, 2018, 6:00 PM UTC

mort96 wrote: post #36

karangoeluw wrote: post #2

Honest q: What home-use workstation would use 32 cores? (Excluding home labs or servers).

I'm literally sitting here waiting for glibc to compile, because I need a version with debug symbols, which the version from Arch Linux' repos lacks. Right before that, I compiled valgrind from the git head, because the current release (3.13) doesn't support glibc 2.28. I have compiled Chromium a couple of times for work. It's very apparent that my 2-core 4-thread i5 5200U is a bit too weak; I'd love to be using a 16…

8-core Ryzen is a huge upgrade from an Intel U and doesn't cost nearly as much as 16- and 32-core machines

Re: AMD Threadripper 2990WX 32-Core and 2950X 16-Core Review

#122

Post by andrewbinstock » Mon, Aug 13, 2018, 6:00 PM UTC

karangoeluw wrote: post #2

Honest q: What home-use workstation would use 32 cores? (Excluding home labs or servers).

One niche application is music composition. When you're writing a score for full orchestra, you need lots of RAM and lots of cores for accurate playback.

Re: AMD Threadripper 2990WX 32-Core and 2950X 16-Core Review

#123

Post by deng » Mon, Aug 13, 2018, 6:05 PM UTC

jdietrich wrote: post #86

deng wrote: post #69

Earlier quoted context omitted.

Not my experience. I've had stutters with Ableton Live on a 4 core machine with 2 cores still idling around. Ableton cannot multi-thread a single track (at least in Version 9), and if you're using a single-threaded VST that does not matter anyway.

If a single instance of a VST can use 100% of a thread, you're using a woefully underpowered processor or a ludicrously inefficient plugin. Many composers regularly work on projects with hundreds of tracks and thousands of plugin instances. Projects of that scale used to require multiple computers and a bunch of DSP accelerator cards, but they're now entirely feasible on one high-end workstation.

It's not hard to find VSTs which will easily max out an i7, especially if they run on Max/MSP. Also, stuff from u-he, like Diva. If you run several instances on high-quality you'll usually need to start freezing tracks.

Re: AMD Threadripper 2990WX 32-Core and 2950X 16-Core Review

#124

Post by sandworm101 » Mon, Aug 13, 2018, 6:17 PM UTC

ReverseCold wrote: post #70

sandworm101 wrote: post #56

Earlier quoted context omitted.

>> Folklore would have it that single-thread performance is the most important aspect of desktop performance There are a few places where it matters. We talk of "gaming" but that is too broad a term imho. Many AAA games do leverage AMD chips well. They know how to split loads across threads to maximize performance. But in recent years the hot games have come from indi developers. Pre-release titles from tiny develope…

KSP 1.1 finally added multithreading btw

When launching a rocket, that rocket's physics engine is still on a single thread. Given that the physics engine is the most CPU-intensive part of KSP, I'd say that KSP's support of multithreading is nominal at best.

Multithreading aside: KSP's approach to physics also doesn't help them. Rather than make creative decisions re what should and shouldn't be calculated, they just hammer the physics thread with lots of calculations that have zero impact on the user experience. In a larger dev these would be optimized away.

Re: AMD Threadripper 2990WX 32-Core and 2950X 16-Core Review

#125

Post by gavanwoolery » Mon, Aug 13, 2018, 6:18 PM UTC

I wonder how many of these tests (if any) are limited by bad multithreading patterns (see: https://www.arangodb.com/2015/02/comparing-atomic-mutex-rwlo...).

Re: AMD Threadripper 2990WX 32-Core and 2950X 16-Core Review

#126

Post by xvf22 » Mon, Aug 13, 2018, 6:42 PM UTC

wtallis wrote: post #99

vesrah wrote: post #96

Earlier quoted context omitted.

You think they'd either pull the review or place a disclaimer at the start about this, but I agree with what you're thinking (throttling).

Ian re-ran all of the affected tests after discovering the mistake, but he also kept the data for later analysis. Those results aren't in the review yet.

Fair enough, mistakes happen but the comment didn't make it very clear which benchmarks were included.

"Me being an idiot and leaving the plastic cover on my cooler, but it completed a set of benchmarks. I pick through the data to see if it was as bad as I expected"

Re: AMD Threadripper 2990WX 32-Core and 2950X 16-Core Review

#127

Post by snuxoll » Mon, Aug 13, 2018, 6:47 PM UTC

Coding_Cat wrote: post #119

deng wrote: post #54

This is the perfect chip for continuous, low-memory number crunching. For everything else... not so much. I mean, this chip consumes 74W when idle, of which almost 90% are spend on the interconnect. That's insane. Most important bit in the review for me: "After core counts, the next battle will be on the interconnect. Low power, scalable, and high performance: process node scaling will mean nothing if the interconnec…

Why do you say low-memory? I got the opposite feeling. I've been drooling over EPYC and TR2 ever since they released the specs for my memory bandwidth limited projects.

The NUMA configuration on TR2 results in two 8-core CCX's without direct memory access, so you still "only" have four memory channels and half the cores have to hop over the infinity fabric to access any of them.

Re: AMD Threadripper 2990WX 32-Core and 2950X 16-Core Review

#128

Post by snuxoll » Mon, Aug 13, 2018, 6:51 PM UTC

znpy wrote: post #91

Hasknewbie wrote: post #41

Earlier quoted context omitted.

"this review shows that the 16-core 2950X compiles Chromium faster than the 32-core 2990WX, presumably this means something other than the thread count becomes a bottleneck after 16 threads" The article mentions that, due to the die packaging, only 16 of the cores have direct access to RAM. So for the 32-core version, half the cores are memory-starved and have to go through the 'connected' cores (also impacting these…

Might memory access model (UMA vs NUMA) play a role here? AFAIK the TR2950wx has configurable model (can be configured to work in either uma or numa mode) whereas the 2950x only has one mode (can't recall which one at the moment)

It's the opposite, the 2950X can be configured in (fake-)UMA ("distributed" mode in AMD's terms) or NUMA mode but the WX chips are NUMA only.

Re: AMD Threadripper 2990WX 32-Core and 2950X 16-Core Review

#129

Post by celrod » Mon, Aug 13, 2018, 6:52 PM UTC

CarVac wrote: post #53

srcmap wrote: post #40

Two additional benchmarks I like to see from Anandtech: 1) some VM base tests for these kind of CPU - 32, 64, 128 VM all running some kind of web/db/redis benchmarks inside. 2) Some compilation testing - time the clean build of AOSP, BSD, some very complex linux app - use max jobs setting for parallel compile and time how long it take to finish the jobs. ( measure the over CPU usages at the same time. )

Phoronix had it compiling the Linux kernel in 32 seconds, compared to 37.5 for the 7980XE.

Wow. I wonder how fast it could do a full bootstrapped build of gcc...

Re: AMD Threadripper 2990WX 32-Core and 2950X 16-Core Review

#130

Post by twoodfin » Mon, Aug 13, 2018, 6:57 PM UTC

gsnedders wrote: post #60

deng wrote: post #44

Earlier quoted context omitted.

They do a compilation of Chrome in the "Office Tests" section. The 32Core takes about an hour per compile, but the 16Core is actually a bit faster (my guess would be that linking takes longer on the 32 core machine as it's very memory intensive).

FWIW, I believe the version they use relies on a version of MSVC where the linker is predominently single-threaded, and linking the final binary ends up as a significant proportion of the time.

Does anyone do multithreaded LTO yet?