The 64 Core Threadripper 3990X CPU Review: In The Midst Of Chaos, AMD Seeks Opportunity
by Dr. Ian Cutress & Gavin Bonshor on February 7, 2020 9:00 AM ESTFrequency, Temperature, and Power
A lot of questions will be asked about the frequency, temperature, and power of this chip: splitting 280W across all the cores might result in a low all-core frequency and require a super high current draw, or given recent reports of AMD CPUs not meeting their rated turbo frequencies. We wanted to put our data right here in the front half of the review to address this straight away.
We kept this test simple – we used our new NAMD benchmark, a molecular dynamics compute solver, which is an example workload for a system with this many cores. It’s a heavy all-core load that continually cycles around the ApoA1 test simulating as many picoseconds of molecular movement as possible. We run a frequency and thermal logger, left the system idle for 30 seconds to reach an idle steady state, and then fired up the benchmark until a steady state was reached.
For the frequencies we saw an ‘idle’ of ~3600 MHz, which then spiked to 4167 MHz when the test began, and average 3463 MHz across all cores over the first 6 minutes or so of the test. We saw a frequency low point of 2935 MHz, however in this context it’s the average that matters.
For thermals on the same benchmark, using our Thermaltake Riing 360 closed loop liquid cooler, we saw 35ºC reported on the CPU at idle, which rose to 64ºC after 90 seconds or so, and a steady state after five minutes at 68ºC. This is an ideal scenario, due to the system being on an open test bed, but the thing to note here is that despite the high overall power of the CPU, the power per core is not that high.
This is our usual test suite for per-core power, however I’ve condensed it horizontally as having all 64 cores is a bit much. At the low loads, we’re seeing the first few cores take 8-10W of power each, for 4.35 GHz, however at the other end of the scale, the CPUs are barely touching 3.0 W each, for 3.45 GHz. At this end of the spectrum, we’re definitely seeing AMD’s Zen 2 cores perform at a very efficient point, and that’s even without all 280 W, given that around 80-90W is required for the chipset and inter-chip infinity fabric: all 64 cores, running at almost 3.5 GHz, for around 200W. From this data, we need at least 20 cores active in order to hit the full 280W of the processor.
We can compare these values to other AMD Threadripper processors, as well as the high-end Ryzens:
AMD Power/Frequency Comparison | |||||||
AnandTech | Cores | CPU TDP | 1-Core Power |
1-Core Freq |
Full Load Power/core |
Full Load Freq |
|
3990X | 64 | 280 W | 10.4 W | 4350 | 3.0 W | 3450 | |
3970X | 32 | 280 W | 13.0 W | 4310 | 7.0 W | 3810 | |
3960X | 24 | 280 W | 13.5 W | 4400 | 8.6 W | 3950 | |
3950X | 16 | 105 W | 18.3 W | 4450 | 7.1 W | 3885 |
The 3990X exhibits a much lower power-per-core value than any of the other CPUs, which means a lower per-core frequency, but it isn’t all that far off at all: less than half the power for only 400 MHz less. This is where the real efficiency of these CPUs comes into play.
282 Comments
View All Comments
ballsystemlord - Saturday, February 8, 2020 - link
Thanks for the article Ian and Gavin!I found no spelling or grammar errors! Reply
Railgun - Saturday, February 8, 2020 - link
So when will the growing backlog of benchmarks be posted into bench? ReplyIan Cutress - Friday, February 14, 2020 - link
They should be in Bench. If not, drop me an email. Replydarealist - Saturday, February 8, 2020 - link
$4000 to ripoff their loyal fanbase. All the shillz be liek "it's a steal!" while typing on their 1600x build ROFL. Replylevizx - Saturday, February 8, 2020 - link
what a stupid dipshit ReplySpunjji - Monday, February 10, 2020 - link
*facepalm*It *is* a steal - for a 64 core CPU.
I can't afford one, I'd never have any use for one, and I don't think anyone who qualifies as AMD's "loyal fanbase" would either. It's basically an industrial tool - the people who need this will buy it based on that need. Reply
StuntFriar - Saturday, February 8, 2020 - link
While it's a little specific, it would be cool to benchmark some Unreal Engine 4 game developer workflows, such as doing a full rebuild/repackage of a game (for Windows, Android, iOS and consoles), rebaking the lighting for a level, importing assets, etc...I'm suggesting UE4 because Epic already has a bunch of freely available demo projects (some are graphical showcases, others are actual playable games that will pass certification on some consoles with a little work) so it's easy to set up a test that other folks can try at work for themselves - which would make it far easier to decide if a CPU upgrade would be worth it.
For fun, you could even do the same tests on Windows, MacOS and Linux to see if there's a tangible difference between operating systems (though the vast majority of developers would be using Windows regardless).
The UE4 Editor seems to be highly parallel in most of its building/compiling tasks and I do wonder they scale up proportionately past 16 cores.
Probably worth doing some Unity Engine benchmarks too since that's the most popular engine on the planet. Haven't used it in over a year, but it seemed to favour higher single-threaded performance for a lot of the building and asset import tasks. But again, it's fairly easy to set up benchmarks that users can replicate at work.
Cheers. Reply
Betonmischer - Saturday, February 8, 2020 - link
Hi Ian! I'd like to chime in on the difference between the Pro and the Enterprise versions of Windows 10 in regards to 128 thread management. Are you absolutely sure that your Pro test system is up to date? I see 2 sockets on the screenshot, which shouldn't happen on either version. Here's the picture of what it looks like on my colleague's test bench. It's Windows 10 Pro, and it's detecting a 128-thread CPU as a single socket. We found no impact on performance either, including the benchmarks that you specifically listed on page 3.https://imgur.com/G2VqgoU Reply
realbabilu - Saturday, February 8, 2020 - link
Since this is targeted to very segmented market like render farm. A bench single cpu tr4 3990x vs clustered cheaper several ryzen 3950x will be fascinating. Replymsroadkill612 - Monday, February 10, 2020 - link
I sometimes fantasise about clusters/arrays of Renoir apuS, each w/ a 2TB NVME of "edge" data, for rendering and AI.What do folks think? Reply