Frequency, 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.


Click to zoom

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.

The 64 Core Threadripper 3990X CPU Review The Windows and Multithreading Problem (A Must Read)
Comments Locked

279 Comments

View All Comments

  • PeachNCream - Monday, February 10, 2020 - link

    Computer viruses predate Lotus 1-2-3.
  • FunBunny2 - Tuesday, February 11, 2020 - link

    the point is: 1-2-3 brought the effort mainstream, by showing how DOS was just a sieve to the hardware. recall that the PC with DOS was only one of three OS available, and PC sales didn't matter much until Corporate figured out that they just had to have 1-2-3. Mitch made Bill rich, not Bill. until 1-2-3, M$ was a legit systems software maker. after that, not so much. Xenix was they're OS of the future.
  • FunBunny2 - Tuesday, February 11, 2020 - link

    ... and, for the PC, not according to this history: https://content.sentrian.com.au/blog/a-short-histo...
    "The first computer virus for MS-DOS was “Brain” and was released in 1986. It would overwrite the boot sector on the floppy disk and prevent the computer from booting. It was written by two brothers from Pakistan and was originally designed as a copy protection."

    learned how to do that from 1-2-3
  • Khenglish - Friday, February 7, 2020 - link

    Here's Unigine Heaven software rendered:

    https://i.imgur.com/0dfV4pd.png
    https://i.imgur.com/CEWhX31.png

    Fun fact: turning on tessellation drops fps by a factor of about 20.
  • Spunjji - Monday, February 10, 2020 - link

    Holy cow, I had no idea.

    I'd be interested (as a purely theoretical exercise) to see where the ideal performance balance of cores / clock speed / memory bandwidth falls when it comes to software rendering.
  • GreenReaper - Saturday, February 8, 2020 - link

    They use DirectX on Windows, and then Microsoft provides the fallback renderer.
  • Mikewind Dale - Friday, February 7, 2020 - link

    That might actually be an interesting test for someone who wants to run legacy games that don't support newer versions of Windows, DirectX, and/or don't have graphics driver support.

    For example, I was trying to play the original Diablo before the GoG version came out. It didn't work on my Radeon RX580, so I had to set up a VMWare Workstation virtual machine, with 3D acceleration support. However, even though VMWare Workstation supports 3D acceleration, it's still using my CPU, not my GPU. It's just that the virtual OS has software DirectX acceleration.

    Anyway, I benchmarked 3DMark2001 SE running in a Window XP virtual machine on my 8-core Ryzen 7 2700X. I actually got scores that were competitive with GPUs from the early 2000s. So my software 3D acceleration on a Ryzen 7 2700X was approximately the same speed as a GPU from circa 2001.

    It would be interesting to see how well a 64 core processor does.
  • Khenglish - Friday, February 7, 2020 - link

    I get 5947 with a 3920xm (full 4c/8t ivb with 8MB cache) at 4.3 GHz. I would expect your 2700x to be a bit more than double that.

    https://i.imgur.com/aeQcFuu.png
  • Mikewind Dale - Saturday, February 8, 2020 - link

    I'm getting about 6800. So perhaps the VMWare Workstation software display device cannot fully take advantage of parallelization?
  • Spunjji - Monday, February 10, 2020 - link

    That or it's not as efficient as Microsoft's software layer at translating DirectX code into something that can run on the CPU. If you had the time, you could try running 3DMark 2001 natively on your system the way Khenglish is and see if there's a difference.

Log in

Don't have an account? Sign up now