Methodology: how we measure power draw
Not only maximum speed, but also maximum efficiency among AMD’s single-chiplet CPUs. That’s the essential characteristic of the Ryzen 7 9700X. While the speed difference from the last generation (and the Ryzen 7 7700X processor) is negligible, zero, or even negative in places, it comes with significantly lower power consumption. And for those who don’t appreciate it, BIOSes with higher TDP are available less than a month from release.
Methodology: how we measure power draw
Measuring CPU power consumption is relatively simple, much easier than with graphics cards. All power goes through one or two EPS cables. We also use two to increase the cross-section, which is suitable for high performance AMD processors up to sTR(X)4 or for Intel HEDT, and in fact almost for mainstream processors as well. We have Prova 15 current probes to measure current directly on the wires. This is a much more accurate and reliable way of measuring than relying on internal sensors.
The only limitation of our current probes may be when testing the most powerful processors. These already exceed the maximum range of 30 A, at which high accuracy is guaranteed. For most processors, the range is optimal (even for measuring a lower load, when the probes can be switched to a lower and more accurate range of 4 A), but we will test models with power consumption over 360 W on our own device, a prototype of which we have already built. Its measuring range will no longer be limiting, but for the time being we will be using the Prova probes in the near future.
The clamps are properly reset before each measurement and connected to a Keysight U1231A multimeter, which samples the current values during the tests via the IR-USB interface and logs them to a spreadsheet at one-second intervals. From this we can then create line graphs of power draw curves. However, we always write the average values in the bar graphs. Measurements are taken in different load modes. The lowest one is idle Windows 10 on the desktop. This measurement takes place on a well “cooled down” system.
Audio encoding (FLAC) represents a higher load, but processors use only one core or one thread for this. Higher loads, where more cores are involved, are games. We test power consumption in F1 2020, Shadow of the Tomb Raider and Total War Saga: Troy in 1920 × 1080 px. In this resolution, the power consumption is usually the highest or at least similar to that in lower or higher resolutions, where in most cases the CPU power draw rather decreases due to its lower utilization.
Like most motherboard manufacturers, we too ignore the time limit for “Tau”, after which the power consumption is to be reduced from the PL2 boost limit (when it exceeds the TDP) to the TDP/PL1 value, recommended by Intel, in our tests. This means that neither the power draw nor the clock speed after 56 seconds of higher load does not decrease and the performance is kept stable with just small fluctuations. We had been considering whether or not to respect the Tau. In the end, we decided not to because the vast majority of users won’t either, and therefore the results and comparisons would be relatively uninteresting. The solution would be to test with and without a power limit, but this is no longer possible due to time requirements. We will pay more attention to the behavior of PL2 in motherboard tests, where it makes more sense.
We always use motherboards with extremely robust, efficient VRM, so that the losses on MOSFETs distort the measured results as little as possible and the test setups are powered by a high-end 1200 W BeQuiet! Dark Power Pro 12 power supply. It is strong enough to supply every processor, even with a fully loaded GeForce RTX 3080, and at the same time achieves above-standard efficiency even at lower load. For a complete overview of test setup components, see Chapter 5 of this article.
- Contents
- AMD Ryzen 7 9700X in detail
- Methodology: performance tests
- Methodology: how we measure power draw
- Methodology: temperature and clock speed tests
- Test setup
- 3DMark
- Assassin’s Creed: Valhalla
- Borderlands 3
- Counter-Strike: GO
- Cyberpunk 2077
- DOOM Eternal
- F1 2020
- Metro Exodus
- Microsoft Flight Simulator
- Shadow of the Tomb Raider
- Total War Saga: Troy
- Overall gaming performance
- Gaming performance per euro
- PCMark and Geekbench
- Web performance
- 3D rendering: Cinebench, Blender, ...
- Video 1/2: Adobe Premiere Pro
- Video 2/2: DaVinci Resolve Studio
- Graphics effects: Adobe After Effects
- Video encoding
- Audio encoding
- Broadcasting (OBS and Xsplit)
- Photos 1/2: Adobe Photoshop and Lightroom
- Photos 2/2: Affinity Photo, Topaz Labs AI Apps, ZPS X, ...
- (De)compression
- (De)encryption
- Numerical computing
- Simulations
- Memory and cache tests
- Processor power draw curve
- Average processor power draw
- Performance per watt
- Achieved CPU clock speed
- CPU temperature
- Conclusion
Article hint: AMD ryzen cpus have taken power also from atx24pin for some non-core rails. iirc memory controller. This is basicly ignored by all reviewers, by hwinfo, by amd ryzen master, by cpu’s own power tracking. I do not know if this is still a case in am5 socket. This has made amd look much more efficient than it has been. Could you do investigating test?
We had measurements of the ATX connector in the in-depth tests of motherboards, but we eventually removed it from them. From my point of view, it didn’t provide all that useful information that would have been helpful for the evaluation. What materials say that the processors are partially powered from the 24-pin connector? Personally, it doesn’t make much sense to me (for something in modern CPUs to be powered by such weak wires), even from the EMC point of view. Rather, I’m worried that it could lead to possibly unnecessary instability. But that’s just a feeling, a layman’s view.
Then there is the other thing, namely that there are other devices on each rail of the ATX connector. For example, PCIe slots (and typically a graphics card) on 12 V, DDR5 memory on 5 V, and 3.3 V should be used to power M.2 SSDs? Well, it probably doesn’t have to be on all motherboards (some may use VRM to change higher voltage from another rail?), but even if we have information about the current drawn through the 24-pin ATX connector, it will be quite difficult to separate which part of it belongs to which device/component within the motherboard. Or? How would you design a methodology for such a test?
Trace out where from VDDIO_MEM3 pins get their power. https://cdn.hackaday.io/files/1733807417889920/AM4%20Pinout%20Diagram.pdf
Thanks for the very nice diagram. When there’s space, we’ll try to study it. In any case, I’m worried about how this would be handled, since the power supply from the 24-pin wires is shared for multiple devices on each rail… I can’t think of a way to separate the devices. Then, with the help of a tool, you can also measure the current directly on the pins of the socket, but this can probably distort the characteristics of the processor as such to a certain extent.
Many reviewers publish “power-at-wall” figures instead of cpu power. In some sense, it is a more relevant measure. What I can remember, Ryzens tend to be more efficient than intels when measured at the outlet too.
Maybe take a closer look one day and compare the power efficiency according to the different measures? Do those mostly agree or not?
Intel is more efficient at idle and low load use **if** a system (firmware/bios) has powersavings configured correctly. AMD is more efficient at full load use – and by a lot.
” **if** a system (firmware/bios) has powersavings configured correctly. AMD is more efficient at full load use”
you can apply the same logic here too: if you powerlimit an intel down to the level of amd it will also be more efficient, for example the apparently most efficient 7800X3D got 17492 points in CB R23 at 83.78W@EPS. My 14700K when limited to 40W got 19266 points, so even if the VRMs wasted 50% it would be still more efficient at that level
Yes, we’ll definitely be looking at the relationship of isolated measurements (only CPU and motherboard VRMs on EPS cables) and system power consumption at some point. It’s a very complex issue. Most reviewers probably measure system power consumption mainly because it’s technically easier to use, but it doesn’t take into account, for example, that different equipped boards have different power consumption of components not related to the CPU per se. When judging the measured values based on system power consumption, it is also important to note that with different CPUs the power consumption of the same graphics card may be different, which is also one of the factors that distort the results. Personally, I find it useful to eliminate these factors. However, it may be interesting to investigate the dependency of system power consumption and isolated power consumption (purely CPU), if in both cases a larger number of model situations with different motherboards and different graphics cards are created.
— „What I can remember, Ryzens tend to be more efficient than intels when measured at the outlet too.“
They are arguably more effective in our tests as well, aren’t they? 🙂
Whether it’s relevant depends on what we want to compare and evaluate. If this was really with a 125W power limit https://www.hwcooling.net/wp-content/uploads/2024/08/gigabyte-b650e-aorus-pro-x-usb4-g262.html it suggests the differences in VRM efficiency can be quite huge. So it would be relevant with regards to motherboards, but deceptive when comparing cpus.