Myth about CPU power consumption

111alan

Member
Mar 11, 2019
98
29
18
Haerbing Institution of Technology
Trying to evaluate the power consumption to see what frequency the CPU actually runs at in rendering. Here's where I noticed something interesting. The 4.0GHz 9900k actually beats 3700x(4.05G) by more than 20w in rendering. The voltage is auto(default p-state frequency-voltage pair).

Here's the list I collected. Already seen a lot of interesting results, like how taxing high-frequency is.Currently still looking for more data, to compare both designs at a fair level.

Here comes the questions:

1. Is 7nm+Zen2 really better than 14nm+SKL at power consumption, at the same frequency&corecnt?
2. Is it really possible for something like 3990x to run at 3.5GHz all core while doing something as intensive as rendering, as some reviews suggest?

pw.png

Test method: Capture both the per-core frequency and the package power during the second half of Cinebench run, when they're stablized.
Reason for this method:
1. both values are stable at that time.
2. almost perfect cpu-bound(dram bound 0.1%-0.8%).
3. other rendering scenarioes(vray, keyshot, corona) yield almost the same frequency under the same TDP limit.
4. The "package power" is known to be reasonably accurate compared to direct measurement on the +12v input lanes subtract the losses on power rails. It's also the deciding factor of what frequency CPU runs at. For example, 8280L will be throttled to a steady-state frequency where "package power" limit was capped at 205W as shown above.

And yes I/O power is considered. They're quite static in the same design. Also the efficiency of F-IVR should be considered as well(this is the main reason why Xeon fall behind in high-load non-simd situations).
 

Attachments

Last edited:

EffrafaxOfWug

Radioactive Member
Feb 12, 2015
1,395
501
113
What is the myth here? Your figures seem to show the 3700X using ~25W less than the 9900K but I don't see any calculation of perf/Watt. If you've capped the 9900K at 4GHz you should also try turning off boost clocks on the AMDs as well and observe the delta - upper limits of boost clocks tend to be very power-hungry.

However, if you're using the power count from a piece of software it's not a very reliable figure of real-world power draw. Just measuring what the motherboard thinks the power usage of the chip is doesn't tell you much, a power meter on the mains is much more useful.
 

111alan

Member
Mar 11, 2019
98
29
18
Haerbing Institution of Technology
What is the myth here? Your figures seem to show the 3700X using ~25W less than the 9900K but I don't see any calculation of perf/Watt. If you've capped the 9900K at 4GHz you should also try turning off boost clocks on the AMDs as well and observe the delta - upper limits of boost clocks tend to be very power-hungry.

However, if you're using the power count from a piece of software it's not a very reliable figure of real-world power draw. Just measuring what the motherboard thinks the power usage of the chip is doesn't tell you much, a power meter on the mains is much more useful.
The "package power" is relatively reliable. Tried measuring the 8pin power with clip-on ammeter on several motherboards before. The result when converted with drmos' efficiency, the difference is at most single-digit. The board used to test the LGA3647 and AM4 CPUs are confirmed with that method. The unreliable number is the "cpu power" sensor. All desktop motherboards seems to have unlocked the turbo duration.

9900K is 0.65GHz higher at the same core count of course it should consume more power. The myth is, despite the overwhelming advertisement of TSMC 7nm used in Zen2, the power consumption isn't showing any noticable difference at the same frequency. Zen2 even loses as a whole due to the added power from the north bridge.

As for performance it's a different question. But from the testing I've done so far all the results are similiar when converted to same frequency and same core count, with AMD better at something runs with almost no dram bound and Intel better at others.

Oh btw I don't know how some reviews get 3990x to 3.5GHz at 280W tdp range. From the tests I've done it's completely impossible. Hope there are other people who can confirm these findings.
 
Last edited: