Discussion PES | Assessing Power and Performance Efficiency of x86 CPU architectures

Page 7 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

BorisTheBlade82

Senior member
May 1, 2020
660
1,003
106
Dear Community,

so this is my first thread here as a long-time lurker - but I felt the desire to share a small hobby-project of mine from the last couple of months with you...

Performance Efficiency Suite - What is it about?
Most Reviewers solely focus on what they consider to be the most important aspect of modern CPUs - the absolute performance. But this is only one side of the equation. Today Power Efficiency is at least as important - or to be more precise: The amount of energy (Wattseconds or Joules) a CPU needs in order to accomplish a given workload. Sadly most Reviewers shy away from the extra mile it needs to assess this aspect. This suite measures the Total Package Power of a CPU while running the Cinebench R23 benchmarks first in single-threaded mode (1 run), then running in multi-threaded mode (for 10 minutes + whatever it takes to finish the last run). The results will be rendered in the provided Results.xlsx Excel file. To combine Efficiency and Performance there is also a score provided called Performance Efficiency Score (how amazingly inspired I am ;)).

In the meantime I was able to aggregate more than 80 samples from members of the 3DC & CB communities (see below).

How-To
  1. Unzip the latest release to wherever you want EXCEPT on your local OneDrive folder.
  2. Open Settings.txt and insert your local Cinebench23 Directory.
  3. Run PES Start - it will ask for Administrator rights as these are needed for measuring Package Power
  4. Wait until the Powershell finishes.
  5. Open the Excel file...
  6. Allow external connections (to the generated CSV-files with the data)
  7. Go to Data -> Refresh all
  8. Enjoy and share your results - just take a screenshot of what the Excel renders.
  9. If you want to do multiple measurements with different settings just copy the Excel file (inside the root-folder) before running and refreshing the data.

Some explanations about the Suite
  • This Suite has been made possible by Michael Möller and his amazing free and open-source Open Hardware Monitor and his .NET Library OpenHardwareMonitorLib.dll - Thanks a lot!!!
    Homepage: https://openhardwaremonitor.org/
    GitHub: https://github.com/openhardwaremonitor
  • The results for the Package Power look pretty accurate compared to the sparse data the internet provides. Seems, that the vendors are much more honest with those sensors than they are with temperature etc.
  • The suite basically consists some powershell scripts and an Excel file for presentation purposes
    • RunAsAdminWrapper.ps1
      This is needed to have a convenient relative path shortcut in the root folder and request admin-rights at the same time
    • Main.ps1
      • After setting up some stuff it basically starts the Cinebench R23 one at a time. It then checks for the "Cinebench.exe" process being active.
      • While this is true it queries the Package Power Sensor data with a lower bound of 10ms (in order to keep CPU-load of the script at bay).
      • After each run the aquired data gets pushed to CSV files located in the LogCsv subfolder.
    • Results.xslx
      • The Excel file basically just does some import, calculations and a hopefully nice presentation of the data.
      • Histogram
        The bold line shows a running average of the last 100 data-points which should be sufficiently accurate. The pale line shows each single data-point.
      • Calculation of Total Package Consumption
        To get that number we need the integral. That is why we first calculate the timeframe between two data-points and then multiply the measured value.
      • Everything else in that Excel is hopefully more or less self-explaining

Online Resources

Disclaimer
I am by no means a Powershell professional or a professional Reviewer. I was just sick of the lack of information and wanted to propose a low-effort solution. Any input for further improvement is highly welcomed. Please feel free to use/extend/rip-off this solution as you wish. But please share your findings to the world.
 
Last edited:

Det0x

Golden Member
Sep 11, 2014
1,027
2,953
136
@Det0x
Might you do me a favor and try this LibreOffice Results file:
https://github.com/BrsVgl/PerformanceEfficiencySuite/raw/main/Results LibreOffice.ods

Just replace the existing one in your folder and try if it works.
Hint: For me it takes a while to import the data, but generally it works.
Yes that actually worked, can view the results myself now :blush:

Btw are you sure its correct to judge/rank results by runtime ?(duration)
Have you tried to compare time vs score ? Does it line up ?

Also alittle surprised by the results.. This is how a high scoring run from me look like:

Can speculate some of the benchmarks have a set runtime like 10sec, and must be judged by the score they achieve in those 10seconds, not the fact they ran for 10sec, if you understand what i mean :)
 

BorisTheBlade82

Senior member
May 1, 2020
660
1,003
106
@Det0x
Great, that the LibreOffice sheet works for you. So I will include that in the next minor release.

As to the nature of GB5: I had always assumed that it is a fixed Workload just like CB. Maybe I am wrong - but I don't know how to prove either way.
But yes, time vs. score at least has the same tendency.
 

BorisTheBlade82

Senior member
May 1, 2020
660
1,003
106
@igor_kavinski
It is technically not possible to run PES on Apple Silicon. The CB numbers for M1 were estimates, look here:

For GB5 so far only @Det0x has submitted any results which makes it hard to realize if the methodology works in general.
 

BorisTheBlade82

Senior member
May 1, 2020
660
1,003
106
So, regarding the nature of the GB5 workloads there is some documentation from Geekbench which can be found here:
  • Some of the explanations are not totally clear and leave some room for speculation.
  • The SQLite workload seems to be the only one which has no fixed processing amount but instead measures throughput in a fixed amount of time.
  • When only measuring the duration, the weighting for the score does not apply. So for example AVX512 supporting CPUs should have an advantage. Generally, a CPU that excels in a bigger workload should have an advantage over another CPU which excels in a smaller one.
What do you think? I guess, we first need more samples in order to judge if the methodology makes sense for GB5.
Another possibility would be to parse the results from the website - but that ATM exceeds my programming skills, to be honest. But hey, as this is open source, feel free to contribute. The GitHub repo can be found in the opening post.
 
Jul 27, 2020
15,741
9,809
106
Dell Optiplex 3000 Core i5-12500

Geekbench 5

Stock
1673108603820.png
Core memory isolation On
1673108700518.png
HT off Core memory isolation On
1673108751463.png
HT off Core memory isolation off
1673108814172.png

Windows build 22621.931

Looks like GB5 loves Alder Lake or ADL is very efficient when running GB5.

RAM is 8GB DDR4-3200 at 2T

1673109047919.png
 
  • Like
Reactions: lightmanek

BorisTheBlade82

Senior member
May 1, 2020
660
1,003
106
  • Updated Rankings and Matrices in the opening posts
  • included the results of @igor_kavinski 's mobile and Desktop Alder Lake (very strong showing of the mobile SKU in CB23 ST)
  • some more results for GB5 now available - for the moment I consider GB5 integration Beta
  • minor update to v0.8.1 - hopefully improved the LibreOffice sheet. Excel still working like a charm.
    Download v0.8.1
This is how the GB5 ranking looks right now - more contribution highly appreciated.
GB5_PES.png
 
  • Like
Reactions: lightmanek

mmaenpaa

Member
Aug 4, 2009
78
136
106
DELL PM3571 I9-12900H/FHD/32GB/512SSD/RTXA2000
- Dell Mobile Precision Workstation 3571
- Intel Core i9-12900H vPro (24 MB cache, 14 cores, 20 threads, 2.50 GHz to 5.00 GHz, 45W)
- 32 GB, 2 x 16 GB, DDR5, 4800Mhz, Non-ECC SODIMM
- 6 cell, 97 WHr standard lifecycle battery

AC:
PM3571_12900HCBR23_AC.jpgPM3571_12900HGB5_AC.jpg

DC:
PM3571_12900HCBR23_DC.jpgPM3571_12900HGB5_DC.jpg
 
  • Like
Reactions: BorisTheBlade82
Jul 27, 2020
15,741
9,809
106
Thanks! Sorry, forgot to mention the system.

That's Core i7-12700K desktop CPU working on the ASROCK PG SONIC Z790 mobo with EXPO RAM DDR5-5600@DDR5-4600 28-30-30-60 timings (can't coax that kit to go any higher in bandwidth).

Using default clocks but mobo power limit is at max 265W (doesn't allow higher than that) with an air cooler.