Question Geekbench 6 released and calibrated against Core i7-12700

Kocicak

Senior member
Jan 17, 2019
982
973
136
I got these scores with my OC 13600K (5500+4200MHz):

geek6b.png



EDIT: When running it for the second time, the application crashed in the Clang part. Upped a voltage a bit and then I got a higher score above.
 
Last edited:

gdansk

Golden Member
Feb 8, 2011
1,944
2,270
136
Sounds purpose designed to disadvantage homogeneous multi-core designs.
On the other hand the longer run time as result of larger data sets is good.
And finally they removed AES/XTS score. :)
 

Kocicak

Senior member
Jan 17, 2019
982
973
136
Dammit, it crashed again! Running these various tasks may be good for quick testing of OC stability.

(I hope that the new version is not buggy itself)
 
  • Like
Reactions: scineram

Kocicak

Senior member
Jan 17, 2019
982
973
136
For a reference, these are scores for a stock 13600K (5100+3900MHz) with 2x16GB 6000MHz DDR5 RAM:

geek6 stock.png


Single core bar graph has a space for ten fold improvement, I wonder how many years it will take... :)
 
Last edited:
  • Haha
Reactions: MangoX

Hitman928

Diamond Member
Apr 15, 2012
5,160
7,595
136
The multi-threaded test appears to be lightly threaded by today's standards, outside of the Clang compile test. Many tests don't even hit 25% utilization across the 24 threads. Only the Clang test seems to load above 50%, at least for any significant length of time that I could see it while monitoring.
 

gdansk

Golden Member
Feb 8, 2011
1,944
2,270
136
Here's my 5900x with PPT lowered to ~120W (I don't remember exactly off the top of my head) and slight negative offsets with curve optimizer.

View attachment 76467

Better score than my 5900X (at default PPT too) score of 2237/12242
 

Kocicak

Senior member
Jan 17, 2019
982
973
136
Coz heavy multithreaded would favor AMD!

WOW. Are they being paid off by Intel?
Lightly multithreaded loads favor AMD, because they have all cores large, fewer threads hit weaker E cores on Intel CPUs, but not that many of them to reap the full benefits of them.

And in spite of this 13600K at stock has 88% ST and 77% MT performance of the 7950X at half the price. Impressive value.
 
  • Like
Reactions: controlflow

Justinus

Diamond Member
Oct 10, 2005
3,165
1,507
136
Last edited:

nicalandia

Diamond Member
Jan 10, 2019
3,330
5,281
136
Ligtly multithreaded loads favor AMD, because they have all cores large, fewer threads hit weaker E cores on Intel CPUs, but not that many of them to reap the full benefits of them.

And in spite of this 13600K at stock has 88% ST and 77% MT performance of the 7950X at half the price. Impressive value.
Compare it to this one..

1676399482552.png

 

Hitman928

Diamond Member
Apr 15, 2012
5,160
7,595
136
Ligtly multithreaded loads favor AMD, because they have all cores large, fewer threads hit weaker E cores on Intel CPUs, but not that many of them to reap the full benefits of them.

And in spite of this 13600K at stock has 88% SST and 77% MT of the 7950X at half the price. Impressive value.

This would only be true if Intel's CPUs had only a few P-cores, but they have 6+. From my (admittedly very brief) testing, the vast majority of MT tests only load 6 or less threads.

Edit: After another quick test where I upped my polling rate, I take back this statement somewhat. Most tests seem to use ~12 threads on my CPU with Clang using ~20. I didn't see many of them before because the higher load only shows up for less than a second, which was what my polling rate was set to. There are still a few that seem to only load 2, maybe 4 threads at most. This is still all just estimates with a higher polling rate though. An actual thread utilization measurement across the tests would be better. This is also just reported utilization, I don't know if the sub test performance actually scales or not.
 
Last edited:
  • Like
Reactions: Kaluan and moinmoin