Post your Cpumark99 scores

[Kosusko]

12900KS ECore - 4.0GHz all cores

In the footsteps of Alder Lake - The rise of SMALL cores

https://imgur.com/m8Ju7RW

https://imgur.com/fj8pY85

Only the 4.3GHz "Raptormont" (sorry Gracemont) with 4MB L2 cache per quadcore cluster result is missing...

 

[Hulk]

View attachment 68570

Ryzen 9 7950X bone stock, but using a custom loop.

Running through WINE, though it doesn't appear to impact the score at all... Frequency was 5.88Ghz during most of the test - ran it three times, very little variation.

Now this is quite impressive! 4.82MHz/CPU99mark

 

[Hulk]

 

[looncraz]

Now this is quite impressive! 4.82MHz/CPU99mark

5.97GHz, but some jumping around of the clock... pretty sure this is the limit of this CPU... since I won't apply extra voltage (in fact, this is -12 curve optimizer and 125MHz PBO boost... and nothing at all else... -15CO fails to boot, but does POST, so we're at the limit)

 

[Hulk]

5.97GHz, but some jumping around of the clock... pretty sure this is the limit of this CPU... since I won't apply extra voltage (in fact, this is -12 curve optimizer and 125MHz PBO boost... and nothing at all else... -15CO fails to boot, but does POST, so we're at the limit)
View attachment 68603

Dead link.

 

[looncraz]

Dead link.

Fixed!

 

[Det0x]

Where can i download this bench ?
Found

Zen4 @ 6ghz = 1310 core

 

[looncraz]

Where can i download this bench ?
Found

Zen4 @ 6ghz = 1310 core

I'm going all core and letting the clocks jump around... seems there's a good chunk of performance left on the table doing that...

 

[Hulk]

Thought this was dead but Zen 4 is throwing down some crazy integer performance.





Might as well fully rock the time machine.

 

[AdamK47]

I'm going all core and letting the clocks jump around... seems there's a good chunk of performance left on the table doing that...

I haven't seen Zen 4 6GHz stable. Most likely another suicide run. If you can call a 23 year old single threaded benchmark at unstable clocks a suicide run.

 

[Hulk]

I haven't seen Zen 4 6GHz stable. Most likely another suicide run. If you can call a 23 year old single threaded benchmark at unstable clocks a suicide run.

Really just used the scores to determine MHz/CPUmark99, the few scores posted here corroborated one another nicely actually.

CPUmark99 primarily tests integer performance if I remember correctly. Zen 4, (like K6) seems to be "ahead of their time."

 

[Hulk]

So if I really need to do some CPUmark99 number crunching I can either set up 5,000 Pentium 90's or 1 7950X.

 

[Kosusko]

• on battery > 355
• on charging mode > 379

A12-8830B

AMD A12-8830B (Excavator) 379 vs 474 Intel Pentium Silver N6000 (Tremont) +25%

 

[Hulk]

for what its worth

balanced power plan
INTEL NUC 11 PRO i5

View attachment 68855

Core i5-1135G7 - 766

Could you have a peek at Harware info at the end of the run and see what the effective clock is? I'm curious as that is a strong score even for a desktop.

 

[Hulk]

Looks like 4.2GHz, which is max turbo frequency. Here we go.

 

[Hulk]

during the running of the benchmark the frequency is mostly 3,7 according to HWMon.

oddity: A12-8830b when running on Windows 7 64bit yields only 309points, but system seems much quicker overall.

All of the MHz/CPUmark99 were based on Windows task manager reported clock, which is generally the "ratio" reported by HWinfo, not the actual or effective you noted. In the Handbrake thread I used effective clock for the computation. Since this one is testing just one core and I have a data base going back to 1999 that does not use effective clock I wanted to keep the basis the same for evaluation.

Not sure why your AMD system "feels faster" but as you know there are other factors at work that the old CPUmark99 does not take into account such as hard drive speed, memory subsystem performance (it probably fits in L1 cache!), and the general state of the software in the computer, meaning how far removed from a clean Windows install.

Also and this is probably the biggest one the 1135G7 really ups the clock for single core operation, which is perfect for this benchmark. Not sure how much your Intel laptop gets to that clock in normal use. Just some thoughts on your observation.

 

[Hulk]

@Hulk, sorry for the confusion, i meant the AMD system Laptop (A-Series Thinkpad A275) feels faster onWindows 7 (score of 309) compared to Windows 11 (score of 379) that is probably because there are a lot fewer processes running on Windows 7 on a new installation; but it still keeps me wondering why the score so low, maybe it is just a driver issue.

the NUC 1135G7 NUC feels snappy, is quite fast, was surprised myself.

Gotcha, and your reasoning makes sense to me.

 

[igor_kavinski]

Intel I9 - 11900 (windows 11) - 706

View attachment 68890

That's really low for an i9. This score would make me sad if I owned this CPU. Beaten by a Ryzen 5600.

 

[igor_kavinski]

after BIOS update, singlecore went down, but multicore up;

Could be some vulnerability mitigation update.

 

[Miramonti]

Where we at?

 

[Det0x]

Where we at?

Seems like around 1430 points @ 6069mhz fmax, not really pushing

6069 / 1430 = 4.24 mhz per cpumark99

No gain from running on X3D CCD


Fire up your Arrow Lake machine @igor_kavinski

 

[igor_kavinski]

Fire up your Arrow Lake machine @igor_kavinski

I would love to but the 8200 XMP kit it loves is in my 9950X3D machine. I will have to run the other sticks I have at 7200 or 7600 (depending on which one I am able to find in my drawer). Will try.

 

[igor_kavinski]

Funny score or a bad omen?

 

[igor_kavinski]

The thread is hopped around by the Windows scheduler a lot.

 

[igor_kavinski]

Weird behavior in Win11. Double-clicked the EXE. It didn't open. While I'm checking around in Task Manager to see what could be wrong, the window appears with the score already calculated.



Ran it again.