• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

CPUZ 1.73 Benchmark Thread

C

csbin

Senior member
DOWNLOAD http://www.cpuid.com/downloads/cpu-z/cpu-z_1.73-en.zip



EgibP.png



Ir3mm.png
 
Excellent!

3570K @ 4.0, stock voltage

5xS4m5X.png


33kFQGH.png


Now that I'm looking at it, I could probably get away with some degree of undervolt, since it's stable at 4.3ghz at 1.168v and 4.4 at 1.192v.
 
2.8GHz "Core 2 Quad" (e5420 with fsb 375) =
418
1224

around the same as a 7850K according to their reference results...

edit: result and reference are for 32bit, 64bit is way higher.
 
Last edited:
Their bench looks like its broken. 2.9x scaling going from 1 to 4 cores.

Results are very weird to say the least. Can't see where an 8350 beats at 4790k stock for stock MT either.

Results are all over the place. 5280k just a hair faster than a stock 4790k.

The 8350 improved over the 8150 by 66% stock vs. stock.

Not at all inline with real world performance.
 
HP Elitebook 840
Core i5-4300U

Windows Power Saver
Single threaded: 515
Multi Threaded : 906

Windows Balanced
Single threaded: 615
Multi Threaded : 904

Windows High Performance
Single threaded: 610
Multi Threaded : 909
 
Well, it depends on what they're testing for. I'm sure it's a valid test of something.

EDIT: I'd buy that my Ivy Bridge i5 is ~50% faster per thread than an FX-83xx, but only around 75% as fast in highly threaded scenarios.

I wouldn't expect perfect threaded scaling in most workloads, but it is a bit fishy that an FX has better scaling than an i5 with thread count. It's probably pretty INT heavy.
 
Last edited:
Phynaz said:
HP Elitebook 840
Core i5-4300U

Windows Power Saver
Single threaded: 515
Multi Threaded : 906

Windows Balanced
Single threaded: 615
Multi Threaded : 904

Windows High Performance
Single threaded: 610
Multi Threaded : 909
Click to expand...

Something seems wrong here.

Very weak stress test. Core power is lower than Cinebench for instance.

Would I expect perfect scaling from real world? No.

But this test isn't real world. Its a purely synthetic test and its not scaling properly.
 
Enigmoid said:
Their bench looks like its broken. 2.9x scaling going from 1 to 4 cores.

Results are very weird to say the least. Can't see where an 8350 beats at 4790k stock for stock MT either.

Results are all over the place. 5280k just a hair faster than a stock 4790k.

The 8350 improved over the 8150 by 66% stock vs. stock.

Not at all inline with real world performance.
Click to expand...
That's the usual problem of benchmarks doing some selected tasks only. It's worse if that are synthetic tasks. Passmark suffered from having the int division weigh so much on the score. Cinebench also does some rather specific stuff relying heavily on SIMD and cache subsystem performance. How does it correlate with general usage scenarios and why is it being used as a ST/MT guessurement standard then (even in c't magazine)? 🙂
 
i5 3550 at stock 3.3GHz = 1200 single / 3646 multi (better than FX 8150 single and equal to FX 8150 multi)
i5 3550 at 3.9GHz = 1425 single / 4220 multi (equal to i7 4750HQ multi).
 
Enigmoid said:
Their bench looks like its broken. 2.9x scaling going from 1 to 4 cores.

Results are very weird to say the least. Can't see where an 8350 beats at 4790k stock for stock MT either.

Results are all over the place. 5280k just a hair faster than a stock 4790k.

The 8350 improved over the 8150 by 66% stock vs. stock.

Not at all inline with real world performance.
Click to expand...
well, hyperthreading is not a second CPU, which the 2cores/module design practically is
 
if I disable HT on an i5 661 it makes the same wierdo result. So its a bug in the bench.

With HT disabled.
ST 964
MT 897

With HT enabled.
ST 960
MT 1762

Who knows what else it messes up.
 
Last edited:
You must log in or register to reply here.

TRENDING THREADS

Back
Top