CPUZ 1.73 Benchmark Thread

[soccerballtux]

6.8x scaling, with HT, for 8 cores on a synthetic benchmark is pretty poor.

that was without HT

 

[YBS1]

6.8x scaling, with HT, for 8 cores on a synthetic benchmark is pretty poor.

*Sigh* Again, look at that last bench I ran and compare it to my original bench at same clocks with HT on in post #50. HT is disabled, 8 cores, 8 threads. The scaling with HT is what is poor. 6.8X scaling with 8 threads (no HT), it only increase to 7.9X with 16 (HT on).

 

[MongGrel]

We'll see. Went downhill from there

Set multiplier for 48x upped vcore 0.020v's booted windows did some quick testing which seemed fine....Hmm.

Went for the illusive 50x upping vcore 0.040v's more which probably was too little....Boot loop from hell is what I got! Bsod and reboot followed by the loop. Clearing cmos was a no go! Pulling battery and clearing cmos was a no go! Pulled 1 stick of ram and she booted with bios warning screen. Set to 42x profile booted and shutdown. Put stick back in boot loops. Swapped to other slots boot loops. Go figure!

Yeah, why I don't push this old thing too far these days.

I pretty much know what it can do more or less all ready, when ya start having to pop the battery etc just to go farther and learn a new rig setup it can be a real PITA.

 

[Dresdenboy]

Yeah, I'm guessing it's just really light on FPU use allowing the FX to really act as 8 pure cores. Eight real Intel cores just demolishes the 4c/8t Intel's.

Yesterday I had a first look at the assembler code found in the hotspots (64b version). The benchmark creates hundreds of threads for the MT test (although only a small number overlapping at any single point in time). The calculations use scalar single precision SSE ops to do something. I'll write down the formulas. Maybe it's some fractal calculation. But there are also things like (integer) a=(b*c)/b.

 

[Enigmoid]

*Sigh* Again, look at that last bench I ran and compare it to my original bench at same clocks with HT on in post #50. HT is disabled, 8 cores, 8 threads. The scaling with HT is what is poor. 6.8X scaling with 8 threads (no HT), it only increase to 7.9X with 16 (HT on).

Whoops.

That's what happens when you comment at 1 in the morning.

Still rather poor scaling considering it scales so well on AMD chips.

 

[SPBHM]

the difference surprised me, they are usually a lot closer on benchmarks

now I know this benchmark is influenced by memory, but the Athlon with single channel has lower latency and copy/read not to far from the pentium with dual channel, so I guess there is something else that makes the k8 so slow...

 

[SlowSpyder]

*Sigh* Again, look at that last bench I ran and compare it to my original bench at same clocks with HT on in post #50. HT is disabled, 8 cores, 8 threads. The scaling with HT is what is poor. 6.8X scaling with 8 threads (no HT), it only increase to 7.9X with 16 (HT on).

Yea, from my ST score to MT score I get ~6.25x scaling. Your CPU scales better with HT off, and of course gets a higher absolute score in both ST and MT.

 

[ShintaiDK]

the difference surprised me, they are usually a lot closer on benchmarks

now I know this benchmark is influenced by memory, but the Athlon with single channel has lower latency and copy/read not to far from the pentium with dual channel, so I guess there is something else that makes the k8 so slow...

Core 2 cache speed, latency and size if I had to guess.

 

[Kenmitch]

Yeah, why I don't push this old thing too far these days.

I pretty much know what it can do more or less all ready, when ya start having to pop the battery etc just to go farther and learn a new rig setup it can be a real PITA.

Only swapped the cpu. Was careful doing so. Leaning towards ram stick went belly up?

Found a 16GB kit for a reasonable price on Amazon. Free same day delivery. We'll see soon enough.

 

[Dresdenboy]

now I know this benchmark is influenced by memory, but the Athlon with single channel has lower latency and copy/read not to far from the pentium with dual channel, so I guess there is something else that makes the k8 so slow...

It also makes heavy use of int and fp division. CPUs with HW int divider have another advantage.

 

[LTC8K6]

T8300 is 720ST and slower, 640 in MT.

 

[Makaveli]

the difference surprised me, they are usually a lot closer on benchmarks

now I know this benchmark is influenced by memory, but the Athlon with single channel has lower latency and copy/read not to far from the pentium with dual channel, so I guess there is something else that makes the k8 so slow...

I think the disadvantage for the athlon is single channel memory and lower frequencies.

I still have my opteron 170 being used as an HTPC. That has DDR1 500mhz models in it if I can run the bench on that we will see if it closes the gap.

 

[TeknoBug]

Just ran the benchmark on my i3 4010U NUC (1.7GHz) with 8GB DDR3L-1600, 641 ST / 1079 MT, didn't expect MT to be that high. lol

 

[LTC8K6]

i3-4360 1350/2350

 

[Denly]

AMD PII X4 B95
443
1280
the bench hanged w10 task manager's CPU % graph lol

C2D E8600
959
939

Xeon E5620
745
2737

 

[LTC8K6]

When W10 build 10525 installed on my test mule today, it said CPU-Z 1.73 was incompatible with the new build.

 

[qookap]

I won't agree those point about K8 but still waiting more bench.
I m finish my work about processor microstructures research yesterday. might be share it later..

My research and benchmark tell me there is no relationship between CPU-Z and DRAM,
it's a success pure CPU performance benchmark tool just like sisoft sandra(before fully commercialize, the best fault as benchmark tool like passmark lol)

I m really thankful for bench data. hope more AMD side bench to help me review the CTP and Flops/Cycle. AMD's CPU never have a clear data like Intel...

now I know the A10-7850K just like Core 2 Quad performace, it's basic 8 single precition floating point opreations per second, 4 DP Flops/sec. Theoretically their performance should very close, streamroller(APU) had 8 DP in 1 module, each module including 2 cores.

------
ps. this is my first post at here and forgive my poor english...

 

[waffleironhead]

athlon2 x4 620
single: 382
multi: 1039

I5 460m
single: 801
multi: 1302

 

[sdifox]

Dell PowerEdge R710 with 2 Xeon L5640 @ 2.27GHz
ST:712
MT:8291

http://valid.x86.fr/kd0bxu

 

[Dresdenboy]

I won't agree those point about K8 but still waiting more bench.
I m finish my work about processor microstructures research yesterday. might be share it later..

My research and benchmark tell me there is no relationship between CPU-Z and DRAM,
it's a success pure CPU performance benchmark tool just like sisoft sandra(before fully commercialize, the best fault as benchmark tool like passmark lol)

I m really thankful for bench data. hope more AMD side bench to help me review the CTP and Flops/Cycle. AMD's CPU never have a clear data like Intel...

now I know the A10-7850K just like Core 2 Quad performace, it's basic 8 single precition floating point opreations per second, 4 DP Flops/sec. Theoretically their performance should very close, streamroller(APU) had 8 DP in 1 module, each module including 2 cores.

------
ps. this is my first post at here and forgive my poor english...

Welcome to this forum! Is "qookap" related with the ever smiling animal Quoka?

Did you collect the bench result in a table? I did and added some first metrics. PM me your mail address to get it.

With your APU you could try CodeXL and measure the bench's hotspots. That tool will show you the asm code of the main loops and subroutines.

 

[Dresdenboy]

Logical cores are not scaling well, physical cores scale like mad, see post #123 above.

As often as the integer code uses idiv, there surely is some fighting for related ressources (HW divider or ex units needed for the ucode) on a hyperthreaded core.

Anyone remembers the int division driven Passmark integer score?

 

[SPBHM]

I think the disadvantage for the athlon is single channel memory and lower frequencies.

I still have my opteron 170 being used as an HTPC. That has DDR1 500mhz models in it if I can run the bench on that we will see if it closes the gap.

as I said, even with these clocks it still gets better latency and not far copy/write memory results on Aida for example

I tested the k8 again, changed the clock to 400 (800) and got no real increase, ST wet to 268 I think under the same conditions, that's a big memory clock jump (300 to 400) and no impact, so I think the problem is not really there...

but, I discovered an interesting thing, keep in mind I disabled power saving features and the clock was fixed at what you see on the CPUz screenshot, but the results I mentioned were from runs with the CPU idling, I decided to run the benchmark with high CPU load from other programs (windows update and some scurity software) like 70-100% load and... the CPU benchmark improves the score, the ST at least, MT is a little lower.

ST with 300 OR 400MHz memory goes up to 280 with another program loading the CPU, I tested it a few times and it seemed consistent, once the other program stops and th CPU load is back to 0-5% the score goes down to 26x, have the program loading the CPU and it goes up to 280... the MT score is actually slightly better when the CPU is at 0-5% when starting the bench.

oh yes, if I enable cool n quiet the ST score goes down to 240-250, but with a background process running it goes over 270.

in any case the result I posted previously is representative of what I got with CNQ off, and idle PC, with 600 or 800 memory clock basically, I don't think it's to memory limited, but seems odd with the other variations;

 

[AtenRa]

FX8150 @ 4.6GHz
win 10 64bit

 

[MagickMan]

5930k @ 4.5GHz (it'll do 4.6 but I don't feel like rebooting right now)

 

[YBS1]

FX8150 @ 4.6GHz
win 10 64bit

Maybe I'm overlooking something....Why is that scoring so much lower than soccerballtux?