Share your CPUZ Benchmarks!

[biffzinker]

I think it might have something to do with the FPU Julia bench it depends on for the frames encoded with the VP8 codec.

Also the instruction set extensions are having and effect.

FPU VP8 Benchmark
The content of the frames are generated by the FPU Julia fractal module.
The code behind this benchmark method utilizes the appropriate MMX, SSE2, SSSE3 or SSE4.1 instruction set extension.

FPU Julia Benchmark
The code behind this benchmark method is written in Assembly, and it is extremely optimized for every popular AMD, Intel and VIA processor core variants by utilizing the appropriate x87, 3DNow!, 3DNow!+, SSE, AVX, AVX2, FMA, and FMA4 instruction set extension.

Edit: Ryzen has a more capable FPU perhaps compared to the previous Bulldozer?

 

[anselmo]

Mostly, FPU benchmarks optimized for AMD chips is what gives it such an impressive score? Can the FPU performance in AM4 chips be better than Intel? I know that AM4 was a breath of fresh, needed air for AMD and that it gave Intel a hard time, but both brands were pretty much side by side on their offerings.

 

[jboydgolfer]

That AIDA 64 is surprising indeed. Can it be due to the core count?

the tests arent comparable between differing Versions of Aida (like the CPU-Z bench tests, which on one ver can be a 2000, then if you update to the following Ver, that same CPU scores a 200, due to restructuring of the scoring system, not the same as Aida i think, but just asa demonstration of how they arent comparable)....i ran it myself, and my 8600k, @ stock beats out the 8700k seems lacking a structured logging of results, atleast structured in the sense where they are categorized by Ver, Frequency, etc. i also ran it again 30 seconds later, and scored lower, then higher again, all within around 100 points, but there seems to be a variance in the scoring that makes it more of a test or basic idea of perf, and not a reliable bench....or atleast thats how i interpreted it. i could very well be wrong

 

[anselmo]

It is that "as is" thing. It works, but most of its default benchmarks are outdated.

 

[jboydgolfer]

It is that "as is" thing. It works, but most of its default benchmarks are outdated.

agreed, they are , but certainly a great tool nonetheless, with a TON of functions.

 

[NSX05]

I do believe. I had a Lenovo T430 (traded it for a disappointing Sony Vaio) before this Toshiba, and that was the last truly upgradable series of Thinkpads and I loved it to bits, even planned on putting an i7 instead of the i5. Old laptops and their upgradeability were also easier to mantain. If the graphics car died (on some, just like the Toshiba A300 series) it could be swapped for another or even upgraded. Or bring some new life by swapping the CPU from an old, tired Celeron that was good for office tasks 10 years ago for a Core 2 Duo that can even support VM's (the P8600 is a great example, I use it to virtualize anything, from XP to Server 2016 or from Ubuntu 10.04 to the Budgie betas, even though, sometimes it could have a bit more horsepower).

I was using my laptop from 2009 when I bought it, until summer of 2016 when a friend offered me the system I currently have at a super cheap price (he had switched to Mac at the time and had no use for it, although he has since gotten rid of his Mac).

My laptop went through four CPU upgrades (Celeron 575 --> Core2 T5900 --> Celeron 925 --> Core2 T9300), although the 925 was purely to test for Penryn microcode support since I had no way of checking the BIOS back then. That's been with me through the entirety of high school and further education, and still in perfect working order. I love that machine; props to ASUS on that one. Who would have thought that a £220 machine (bought second-hand via eBay) could be upgraded with a $530 (2008 RRP) CPU? Memory is also maxed out and there's a WD Black in there. I spent £35 on the T9300 in 2013, by the way.

The only way I could perhaps make it last longer, is to get an SSD (SATA-II speeds, though...), or to buy some very specific memory to get 4+ GiB bootable.

[...] seems lacking a structured logging of results, atleast structured in the sense where they are categorized by Ver, Frequency, etc.

I think it suffers from the same problem as Geekbench, PassMark and UserBenchMark in that results are highly skewed rather than setting actual baselines independently like CPU-Z appears to have.

 

[jboydgolfer]

I think it suffers from the same problem as Geekbench, PassMark and UserBenchMark in that results are highly skewed rather than setting actual baselines independently like CPU-Z appears to have.

I agree, just like with any benching result, it has to all be done in a controlled "like" environment ,otherwise what's the point ? I don't know for certain, but it seems like many of the programs you mentioned don't categorize or at least don't log their results in a discerning ,or accurate manner. So you end up with an apples versus oranges situation, or rather a green apples versus red apples situation. I think it's good for a loose or basic idea of performance, and it's also a good way to stress certain aspects of a persons processor but other than that I don't put very much stock in it

 

[anselmo]

I was using my laptop from 2009 when I bought it, until summer of 2016 when a friend offered me the system I currently have at a super cheap price (he had switched to Mac at the time and had no use for it, although he has since gotten rid of his Mac).

My laptop went through four CPU upgrades (Celeron 575 --> Core2 T5900 --> Celeron 925 --> Core2 T9300), although the 925 was purely to test for Penryn microcode support since I had no way of checking the BIOS back then. That's been with me through the entirety of high school and further education, and still in perfect working order. I love that machine; props to ASUS on that one. Who would have thought that a £220 machine (bought second-hand via eBay) could be upgraded with a $530 (2008 RRP) CPU? Memory is also maxed out and there's a WD Black in there. I spent £35 on the T9300 in 2013, by the way.

The only way I could perhaps make it last longer, is to get an SSD (SATA-II speeds, though...), or to buy some very specific memory to get 4+ GiB bootable.

That's the beauty of it, older machines were slightly better. Speed and energy efficience, no, but upgradeability and (arguably) portability in some cases. I dearly miss those old laptops that had those weird screens that folded into horizontal, vertical, whatever positions.

 

[NSX05]

That's the beauty of it, older machines were slightly better. Speed and energy efficience, no, but upgradeability and (arguably) portability in some cases. I dearly miss those old laptops that had those weird screens that folded into horizontal, vertical, whatever positions.

Considering its age, it's more than capable for every day tasks. Office tasks would be fine. What really lets those platforms down, is the low memory capacity limit. And honestly, that's my only complaint about my laptop. The Core2 itself is very snappy. Needless to say, I noticed the jump from 3 GiB to 16 GiB.

 

[Ruru]

Last time with 7600K, I hope that I'll get 7700K today. Not mine, just a typical 7600K stock result as a reference.

 

[jboydgolfer]

Last time with 7600K, I hope that I'll get 7700K today. Not mine, just a typical 7600K stock result as a reference.

Am I the only one who can't read the results in the pictures in this post? They're so tiny....maybe its a mobile issue. Ill try DT

 

[Ruru]

Am I the only one who can't read the results in the pictures in this post? They're so tiny

Double click when fullscreen

 

[jboydgolfer]

Double click when fullscreen

I tried that, it just makes it double blurry. I'll have to try on my desktop. It must be the aspect ratio ,because other images posted in this thread I can see fine on my iPhone ,it's just that one's a wider format i think.

yeah, i was right. it is the combination of my iphone, and the format....see it on DT, its just smaller is all.

 

[NSX05]

@anselmo; out of curiosity, do you still have access to that Pentium II machine?

@Chloe Price; are you wanting me to add that result?

 

[Ruru]

@Chloe Price; are you wanting me to add that result?

If I can have two results, with a 7600K and with a 7700K, then add! If not, then wait for a day or two that I get my sweet 7700K

 

[jboydgolfer]

@ stock frequency.

 

[NSX05]

If I can have two results, with a 7600K and with a 7700K, then add! If not, then wait for a day or two that I get my sweet 7700K

You can have as many as you like!

 

[Ruru]

You can have as many as you like!

Then feel free to add.

 

[NSX05]

Okay guys; here are the new graphs. Still finalizing a few things before the webpage version is done.

CPU-Z 1.8.4.x — Single | Multi

 

[Ruru]

Okay guys; here are the new graphs. Still finalizing a few things before the webpage version is done.

CPU-Z 1.8.4.x — Single | Multi

You are getting new shit when I get my 7700K today and I OC it

One dude from my streams promised to borrow a delid tool, great.

HOLD ON A FUCKING MINUTE?!*

I'm on first on single threaded? DAMN!

*Reference from 8 Mile..

... just wait that I get that 7700K!

 

[anselmo]

@anselmo; out of curiosity, do you still have access to that Pentium II machine?

I do, why? It is mine, after all.

 

[NSX05]

You are getting new shit when I get my 7700K today and I OC it

One dude from my streams promised to borrow a delid tool, great.

HOLD ON A FUCKING MINUTE?!*

I'm on first on single threaded? DAMN!

*Reference from 8 Mile..

... just wait that I get that 7700K!

You are indeed on top.

In before an 8700K owner walks in and steals that spot.

I do, why? It is mine, after all.

Would you be able to open CPU-Z and screenshot the entire window? The version number is in the bottom-left. I ask because it has come to my attention that Windows 98 versions exist for all of the latest versions as well. Upon initial inspection of the site, it appeared to me that Windows 98 stopped being supported with version 1.7.8, but I was wrong. I would like to add you to the correct graph.

 

[Ruru]

Won't post yet, now I have 7700K kickin', but since I didn't have any thermal paste, this shit throttles.

Roommate's brother will probably bring some.

 

[NSX05]

Won't post yet, now I have 7700K kickin', but since I didn't have any thermal paste, this shit throttles.

Ah yes... The dreaded Intel-branded toothpaste.

 

[jboydgolfer]

Okay guys; here are the new graphs.

im curious if the scoring of this newer CPUZ is comparable to the older Ver from when i had 1st place in CPUz ST bench?? My 2c/2t G3258 scored 516, and beat out 3960 extreme , but i know some times the CPUz results arent comparable *edit* i ran 1.79 VS 1.84 and they are using the same scoring system

just in case. G3258 @ 4.7Ghz. (wish i had the results from my 4.9Ghz run, but i never Screen capped it apparently )
i no longer have the chip, so if it isnt valid for submission, its all good, another member own's it now