CPU-z Multi Thread + Single Thread competition (all welcome)

[storm-chaser]

Boards updated 6:40pm EST 8/22

I am well aware that CPU z isn't the most reliable standard when assessing CPU system performance. I want to say that right up front. But pros list, everyone has it and we all play by the same rules so it should make for a relatively painless, competitive experience that nearly all here on the forum can get involved with.

Very simple, run the CPU z "bench" tab and post that screenshot here - included in the snip should be the initial tab for system spec. Your result should look nearly identically to this, without out any superfluous data, focusing solely on the CPU and Bench tabs ONLY! Failure to follow adherence to protocol might result in your sub being rejected!

**************************
CPU...…….2x Intel Xeon E5 2673s @ 3.6 all core + 4.0GHz single core turbo
Memory...64GB Octal channel DDR3 @1866mhz
Cooling...Stock (liquid)
GPU...…..RX 5700 XT

*************************
Boards updated 6:40PM 8/22

Multi Core Leaderboard:



Single Core Leaderboard:





Thread Count Leaderboard:


Core monsters, please join the club and challenge storm-chaser's total thread count of 48. I know you are out there!

Slow motion (red lantern) leaderboard:


Scoring
We will look at CPU single thread as well as CPU multi thread in scoring. There will be a leaderboard for Single Core and Multi Core divisions... Not quite sure what to do with the multi-Thread ratio result, but it's posted anyway with the single core results.

 

[ereko]

Otherwise good idea, but isnt it better test with avx2 beta ?

 

[Mr.Scott]

<o><o><o><o><o><o><o><o><o><o>

CPU...…….2x Intel Xeon E5 2673s @ 3.6GHz
Memory...64GB Octal channel DDR3 @1866mhz
Cooling...Stock (liquid)
GPU...…..RX 5700 XT




<o><o><o><o><o><o><o><o><o><o>

Your screen shot says 3.9, not the 3.6 in your specs.

CPU...……. 2x Intel Xeon 5675s @ 4.3GHz
Memory...12GB Triple channel DDR3 @1870mhz @ 7-7-7-x
Cooling... Air
GPU...….. (doesn't matter)

 

[robal]

 

[jlewis02]

Will try to get higher this weekend.

 

[ppn]

 

[storm-chaser]

Otherwise good idea, but isnt it better test with avx2 beta ?
View attachment 163951

Sure, Im not familiar with some of the CPUz benching modes so I selected what was just the default. If you think we should move to AVX2 we can definitely still do that.

Guys we need to talk about multi thread ratio and how it's determined. It appears to strike at the very heart of processor efficiency. Mathematically, it's determined by dividing your multi core score by single core result. Can anyone explain this in more depth? Im just looking in. Believe it or not there's not a whole lot of data out there.

Your screen shot says 3.9, not the 3.6 in your specs.

CPU...……. 2x Intel Xeon 5675s @ 4.3GHz
Memory...12GB Triple channel DDR3 @1870mhz @ 7-7-7-x
Cooling... Air
GPU...….. (doesn't matter)

View attachment 163960

I generally post the processor base speed for my submissions... This chip actually does go to 4.0GHz in single core turbo but in multi thread tests like this, the all core turbo is 3.6GHz, so this frequency acts like a base clock as well.

The three primary chips that have eight cores and HT in the Xeon e5 2600 family:
E5 2667 v2 - 3.3.GHz base vs 4.0 Boost - 130w
E5 2673 v2 - 3.3GHz base vs 4.0 Boost - 110w
E5 2687w v2 - 3.4GHz base vs 4.0 Boost - 150w

Combine this in a dual core chassis and you have 300w TDP for the 2687
or 260w for the two 2673s. Same performance and clocks as Xeon's best in for Ivy bridge best in class? for 80 watts less? I could use without the extra heat. And since they also use the identical turbo specs, all chips listed will actually run at a base clock of 3.6GHz.

 

[Outback Bronze]

Grabbed this from the CPUZ Benchmark thread.

Single Thread is pretty good : )

 

[Hugis]

 

[biffzinker]

CPU...…….Ryzen 7 3800X @ stock except I nudged the base clock up to 100.5 MHz
Memory...32GB Dual Channel @ 3333 MHz
Cooling...Hyper 212 Black Edition
GPU...…..MSI RTX 2060 GamingZ

Edit: Improved scores
Hot outside today, bedroom was a bit warm.

 

[alxeus]

CPU ---
i7-4790K @~1.295V up to 1.35V (only while turbo)
4 core |3 core | 2&1 core
4600 MHz|4700 MHz|4800MHz
Cooler ---
Arctic liquid freezer II 280
RAM ---
DDR3 10-12-12-32 @ 2133MHz (perhaps 2200 would be ok too)
GPU ---
5700XT and Vega64

Note: 19.01.64 AVX2 perhaps I hit my set TDP limit of 125W while all core. Intel's 4th gen toothpaste do not get much further, even with liquid cooler.

EDIT:
-forgot cooler

 

[storm-chaser]

Your screen shot says 3.9, not the 3.6 in your specs.

CPU...……. 2x Intel Xeon 5675s @ 4.3GHz
Memory...12GB Triple channel DDR3 @1870mhz @ 7-7-7-x
Cooling... Air
GPU...….. (doesn't matter)
View attachment 163960

Still loving that triple channel memory kit of yours? So long as it's above dual, you can just keep your head above water.

Remember Mr. Scott, you have nobody to blame but yourself for getting me into dual socket rigs and endless rounds of eight channel memory in the first place.... this was all your fault! lol I have some 5790s? Xeon CPUs in storage and I made a couple runs with that rig as well and if I remember correctly that was six channel in total. I need to update my inventory first.

As for the benchmark including extra data, by listing things such as GPU system, by all means I say we can proceed with graphics system information. Sometimes there is good data there regardless and you can get a better feel for the project with a GPU choice mentioned in the spec section.

Grabbed this from the CPUZ Benchmark thread.

Single Thread is pretty good : )

View attachment 164003

Indeed, very impressive result so far.

Good to see some honest turn out here. I will get a rough leaderboard up tonight and then we can refine from that point on. I may split results in three categories (Single bench result, multi thread result, and multi thread ratio)

Worked to improve single core performance slightly by turning off background tasks and closing things that may be using the CPU to perform other work which has a tendency to undercut benchmark results for obvious reasons. Slight improvement all around.

Slight improvement all around after disabling background tasks but I will doubt to see much more than this. And while the z820 cooling system is absolutely sick, the processors are locked down and no changes to BCLK or turbo control can be made. I heard someone say they were actually able to disable the high wattage limit for this chip. Giving you better overclock margins, but take with a grain of salt...

CPU ---
5700XT and Vega64

Good choice on the 5700 XT. What exact card did you get? How are temps when gaming?

 

[ereko]

I have too good score in the list. This is the score with 17.01.64. Not that it bothered me

 

[Arctucas]

Which benchmark test are we supposed to use?

 

[Hugis]

@storm-chaser seems your number 1 slot is using avx test no?

 

[storm-chaser]

Which benchmark test are we supposed to use?

Well I wanted to have a discussion about that. But nobody really responded. I guess since most people have already left it stock, lets just run 17.01.64 going forward, even if it is not ideal. So thats the data we already have, thats the data we will work with. In other words, open CPUz and run the benchmark on stock settings.

@storm-chaser seems your number 1 slot is using avx test no?

yes I have to modify that. I forgot to go back and do that and then I uploaded the data without checking. But since we are not getting much going here, I might just call it off. see how it goes for the next couple days..

 

[Hugis]

Well I wanted to have a discussion about that. But nobody really responded. I guess since most people have already left it stock, lets just run 17.01.64 going forward, even if it is not ideal. So thats the data we already have, thats the data we will work with. In other words, open CPUz and run the benchmark on stock settings.


yes I have to modify that. I forgot to go back and do that and then I uploaded the data without checking. But since we are not getting much going here, I might just call it off. see how it goes for the next couple days..

Oh and my clock rate is 4.4Ghz FYI

 

[Arctucas]

CPU - 9900K
Memory - 16GB DDR4 @ 4000MHz
Cooling - custom loop
GPU - RTX2080

 

[storm-chaser]

CPU - 9900K
Memory - 16GB DDR4 @ 4000MHz
Cooling - custom loop
GPU - RTX2080

View attachment 164313

Nice run. Are you running windows 7 on this rig? If my memory serves me right...

 

[Arctucas]

Nice run. Are you running windows 7 on this rig? If my memory serves me right...

Thanks.

Yes W7.

 

[lsevald]

I'm getting 2-300points more in win10 over win7 for the multi thread test, single thread no difference.

CPU...…….9900k@5.3GHz
Memory...32GB G.Skill 3600CL15@3900CL15
Cooling...Custom water
GPU...…..1080ti

 

[Mr.Scott]

CPU - 9900K
Memory - 16GB DDR4 @ 4000MHz
Cooling - custom loop
GPU - RTX2080

View attachment 164313

Nice clock speed.

 

[storm-chaser]

Nice clock speed.

He's already cleaned house around here a bunch of times with that 9900K and whatever type of sick cooling system he's got under the hood. We should all be concerned. But I do have something slightly more powerful I can pull out of storage if my multi thread score comes under attack...

 

[BlaezaLite]

Gigabyte GA-Z77PD-3
Xeon 1240V2 @3.62
Arctic Freezer 7 Pro
16gb samsung ddr3 @1800
Sapphire RX570 pulse

 

[sam_86314]

Main PC
CPU: R5 2600X (PBO off, XFR on, -0.15v vcore offset)
RAM: 2x8GB DDR4-3000
Cooling: Cryorig R1 Ultimate
GPU: GTX 1070

Main Laptop (HP EliteBook Revolve 810 G2)
CPU: i5-4310U (stock speed)
RAM: 2x4GB DDR3-1600
Cooling: single heatpipe laptop heatsink
GPU: HD 4400

Gaming Laptop (Clevo W650KK1)
CPU: i5-6500T (stock speed)
RAM: 2x8GB DDR4-2133
Cooling: dual-heatpipe laptop heatsink
GPU: GTX 1050 Ti 4GB