CPUZ Benchmark (all welcome)

[storm-chaser]

 

[storm-chaser]

Aye, showin their age now tho LOL

Respect your elders!

 

[storm-chaser]

Probably slush in the lines... As for me, no go on matching your MHz result at the moment. I did get to 5.7GHz but it crashes half way through the benchmark. So I think it's most likely due to the ambient temps here right now and the el cheapo waterblock which I picked up just for kicks. That combination is not letting get any higher than 5.7GHz.

Just wanted to let you know I am waiting on a new waterblock and proper thermal paste, when those arrive, I will give it another go.

The problem is not flow, I have 3 pumps in total so that is not a factor here yet she's still running a little hot, and the only other thing that changed was the application of a substitute for thermal paste, so that's probably the culprit.

 

[storm-chaser]

I just ordered some liquid metal as a thermal medium, instead of conventional thermal paste.

We will see how the heatkiller responds, however I am on the hunt for a new waterblock that is less restrictive. If I can lower the temps by just a few degrees, I can probably bench at 5700Mhz and hold a little more stability.

 

[storm-chaser]

The velocity2 has good flow but i really dont like its mounting mech, much prefer the old style of springs etc

Danger close (If we ever get another cold night)

So the liquid metal showed up today... and had massive gains bro!

How is it possible? During a quick torture test I shaved 25*F with the fans on idle
And shaved 30*F with all the fans on full throttle.
That is an insane improvement, anyone else getting numbers like this?

It's also allowing me to run 5.5GHz much easier than before, and without the chiller attached.

 

[storm-chaser]

c'mon then wheres that 5.7g???? nice gains in temps, my last run was with liquid metal. 3 hours spent benching and everything is stained!!

Unfortunately it has not been cold enough here to make 5.7 viable. I guess I have to just keep hoping for some good cold weather later in the month to take advantage of.

 

[storm-chaser]

wow! will give that a go ... any specific type? might also bin cheapo 240mm aio and get a quality 360mm, whats likely improvement?

I actually used a cheaper version known as silver king liquid metal. Made by thermalright. I bet they all work prettty well tho. Although this brand works for me they advertised one gram of liquid metal but it was more like 2/3gram. Still, enough for one CPU.

Yes. get that done on your rig post haste. You will never go back thermal paste.

 

[storm-chaser]

wow! will give that a go ... any specific type? might also bin cheapo 240mm aio and get a quality 360mm, whats likely improvement?

What CPU are you running here?

 

[storm-chaser]

i9 10850k

I would say yeah, go for it. You might only notice a couple degrees difference between the two, but it could make the difference in some cases.

I would get an 360 AIO with G1/4" fittings so you can add more radiator later or modify to your liking.

 

[storm-chaser]

First time on Overclockers, so apologies for any faux pas with rules and posting and stuff.

Just run CPU-Z for the first time and submitted my results, i7-12700k, here: https://valid.x86.fr/h74twq. It's turned out pretty well, but how can I see how other i7s faired? I guess by going through this forum right?

I usually post the update leaderboard after new subs come in, so if you look back a page or two your will see the most current leaderboard available.

EDIT: thanks for your sub. Get u added today most likely, or tomorrow if I run out of time.

 

[storm-chaser]

c'mon then wheres that 5.7g???? nice gains in temps, my last run was with liquid metal. 3 hours spent benching and everything is stained!!

Have some cold weather coming up... I lost my barrow D5 pump last week, so it will be down to two pumps, but that should be fine. Only reason for the third pump was because it was imploding the lines on the outtake (or post) from the CPUz, because
A) The heatkiller IV is very restrictive
B) to much pull, not enough push

I am putting together a new rig right now based on the HP Z840 rig, so I will focus on multi core result as well because it will have two processors for a total of 36T/72T and I do believe these CPUs can be turbo hacked.

 

[storm-chaser]

Good GOD man. How did you pull that one off?

 

[storm-chaser]

That score is absolutely nutty

The CPU speed only says 5400MHz, but I suspect it is much higher than that. Because we've benchmarked many other 12900Ks in this test and nothing in the 5.4GHz "realm" comes even close.

 

[storm-chaser]

see post #180 single core speed 5400 and multi 5100 this is on regular 12900k, KS arrives tomorrow

Just for reference purposes here is my 9600KF running AVX with a moderate "spring" OC. lol

 

[storm-chaser]

Nice boost from your original, no need to wait for the weather now!!!

But can these scores be translated accurately from earlier versions?

 

[storm-chaser]

not comparable to be honest but neither are the Epic and threadripper cpu's!! you could add a separate avx box or just state original bench only. Still need a 5950X user to compare @MadMatty @kitfit1

Wonder how my 36 core Xeons are going to compare with your results.

 

[storm-chaser]

HP Z840 with two Xeon E5-2696 v3 processors 36c/72t, liquid cooled, liquid metal thermal paste (Ill try with 19.01 a bit later)

 

[storm-chaser]

19.01

 

[storm-chaser]

 

[storm-chaser]

Nice boost from your original, no need to wait for the weather now!!!

I just buy my way out of trouble by putting together this 36c/72t beast together. lol

From preliminary results, it appears to hang with a ~5500MHz ADL 12900K in CPUz benchmarking
Appears to hang with a stock 5950X in Y cruncher.

Still have to figure out how to hack this turbo but once that is done Im going to put a 360mm radiator in and try to throttle all cores up to 3.8GHz.
Still working on tuning via throttlestop.