• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Sem

Sem

Soldato
Joined
18 Oct 2002
Posts
4,602
Location
London
well been testing the chip for an hour in my ITX rig and i think ive seen enough

all at stock cinebench is fine with all cores boosting to 4.7 with a score around 1950 CB (1.248v) auto

the moment i run any real stress tests like Realbench or AIDA it drops to 4GHz all core with max temp reaching 91c core voltage drops to 1.050v

tbh i didnt expect much for an ITX 370 board and a 120MM AIO my 8700K did 5Ghz in the same system AVX offset of -3 with sub 80c temps

well i guess its time to put the chip in my z390 (M XI Formula) and custom loop and see what she can really do
 
Associate
Joined
30 Aug 2018
Posts
2,483
very nice cpu you have there. have seen a similar one for sale 5.3 @1.296v price is v.high, he will sell it but maybe not at the price asking
I saw that too. It's delidded, does that make any difference to vcore do you think?

He's taking his vcore measurements from CPUz, no mention of what he put into bios or hwinfo etc. For me CPUz shows a much lower voltage than hwinfo.
 
Associate
Joined
30 Aug 2018
Posts
2,483
well been testing the chip for an hour in my ITX rig and i think ive seen enough

all at stock cinebench is fine with all cores boosting to 4.7 with a score around 1950 CB (1.248v) auto

the moment i run any real stress tests like Realbench or AIDA it drops to 4GHz all core with max temp reaching 91c core voltage drops to 1.050v

tbh i didnt expect much for an ITX 370 board and a 120MM AIO my 8700K did 5Ghz in the same system AVX offset of -3 with sub 80c temps

well i guess its time to put the chip in my z390 (M XI Formula) and custom loop and see what she can really do

Have fun :D

I've heard horror stories from the clickbait youtubers about the asus VRMs and while they might be true for the hero with zero airflow, they certainly arent for the formula. My VRMs are always in line with my loop temps so far.

I'm hoping when i finish the loop that extended stress testing at higher core clocks wont see VRM temps rise too much.
 
Soldato
Joined
15 May 2012
Posts
5,812
Location
Louth, lincs
I saw that too. It's delidded, does that make any difference to vcore do you think?

He's taking his vcore measurements from CPUz, no mention of what he put into bios or hwinfo etc. For me CPUz shows a much lower voltage than hwinfo.

It will make a slight difference yes, with a 10c water drop I can lower volts by approx 0.02v.
Probably shot himself in the foot though as most extreme OC guys will want the solder on as TIM can be a pita due to the crack, basically pot losing contact with TIM/die.
 
Associate
Joined
23 Feb 2009
Posts
1,021
More playing:

5.2ghz core 4.7ghz cache
1.320 vcore
Realbench 30 mins hottest core average: 80 spike 84
OCCT large 1hr hottest core average 59 spike 63

No long stress test (will do next weekend to finally find my 24/7), just playing games. No crashes
 
Associate
Joined
23 Feb 2009
Posts
1,021
Nice, another good CPU, retail or tray?

Tray from here :) Cooling is 2x360 + 240, so that helps out a lot.

Funnily the heat of these chips has actually made my watercooling feel better value now. I put it together for aesthetics as its been a long time since WC gave me meaningful higher clocks than air.
Now it's actually being used for what its meant to do :D
 

Sem

Sem

Soldato
Joined
18 Oct 2002
Posts
4,602
Location
London
did some quick testing in my caselabs rig
reset the bios to default and only changed 2 settings Multicore enhancement to enabled and AVX offset to 0
apart from gaming i mainly use my rig for x264/5 encoding so i am only interested in AVX stability (which should also mean gaming stability)

So far chip does 5ghz avx at around 1.312v (hwinfo) not changed any of the voltages or LLC or anything all on auto and passed 30mins real bench
will try some more in the morning but i'm aiming for 5ghz avx for 24/7 stability might do few suicide runs up to 5.3 or something to see how high i can get it cinebench stable tho

Esy87pB.png
 
Associate
Joined
23 Feb 2009
Posts
1,021
More testing:
5.2ghz no crashes in realbench, aida64, cinebench, OCCT all at 1.32v. 0 crashes whilst gaming the last couple of days.

But, prime95 avx small tft and its instant shutdown. Threw 1.37v and still crashing very quickly. Any tips?
Seems this is the reason why I reported only a 0.02v jump from 5.0 to 5.2ghz. I hadn't factored in prime avx stress (was part of my 5.0 test suite).

So what I need to answer for myself. Is 5.2 good for me, but i'll always know that im definitely not prime95avx stable. Is this a silly stress test? As long as I am not crashing in all other stress tests, should I forget about prime95 avx?
I don't want to use offset because other avx stress tests i'm good. So i'll be concerned that my offset will apply in situations with avx my comp can handle.
 

Sem

Sem

Soldato
Joined
18 Oct 2002
Posts
4,602
Location
London
not sure if there's a bug in the voltage reading cause this seems too good to be true but currently at 5.2 AVX realbench 15mins stable at only 1.28v
tired now will test more and tweak more settings in the morning
J6fhcdW.png
 
Associate
Joined
30 Aug 2018
Posts
2,483
More testing:
5.2ghz no crashes in realbench, aida64, cinebench, OCCT all at 1.32v. 0 crashes whilst gaming the last couple of days.

But, prime95 avx small tft and its instant shutdown. Threw 1.37v and still crashing very quickly. Any tips?
Seems this is the reason why I reported only a 0.02v jump from 5.0 to 5.2ghz. I hadn't factored in prime avx stress (was part of my 5.0 test suite).

So what I need to answer for myself. Is 5.2 good for me, but i'll always know that im definitely not prime95avx stable. Is this a silly stress test? As long as I am not crashing in all other stress tests, should I forget about prime95 avx?
I don't want to use offset because other avx stress tests i'm good. So i'll be concerned that my offset will apply in situations with avx my comp can handle.
Prime with AVX is a monster. I don't know what you can tweak to make the system stable.
 
Associate
Joined
30 Aug 2018
Posts
2,483
not sure if there's a bug in the voltage reading cause this seems too good to be true but currently at 5.2 AVX realbench 15mins stable at only 1.28v
tired now will test more and tweak more settings in the morning
J6fhcdW.png
What does it say for vcore in hwinfo? I found that CPUz was reporting much lower than HWINFO for that particular reading.

What have you got it set to in Bios?
 
Back
Top Bottom