Overclock.net banner

261 - 280 of 294 Posts

·
Registered
Joined
·
2,507 Posts
Was that on prime? Your chip seems to be almost identical to mine when testing Prime95 blend for hours. 4.3 stable but 4.4 at 1.33v fails after a few hours. (I tried again last night and it failed after the 6 hour mark this morning too)\


Edit: Seems like my setup is very sensitive to SOC voltage.. I had it at 1.125 in bios which was stable for the few hours mentioned above but failed. I tried higher 1.135v but that fails in 5 minutes. I will try lower and hopefully it's stable.
It was Prime, i think i need to research how to overclock with this board better, I feel like 4.4 is obtainable.
 

·
OCN is bad civilization
Joined
·
3,541 Posts
Finally got fuma 2. Date on cpu was 2010, first week of march 2020.
As for OC, I can definitely boot at 4.6ghz with 1.32v but system crashes almost immediately if I try to run cb r20. On 4.5ghz 1.32v, cb r20 crashes 5 seconds into the benchmark. I'd guess I could get it stable at 4.5 if I tweak settings a bit but I really can't be bothered.
 

·
AMD OVERLOCKER
Joined
·
727 Posts
Finally got fuma 2. Date on cpu was 2010, first week of march 2020.
As for OC, I can definitely boot at 4.6ghz with 1.32v but system crashes almost immediately if I try to run cb r20. On 4.5ghz 1.32v, cb r20 crashes 5 seconds into the benchmark. I'd guess I could get it stable at 4.5 if I tweak settings a bit but I really can't be bothered.
How getting a crash at 4.5ghz shows that it could be stable with some settings?
Your description is about a 3600 which probably can be stable at 4.4ghz.

Sent from my ONEPLUS A6003 using Tapatalk
 

·
Registered
Joined
·
883 Posts
How getting a crash at 4.5ghz shows that it could be stable with some settings?
Your description is about a 3600 which probably can be stable at 4.4ghz.

Sent from my ONEPLUS A6003 using Tapatalk
Maybe he means stable for some light gaming ??

:p
 

·
AMD OVERLOCKER
Joined
·
727 Posts
Maybe he means stable for some light gaming ??



:p
I really can't tell but it seems that people are confused about what stable really means.

And this specific CPU doesn't seem to behave like the great new units that can handle 4.5ghz with no issues.

Sent from my ONEPLUS A6003 using Tapatalk
 

·
Registered
Joined
·
112 Posts
My new 3600 (2018 batch number) is stable on 4.5 1.275v. But I couldn't make it stable on 4.6, even on 1.35v. I guess I can make it stable but I am too lazy to bother with all the settings for just 100mhz. I am more than happy with this new cpu, my old 3600 barely reached 4175 on 1.45v lol.
 

·
Registered
Joined
·
165 Posts
I think you people are exaggerating with 4.5ghz being stable at sub 1.3v...its not truly stable. I myself have one of these new 3600's and i was thrilled and thought after passing 2 hours on aida64 and realbench im g2g with [email protected] 1.26v, but believe me, if youre gonna perform cpu intensive tasks for long enough , youll find out its gonna give you an occasional funny behaviour/crash. Instability can sometimes reveal itself only occasionally, in supposedly random times with variable intervals in between failures. Gamers nexus made a video pretty much disputing the common nowadays claims that the new batches can oc this high stably. Its true the silicon is way better with the new batches, but getting a truly stable ~4.5ish in sub 1.3v is just an iilusion, unless stable means super light loads and that's it.
 

·
Registered
Joined
·
4 Posts
I just got a 3600 to replace my 3700x in my htpc (CPU went into my new Eluktronics Thicc 15).

Running 4.6ghz at 1.35v.

I posted this on Reddit and everyone went nuts.

Apparently I'll raise Satan from the dead??
 

Attachments

·
AMD OVERLOCKER
Joined
·
727 Posts
Question about reaching the limits.

I get my 3600x up to 4.450mhz and runs Cinebench R15 just fine at 1.46v

Then even at +25mhz I get a reboot immediately when I launch the test.
I have increased the vcore at 1.48v but no difference.

Is there anything else that I should try?

Sent from my ONEPLUS A6003 using Tapatalk
 

·
Registered
Joined
·
1 Posts
Hello, I've recently joined Ryzen family, I bought 1 month ago a R5 3600 produced in Nov.19', and I couldn't even reach 4.3GHz with a <1.38Vcore. I had it at 4.2GHz at 1.262V manually. So I've decided to send it back and bought another one, a 2017SUS (Produced in April)

Currently I'm at 4.5GHz all cores manual OC 1.262V, LLC set to "turbo" in my Aorus Elite X570 BIOS, 3200/1600IF

My system is "stable" for my daily tasks (gaming / rendering in After Effects-Sony Vegas), I didn't try any kind of heavy stress test but only these benchmarks (CB15,CB20,CPUZ,Firestrike,Timespy)

Never had a single crash/bsod, but if I change my LLC to "high" or a lower preset, CB20 test will not complete.

I'm really happy.

 

·
Registered
Joined
·
472 Posts
Did you try looping cinebench r20? I found that to be as stressful on OC's as aida64 or realbench. Sounds like your 3600 is very similar to my 3600. Mine is p95 small fft stable [email protected]/1.15v under load, I couldn't get 4.4ghz stable with 1.25v/1.2v underload at p95 large fft.
Mine is stable in cinebench r20/aida64/realbench stable [email protected]/1.16v under load.
 

·
AMD OVERLOCKER
Joined
·
727 Posts
Did you try looping cinebench r20? I found that to be as stressful on OC's as aida64 or realbench. Sounds like your 3600 is very similar to my 3600. Mine is p95 small fft stable [email protected]/1.15v under load, I couldn't get 4.4ghz stable with 1.25v/1.2v underload at p95 large fft.
Mine is stable in cinebench r20/aida64/realbench stable [email protected]/1.16v under load.
Any advice for passing CB15 at a bit higher cpu clocks, as I have mentioned 2 posts earlier?

Sent from my ONEPLUS A6003 using Tapatalk
 

·
Here longer than mods lol
Joined
·
3,795 Posts
Beautiful buddy, congrats :)
I've a 3700x that can do 4.6Ghz at 1.35 but keep it at 4.4Ghz 1.3v daily use
My single core is 537 too so just ahead lol
 

·
Registered
Joined
·
472 Posts
Question about reaching the limits.

I get my 3600x up to 4.450mhz and runs Cinebench R15 just fine at 1.46v

Then even at +25mhz I get a reboot immediately when I launch the test.
I have increased the vcore at 1.48v but no difference.

Is there anything else that I should try?

Sent from my ONEPLUS A6003 using Tapatalk

Lower memory speed, undervolt SOC. That could give you more frequency maybe but it sounds like you're maxed out as is. I can run 3400mhz with 0.9v SOC 0.825v vddg and 0.750v vddp.
 

·
AMD OVERLOCKER
Joined
·
727 Posts
Lower memory speed, undervolt SOC. That could give you more frequency maybe but it sounds like you're maxed out as is. I can run 3400mhz with 0.9v SOC 0.825v vddg and 0.750v vddp.
What about the LLC?
Do u think it can affect the higher clock stability for benchmarks?

Sent from my ONEPLUS A6003 using Tapatalk
 

·
Registered
Joined
·
472 Posts
What about the LLC?
Do u think it can affect the higher clock stability for benchmarks?

Sent from my ONEPLUS A6003 using Tapatalk

I've heard of people enabling liquid nitrogen mode(I think it changes the PLL and VPP voltages) to get 1933mhz fclock, maybe it would help you get some extra core speed too?
 

·
AMD OVERLOCKER
Joined
·
727 Posts
I've heard of people enabling liquid nitrogen mode(I think it changes the PLL and VPP voltages) to get 1933mhz fclock, maybe it would help you get some extra core speed too?
Good tip. Will check if it's on or not.

Sent from my ONEPLUS A6003 using Tapatalk
 

·
AMD OVERLOCKER
Joined
·
727 Posts
Also try a ccx overclock, maybe 1 of the ccx's can go 25-50mhz higher.
Is that working from Ryzen master?
I have tried to oc with different clocks for different ccx and never managed to make it run correctly.

Sent from my ONEPLUS A6003 using Tapatalk
 
261 - 280 of 294 Posts
Top