Apparently I need help with AMD 1090t overclock

[Phate.exe]

1 hour ago, Mothballs said:

Here's the new kicker. Moved it across the room to a different outlet and Powerbar. Dropped voltages by .03 under load. I: The outlets are on different circuits, etc.

I think I'm gonna get a UPS as opposed to a power bar, my power is shit here anyways.

Dirty power will seriously screw with a PC.  I helped my old roommate chase down all sorts of issues on his PC (FX9590, R9 290, basically all the most power-hungry monster amd parts) because it would go from perfectly functional to random crash/shutoff.  Multiple OS reinstalls, slight underclocking, fans set to max, all drivers reinstalled, etc.  Moved it to my room, and everything was rock solid.  After hooking up a UPS in that room at a later date, we noticed it would click on and off periodically.  Because why the hell would the problem be the outlet?

 

And that's how we learned that the wiring in my old apartment couldn't handle the power draw from AMD's hungriest components, and why I won't be at all surprised if that building burns down in the future.

[Mothballs]

6 hours ago, Phate.exe said:

Dirty power will seriously screw with a PC.  I helped my old roommate chase down all sorts of issues on his PC (FX9590, R9 290, basically all the most power-hungry monster amd parts) because it would go from perfectly functional to random crash/shutoff.  Multiple OS reinstalls, slight underclocking, fans set to max, all drivers reinstalled, etc.  Moved it to my room, and everything was rock solid.  After hooking up a UPS in that room at a later date, we noticed it would click on and off periodically.  Because why the hell would the problem be the outlet?

 

And that's how we learned that the wiring in my old apartment couldn't handle the power draw from AMD's hungriest components, and why I won't be at all surprised if that building burns down in the future.

Well, I'm debating holding off any more stress testing until I have a UPS in here.

Any recommendations on that front? I do get occasional Brownouts.

Also, I'm in the clear running prime until about 8 hours in, then I get "Prime stopped working correctly" error and the program just terminates. Doesn't BSoD, which is nice, but It just decides to quit at 8 hours.

[Mothballs]

Going to go back at it tomorrow. I'm thinking this is more indicative of ram. From what I've seen so far, RAM tends to have a program stall and things freeze, where cpu sends the computer into a crash or a worker failing.

 

There are three things that I think should work. The first is just bumping ram voltages a tiny bit. On auto it wants to run 1.7-1.8 and it's supposed to be 1.5. Bringing it up a tiny bit more should be fine, seeing it ran for months at 1.7v.

 

Second thing to do is to slightly bump the cpu/NB voltage again. If it's ram related, that should also help in theory

 

One that's literally a last ditch effort that I'm going to hold of on until I've tried those is bumping HT voltages. Hopefully not the solution.

 

 

Part of me wishes this was a Deneb, because then all it cares about is vcore, multi and temp.

 

Though this is a challenge. And i like that.

[Phate.exe]

14 hours ago, Mothballs said:

Going to go back at it tomorrow. I'm thinking this is more indicative of ram. From what I've seen so far, RAM tends to have a program stall and things freeze, where cpu sends the computer into a crash or a worker failing.

 

There are three things that I think should work. The first is just bumping ram voltages a tiny bit. On auto it wants to run 1.7-1.8 and it's supposed to be 1.5. Bringing it up a tiny bit more should be fine, seeing it ran for months at 1.7v.

 

Second thing to do is to slightly bump the cpu/NB voltage again. If it's ram related, that should also help in theory

 

One that's literally a last ditch effort that I'm going to hold of on until I've tried those is bumping HT voltages. Hopefully not the solution.

 

 

Part of me wishes this was a Deneb, because then all it cares about is vcore, multi and temp.

 

Though this is a challenge. And i like that.

I've never touched a deneb, but yeah those do seem to be a bit siimpler to deal with.

 

At least until I get around to a repaste I'm stuck at just under 4.1GHz on mine, but I'm probably getting close to the wall anyways.

[Mothballs]

14 minutes ago, Phate.exe said:

I've never touched a deneb, but yeah those do seem to be a bit siimpler to deal with.

 

At least until I get around to a repaste I'm stuck at just under 4.1GHz on mine, but I'm probably getting close to the wall anyways.

I checked error logs, and it kicked out another small cpu error, but didn't crash or hiccup so I bumped it up a little on voltage.

 

If you have a water loop that's good you should be fine to punch towards 4.5ghz. I'm just at the edge of 52c with a 212 and it's around 4.06- .09 so far. seems to me like you should be fine. Try messing with the cpu/nb, it really does seem to add a ton of stability as long as you keep it proportional to the cpu speed.

 

 

 

More to come.

[Mothballs]

I dug this up in my research. Could be handy to anyone interested in this old CPU.

[Mothballs]

A bit of digging brought up possible issues with bumping up Baseclock and Data corruption/SSD damages. While a very minimal risk, I Decided to err on the side of caution. Trying for just a straight Multi Overclock, bust saved my last known good settings. Doesn't look promising.

Does anyone else on here not in the Drive section have any idea if there's truth to Baseclock causing issues with SSDs and Data corruption? I'd like some more oppinions.

[Mothballs]

Went with the baseclock anyways. If so many people use it it should be fine is properly declared stable.

 

That said, WHEA Error Cache hierarchy error at 21 hour. So close you can taste it...

 

Threw this when I had the memory bumped too high. Not sure if it's RAM related or... Regardless, going to drop the baseclock a notch, temps are fairly high so It's gotta come down...

[Mothballs]

Got there. After a thousand years I rise.