MSI Unify 490 and i9-10900k overclock questions/settings/trials and frustrations? High temps?

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
I've got this board and cpu combo coming, along with pc4-3200 4000mhz cas17 gskills (16x2).

I'm curious as i've seen many successfully overclocking to around 5-5.2ghz on average, has anyone tried doing the per core to keep the (two?) 5.3ghz cores and lower all the rest to 4.9 to 5.0 at most? Any reason this cant be done or does that cause stability issues to not keep them all the same?

I was thinking that would optimize a few cores for gaming (sims) that could benefit from the higher clock speed. I guess ill tinker once the equipment gets here, but just wondering.

Thanks in advance
 

Darkswordz

Gawd
Joined
Jan 5, 2016
Messages
655
I have the same combo, and had the same thoughts. However, when I tried to do per-core overclocking in the BIOS it wouldn't take. I also tried the Intel XTU but I couldn't get per-core working in it either.

Maybe I was doing it wrong. Eventually, I just settled on a 5.1 all core overclock and felt the per-core overclocking wasn't worth the effort since I game at 4K and am GPU limited anyway.
 
Last edited:

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
I have the same combo, and had the same thoughts. However, when I tried to do per-core overclocking in the BIOS it wouldn't take. I also tried the Intel XTU but I couldn't get per-core working in it either.

Maybe I was doing it wrong. Eventually, I just settled on a 5.1 all core overclock and felt the per-core overclocking wasn't worth the effort since I game at 4K and am GPU limited anyway.
By wouldnt take do you mean the ratio never gets hit when your stress testing or gaming?

Im getting close to some values that work, however:
as of now
1.3 volts bios (lowered from 1.31), have set to auto cpu vcore (not override one), have turbo boost on (wont go above 37x if not on), eist is on.
I set the rato on two cores to 53x and the rest to 51x. For the ram i have it set now to 4200 (18,18,18, 40 i think are the timings) and 1.5 (i may try to lower these volts). CIO/SA are still at I think 1.2 for now.

When i do all this, i test fs2020 as one test, eglc in mr20 carenado runway 27 clear skies, 4k setting, high and i get 38 fps (i was getting 40 with my 8700k at 5ghz ram at 3200), so im a bit bummed.. i noticed in sim the ratio never exceeds 47x (never looked before so not sure if normal), this with cpu 19-25%

Then occt medium no avx test, im only seeing a max clock on each core of 5001.2mhz, max temp 78C over 17mins (and errors too, so lowering to 1.30 volts was probably a bad idea)

Im not clear why the ratios arent hitting 53 on two and 51 on the rest?
 

Dan_D

Extremely [H]
Joined
Feb 9, 2002
Messages
57,036
All core, these typically do 5.0GHz to 5.1GHz. Some people are a little luckier, and might be able to get 5.2GHz with the right cooling. The reason why you aren't seeing 5.3GHz on two cores and 51x on the rest comes down to the fact that these are a lot like Ryzens. All core loads aren't going to get 5.3GHz on two cores and less on the rest. It will settle on an all core clock of around 5.0GHz or less depending on the workload.
 

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
All core, these typically do 5.0GHz to 5.1GHz. Some people are a little luckier, and might be able to get 5.2GHz with the right cooling. The reason why you aren't seeing 5.3GHz on two cores and 51x on the rest comes down to the fact that these are a lot like Ryzens. All core loads aren't going to get 5.3GHz on two cores and less on the rest. It will settle on an all core clock of around 5.0GHz or less depending on the workload.
After i changed from per core to all core ratio and 51x, it now shows 51x in occt. so i guess its actually the per core setting that had an issue, individually.
 

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
Updated now with pretty good success:

Survived 2 hours on Occt Medium and about 4 hours on Occt Large
Max temp 81C

Bios 1.32 and auto mode (not override)
All Core, 51x
Eist disabled greyed out/Turbo greyed out
Ring Set to auto (43x)
Thermal velocity Boost is set to defaults auto
AVX offset -4
SA 1.26 (possibly a hair higher than safe 24/7? 1.25)
IO 1.26
Ram 1.48
Ram xmp and 4000
LLC4
LN2 Extreme
Cstate Enabled
Speed shift Disabled
PCIe Spectrum Enabled

Im now trying to boost the Ring ratio, tried 49x and it froze on boot, so now trying 47x. Once i get that set ill walk back on the vcore and/or cio/sa values slightly (though i think before when i tried this it was error city for occt medium in minutes)

47 ring made it 3hrs med occt. I think the ram temp hit 55c.overall max still around 84c, i was then able to do 1.30v 81C (ac unit blastin to keep the ambient lower) for 1hr 16 med occt at 47ring/51x. Testing 1.25 cio/sa now, trying to not exceed 1.25 (may try for 4100 or 4200 ram again though, everything else held equal).

Tried 52x at 1.31, got a bsod on med occt in 10 sec, i guess 52x is a no go, not without more volts and too much heat.
I tried going down to 1.29, had an insta fail there.

**So at this point i think the sweet spot for 51x is 1.30v-1.31 volts. Avx offset is still -4
When set to 1.305v in bios, hwinfo vcore shows 1.298 for the max (core temp has vid as high as 1.3589v).
If i set the power management to balanced, it appears the cpu is throttling down (since i have it on auto not override for vcore i guess).

I need to test an avx load, but p95 with avx the cpu gets too hot, may try just realbench for now a few times.
 

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
My real word temps now that i'm using the above settings are about 72-80c max for general windows use, 83C during some games (others less) and 87C for rendering (avx involved i think). My 8700k delidded at 5ghz was around 78c at worst on anything.

I think the temps are too toasty? Even for the thermaltake 360 unless its not working that well anymore.

My option is to use a delidder i have coming, but even then i might only gain 5c.

Anyone else with the 10900k can comment on your typical load temps?
 
Last edited:

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
Well it turns out that my oc probably wasnt stable. I passed P95 blend for 12 hours, but random watchdog violations keep occuring. Retested occt small (95c btw) doesnt pass about 15 mins, lowered to 50x still 95c same volts, making it further on small occt. Probably need small occt for an hour to be safe. update: now making it about 1 hour on small occt but at 50x, probably needs a bit more than 1.305v to stay at 51x , guessing 1.315 or so

Temps at this voltage level i now have real data, the worst is realbench, 88C, but integration rendering software only hits about 78, general gaming maybe 75 max (or 83 on a load sequence)) if you ignore realbench i'd probably fine to NOT attempt a delid, maybe just try a different cooler, perhaps my fluid in mine after 3 years has reduced.
 
Last edited:

sirmonkey1985

[H]ard|DCer of the Month - July 2010
Joined
Sep 13, 2008
Messages
22,228
My real word temps now that i'm using the above settings are about 72-80c max for general windows use, 83C during some games (others less) and 87C for rendering (avx involved i think). My 8700k delidded at 5ghz was around 78c at worst on anything.

I think the temps are too toasty? Even for the thermaltake 360 unless its not working that well anymore.

My option is to use a delidder i have coming, but even then i might only gain 5c.

Anyone else with the 10900k can comment on your typical load temps?

10900k isn't worth delidding since the die's sanded down(e.g. doesn't have the problem the 9900K has) and it's soldered so at best maybe you see 2-3c difference. you'd see more gains out of lapping the IHS and your block then delidding it and even that's probably not going to be worth a whole lot other than tightening up the individual core temp delta's.
 

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
10900k isn't worth delidding since the die's sanded down(e.g. doesn't have the problem the 9900K has) and it's soldered so at best maybe you see 2-3c difference. you'd see more gains out of lapping the IHS and your block then delidding it and even that's probably not going to be worth a whole lot other than tightening up the individual core temp delta's.
I think the average drop is more like 7 from what I saw, but bringing it down from a max of 87C sounded like it was worthy, however i guess its not that bad, but...

I spotted the rocketcool delid kit with direct die option, looks fairly simple to achieve and i'm going to give this a go once it gets here, should drop things 10-12c hopefully. First however, i'm testing another AIO soon to see if it works better than my 3 year old one.
 

somebrains

[H]ard|Gawd
Joined
Nov 10, 2013
Messages
1,361
Do you have to be pegged 24/7?
Adaptive voltage and real world clock thresholds are what I'd use after finding my boundaries.

Have you considered an open loop if delidding and bare die cooling are on the table?
If running pegged 24/7 was my plan, I'd look to as much thermal capacity as I could cram in.
 

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
Do you have to be pegged 24/7?
Adaptive voltage and real world clock thresholds are what I'd use after finding my boundaries.

Have you considered an open loop if delidding and bare die cooling are on the table?
If running pegged 24/7 was my plan, I'd look to as much thermal capacity as I could cram in.
I really dont want it pegged 24/7, i was going to explore that once stable.
I have it on fixed cpu ratio mode now, i think dynamic might be a better choice there and maybe adaptive voltage as well, but i forget how to figure that one out.

I probably wont go open loop, not yet, i can hit 89c during realbench at 1.30 ish, if i can bring that down a tad or even safely delid and get it to 82 i'd be happy, though its not too bad.

I had everything working at 1.29, played around with ram speeds to try to boost things, failed on that, so i put back the working 1.29 (hours of success on realbench, prime95 blend, occt medium etc), now realbench will crash each time after 15-30 mins with page fault in non paged area.

Back to drawing board for some odd reason. I may even swap out the msi board for the asus extreme II to rule that out, though i'm pretty sure its just a flaky chip and there arent other 10900ks avail to swap with now to test that idea, even 10850k's are hard to find, though i think i could get one here in a week.
 

Dan_D

Extremely [H]
Joined
Feb 9, 2002
Messages
57,036
After i changed from per core to all core ratio and 51x, it now shows 51x in occt. so i guess its actually the per core setting that had an issue, individually.

I didn't have much luck with per core overclocking on the Core i9-10980XE either. I've never tried it on the 10900K.
 

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
So i'm still unsure how i managed stability at 1.29 previously, but now that i relied on auto volts i've found real stability:

Auto volts in bios on all, LLC auto, xmp and 4000 for memory and auto volts there too. 51x and auto ring (if i set ring to 47 windows wouldnt boot even on auto volts)This results in an idle of around 1.301 initially in windows. Under extreme load this is 1.315 (sa 1.321 and cio 1.242).

Realbench initially hits 81C by 3min 89C, by the end 94C (too hot), this is with the case side off too. Now if i blast the ac window unit i can get it down to around 90C.
I was "realbench stable" all night, about 6 hours worth. I consider this ok now.
I rebooted into normal windows (I use an isolated test windows for stress tests) and tested a few activities, ambient is slightly cool right now though.
Xplane 11 sim i only hit about 78C (probably 82 when room is warmer). Fs2020 only 65C. A quick 4k video render, about 75C. An astro 100 sub integration (avx) hits 91C with ambient up by the case at 24.8C (76.6F).

*ambient where the case sits above the desk is around 76-77F, so around 25iC ish (fans on 360 blow out the top, 1080ti hybrid rear and 2 120's in the front pull air in, in the v51 thermaltake case) , it would be about 23c (74f) if i had the room to put it more floor level

Cinebench : 6506 (default clocks 6350 at 49x)
Passmark cpumark: 27868 (default clocks 27031 at 49x), no change in fs2020 or xp11 fps regardless here, my integration time did go from 12 min to 9 minutes though (avx).

I guess Ill be leaving cio/sa at auto, seems ok. I will probably move the cpu to fixed 1.315 (the most under load) and actually try to figure out adaptive/offset, guessing i set adaptive to 1.301 and +.014 offset. Tho the cpu mode is set to "auto" not override or adaptive right now, i believe it just still down volt under no load regardless.

I'm going to try the krakken 360 z73 before delidding/and/or another version of the chip.
 
Last edited:

Furious_Styles

[H]ard|Gawd
Joined
Jan 16, 2013
Messages
1,947
So i'm still unsure how i managed stability at 1.29 previously, but now that i relied on auto volts i've found real stability:

Auto volts in bios on all, LLC auto, xmp and 4000 for memory and auto volts there too. 51x and auto ring (if i set ring to 47 windows wouldnt boot even on auto volts)This results in an idle of around 1.301 initially in windows. Under extreme load this is 1.315 (sa 1.321 and cio 1.242).

Realbench initially hits 81C by 3min 89C, by the end 94C (too hot), this is with the case side off too. Now if i blast the ac window unit i can get it down to around 90C.I was "realbench stable" all night, about 6 hours worth. I consider this ok now.I rebooted into normal windows (I use an isolated test windows for stress tests) and tested a few activities, ambient is slightly cool right now though.Xplane 11 sim i only hit about 78C (probably 82 when room is warmer). Fs2020 only 65C. A quick 4k video render, about 75C. AAn (avx) integration hits 85C in the somewhat cool morning ambient edit 89c with case side on. I dont know many other workloads to test, but given this it might be more than fine for what i do.

*ambient where the case sits above the desk is around 76-77F, so around 25iC ish (fans on 360 blow out the top, 1080ti hybrid rear and 2 120's in the front pull air in, in the v51 thermaltake case) , it would be about 23c (74f) if i had the room to put it more floor level

Cinebench : 6506 (default clocks 6350 at 49x)
Passmark cpumark: 27868 (default clocks 27031 at 49x), no change in fs2020 or xp11 fps regardless here, my integration time did go from 12 min to 9 minutes though (avx).

I guess Ill be leaving cio/sa at auto, seems ok. I will probably move the cpu to fixed 1.315 (the most under load) and actually try to figure out adaptive/offset, guessing i set adaptive to 1.301 and +.014 offset. Tho the cpu mode is set to "auto" not override or adaptive right now, i believe it just still down volt under no load regardless.

The last debate is whether to try the asus ROG Maximus XII Hero which i ordered to arrive tomorrow in case i couldnt figure things out. I'm not convinced that the "better" board would yield lower temps under load, but maybe it would. If temps are higher than this for any other activity (IE: above 88C, ill probably delid, i feel i can do it safely and drop 7c). edit 89c with the case side on during one process, thats borderline too much.

Your method is all over the place. You keep changing stuff so it's hard to keep track of what the main issue is. And I don't know what your goal is either. Dropping 7C will net you almost nothing significant performance-wise nor will buying a new mobo unless you think your current one is defective.
 

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
Your method is all over the place. You keep changing stuff so it's hard to keep track of what the main issue is. And I don't know what your goal is either. Dropping 7C will net you almost nothing significant performance-wise nor will buying a new mobo unless you think your current one is defective.
Admittedly it was a mess at first but now I have the voltage nailed at 1.315

Issue is my worst real world temp is 91c during avx integration routine. Which is less often ran. I feel better at 85c. Yea the board swap won't help. But a 7c delid would. Or maybe the kraken z73 VS the thermaltake I have now
 

Furious_Styles

[H]ard|Gawd
Joined
Jan 16, 2013
Messages
1,947
Admittedly it was a mess at first but now I have the voltage nailed at 1.315

Issue is my worst real world temp is 91c during avx integration routine. Which is less often ran. I feel better at 85c. Yea the board swap won't help. But a 7c delid would. Or maybe the kraken z73 VS the thermaltake I have now

Yeah but I'm just also wondering is dropping $500+ for 100mhz that worthwhile? I also noticed you are using FS2020 to test, a notoriously unoptimized pos that should be avoided.
 

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
Yeah but I'm just also wondering is dropping $500+ for 100mhz that worthwhile? I also noticed you are using FS2020 to test, a notoriously unoptimized pos that should be avoided.
No, going from 8700k 5ghz to this isnt worth the price. Unless you are me and doing integrations (astronomy) that feed off of as many cores as possible and had a secondary living room htpc that had a 4790k cpu with 1080 in it. I moved my 8700k to that machine ftw. Gained cores for rendering.
Yeah fs2020 doesnt use much cpu, xp11 doesnt use much extra but a tad more so. My two primary sims. If i can find the right 50x voltage and lower the rendering from 91c to 87C ill probably just leave alone and not try switching coolers or even testing another cpu (or delidding).
 

Furious_Styles

[H]ard|Gawd
Joined
Jan 16, 2013
Messages
1,947
No, going from 8700k 5ghz to this isnt worth the price. Unless you are me and doing integrations (astronomy) that feed off of as many cores as possible and had a secondary living room htpc that had a 4790k cpu with 1080 in it. I moved my 8700k to that machine ftw. Gained cores for rendering.
Yeah fs2020 doesnt use much cpu, xp11 doesnt use much extra but a tad more so. My two primary sims. If i can find the right 50x voltage and lower the rendering from 91c to 87C ill probably just leave alone and not try switching coolers or even testing another cpu (or delidding).
Any reason you didn't go something like the 3950x? Seems like core count is more beneficial for your workload.
 

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
Any reason you didn't go something like the 3950x? Seems like core count is more beneficial for your workload.
Mainly out of bias for intel :D More so clock speed, i assumed i could hit 5ghz more easily with the intel and since i do alot of simming, clockspeed wins, but yes that amd is a beast too.
 

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
Update on things, so at this point i was stable at 1.285 bios volts for 5ghz.

I ended up delidding, I gained an average of 5 or 6C as expected.

I decided to redo going for 5.1ghz.

I had to settle on this setting so far:

1.330 bios LLC4, load of about 1.320, 51x, ring at auto (43x).
Cinebench test example: 73F ambient/case on: 83C, score 6403

I think that number before was about 88C

I still think compared to other 10900k reports i've seen at similar load volts, the 83C is high, others seem to be around 77C but maybe a fluke or maybe the whole variance of 10C for this chip.
edit: another comparison i found was 84C cinebench with coolant at 29C (i cant read mine, no usb), ambient about 24C, so he was about the same, but not delidded, so i'm maybe off 5C unless chip temp lottery. (edit he was using a custom loop so i think my numbers are about right then)
 
Last edited:

Dan_D

Extremely [H]
Joined
Feb 9, 2002
Messages
57,036
Don't forget, altitude effects temperature readings. Also, the method for reading the CPU temps make a difference as well. I don't typically worry about who is off by 7c or whatever. As long as the system is stable and you aren't seeing evidence of throttling in benchmarks you should be good to go.

Buying a different board isn't going to change your results. Generally I can get the same overclocking results on any motherboard. The main difference is how much effort it takes to achieve that. Now, long term life span of the board in question is another matter. The cheap ones I can't imagine surviving for years with their VRM's at near capacity under hard use.
 
Last edited:

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
Don't forget, altitude effects temperature readings. Also, the method for reading the CPU temps make a difference as well. I don't typically worry about who is off by 7c or whatever. As long as the system is stable and you aren't seeing evidence of throttling in benchmarks you should be good to go.

Buying a different board isn't going to change your results. Generally I can get the same overclocking results on any motherboard. The main difference is how much effort it takes to achieve that. Now, long term life span of the board in question is another matter. The cheap ones I can't imagine surviving for years with their VRM's at near capacity under hard use.
Yeah, i think its all fine now. Delidding certainly helped my sanity on keeping them to my own personal comfort level, though my Heart Rate went way up during the delid process :D (I typically get from silicon lottery but used the one d8railer tool this time).
 

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
My 51x settings previously were not quite stable, i was getting L0 cache errors in hwinfo.

I was able to use occt linpack to quickly test and get rid of those errors (along with the std tests) so now i'm at these for 51x:

51x, ring auto (43), 1.35 bios , llc4, avx auto (51x)
Load 1.339
My astro avx integration hits 85C (at 73F)
Cinebench 84C and 6400 score

**if i up the ring to 47x
My astro avx integratin hits 88C (at 73F)
Cinebench 85C and 6560 score

Passmark at 47x ring is 27,787

Again all these temps are about 5 or 6C cooler after delidding, i may just leave it at 47X.
 

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
Another update/odd situation:

Not sure if its related to the overclock which can pass every test at this point, but im randomly losing my usb connections, keyboard, everything, non responsive, try different port same deal. Might also be related to sleep mode and usb selective suspend, unsure.
Example, PC was asleep (have to wake up via power button, keyboard taps wont wake it for some reason), everything functioning normally after awake, ran fs2020, exited, ran xp11, then keyboard lost and tried other ports most everything else usb wise was "down" but showing in dev manager.

edit : fixed the wake with keyboard issue, however i think the usb issue is more hardware and not related to overclock, i reset things back to default auto settings and im losing the keyboard and other peripherals randomly, though usually from unplugging one and plugging back in another. Not sure if it matters but i do have a usb going into the bios usb port on the motherboard. When this happens even putting the keyboard into the 3rd party usb card doesnt work.
 
Last edited:

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
I seem to have overcome the usb issue by moving things around, however

My os has bsod a few times, i cant seem to figure out why. 5.1ghz 1.345 (now 1.35 after last bsod) either way load of 1.339, avx 0, ring 43, ram 4000 at xmp and autos.
I can pass 12 hours prime 95 blend (avx fma set to off), i can do 6 hours small ffts (avx/fma off), i can do hours of occt small fft, i can do realbench for 4 hours etc, still a random bsod.

The last one seemed to be related to a western digital driver/hdd (Wdf01000.sys 0x0000010d)
. Before that it was kmode_exception_not_handled.

Hoping that raising to 1.35 prevents this, though it passes the tests all the same and load didnt change (read that sometimes it may like a higher vdroop), so .90mv vs .95mv i guess. I also didnt reinstall the OS after swapping motherboards, so unsure if its related. I may try running the same stress tests in the culprit OS (i run on a clean OS/isolated stress tests).
 

somebrains

[H]ard|Gawd
Joined
Nov 10, 2013
Messages
1,361
I think you are finding stability issues back when guys clocking would use ps2 peripherals.
Driver issues can be a problem, mine usually come from the fact I make a living using HyperV, and it doesn't like a system pushed to the edge all core.

I had usb issues like you describe back in the Core era and prior chipsets when clocking.
If you feel dive 15 years ago you'll find guys with their own blend of custom IRQ mapping.
 

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
I think you are finding stability issues back when guys clocking would use ps2 peripherals.
Driver issues can be a problem, mine usually come from the fact I make a living using HyperV, and it doesn't like a system pushed to the edge all core.

I had usb issues like you describe back in the Core era and prior chipsets when clocking.
If you feel dive 15 years ago you'll find guys with their own blend of custom IRQ mapping.
My usb issues i dont think are related to the overclock. I put things back to default and they still occurred (though i left ram on xmp for defaults).

I'm now suspecting my keyboard + extension is somehow causing an issue, or the keyboard has a glitch, its a very old ms ergonomic.

However, back to the overclock at 5.1 and the delid, despite the delid, my blender benchmark temps after only 2 of 6, hits 89 to 91C depending on room temperature and my real world avx software that i use hits about 86C on average. I'm ok with the 86c, not so much on the 91C but that is blender however. I may drop back to 5.0 and 1.285 volts rather than the 1.35 bios volts at 1.339 load im at now.
 

Dan_D

Extremely [H]
Joined
Feb 9, 2002
Messages
57,036
I seem to have overcome the usb issue by moving things around, however

My os has bsod a few times, i cant seem to figure out why. 5.1ghz 1.345 (now 1.35 after last bsod) either way load of 1.339, avx 0, ring 43, ram 4000 at xmp and autos.
I can pass 12 hours prime 95 blend (avx fma set to off), i can do 6 hours small ffts (avx/fma off), i can do hours of occt small fft, i can do realbench for 4 hours etc, still a random bsod.

The last one seemed to be related to a western digital driver/hdd (Wdf01000.sys 0x0000010d)
. Before that it was kmode_exception_not_handled.

Hoping that raising to 1.35 prevents this, though it passes the tests all the same and load didnt change (read that sometimes it may like a higher vdroop), so .90mv vs .95mv i guess. I also didnt reinstall the OS after swapping motherboards, so unsure if its related. I may try running the same stress tests in the culprit OS (i run on a clean OS/isolated stress tests).

Try some Handbrake encoding or BlenderBench. I've found these things will cause crashes when many stability testing programs gave the impression a machine was stable. Additionally, take the ring stuff out of the equation and set that back to auto. Messing with that can cause stability issues real quick. If everything is solid, then you can start messing with that stuff to fine tune performance. I'd also look at your load-line calibration settings. You might want to see where its at and increase it slightly.

Also, if you are having USB and other issues, that could be related to anything that causes instability. However, check your base clock and make sure it's not running over 100MHz.
 

markm75

Gawd
Joined
Sep 12, 2008
Messages
747
Try some Handbrake encoding or BlenderBench. I've found these things will cause crashes when many stability testing programs gave the impression a machine was stable. Additionally, take the ring stuff out of the equation and set that back to auto. Messing with that can cause stability issues real quick. If everything is solid, then you can start messing with that stuff to fine tune performance. I'd also look at your load-line calibration settings. You might want to see where its at and increase it slightly.

Also, if you are having USB and other issues, that could be related to anything that causes instability. However, check your base clock and make sure it's not running over 100MHz.
Yeah i probably posted the same time you did, but updated to say i (briefly) tested blender benchmark (2 of 6) and was noting mainly my temps at running avx at 5.1x. 89-91C (delidded temps, case on, 74-77F room temp)

I thought i read maybe avx should be -3 in fact (meaning 48 or 49 if -2), but i swear when i tested things on auto it was running at the clock speed and was set to zero, unsure on that. I could drop temps by going negative on avx slightly. Ring has been on auto (43x), i havent tried going up yet on that. I think the usb issue is a faulty old keyboard + extender or similar, though i cant wake the machine by keypress despite setting that bios option, no matter which keyboard. I have just about everyting on auto and things like turbo and eist set to auto/enabled at this point, so possibly bclk is not 100 fixed (i have the cpu on dynamic too, but it stays at 5.1ghz under any loads).

Today i'm also swapping out my cpu aio cooler from the thermaltake 2017 version to the x73 kraken aio 360. Mainly because the fans they have on the old TT arent being made anymore and any others cant interface with the controller. Maybe ill get lucky and the new one will cool slightly better.
 
Last edited:
Top