• 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.

ASROCK X570 TAICHI - THE RYZEN 3000 MOTHERBOARD GIBBO & 8 PACK RECOMMENDS

Bios 3.91 just realised that under PBO limits it now shows a disabled option which i am pretty sure was not there on previous versions?

Anyway just selected this and all cores of my 5950X now hitting 4950 and above.
 
Ok lets rephrase this.

AMD overclocking PBO limits Auto, disabled, motherboard, manual.

BUT just ran cinebench and back down to 26K plus with 61C peak
No load shows all cores peaking above 4950 which of course aint worth a damn really.

With manual all at 300 PBO limits max temps 80C and scores over 29K Cinebench R23.
I shall be going back to this I think.

Unless of course theres something else i am missing.
 
Spoke too soon, the problem occured again with the memory set to xmp 1.35v 3600mhz 1800 IF. Could be a problem elsewhere, not the gpu as it's been changed but I use phanteks sleeved cables to power it, could potentially cause issues? although it does run for about 30 mins before crashing. Failing that it's going to be setting up memtest 86, maybe looking if cpu or board may be faulty? Power supply is quite new incidentally, was gearing up for a 3080 with a superflower 750W leadex 3, don't know if that could be an issue?

edit-the error that comes up is 'The UE4 oakgame game has crashed and will close'
'Fatal error'
'unhandled exception:exception_access_violation writing address 0xffffffffffffffcf'
 
Shac said:
Ok lets rephrase this.

AMD overclocking PBO limits Auto, disabled, motherboard, manual.

BUT just ran cinebench and back down to 26K plus with 61C peak
No load shows all cores peaking above 4950 which of course aint worth a damn really.

With manual all at 300 PBO limits max temps 80C and scores over 29K Cinebench R23.
I shall be going back to this I think.

Unless of course theres something else i am missing.
Click to expand...

I found manual limits is the only way to get PBO to work and even then it doesn’t work as well as on other boards.
 
Finally set up memtest, about 2 hours in it showed 3 errors at the 3600mhz. Explains the intermittent fault as it took the 3rd pass for it to show up. The good news is it's hopefully the easiest fix, as long as a new set is stable I can carry on with the plan to put a 5800x in. I just needed to get things 100% working before commiting to an upgrade. At least as part of the troubleshooting the latest bios is on there now ready for it.
 
it's in a lian li 011 dynamic with 3 side fans, the gpu is obviously right under them but I wouldn't have thought just on memtest that could cause a problem, the sticks are slightly obscured by the 24 pin motherboard connection but there is a clear path from the fans right next to them. Might be unrelated but i've got the fan curve set up on the south bridge fan so it kicks in at 65 deg.
edit- the radiator for the cpu is at the top of the case as exhaust, so only cool air coming in. ambient temp 20 deg.
 
I saw 1 error on mine after 3 hours. A fan stopped that error. During 'normal' usage, the RAM will not get that hot - I have stuck probes on my b-dies. I am putting 1.5v into mine though. I have the XL, so air flow sh (sould be the same ish - side in, bottom (rad) in top (rad) out.
 
It could help, but the main problem I have is during gaming and it has been known to happen a couple of minutes after launching. At the same speeds it takes 2 hours to error in memtest. Other times it can go about 6 hours of on and off gaming when it's set to 3200mhz, but it still crashes. That's what makes me think the rest of the heat from the system isn't causing the problem. 3000mhz seems to be much safer but obviously I didn't get 8 pack ram for that!
 
ok ill try at the 1.4V, that's the max it says on the box, on the back it says 1.2-1.4 ultra low working voltage
Edit-1.398 was as high as the board allowed it to go, there is probably some way of overclocking but I'll test at that for now
 
Last edited:
It just crashed at that anyway, about 30 mins in. Looking on the bios, typing in anything more than 1.398 makes it turn red and it just defaults back to that maximum. I can't see anywhere else to change the voltage. On the dram info it has the xmp profile set as 1.35v, the standard jedec are at 1.2. Part number is TEAMGROUP-UD4-3600 so it should do it, the bandwidth is down as pc4-19200 (1200mhz).
So if I return it and its not faulty, I'll just have to pay the shipping and at least I narrow it down.
edit- shouldn't make a difference if the dimms are in slot 2 and 4? the user manual says to have them spaced apart for dual channel
 
Last edited:
You are correct, anything over 1.398v and the text turns red. You are permitted to use that though. I think your BIOS is the problem.

Have you tried a complete flash back. If you are on Ryzen 3000 I found 3.61 the best for memory clocking, anything newer and it would crash for me.

Also, have you confirmed that your chip can run 1800 if?
 
Chip can run 1800 if no problem, as long as the speed is 3000mhz or less. OK ill try another bios tomorrow, I started on 1.6 now this one is 3.8, i'll try the one you have suggested
 
You must log in or register to reply here.
Back
Top Bottom