Advice on Max safe voltage on Mem

Had an interesting day today. Quite helpful working at home as can make the odd BIOS a tweak now and then and leave memtest to it :)

Turns out too many volts was holding me back. I was 1.45v RAM and 1.175v SOC. Stable only at 14-14-14-14-38 with 56 tRC and 350 tRFC.

Dropping to 1.4v and 1.125v I’m now 400% HCI stable at 14-14-14-28 with 46 tRC and 270 tRFC which is a considerable gain.

No idea if properly stable and needs cycling as doubtless F9 hell remains but interesting nonetheless!
 
Interesting... keep me posted... would love it if they have come up with a memory training process that doesn't F9 so easily...
 
With B die sometimes lower volts is helping for 24-7. Some ofcourse scale others do not. Always worth trying different voltage combinations if you have time.
 
EL_JOCK said:
good work, defo interesting with the volts.....Just put F8 bios on so might give it a look
Click to expand...

F9 now out and a couple of reports of people getting 3466 where before they were stuck lower. Might give it a go. Did you have any luck with F8?
 
F8 was much the same, will have a look at F9 next week when i have a bit of time before starting university,
 
Put Bios F9a on today, my 3333 settings went in with no issues and seems to be running a little better.

Going for 3466 seemed better but still unable in my quick testing to get it stable but it was running better i.e no mass F9 errors. The only issue would be the slacker timings for 3466 vs 3333 might be counter productive.
 
Cool. I shall try it tomorrow AM and report back. I’m stable in HCI so if I can just get rid of F9 hell I might be good at 3466.

Also heard that the voltage drop under load is fixed so might be able to run my 4ghz...

Gigabyte rep also just announced that P states are coming in the next BIOS so if that all works the board is in a good place.

Bet it doesn’t though :D
 
Seems less F9 prone at 3466 for me. Not perfect but an improvement. Will tweak some more and report back.

vcore and vdrop not improved although 1.5ghz bug fixed
 
Got 3466c14 HCI stable with some seriously tight timings but F9 whilst better still isn’t good enough sadly. Likely back to 3333 although I may try some further tweaking when I get time.
 
at least the BIOS updates seem to be making it better, should not be long before the next aegis update and maybe P state overclocking as well
 
You must log in or register to reply here.
Back
Top Bottom