This morning, one GPU started mining at 77 MH / sec on its own. Otherwise it works at about 40 MH / sec.
Nothing has changed in OC.
OS version 0.6-210@211018-2
Deiver 460.39
MiniZ 1.8y4rc1
It lasted from 7:30 to 09:00. At one GPU hasherate it was 78 Mh / s. Now the question arises.
Is it possible that this hasharate is already possible, so something is blocking it in the Hiveos software itself, or maybe in Miniz, so something unplanned happened so that it came to the surface.
Again this morning. Between 03:40 and 06:00 one GPU on its own raised the hash rate from 40 MH /s to 78 Mh /s.
Does anyone have an explanation for this
Is this hashrate also visible on the mining pool side?
Yes it’s visible
Damn, this could make these cards really expensive really soon, if someone can check what is “wrong” with his card. It kinda makes sense, 12gb memory…
I’ve had this on rig restart…3060ti bounced to around 72mhs…only lasted a couple of minutes then went back to its 46mhs.
Did make me think maybe it’s possible to get it Unlocked futher
has anyone tried upgrading the cooling pads in their cards? (eg gelid ultra)
i have other higher spec RTX card that exhibit similar behaviour ie automatic throttling back once they get hot.
put it outside on a really cold day and see what happens…
I have possibilty to put cards outside below 0 celsius. But where is that “throttlig back” setting?
3060 LHR ASUS
upd: i have LHR v2, put my rig to outside, 3060 was 10 degree celisuis, no effect. 35 mh.
yep… I changed the thermal pads… I also placed thermal pads between the backplate and the board and fitted aluminum radiators on it with thermal double-sided tape… and no change whatsoever… just a waste of money… I used Thermalright Odyssey pads and Arctic MX-5 for GPU…
if interested, for the front of the card, you need 2mm for VRAM and 1.5mm for MOSFETs… for the back, you need 3mm pads
thanks for the pad info: im doing them on 3080 and 3090 so, if my research is correct, other width of pad are use !?!?!
to earlier commenters: the throttling back is automatic and cannot be controlled afaik, other than with awesome cooling and the already known hacks that are inbuilt to the miners. I vaguely remember reading there was a ddr5/6 hack setting in hiveos??? my cards have ddr6x so it doesn’t work on them
FYI: on my 3090s the cards will almost always start at 123MH. then they throttle back relatively quickly to, say 110 or 113, sometimes lower and then stabilise. one card was even 80-something MHS…that went back. Two of my 3090s are perfectly ok from new and a third is nearly OK, overheats but still works at nearly 123MHs. you will find MANY factors affect them. for example mine perform differently on a rack depending on which other card is next to them and venting out warm or hot air directly onto the 3090. I put a 3090 into a rack chassis with awesome cooling…guess what…one of my dodgier cards was simply fixed by doing that as i had very good push and pull fans ie good airflow
then you realise that it’s now the middle of winter and you are going to have a heck of a problem in 6 months time when the air is significantly warmer. good luck!
yeah outside with a massive fan. then that would give you a pretty good indication on the status of your cooling pads!
I’ve done it for 3060s… for others like 3080, 3090, there are different requirements for pad thickness
Hey you three. What does this have to do with the topic of the basic post. Can you read. I ask along the way.
just following
Following what ? 3060 and 70 Mh/s ?
please, I have same cards and the maximum I’ve reached is 38/mhz, how did you reach 40MHz? can you share the miner configuration and the versions? thanks in advance
This topic was automatically closed 416 days after the last reply. New replies are no longer allowed.