Back to ETH after 3 months on ERG; I guess this could be the final push so I decided to re-do all my software and overclocks for ETH…
Very successful almost first time 
; 25x Sapphire Nitro+ Vega 64 Cards (Samsung Memory) with 56 BIOS flashed - across 4x Rigs keeping it as simple as I can:
This is a B-mode configuration (a little harder to stabilise).
- HiveOS: 20.40 (5.11.0701)
- TeamRedMiner: Extra config: --eth_config=B
- B-mode OC’s: Core Clock = 1050Mhz, Core Voltage = 812mV, Memory Clock = 1020Mhz
(Memory Clock = 1010Mhz for the card that crashed) - Advanced config: amdmemtweak --CL 20 --RC 37 --RP 11 --WR 14 --CWL 8 --FAW 12 --RAS 20 --REF 65535 --RFC 248 --RTP 5 --RRDL 6 --RRDS 3 --WTRL 9 --WTRS 4 --RCDRD 12 --RCDWR 12
(–RC 38, for the card that crashed) - One card that I had to do both on to get stable; Memory Clock = 1010Mhz and --RC 38…
My Vega 64’s are putting my 5700 XT’s to shame for Cost vs MH
Hola tengo una gigabyte con memoria samsung , vega 56 claro cual seria la correa?
Guys any help for invalid shares in GPU 0 ? the invalid keep building up to 5 invalid per days
it’s give me a headache, I’ve done :
- up RC from 37 to 38
- up RCDRD from 12 to 14
- decreased MEM from 1020 to 1010
- decrased CORE from 1050 to 1030
amdmemtweak --CL 20 --RC 38 --RP 11 --WR 14 --CWL 8 --FAW 12 --RAS 20 --REF 65535 --RFC 248 --RTP 5 --RRDL 6 --RRDS 3 --WTRL 9 --WTRS 4 --RCDRD 14 --RCDWR 12
should I lower the OC settings more ?
Hello, So i have a powercolor vega oc 64, not sure what the original bios was but this one in use is 016.001.001.000.008769.
I would like to flash to 56 bios to get higher than 36mh max…
UPADTE: after messing around with tweaks, i got up to 54 mhs with the 64 bios. After reading vegas 64 oc 8gb with samsung mem dont get much pass 50mh, im gonna leave it alone. unless someone has any advice to boost it up any.
After some more tweaking got it to 50mh with 73 watts(seems low, idk how), that good overall, or better to try and boost mh?
New to hiveos, last build was in 2018, so i ask noob questions.
WELL, of course i tried to get more MH and gpu now assumed bricked?
Fan doesnt spin, but starts to when turned on but stops before 1 spin, tried to use switchs on bottom of card, no luck… checked risers, slots, plugged in only that card, etc…
Any ideas? Thanks for any help.
try to set amdmemtweak as image below and change the RC to 38
last resort try to flash using 56 bios, because you gain more hash and lower watt in 56 bios
Cant flash bios, cant see card with atiflash or HWiNFO. But can see card with Display Driver Uninstaller DDU. Have another card that was bricked and had vega bios and 4, 570 bios recognized, ugh.
Anyways, got a couple techniques to try, reflash from dos, connect pins on pcb… reflash emproms wwith j-link or something.
Hey anyone has an idea why some of my cards are getting GPU DEAD after like 12 hours or mining? I’ve tried upping the voltage even by 50 still the same…
Do u have any changes in ambient temp during 24hrs ? Check stats on that problem gpu and see if there is any qlues there at the time when that card dies…power, fan speed,temps. Maybe try swap riser´s with another card and see if issue continues and go from there.
Hi guys, need some help with 3x vega56 hynix memory, no matter what straps i apply and oc invalid shares are nightmare !(ethash) Screenshot_20211003_004137_com.android.chrome_edit_345322760176993|690x114
15 invalids so far, second one and third one
any solution?
What wattage are you using seems a little low
pl is blank
Well, fixed one “bricked” card. One of the power wires was melted, so gpu lite up but didnt work.
2nd bricked vega card, that was not even recognized before, now runs for 2mins then crashes in hivesos, in win 10 runs for 2-20mins the fan goes crazy up and/or down constantly like every half second randomly.
Atiflash says the bios is one i flashed, because ddu said it was a microsoft card before, but ddu still says Microsoft display adaptor card/bios while atiwebflash says it different bios/vega card?
Maybe wrong BIOS??? any thoughts?
Οκ . If you managed to 51mhs with hynix i believe that you are a god .
Give us all information .timings etc
Is there a value ? I dont now what to do
Well, added amd drivers to the card and now it actually mines for 1-3 blocks then crashes. I got picture of log. But it seems like it has to do with something with the fan and saying it is over heating, but it is not. It will run idle no problem and temp is fine.
And ideas?
so I was able to get it back up to 160 by reinstalling hiveos and resetting the whole system.
But now the card is a bit unstable and it’s rebooting a lot. I’m slowly in the process of lowering mem clock and core clock to test reliability. I’m wondering if people have experienced similar with the Vega 64 Founders edition.
Since my lowering of the core + mem clocks it looks like the card is a little more stable (haven’t seen it drop out for 5 hours) and it drops to 158 mhs. Is it unsafe for it to reset like that or is that just what vegas do?
Thanks!
vega probs|434x274
what are your straps ? core mV is good if its not crashing, with my few hynix i can run them ~800mV
Btw. you can remove DPM state values when you have core mV set.