great thanks. asus 64 flashed to 56 working great. what settings you have for fans?auto? how to set fan to set level? what ever i put in overclock settings fan does whatever it wants
if using trm, Che k their doc’s, on usage doc there is fan settings if u want to set manually.
yes, i have one gigabyte vega 64 flashed to 56, cant get over 45mhs also 
try this straps with your amd/ati
amdmemtweak --CL 20 --RC 38 --RP 11 --WR 14 --CWL 8 --FAW 12 --RAS 20 --REF 65535 --RFC 248 --RRDL 4 --RRDS 3 --WTRL 9 --WTRS 4 --RCDRD 12 --RCDWR 12
Thanks to this forum I was able to reach almost 55 mh/h on a vega 56, stable with 1 invalid each 4 hours, i can live with it
1 Like
Anyone having trouble with the new update?
My OC wont kick in whatever i do. Did downgrade.
i would say under 80C is good. I set autofan on 79C and let software do the speed on my fans.
yes, but im not sure if its probleme,
i lost the internet connection to the hiveos serveur
Did you also change the thermal pads? If so, could you tell me what thickness you used.
Thanks
for ref 56 1mm pads, one small component needs 1.5mm. scroll up thread and u find all info, pictures etc…
3 Likes
Sorry, been out for a while 
My autofan is set to 60°C for the “target MEM TEMP” (and 50°C for core); however when ambient temperatures go >25°C the HBM’s go up to 70°C with fans at 90-100%.
If they go over 70°C I downtune them to lower clocks as described earlier in this tread.
Yes, there are most likely better combinations of core clocks, vdd’s, mem clocks and straps in this thread. It’s a huge read to scroll thru all posts but definately worth it.
And the B mode is not a magical button; it doesn’t lower the wattage by just turning it on. When you turn it on, it allows you to lower core clock and then also vdd all with the same mem clocks and straps, resulting in lower wattage for the same hashrate. Again, trial and error to lower the clock by steps of 10 until hashrate goes down, then up the clock by steps of 5 until the hashrate is normal again; then lower the vdd by steps of 10 until you get dead gpu, then up the vdd by steps of 5 again until it doesn’t die anymore.
The cores of my samsungs go down til 1050 and my hynix til 1000 (maybe here and there a small loss of hashrate but I don’t care).
The vdd’s however have mixed results, going down as low as 760 to high as 840.
And I also have 2 gpu’s that don’t want to work in B-mode at all as described earlier in this thread; I still use A-mode for them.
I’ve also described the difference in wattage between A and B-mode on my 13xVega rig above.
Nevertheless, if you have a Gigabyte Gaming OC card, it display the wattage differently on hiveos compared to other vega’s; so 150W on Gigabyte might well be very good compared to 110W on other Vega’s.
what hiveos version, what bios, what core, vdd, mem, straps, …?
I have only changed the thermal pads on one troublesome Vega, and it didn’t improve a lot. I only change the thermal paste of the Core/HBM.
Thanks for the information 
I have been struggling with my 5x vega56 cards.
Bought them second hand, 3 years old. I have tried everything in this forum and i cant get it to run stable for more than 3h.
The Hiveos suggested straps works the best, but still dies after 3h due to 511 error or just GPU dead.
Any suggestion?
Tried changing risers, cables and the basic stuff.
GPU 2 is the one always dying for the most pars. The wattage tropps to 30-40w and then reports dead.
Picture settings is the most stable ones ( 3h )
I hate this vega 56 Gigabyte OC card, my literal achilles heel. It will always die on me no matter what strap I put on it. Running in B mode, 1000/830/960(or 930) I can only get 44.5 MH/s. Tried your OC and strap, but not working. I’ll try to raise RCCRD and lower mem, but this has got to be my 50th time trying to make it work with a strap! I have a different Gigabyte VEGA 64 water cooled that luckily only had Gigabyte branding (Reference Water Cool model). Runs good at 55 MH/s.
Feeling very generous today for this great thread, takes some searching though! I like this setup rn, haven’t tweaked it in 4 months, but now I tweaked and boosted 10 MH/s here and there.
0-1: VEGA 64 Flashed to VEGA 56 BIOS
2-3: VEGA 64 Reference Water Cool flashed to ATI/AMD VEGA 56 BIOS
4-7: Stock VEGA 56, flashed all reference models to MSI, MSI is best ROM for all reference models (not for water cooled though!).
Dont kill me over the temps haha 
. This is running outside!
STRAPS:
0,1: 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
2: 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 13 --RCDWR 12
3: 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 14 --RCDWR 12
4: amdmemtweak --CL 18 --RC 34 --RP 12 --WR 13 --CWL 6 --FAW 12 --RAS 22 --REF 65353 --RFC 164 --RTP 4 --RRDL 5 --RRDS 3 --WTRL 9 --WTRS 4 --RCDRD 15 --RCDWR 4
5, 6: 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
7: amdmemtweak --CL 18 --RC 34 --RP 12 --WR 13 --CWL 6 --FAW 12 --RAS 22 --REF 65353 --RFC 164 --RTP 4 --RRDL 5 --RRDS 3 --WTRL 9 --WTRS 4 --RCDRD 12 --RCDWR 4
Generous?? 
: ETH (NO BEP): 0x7A4bA30Ddc73dC63460d491cE700827CAf08119f
1 Like
Most my Samsung Vegas will crash with RC 36, might want to try RC 37 on those cards that are crashing.
3 Likes
Thank you! i will try that