try this:
CoreClock:1375 or 1404
CoreVoltage:900
MemoryControllerVoltage:
MemoryClock:875
MemoryVoltage:
Fan:80%
PowerLimit:
amdmemtweak --CL 20 --RC 36 --RP 11 --WR 14 --CWL 8 --FAW 12 --RAS 20 --REF 65535 --RFC 248 --RTP 5 --RRDL 5 --RRDS 3 --WTRL 9 --WTRS 4 --RCDRD 12 --RCDWR 12
This is with 1375 cClock:
with 1404 cClock GPU dies instantly.
I will check now if that hashrate and power increase, it´s worth it as well with the louder fan, as with 65% fan you can’t even hear it, but with 80%… noise it´s insane 
Thanks!
EDIT: Yeah, I would earn 3$ more / month, but with a louder fan. Well, I’ll think about it 
for me it’s 1404 and 875 memory and vvd for the worst card is 910mV = 187.5 MH/s
Try some number from 1404 down like 1400 or 1390 depends how the clock will work with your chip
I use autofan in hiveos min 25% max 90% 
at least the nights are a little more quiet
This is an amazing guide — thank you. Was able to stabilize my card and increase the hash rate by about 50mhs. Only thing I’m noticing is that these settings without the memtweak brings my MEM temp up to about 87. With the memtweak it exceeds 90. I have good ventilation and a room temp room.
with ERG mem usage is lower than with ETH 87deg is quite high temp anyways, what card is that…ref or some twin/triple fan? what temp is core on that card when runing mem 87c ?
IMHO would be good idea to change pads to that gpu…if u havent done it yet 
Hi all, firstly thanks to all for contribution. I went thru the whole topic 
I have rig with 5x Vegas. 2x with samsungs - setted up easily to 187-196 mh/s according to first post. And 3x cards with Hynix memory and this was PITA to set up so i would like to share my setup. Finally find this set up stable:
with this strap for Hynix memories:
amdmemtweak --CL 20 --RC 36 --RP 13 --WR 13 --CWL 7 --FAW 12 --RAS 23 --REF 65535 --RFC 248 --RTP 6 --RRDL 5 --RRDS 3 --WTRL 9 --WTRS 4 --RCDRD 20 --RCDWR 11
Whenever I am trying to add core clock (and voltage) cards are crashing. I have tried many combos.
Hynix:
GPU 1&3 are Air Boost MSI
GPU 2 is Gigabyte
Samsungs
GP0 is 64 Nitro+ flashed to 56
GP4 is Sapphire 56
I am pretty surprised with low temperatures. Rig with sapphires 5700xt (close to this one) is now ~84-86°C on ETH.
I cranked up the fan speed and the temp dropped to about 77. It’s a single fan GPU. I did buy this GPU used, and it arrived dusty, so that can’t be helping. I plan to clean it up. This noob will have to be extra careful with the changing of the pads lol. Good suggestion though. Thanks
very envious of your power to hash ratio!
I can’t get my hynix vega 56 to start up at your settings, what memory strap do you use? do you have SOC set at all?
I can get it down to 130watts but max of 157Mhs 
i have 3 vega 56 (one flashed from a vega 64 ) and total power on Hive is 500watts, and about 535 Mh/s total. I know I can do better!
Imo that pulse is great gpu, better than ref / turbines
hey guys. any idea why would my pulse crash every few hours?core 1350 and voltage 850-870 tried with memory 960 and alwas crashing with time. other pulse same settings no issues. samsung memory
Something similar happens to me.
Every “x” hours, I get the message: GPU Dead and rig restarts. It´s just 1 min or so, but it restarts and I would like to know the reason, as I think it´s not about temperatures and OC is a normal one:
CoreClock:1340
CoreVoltage:870
MemoryControllerVoltage:0
MemoryClock:880
MemoryVoltage:–
Fan:65%
PowerLimit:200
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
Any idea what could be the issue guys?. Thx
i have to exact same cards. other one running fine at 1350 core clock. one that keeps dying im testing core clock maybye reason is dying. at them momen running fine at 1300. will update after a day. nothing to do with strips as was dying even without them
Nah, hours later again DEAD and it resets and continues. I don’t know if I should lower any value mmmm.
[2021-09-07 17:35:29] ----------------------------------------- GPU Status -------------------------------------------
[2021-09-07 17:35:29] GPU 0 [48C, fan 87%] autolykos2: 178.0Mh/s, avg 174.2Mh/s, pool 181.1Mh/s a:400 r:0 hw:0
[2021-09-07 17:35:29] Total autolykos2: 178.0Mh/s, avg 174.2Mh/s, pool 181.1Mh/s a:400 r:0 hw:0
[2021-09-07 17:35:29] ----------------------------------------- Pool Status ------------------------------------------
[2021-09-07 17:35:29] ergo-eu1.nanopool.org autolykos2: 174.1Mh/s, avg 170.7Mh/s, pool 181.1Mh/s a:400 r:0
[2021-09-07 17:35:29] ergo-eu2.nanopool.org autolykos2: 0.000 h/s, avg 0.000 h/s, pool 0.000 h/s a:0 r:0
[2021-09-07 17:35:29] ------------------------------------------------------------------------------------------------
[2021-09-07 17:35:35] GPU 0: detected DEAD (07:00.0), will execute restart script watchdog.sh
[2021-09-07 17:35:35] Watchdog script executor thread executing script ‘watchdog.sh’
I increased fan to 90% just to test a few hours.
Nah, GPU Dead. I will keep it with fan 65%… I don’t know why it dies. Whatever, I need winter to comeeeeeeeeeeeeeee xD
I changed thermal paste and well, now it looks more stable. Let´s see how it goes.
GPU0 = Powercolour Red Dragon Vega 56
GPU1 = Sapphire Nitro+ Vega 56 (hynix)
GPU2 = Gigabyte Vega 64 OC Flashed to 56.
Any tips to reduce power consumption on GPU0? It easily goes over 200MHs but trying to make my rig a bit more efficient.
Any tips to increase hashrate on GPU1? - I’ve tried lots of things in this thread, the actual core clocks are way off and I can’t seem to get to memclock state 3
Any tips to increase hashrate on GPU2? - I understand this card shows higher power consumption than other brands but another 10 MH/s would be nice
This thread is amazing source of info, I managed to get the power consumption down a lot, but i see lots of people running at 140-150watts with 180-190MH/s.
I guess some of it could be down to the silicon lottery
Any tips are greatly welcomed
Hi, I’m new to flashing, so i just need advice for chosing bios.
I have:
1x Sapphire RX Vega 64 8GB
113-D0500110-O01
and
5x Asus RX Vega 64 8GB
115-D050PIL-100
Found bios for sapphire:
And for strix:
Is this bioses safe to flash on this cards?
Thank you in advance
Hello All,
I have been reading all the information on here and it’s been a big help.
I am having a few issues with stability and looking for advise.
Gpu 0 is unstable. It crashs a few times a day. I have adjusted the memory and core down and up but it did not resolve it. I don’t know enough about the straps to change them so any input would be great.
Gpu 3 is a real pain. It’s a 64 flashed to 56 and tempermental. If I make and adjustment on any card Gpu 3 will crash. No adjustments no crash… You tell me.
Last is my new card. Gpu 6. It has hynix memory. I would like to get it up into the 190’s if possible.
I have noticed that any adjustments on any card effects the hash on the others next to it.
Does anyone else have that issue?
for GPU 0 I would try out RC 37 to help with the crashing, if that doesn’t help just add more vdd.
GPU 3, in my experience there is usually one card that crashes when you make a change to OCs while the miner is running, only fix I have found is to stop the miner before updating OCs.
GPU 6 I have not been able to get my hynix Airboost to 190, but if its changing performance of the other cards you many be approaching, or past your psu’s limits
Thanks for the input. I changed the RC to 37. If that doesn’t help I will turn up the power. I added another psu for GPU 6 but I was having the issue before adding that card on the system. I may put another card on the new psu to see if it help.
Hi All,
I’ve been having issues with my vega 56 within seconds I get " GPU 0: detected DEAD (03:00.0), will execute restart script watchdog.sh" I’ve been reading everywhere for a couple of days trying to fix it and though it was my power supply so I got a new one and the problem is still happening. I’ve tried giving it more core voltage and lowering core and memory clock but still the problem keeps happening. I tried flashing it back to a vega 64 and flashing it back to a vega 56. I have no idea what to do now. These are my settings. The gpu works when not mining.
