RX580’s are not as complicated as Vega’s. If you do some research you’ll quickly find out how to mod the vbios yourself.
But as this is a Vega topic, please post your RX580 questions in an RX580 topic 
99.7% accepted is pretty good already.
Is it Samsung or Hynix? Mem seems high for Hynix, RC low for Samsung.
Oni it’s samsung memory
I have this one running pretty stable on my Samsung 56’s (maybe 1 error every few days):
amdmemtweak --RC 37 --RP 11 --WR 13 --CWL 8 --FAW 12 --RAS 22 --REF 65535 --RFC 239 --RTP 6 --RRDL 5 --RRDS 3 --WTRL 9 --WTRS 4 --RCDRD 12 --RCDWR 12
I’ve been reading and trying various things from this thread and so far the results have been superb so thanks to everyone for contributing.
I do have a question though as I am really struggling to get the Vega 64 Nitro+ that I purchased a few days ago to work. I’ve tried a few different bios flashes (to 56) as well as various OC settings and strap configurations but I just cannot get it to remain stabble for much longer than a few minutes (a mixture of the DEAD gpu error, the 511c temp thing or both).
Is anypone using this card and if so can you share OC settings, straps AND a link to the BIOS that you are using?
I think 511 is usually an issue with cables (both power and riser usb) or riser itself. Unplug all, check for damages, plug them back in firmly.
If not working, try other cables and/or riser.
thanks, I’ve tried that but to no avail. I’ve even tested using cables from other GPU’s on my other rig that works and still nada 
And putting this gpu in that other rig?
yeah, tried that too (although that one is an NVIDIA rig with 6 x 1080 TI’s)
Do you use latest OS ? if so, or even if not, try to roll back few versions and see if issue persisit.
I have had varios errors/issues that have got solved simply just by downgrading OS/Miner, btw, u do use TRM in AMD rig, right?
Currently i dont pay too much attention to new releases, maybe i take quick look if there is some major Team Red Fix…tho as long miner is working solid so im not waiting or gonna install any fixes.
I have 56 nitro+, i belive im running that with same vbios than my sapphire Pulse 56
Atm im with OS 0.6-204@210608 on my Vega rig
I’m using TRM on the AMD rig. It is more stable when I remove all straps but it still does tend to do the same thing randomly within a 5 minute to 3 hour timeframe.
I think I’ll try your suggestion later and roll back the the OS/Miner and see if I can find a pairing that wokrs. I’ll start with what you are using, what straps are you using for that as I’d be interested to see if I can replicate the results?
Hello everyone, I have just received Vega 56, Originally it runs at 28MH on ETH with GMINER, the configuration procedure that I do not understand, by or started? the OC parameter does not work reduced even the result! Thank you for your help
These Nitro+ and Pulse are great cards, they run high hash with low temps and quite low wattage.
these are my starps for them.
amdmemtweak --RC 37 --RP 11 --WR 13 --CWL 8 --FAW 12 --RAS 22 --REF 65535 --RFC 239 --RTP 6 --RRDL 5 --RRDS 3 --WTRL 9 --WTRS 4 --RCDRD 12 --RCDWR 12
I have one 64/56, that runs stable with,
1100-825-1020. +55mh.
With these clocks it consume like 124w, so i usually use it with lower oc s.
1000-820-1000 gives +53mh @117W.
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
good luck with tuning
Hello, today I succeeded in configuring the card, I was using the wrong miner (Gminer) I switched to (teamredminer) on ETH with HiveonPool and it’s perfect I have not yet tried several OC settings but the first test gives good result over 53.50MH, happy to read your sharing and experience which we solved my problem, Good job everyone
Hello all, i have two rx vega 56.
one is Sapphire Samsung HBM2, and the other ASUS Hynix HBM2.
i’m using teamredminer with hiveos
i’m trying settings all day to improve the asus, but always crashing, giving a lotttttt of errors. i left for 1-2h, and got 123 invalid shares.
i already tried a lot of settings at this page, but i cant get the correct settings…
right now i have the gpu mining for 1-2h, and is stable, but the hashrate is too low…
the saphire is stable at 51.74mh
any advices to improve the two gpu’s?
thanks
Hynix or Samsung memory? My Pulse with Hynix is a headache!
I have both, Samsung is naturally hashing bit more but Hynix has been solid allso.
here is Hynix Pulse,
For that Sapphire Samsung, try different settings from these straps. i have 2 gpu´s with that same bios and they both run 1045-825-1130 – 56.14mh @115W / 56.15mh @116W
amdmemtweak --RC 37 --RP 11 --WR 13 --CWL 8 --FAW 12 --RAS 22 --REF 65535 --RFC 239 --RTP 6 --RRDL 5 --RRDS 3 --WTRL 9 --WTRS 4 --RCDRD 12 --RCDWR 12
There is many values that i would change on your Sapphire, starting from RC36 to 37, that will give u room to tighten other values, you can compare yout starps and try one by one what is working for you, like changing RFC,REF,RP values could be worth trying.
That Hynix bios xxx PIL-100. About clock´s, i would try to stay with mem 960 core 1050
amdmemtweak --CL 20 --RC 36 --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 is too tight imho, go with 36-37,REF 65535,go up with your RFC, like with Sapphire…trial and error. U should expect for sure +50mh from that Asus.
Start tuning your gpus with generously VDD, like 830…850 and dont pay too much attention to wattage at the beginning, when u get gpu stable with hash that u are happy, start going down with vdd with oo5 at the time and let them run long time with new core mV, if card start´s to act up with new core voltage, adjust core/mem/straps…or back up with your VDD.
Tuning these cards takes long time. its pain to get all values right, u need to change one value at the time run miner for some or long time to verify your straps/oc´s…some cards fails or starts to spit invalids after 20min´s others after several hours and some start´s after 1 day and the combination of values needs to be right…compared to your clock´s and vdd…and things will change when ambient temps go up/down, if u can keep gpu´s mem under 70c deg with good air flow around gpus you can expect good results, usually HBM2 will start to give issues above 70…75c deg, tho it will still run just fine with lower oc´s etc.
Good luck and “happy” mining 
thanks, but didnt work… always crash…
okay. fast test for all configs…
sorry to hear that
