Any word on amd vega hash rates?

[Yerba]

As far I understand then 4 x Vega56 are max per rig? Otherwise it wont work at all or wont work stable.

Is there some difference from manufacturers also - what would you suggest? Gigabyte, Sapphire or MSI ?

[STT]

There is no difference across manufacturers at present, there is one factor you can look out for which is the revision number in the top right corner of GPU-z.   I think people were saying the liquid cooled cards did not have the best revision at least on launch day, maybe they were graded as not suitable for Frontier.   I forget the exact detail but thats the only real variation I've heard of being important.

only thing is the HBCC reset that is causing some...problems.  lose power, reset.  anyone have any ideas for that ?

To reset the driver only requires to log out I believe.   Its possible this could be a scripted process when a miner exits from stalled hash rate maybe

[MingMining]

i had order new one RX Vega 64,its price 681USD
too much,but i cant found under 599$.

https://www.newegg.com/Product/Product.aspx?Item=N82E16814202300&cm_re=vega_64-_-14-202-300-_-Product

$619

[Toughit]

I am familiar with the polaris mining and bios editing scene, but new to the Vega scene. Couple questions:

How does vega 64 compared to 56 for cryptonight (monero) mining? (hashrate and power consumption)

Also, is it possible to edit the bios (mem strap timings) of vega, similar to polaris, to increase cryptonight hashrates?

Read backwards a few pages...
1) 56 maybe 100H less that 64 (close to 2000) at less power
2) cant mod bios, but can mod registry

[astark052970]

Is anyone successfully running 3+ V56 for extended periods of time? Using the usual settings I was easily able to get around 1800 H/s running on 3x V56 for a couple days with no issues. Got the fourth board setup and although it worked well enough at first the rig crashed after about 14 hours. Have it running again now with slightly lower memory clocks. If that doesn't work I'll probably try messing around with HBCC amount. Anyone else having stability issues with many GPUs?

If you give me (PM) all of your configs I will try on my 6 card vega 56 rig.

Nothing special really basically the settings discussed in this and other threads. Core freq: -30%, Mem freq: 920mhz, Core/Mem voltage: Auto, Power level: -20%, Minimum fan speed: 2500rpm. HBCC on at 16080GB RAM. Blockchain driver. Running xmr-stak-amd with 8 threads (2/board) and gpu_threads_conf like this:
{ "index" : 0, "intensity" : 2016, "worksize" : 8, "affine_to_cpu" : false },
{ "index" : 0, "intensity" : 1600, "worksize" : 8, "affine_to_cpu" : false },

These settings on 3 boards was basically set it and forget it. With the fourth board it crashed after a few hours. Lowered mem freq down to 900 and it lasted longer but eventually crashed. My current theory is the new board just can't take the RAM overclock. Running at 890 now for about 8 hours. If it makes it through tomorrow hopefully I'll have my answer and it has nothing to do with 4 boards but rather the fourth board just can't overclock as far. These boards are great miners but definitely not as easy to work with as the 1070s.

[fr4nkthetank]

Yeah its a bit more work with 4 VEGA's, havent tried more than 4.  Dont think I will.  Now what would be the fun in mining if everything worked perfect the first time you plug it in ?   

[Shnikes101]

Got back around to messing with softpowerplaytable.

Running 3 64's around 1950ish a card right now. I need to bring down the power a bit. I'm currently at 155w per car at the wall.

How low are y'all running your core voltage? 900? 875?

[cryptonist]

Got back around to messing with softpowerplaytable.

Running 3 64's around 1950ish a card right now. I need to bring down the power a bit. I'm currently at 155w per car at the wall.

How low are y'all running your core voltage? 900? 875?

Can you PM your registry edits assuming you made changes? I seem to run into 3 64s on my end with stability. Want to match my settings with someone else's. Thanks!

[hepa_s]

I have 4 pcs of Vega 64 GPUs and I tried to use Oldcomer version of softpowerplaytable mod. Three of those Vegas worked well. I got about 1850 H/s per GPU, but one GPU has somekind of problem. If there is softpowerplay mod in windows register with that specific GPU, then I got blue screen with "THREAD STUCK IN DEVICE DRIVER" message at windows start up. On start up all other GPU initialized correct way with fast acceleration and decleration of blower. This problematic GPU blower accelerates and keep running on high speed till windows automatically restarts.

Everything worked well after I boot windows to safe mode and remove "PP_PhmSoftPowerPlayTable" instance from GPU 0004. I even try to change HBM clock rate from 1100 MHz to 1000 MHz and still get stuck with this one GPU.

I'm using integrated intel GPU for monitor. This is 0000 instance in [HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Class\{4d36e968-e325-11ce-bfc1-08002be10318}\

0001 - 0004 are vega 64 GPUs.

I got about 7200 H/s with three vegas undervolted and this problematic one with stock voltage and clock settings.

[davith75]

@hepa_s

I have the boot issue you're having, but on a "main" Vega which is plugged to the monitor itself.

By putting it behind an intel GPU like you did, the Oldcomer softpowerplaytable boots fine.

Instead of putting it stock did you tried to overvolt a bit in a custom powerplaytable ?
I think if you put 920-930mv in each states it should help the card booting, or just adding like +10% power target.

Do you use xmr-stack-amd, 2 threads/cards like it's explained page26 (2016-1600) for the 1850h/s ?

Haven't tried to monero yet, was already happy to have 4 R56 (flashed at 64) softpowermodded on main ETH at 173MH @750w at the wall.

[bolcoin]

I have 4 pcs of Vega 64 GPUs and I tried to use Oldcomer version of softpowerplaytable mod. Three of those Vegas worked well. I got about 1850 H/s per GPU, but one GPU has somekind of problem. If there is softpowerplay mod in windows register with that specific GPU, then I got blue screen with "THREAD STUCK IN DEVICE DRIVER" message at windows start up. On start up all other GPU initialized correct way with fast acceleration and decleration of blower. This problematic GPU blower accelerates and keep running on high speed till windows automatically restarts.

Everything worked well after I boot windows to safe mode and remove "PP_PhmSoftPowerPlayTable" instance from GPU 0004. I even try to change HBM clock rate from 1100 MHz to 1000 MHz and still get stuck with this one GPU.

I'm using integrated intel GPU for monitor. This is 0000 instance in [HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Class\{4d36e968-e325-11ce-bfc1-08002be10318}\

0001 - 0004 are vega 64 GPUs.

I got about 7200 H/s with three vegas undervolted and this problematic one with stock voltage and clock settings.

Same problem with 2xVega 64. Let's try to boot from Intel GPU like davith75 suggested

[badfad]

I am familiar with the polaris mining and bios editing scene, but new to the Vega scene. Couple questions:

How does vega 64 compared to 56 for cryptonight (monero) mining? (hashrate and power consumption)

Also, is it possible to edit the bios (mem strap timings) of vega, similar to polaris, to increase cryptonight hashrates?

The answer to your questions is literally 2-3 pages back, hard to read?

[mutual.consent]

I have 4 pcs of Vega 64 GPUs and I tried to use Oldcomer version of softpowerplaytable mod. Three of those Vegas worked well. I got about 1850 H/s per GPU, but one GPU has somekind of problem. If there is softpowerplay mod in windows register with that specific GPU, then I got blue screen with "THREAD STUCK IN DEVICE DRIVER" message at windows start up. On start up all other GPU initialized correct way with fast acceleration and decleration of blower. This problematic GPU blower accelerates and keep running on high speed till windows automatically restarts.

Everything worked well after I boot windows to safe mode and remove "PP_PhmSoftPowerPlayTable" instance from GPU 0004. I even try to change HBM clock rate from 1100 MHz to 1000 MHz and still get stuck with this one GPU.

I'm using integrated intel GPU for monitor. This is 0000 instance in [HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Class\{4d36e968-e325-11ce-bfc1-08002be10318}\

0001 - 0004 are vega 64 GPUs.

I got about 7200 H/s with three vegas undervolted and this problematic one with stock voltage and clock settings.

I have a single Vega 64 that had the same issues. I ended up using the xslm spreadsheet that was posted a few pages back to create a series of custom .reg files and everything seems to be working well now - albeit at 925mV. Happy to share the settings I used if you pm me (saves cluttering the thread). Standard disclaimer of use at your own risk applies though - I take on no responsibility for failures.

[hepa_s]

Now I tried to use first 950mV and then even 1000mV for P0 core V - P3 core V others are little bit higher. Same problem again. Next I have tried to lower MEM clock from 1000MHz to 945 MHz and still same problem. Again the problem was solved when I remove "PP_PhmSoftPowerPlayTable" instance from GPU 0004 with regedit.

I don't know what is wrong there or do I have bad luck in silicon lottery 

I use xmr-stak-amd with 2016/1600 intensity.

[Ystss]

I have 1 pcs of Vega 64 GPU and I use only wattman without regedit,-30 gpu, 1000 mem and xmr-stak-amd with intensity 2016 + 1800. HBCC on at 12GB RAM + 30GB swap file. On the pool I see this:
https://i.imgur.com/kcBewjN.jpg
Does anyone have something like that? Why is the hash rate unstable?

[coke15]

hi

i have 2x 56 cards.i flashed with 64 bios.

i put -20% /1100/-20%,hbcc ok--->1800/1850 h/s per 1 card,connect to my screen with hdmi.

when i put the second card,same tweaks, the card connected to screen lose 400h/s ?

is there a way to fix this?

thanks

edit : bios 64 drain more watts that the 56?

[miner1337]

6x VEGA 64

Using https://github.com/halsafar/Vega64SoftPowerTableEditor
Modify registry:
- 1408mhz for P5,6,7.
- 1050mhz for mem
- 950 for core voltages

HBCC only works for 4/6 cards so I am short ~1000H.

Getting 10500H @ 1150W (total draw at the wall).

Total draw at the wall when idle is 250W.  

So that is ~150W per card.

Stable for 72 hours.  99% submission rate on nano pool.  My average on the pool is expected, 10500.

This is huge given if I try the same setup using stock settings the total draw is ~2000W.

[Lithoss]

Question for those using vega 56 with no bios mod. When you look at using the softpowerplaytable mod, is there a registry entry "PP_PhmSoftPowerPlayTable" already in your 000X folder, or is this something new you add that wasnt already there?
I've looked on two different computers with vega 56 with blockchain drivers and neither have a  PP_PhmSoftPowerPlayTable folder present.

Thanks

[STT]

hi

i have 2x 56 cards.i flashed with 64 bios.

i put -20% /1100/-20%,hbcc ok--->1800/1850 h/s per 1 card,connect to my screen with hdmi.

when i put the second card,same tweaks, the card connected to screen lose 400h/s ?

is there a way to fix this?

thanks

edit : bios 64 drain more watts that the 56?

The 64 bios has higher voltage limits and on the HBM memory its most useful.   HBM isnt using large amounts of power in any case, its like 10w difference
Theres a theory that the 56 actually overclocks and runs at high end better then the 64, maybe because it has less shaders.    That would be ironic, I guess mining is not using those extra shaders I dont know.  Its mostly about memory bandwidth I read.

With your setting of the power limit.  You want to lower the volts on left and right, lower the power limit yes but not long after you raise the power limit back up and this enables the best performance and you should retain the lowest voltage usage in any case.  It may wobble back and forth a bit, 120 to 140 watts I see but mostly its just showing a reading of 130w.   Some people no doubt have got lower power usage then that even, there is a correct sequence to get the best results though.  If driver crashes, log out and start over again as it wont run as well in my experience

[vg54dett]

Stable for 72 hours.  99% submission rate on nano pool.  My average on the pool is expected, 10500.

Be careful with nanopool, it seems the payout are 20% lower than expected these days ...