Hi all, can anyone confirm that: ethOS 1.2.4 changelog (released September 3rd 2017)
Major Changes -Added ability to set voltage for RX series GPUs.
is realy working for RX GPUs by using vlt parameter? I see vlt parameter in http://ethosdistro.com/pool.txt. Just thinking about buying ETHOS instead of using SMOS.
|
|
|
IS there anyway to undervolt the core or memory from the command line in SM or through the settings? Or is the only way to flash bios with lower voltage setting
It's not possible to undervolt. You can only change power stage, memclock, coreclock. It's AMD Linux driver limitation.
|
|
|
@tytanick, default config for XMR (monero) mining with SMOS is: Default config: -s xmr-eu1.nanopool.org:14444 -w 42MmCH6p1B3P7BaC3y8KcXayNcKa1jaU9RnUPKFy7SMAJrGrgMajd17WbwaVe4vUMveKAzAiA4j8xgUi29TpKXpm3xUCcQ2.$rigName -p z -i 1024 I have XMR wallet on Bitfinex and they said: Depositing Monero to Bitfinex requires BOTH a deposit address and a deposit Payment ID.
To make a Monero deposit, you need to use both:
Address Payment Id On the following page, the deposit address is shown at the top, and then below you will see a deposit Payment ID for each of your wallets. To make a deposit to your Exchange wallet, for example, you will need to use the deposit address shown near the top and the deposit Payment ID next to your Exchange wallet.
Do NOT send Monero to Bitfinex unless you know how to indicate a Payment ID. If you fail to include the Payment ID with your deposit the funds will be lost forever. What should be the correct config string for SMOS if for example: My XMR address is: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa My XMR payment ID is: bbbbbbbbbbbbbbbbbbbbbbbbb Edit: in this way: -s /xmr-eu1.nanopool.org:14444 -w aaaaaaaaaaaaaaaaaaaaaaaaa.bbbbbbbbbbbbbbbbbbbbbbbb.rig1.email -p z -i 1024 is ok?
|
|
|
I'm getting currently 75 Mh/s from each rig of 3 RX 480... What hash rate can I get under 500 Watts through 3 RX 570/580 if I exchange my 480's with 570's by paying something extra (I've an offer for that)? Should I exchange? Will that provide me a big increase in hashing? As a reference, I'm getting 7x 30.5MH/s (213.5MH/s) from 7x RX580 with modded BIOS. If 30 Mh/s is all that RX 580 is providing, I'm better off with my RX 480. Thanks for quick reply! My RX580 can mine with stable 31.5MH/s but power consumption is too high for me. But 30.5MH/s is acceptable.
|
|
|
I'm getting currently 75 Mh/s from each rig of 3 RX 480... What hash rate can I get under 500 Watts through 3 RX 570/580 if I exchange my 480's with 570's by paying something extra (I've an offer for that)? Should I exchange? Will that provide me a big increase in hashing? As a reference, I'm getting 7x 30.5MH/s (213.5MH/s) from 7x RX580 with modded BIOS.
|
|
|
Perfect solution for SMOS would be to have additional two fields (OC page) related to mem/core voltage but as I understand it's not possible. Right?
Its a limitation of the AMD Linux drivers, nothing tytanik can do about it. Me and cryptoscum will be working with this computing kernel and driver to meaby fix this power wattage issues. I am starting to love nvidia over amd but nvidia also has some stupid limitations Hi @tytanick, any news regarding works on wattage issues?
|
|
|
@tytanick, I just loaded above image (simpleminer-RX-v1118-17.30-dagfix-computekernel.img) to make some tests with AMD 17.30 GPU drivers. What I see on dashboard is AMDGPU 16.60 (see screen shot).
|
|
|
Any idea what could be wrong? Claymore 10.0 and latest Windows Blockchain drivers: 15:48:35:952 504 args: -epool xxxxxxxxx -ewal xxxxxxxxxxxxx -eworker rig1 -epsw x -esm 0 -allcoins 1 -mode 1 -fanmin 40 -fanmax 80 -cclock 1250 -mclock 2200 -cvddc 900 -mvddc 900 15:48:35:965 504 15:48:35:981 504 ÉÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ» 15:48:35:987 504 º Claymore's Dual ETH + DCR/SC/LBC/PASC GPU Miner v10.0 º 15:48:35:997 504 ÈÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍͼ 15:48:36:004 504 15:48:36:211 504 ETH: 1 pool is specified 15:48:36:222 504 Main Ethereum pool is 79.98.144.27:8008 15:48:36:977 504 OpenCL platform: Intel(R) OpenCL 15:48:36:982 504 OpenCL platform: Intel(R) OpenCL 15:48:36:993 504 OpenCL platform: AMD Accelerated Parallel Processing 15:48:37:003 504 OpenCL initializing...
15:48:37:009 504 AMD Cards available: 7 15:48:37:016 504 GPU #0: Ellesmere, 8192 MB available, 36 compute units 15:48:37:021 504 GPU #0 recognized as Radeon RX 480/580 15:48:37:026 504 GPU #1: Ellesmere, 8192 MB available, 36 compute units 15:48:37:031 504 GPU #1 recognized as Radeon RX 480/580 15:48:37:035 504 GPU #2: Ellesmere, 8192 MB available, 36 compute units 15:48:37:039 504 GPU #2 recognized as Radeon RX 480/580 15:48:37:044 504 GPU #3: Ellesmere, 8192 MB available, 36 compute units 15:48:37:047 504 GPU #3 recognized as Radeon RX 480/580 15:48:37:050 504 GPU #4: Ellesmere, 8192 MB available, 36 compute units 15:48:37:053 504 GPU #4 recognized as Radeon RX 480/580 15:48:37:057 504 GPU #5: Ellesmere, 8192 MB available, 36 compute units 15:48:37:061 504 GPU #5 recognized as Radeon RX 480/580 15:48:37:064 504 GPU #6: Ellesmere, 8192 MB available, 36 compute units 15:48:37:067 504 GPU #6 recognized as Radeon RX 480/580 15:48:37:070 504 POOL/SOLO version 15:48:37:076 504 b415 15:48:37:080 504 Platform: Windows 15:48:37:107 504 start building OpenCL program for GPU 0... 15:48:37:393 504 done 15:48:37:603 504 start building OpenCL program for GPU 1...
and and this moment EthDcrMiner64.exe crashed.
|
|
|
Perfect solution for SMOS would be to have additional two fields (OC page) related to mem/core voltage but as I understand it's not possible. Right?
Its a limitation of the AMD Linux drivers, nothing tytanik can do about it. I think so but just mentioned what would be the best
|
|
|
Underolting in linux works only when you set lower powerstage but this works not always ! For sure it's working for my rig but with very limited range Or when you have custom bios with addedd undervolting offset I would avoid additional BIOS modding except what is required to get 30MH/s. Perfect solution for SMOS would be to have additional two fields (OC page) related to mem/core voltage but as I understand it's not possible. Right?
|
|
|
You should be aware that we need real AMD Linux driver with fixed DAG issue. This driver also should allow to undervolt!!!Just did a test with my rig (H110 Pro BTC+, modded 7xRX580 8GB): SMOS image simpleminer-RX-v1118-17.30-dagfix-computekernel.img.zip & simpleminer-RX-v1118_dagfix_computekernel.img.zip: - every card stable 30.5MH/s - total from wall 1500W!!!- changing power stage does not help a lot - NO UNDERVOLTING possibleWin10 + latest Blockchain drivers: - every card stable 30.5MH/s - total from wall - 980W!!!- undervolted in Claymore 10.0 (-fanmin 40 -fanmax 90 -cclock 1250 -mclock 2200 -cvddc 900 -mvddc 900) For now Windows is the winner
|
|
|
I will release new version within a couple of hours.
@Claymore, thanks for sharing.
|
|
|
@Claymore, any ETA for Claymore 9.9???
|
|
|
Does it suport AMD (RX 580 or Vega) overclock? Or which AMD driver this distribution uses? Thank you so much.
It supports only 580. For vega support i will make soon new image with new drivers. H110 Kindly please do not forget about this:
Cant do anything about this one. The kernel is what it is and some motherboards have problem. I heard that plugging 8GB RAM solve problem, can you check it ? Anyway i need fixed driver from amd to make it work on all motherboards Still no driver, only this strange kernel fix. @tytanick, currently I have 2x 4GB (8GB) DDR4 installed. Seems I have to back to simpleminer-RX-v1118-kernel4.11-16.60-noaer.img
|
|
|
@tytanick : any ideas?
Need to answer all emails then i will look into it. I have too much work overdue right now. Been feeling not so good last 1-2 weeks so my efficiency was fucked Kindly please do not forget about this: simpleminer-RX-v1118_dagfix_computekernel.img.zip DIDNT SEE RX580
I can only confirm above. simpleminer-RX-v1118_dagfix_computekernel.img.zip does not work with my rig (RX580 and AsRock H110 Pro BTC+). SEE SCREEN SHOT. It was working fine on simpleminer-RX-v1118-kernel4.11-16.60-noaer.img
|
|
|
@tytanic - my rig (4x RX580) was working FINE with simpleminer-RX-v1118-kernel4.11-16.60-noaer.img BUT same rig with simpleminer-RX-v1118_dagfix_computekernel.img does not work. SEE SCREEN SHOT.
|
|
|
simpleminer-RX-v1118_dagfix_computekernel.img.zip DIDNT SEE RX580
I can only confirm above. simpleminer-RX-v1118_dagfix_computekernel.img.zip does not work with my rig (RX580 and AsRock H110 Pro BTC+). SEE SCREEN SHOT. It was working fine on simpleminer-RX-v1118-kernel4.11-16.60-noaer.img
|
|
|
@tytanick, have you seen this post: Here is the solution for linux amdgpu blockchain fix: The new ROCm Kernel works great, my hashrate went from 155 > 167Mhs/s with 6*RX470 with Ubuntu 16.4 #Install new ROCm Kernel: wget -qO - http://repo.radeon.com/rocm/apt/debian/rocm.gpg.key | sudo apt-key add - sudo sh -c 'echo deb [arch=amd64] http://repo.radeon.com/rocm/apt/debian/ xenial main > /etc/apt/sources.list.d/rocm.list' sudo apt-get update sudo apt-get install rocm #edit grub sudo nano /etc/default/grub #2MB fragments for Ellesmere are enabled with a grub option: GRUB_CMDLINE_LINUX="amdgpu.vm_fragment_size=9" sudo update-grub sudo reboot choose new Kernel and its done! Thx to gstoner from ROCm Github. Feel free to donate if you happy ;-) ETH:0xccad3d66b29ae2924df835df1417169ca59943b5 BTC:12F7uXSpTwVhf4ppdh92uieAvrevZ4ShEf Greetz Does it apply for SMOS too? has this been confirmed ? Just did it and getting increase. This is cool. If you follow these instructions, just append amdgpu.vm_fragment_size=9 to what's already in the grub line so it's GRUB_CMDLINE_LINUX="net.ifnames=0 biosdevname=0 consoleblank=0 amdgpu.vm_fragment_size=9" Before After @clintar2, may I kindly ask you for step by step instruction for linux noob user (me)? I know how to login to SMOS with root - that's easiest part but the rest of linux commands are not known for me. Should I just type each below line and press enter after each one? wget -qO - http://repo.radeon.com/rocm/apt/debian/rocm.gpg.key | sudo apt-key add - sudo sh -c 'echo deb [arch=amd64] http://repo.radeon.com/rocm/apt/debian/ xenial main > /etc/apt/sources.list.d/rocm.list' sudo apt-get update sudo apt-get install rocm sudo nano /etc/default/grub GRUB_CMDLINE_LINUX="amdgpu.vm_fragment_size=9" sudo update-grub sudo reboot What does it mean: choose new Kernel and its done! How to do that? When/where?
|
|
|
Just did it and getting increase. This is cool. If you follow these instructions, just append amdgpu.vm_fragment_size=9 to what's already in the grub line so it's GRUB_CMDLINE_LINUX="net.ifnames=0 biosdevname=0 consoleblank=0 amdgpu.vm_fragment_size=9" Before After @clintar2, did you do that in SMOS with root login? yes Just asking for it because I'm not a linux user Doing anything with linux (without icons/windows) is a nightmare for me. Just lack of proper knowledge. Did you just logged on with root and then did all described text commands one by one exactly as they was described?
|
|
|
Just did it and getting increase. This is cool. If you follow these instructions, just append amdgpu.vm_fragment_size=9 to what's already in the grub line so it's GRUB_CMDLINE_LINUX="net.ifnames=0 biosdevname=0 consoleblank=0 amdgpu.vm_fragment_size=9" Before After @clintar2, did you do that in SMOS with root login?
|
|
|
|