SRBMiner-MULTI AMD & NVIDIA & INTEL & CPU miner

[lncm]

MSR isn't enabled by default? I see a huge performance drop in MinotaurX with this version on a Ryzen 3900x.

Going back to 1.0.7 binary fixes it.

[doktor83]

MSR isn't enabled by default? I see a huge performance drop in MinotaurX with this version on a Ryzen 3900x.

Going back to 1.0.7 binary fixes it.

Read the release notes : https://github.com/doktor83/SRBMiner-Multi/releases/tag/1.1.1

[lncm]

MSR isn't enabled by default? I see a huge performance drop in MinotaurX with this version on a Ryzen 3900x.

Going back to 1.0.7 binary fixes it.

Read the release notes : https://github.com/doktor83/SRBMiner-Multi/releases/tag/1.1.1

Thanks, I did miss that part.

[doktor83]

V2.0.0
+ Added support for NVIDIA architectures : Pascal, Turing, Ampere, Ada Lovelace
+ Added parameters '--disable-gpu-amd', '--disable-gpu-nvidia'
+ Added GPU OC parameters '--gpu-cclock' and '--gpu-mclock' [NVIDIA Volta and newer]
+ Added GPU efficiency to stats & API
+ Removed kernels for dual ETHASH/ETCHASH/AUTOLYKOS2 + BLAKE3_IRONFISH

This is the first release that supports NVIDA GPU's, so please report any bug/s you find in the discord support channel.

***
Algorithms currently supported on NVIDIA GPU's:

- autolykos2
- etchash
- ethash
- sha512_256d_radiant
- kaspa
- heavyhash
- lyra2v2_webchain
- blake3_alephium
- sha256dt
- kawpow
- evrprogpow
- firopow
- progpow_epic
- progpow_sero
- progpow_zano
- progpow_veil
- progpow_veriblock

Dual mining :

- ethash + kaspa
- ethash + blake3_alephium
- ethash + heavyhash
- ethash + sha512_256d_radiant
- ethash + sha256dt

- etchash + kaspa
- etchash + blake3_alephium
- etchash + heavyhash
- etchash + sha512_256d_radiant
- etchash + sha256dt

- autolykos2 + kaspa
- autolykos2 + blake3_alephium
- autolykos2 + heavyhash
- autolykos2 + sha512_256d_radiant
- autolykos2 + sha256dt


***
Known issues on NVIDIA:
- autotune (--gpu-auto-tune parameter) isn't working for every algorithm

[ybelevi]

Hello
ver 2.0 1050ti cards on the system
miner says

No worker devices found

[doktor83]

Hello
ver 2.0 1050ti cards on the system
miner says

No worker devices found

yeah, at the moment you need at least 526 drivers. im gonna change that in future release/s.

[doktor83]

V2.0.1
+ Added support for NVIDIA 'Hopper' architecture
+ Added NVIDIA support for algorithms : UBQHASH, DYNAMO, BLAKE3_IRONFISH
+ Added GPU OC parameter '--gpu-plimit' [ NVIDIA only ]
+ Re-added kernels for dual ETHASH/ETCHASH/AUTOLYKOS2 + BLAKE3_IRONFISH because of testnet3 [ NVIDIA support added ]
+ Fixed detection of NVIDIA GPU's
+ Fixed compatibility with older NVIDIA drivers
+ Various fixes for mining on nicehash pools / nicehash mode
+ Minor bug fixes

***
--gpu-plimit sets a limit on the maximum power a GPU can use [ defined in W ]

Available parameters for NVIDIA OC:
[separate value with ! for multiple GPU's]

--gpu-cclock (set fixed core clock)
--gpu-mclock (set fixed mem clock)
--gpu-plimit (set power limit)


HiveOS update script from v2.0.0 to v2.0.1:

Code:

cd /tmp && wget https://github.com/doktor83/SRBMiner-Multi/releases/download/2.0.1/SRBMiner-Multi-2-0-1-Linux.tar.xz && tar -xvf SRBMiner-Multi-2-0-1-Linux.tar.xz && cd SRBMiner-Multi-2-0-1 && miner stop && cp SRBMiner-MULTI /hive/miners/srbminer/2.0.0 && miner start

[doktor83]

V2.0.2
+ Improved mining performance on 'sha512_256d_radiant' for AMD RDNA/RDNA2 GPU's and NVIDIA Ampere GPU's
+ Improved mining performance on dual ETCHASH/ETHASH + SHA512_256D_RADIANT for AMD RDNA2 GPU's
+ Added experimental support for NVIDIA architectures older than 'Pascal'
+ Added GPU OC parameters '--gpu-coffset', '--gpu-moffset' [ NVIDIA - LINUX only ]
+ Fixed bug with --gpu-cclock , --gpu-mclock and --gpu-plimit getting applied only in dual mining mode
+ Removed obsolete algorithms: cosa, circash, 0x10, frkhash, balloon_zentoshi
+ Minor bug fixes

***
--gpu-coffset sets the core offset of the GPU [ defined in MHZ ]
--gpu-moffset sets the memory offset of the GPU [ defined in MHZ ]

HiveOS update script from v2.0.1 to v2.0.2:

Code:

cd /tmp && wget https://github.com/doktor83/SRBMiner-Multi/releases/download/2.0.2/SRBMiner-Multi-2-0-2-Linux.tar.xz && tar -xvf SRBMiner-Multi-2-0-2-Linux.tar.xz && cd SRBMiner-Multi-2-0-2 && miner stop && cp SRBMiner-MULTI /hive/miners/srbminer/2.0.1 && miner start

[jp_math]

Hi.

I've a rig with 7 different GPUs.
I want to dual-mine ETC/ZIL+ALEPHIUM on 6 of them (they are AMDs), with different intensities between etchash & blake3
I'm a bit confused about how to declare that :

Code:

SRBMiner-MULTI.exe --disable-gpu-nvidia --algorithm etchash;blake3_alephium --gpu-intensity 23!23!23!23!23!23;7!0!9!7!7!1 

doesn't work, all gpus have the same intensities : 23;7.
Is it possible to tune that per GPU?

[doktor83]

Hi.

I've a rig with 7 different GPUs.
I want to dual-mine ETC/ZIL+ALEPHIUM on 6 of them (they are AMDs), with different intensities between etchash & blake3
I'm a bit confused about how to declare that :

Code:

SRBMiner-MULTI.exe --disable-gpu-nvidia --algorithm etchash;blake3_alephium --gpu-intensity 23!23!23!23!23!23;7!0!9!7!7!1 

doesn't work, all gpus have the same intensities : 23;7.
Is it possible to tune that per GPU?

If using a dual algorithm (like etchash/blake3) miner will find the best intensity for the second algorithm based on your current OC settings.
You can define them manually too, but you first need to define which gpu's you wanna use with --gpu-id. so:

Code:

--gpu-id 0!1!2!3!4!5 --gpu-intensity 23!23!23!23!23!23;7!0!9!7!7!1

[doktor83]

V2.1.0
+ Added support for INTEL Alchemist architecture (INTEL ARC 300/500/700)
+ ZIL is now mineable in any single/dual algorithm combination [triple mining]*
+ Added ZIL parameters '--zil-enable', '--zil-pool', '--zil-wallet', '--zil-password', '--zil-esm', '--zil-epoch', '--zil-oc-script'
+ Added ZIL OC parameters '--zil-cclock', '--zil-mclock', '--zil-coffset', '--zil-moffset', '--zil-plimit'
+ Added auto buffer mode for ZIL mining*
+ Added parameters '--disable-gpu-intel', '--gpu-disable-auto-buffer'
+ Added OC parameters '--gpu-cclock1', '--gpu-mclock1', '--gpu-coffset1', '--gpu-moffset1', '--gpu-plimit1'
+ Removed parameters '--gpu-ethash-dag-cache', '--a0-is-zil'
+ Renamed parameters '--a0-start-script' to '--a0-oc-script' , '--a1-start-script' to '--a1-oc-script', '--start-script-delay' to '--oc-delay'
+ Added '--esm 2' which indicates that 'EthereumStratum/1.0.0' + nicehash mode is used for communication with the pool*
+ Allowed separator ',' (comma) which can be used as an alternative for the current '!' separator in all parameters except in '--password'
+ Added custom miner version for Hiveos so you can run 2 SRBMiner instances on the same rig*
+ Minor bug fixes

This is the first release that supports INTEL Alchemist GPU's, so please report any bug/s you find in the discord support channel.

***
Algorithms currently supported on INTEL Alchemist GPU's [more algorithms soon]:

- kaspa
- sha512_256d_radiant
- sha256dt
- blake3_alephium
- heavyhash

No readings from sensors yet (clocks, temp, etc.)



***
ZIL mining parameters

--zil-enable
--zil-pool
--zil-wallet
--zil-password
--zil-esm
--zil-epoch

Every parameter except '--zil-enable' and '--zil-epoch' can have multiple values separated with ! (or now also ,)
You can define multiple ZIL pools, with different wallets for example.

'--zil-epoch' parameter isn't needed now, but it can be used in the future if Zilliqa decides to change the work epoch from 0.


***
GPU OC parameters [NVIDIA ONLY]

First algorithm [A0]:

--gpu-cclock0
--gpu-mclock0
--gpu-coffset0
--gpu-moffset0
--gpu-plimit0
--a0-oc-script

Second algorithm [A1]:

--gpu-cclock1
--gpu-mclock1
--gpu-coffset1
--gpu-moffset1
--gpu-plimit1
--a1-oc-script

ZIL:

--zil-cclock
--zil-mclock
--zil-coffset
--zil-moffset
--zil-plimit
--zil-oc-script

OC parameters are used to set GPU clocks, offsets, etc..
SRBMiner supports algo switching (--multi-algorithm-job-mode 3 and ZIL), so these parameters can be used to change GPU's OC on every algo change!

For example if you mine ETC+ZIL, miner will set OC for ETC on miner start from --gpu-...0 parameters.
When ZIL window opens, miner will apply the OC settings you defined with --zil-.. parameters.
When ZIL window closes, miner will apply OC settings you defined with --gpu-...0 parameters again.

--zil-oc-script :
Path to an executable script (.sh or .bat) which can call external tools to set GPU OC parameters right before switching to ZIL mining.
When ZIL mining period is finished, --a0-oc-script will be called (if user set it) before switching back to main algorithm mining.

--oc-delay can be used to define a delay (in seconds) after the OC settings were applied. Default value is 3 sec.

***
New functionality : auto switch on some algos to '--gpu-buffer-mode 1' on GPU's that probably wont have enough VRAM for ZIL dag and the main algo memory requirements.
Works only if the auto tune parameter is NOT used.
To disable this functionality use : '--gpu-disable-auto-buffer' parameter.
If you did not disable auto buffer mode, but miner crashes on algorithm switching (probably because of the mentioned reason - not enough VRAM), just add manually the --gpu-buffer-mode 1 parameter



***
Examples on how to set up ZIL mining in different combinations:

TIP : if your GPU doesn't have enough memory to hold the buffers of every algorithm you use, try adding '--gpu-buffer-mode 1' parameter

1. Mine ETHW+ZIL on ezil.me :

Code:

--algorithm ethash --pool eu.ezil.me:50100 --wallet ETHW-WALLET.ZIL-WALLET

2. Mine RADIANT+ZIL, RADIANT on rplant.xyz, ZIL on crazypool.org :

Code:

--algorithm sha512_256d_radiant --pool stratum-eu.rplant.xyz:7086 --wallet RADIANT-WALLET --zil-enable --zil-pool eu.crazypool.org:5005 --zil-wallet ZIL-WALLET --zil-esm 2

3. Mine ERGO+NOVO+ZIL, ERGO on herominers.com, NOVO on mynewmining.com, ZIL on crazypool.org :

Code:

--algorithm autolykos2 --pool de.ergo.herominers.com:1180 --wallet ERGO-WALLET --algorithm sha256dt --pool pool.mynewmining.com:3091 --wallet NOVO-WALLET --zil-enable --zil-pool eu.crazypool.org:5005 --zil-wallet ZIL-WALLET --zil-esm 2

4. Mine EPIC+KASPA+ZIL, EPIC on 51pool.online, KASPA on herominers.com, ZIL on crazypool.org :

Code:

--algorithm progpow_epic --pool 51pool.online:4416 --wallet 51POOL-USERNAME --algorithm kaspa --pool de.kaspa.herominers.com:1206 --wallet KASPA-WALLET --zil-enable --zil-pool eu.crazypool.org:5005 --zil-wallet ZIL-WALLET --zil-esm 2 --multi-algorithm-job-mode 3

Now comes the fun part. Adding GPU OC parameters. For simplicity the example will have 2 GPU's (id's 0 and 1) that use different OC settings.

2. Mine RADIANT+ZIL, RADIANT on rplant.xyz, ZIL on crazypool.org :

Code:

--algorithm sha512_256d_radiant --pool stratum-eu.rplant.xyz:7086 --wallet RADIANT-WALLET --zil-enable --zil-pool eu.crazypool.org:5005 --zil-wallet ZIL-WALLET --zil-esm 2 --gpu-id 0!1 --gpu-cclock0 1600!1800 --gpu-mclock0 810!810 --zil-cclock 1200!1300 --zil-mclock 5001!8000

4. Mine EPIC+KASPA+ZIL, EPIC on 51pool.online, KASPA on herominers.com, ZIL on crazypool.org :

Code:

--algorithm progpow_epic --pool 51pool.online:4416 --wallet 51POOL-USERNAME --algorithm kaspa --pool de.kaspa.herominers.com:1206 --wallet KASPA-WALLET --zil-enable --zil-pool eu.crazypool.org:5005 --zil-wallet ZIL-WALLET --zil-esm 2 --multi-algorithm-job-mode 3 --gpu-id 0!1 --gpu-cclock0 1200!1300 --gpu-mclock0 5001!8000 --gpu-cclock1 1700!1800 --gpu-mclock1 810!810 --zil-cclock 1200!1300 --zil-mclock 5001!8000

***
--esm values :
0 - getwork/ETHPROXY
1 - EthereumStratum/1.0.0
2 - EthereumStratum/1.0.0 + nicehash mode


***
Custom miner for Hiveos

To use it you need to select “Custom miner” when you create Flight Sheet and paste this into 'Installation url' : https://github.com/doktor83/SRBMiner-Multi/releases/download/2.1.0/srbminer_custom-2.1.0.tar.gz
After that you must set everything manually through the 'Extra config arguments' field by using SRBMiner parameters [ https://github.com/doktor83/SRBMiner-Multi/blob/master/Parameters ]

***
HiveOS update script from v2.0.2 to v2.1.0:

Code:

cd /tmp && wget https://github.com/doktor83/SRBMiner-Multi/releases/download/2.1.0/SRBMiner-Multi-2-1-0-Linux.tar.xz && tar -xvf SRBMiner-Multi-2-1-0-Linux.tar.xz && cd SRBMiner-Multi-2-1-0 && miner stop && cp SRBMiner-MULTI /hive/miners/srbminer/2.0.2 && miner start

[BoozyTalking]

Hi, GTX1660 not supported?

Code:

Detecting GPU devices...

GPU0 : nvidia_geforce_gtx_1660               [turing       ][MEM:  6 GB][CU:  22][BUS: 06]

======================================================================
SRBMiner-MULTI 2.1.0

Press 's' to display stats
Press 'h' to display hashrate
Press 'p' to switch to the next pool
Press 'o' to switch to the previous pool
Press 0-9 to disable/enable GPU 0-9, shift+0-9 for GPU 10-19
Press + or - to change gpu boost value
======================================================================

[2023-02-01 11:41:15] No worker devices found
Press any key to continue . . .

[deedeeranged]

Hi, GTX1660 not supported?

Code:

Detecting GPU devices...

GPU0 : nvidia_geforce_gtx_1660               [turing       ][MEM:  6 GB][CU:  22][BUS: 06]

======================================================================
SRBMiner-MULTI 2.1.0

Press 's' to display stats
Press 'h' to display hashrate
Press 'p' to switch to the next pool
Press 'o' to switch to the previous pool
Press 0-9 to disable/enable GPU 0-9, shift+0-9 for GPU 10-19
Press + or - to change gpu boost value
======================================================================

[2023-02-01 11:41:15] No worker devices found
Press any key to continue . . .

Windows/Linux drivers with cuda installed?

[BoozyTalking]

471.68

[sxemini]

471.68

Try newer driver maybe. Or you try to mine the wrong algo. Nvidia is not supported for all algos at the moment.

[Kareli]

Even INTEL came but FLUX algorithm did not come.
We wait by digging in Lolminer

[BoozyTalking]

471.68

Try newer driver maybe. Or you try to mine the wrong algo. Nvidia is not supported for all algos at the moment.

RTX3060 work fine, but with newer driver, will try new driver version.

[hbouma]

Cause you are using the old blockchain driver which can't allocate big buffers, and the DAG for Firo is 4+GB.
You need to use newer drivers, 19.x or newer

I've ran into this same problem with the latest SRBMiner-MULTI using the 20.4 drivers which is newer than 19.x with FiroPow. It seems to think my max buffer size is still 4GB and I'm not sure what to make of the 3004.80 driver version:

[2023-02-18 07:55:33] Starting init of worker threads
[2023-02-18 07:55:33] GPU DeviceID 0 [BUS:03] driver version: 3004.80
[2023-02-18 07:55:33] GPU DeviceID 0 [BUS:03] reported total memory: 8192 MB
[2023-02-18 07:55:33] GPU DeviceID 0 [BUS:03] reported free memory: 8074 MB
[2023-02-18 07:55:33] GPU DeviceID 0 [BUS:03] max buffer size: 4040 MB

If newer drivers are now required, which ones are good? The latest ones really hate bifurcation and everything freezes up for a minute when trying to enumerate them so hopefully I can use older ones that don't freeze up.

Also, I have this problem when trying to use --password with firopow. It says more passwords than pools given, even if I only just set a password and not try to pass in any parameters.

Thanks for all your help and writing this miner for all these years!

Hal

[matejbilahora]

It is good to have pufferfish2bmb algo for Pandanite mining. Thank you for your work..

[doktor83]

V2.2.0
+ Added algorithm 'dynex' (DYNEX coin) for GPU mining, fee 3.0%* [AMD/NVIDIA/INTEL]
+ Added INTEL ARC support for algorithms : autolykos2, etchash, ethash, ubqhash, blake3_ironfish
+ Added INTEL ARC support for dual mining : autolykos2/kaspa, autolykos2/heavyhash, autolykos2/sha512_256d_radiant, autolykos2/sha256dt, autolykos2/blake3_alephium, autolykos2/blake3_ironfish
+ Added dynex specific parameter '--mallob-endpoint'
+ Fixed 'blake3_ironfish' algorithm [also for DUAL modes]
+ Fixed 'sha256dt' algorithm for Nvidia ADA and HOPPER architectures
+ Added parameter '--disable-ptx-check' [nvidia]
+ Fixed bugs with NVIDIA OC parameters

Dynex:

Multiple pools & mallobs can be used.
There is no 'adj' parameter, to manually set the number of chips use '--gpu-intensity' parameter.

Dynex can be 'dual' mined with ZIL :

Code:

--disable-cpu --algorithm dynex --mallob-endpoint mallob.neuropool.net:8081,https://dnx.eu.ekapool.com --pool dynex.neuropool.net:19331,dnx.eu.ekapool.com:19666,dnx.mineradnow.space:18000 --wallet dynex-wallet-here --zil-enable --zil-pool eu.crazypool.org:5005 --zil-wallet zil-wallet-here --zil-esm 2

You can define multiple pools with the same/different wallets, miner will automatically failover when needed:

Code:

--pool dynex.neuropool.net:19331,dnx.eu.ekapool.com:19666,dnx.mineradnow.space:18000 --wallet pool1-wallet,pool2-wallet,pool3-wallet 

Windows users [ AMD GPUS ]:
Dynex POUW computations are long, so you should apply TDR fix [can be found in the Help folder], or else driver crashes might occur.