SRBMiner-MULTI CPU & AMD GPU Miner 0.4.4
Windows & Linux
**Download links: mega.nz/#F!tGoFHRbS!xJaRZ_54xiAYiZmBihxPWg
[CPU & GPU]
MTP scratchpad creation is very slow at the moment, it is waiting on a big rewrite. This has an impact on poolside hashrate if a lot of new works are sent out.
- VEGA 56/64/FE/VII
- RX 460/470/480/550/560/570/580/590
- R9 285/285X/380/380X
- R9 290/290X
- R9 Fury/Nano
Linux : amdgpu-pro drivers
- Guided setup mode
- Algorithm switching capability without external application
- Run in background without a window
- Hashrate watchdog that restarts miner on GPU error
- Monitoring of GPU temperature, and auto turn off if temperature is too high
- System shutdown on too high GPU temperature
- Miner auto restart on too many rejected shares
- Startup monitor which ensures your miner starts working normally
- API for miner statistics
- Web based GUI interface for miner statistics
- Multiple pools with failover support
- Add new pools on the fly without restarting miner
- Difficulty monitor, reconnects to pool if difficulty is too high
- Job timeout monitor, reconnects to pool if no job received for a long time
- Switch AMD video cards to compute mode easily
1. Disable CPU mining, use only GPU mining
SRBMiner-MULTI.exe --algorithm keccak --pool your-pool-here --wallet your-wallet-here --disable-cpu
2. Run miner in background without console window, with API enabled on port 17644 (127.0.0.1:17644(127.0.0.1:17644/))
SRBMiner-MULTI.exe --algorithm keccak --pool your-pool-here --wallet your-wallet-here --background --api-enable --api-port 17644
3. Disable GPU mining, use 7 CPU threads with extended logging enabled and saved to Logs\log.txt file
SRBMiner-MULTI.exe --algorithm keccak --pool your-pool-here --wallet your-wallet-here --disable-gpu --cpu-threads 7 --log-file Logs\log.txt --extended-log
4. Full example for CPU & GPU mining set only from cmd (4 GPU/s used and 7 CPU threads on K12 algorithm)
SRBMiner-MULTI.exe --algorithm k12 --gpu-id 0,1,2,3 --gpu-intensity 26,25,26,26 --gpu-worksize 256,256,256,256 --gpu-threads 1,1,1,1 --cpu-threads 7 --pool your-pool-here --wallet your-wallet-here
5. Disable GPU mining, use 15 CPU threads with extended logging enabled and saved to Logs\log.txt file, start mining Randomx from block height 1978433 and run miner in background
SRBMiner-MULTI.exe --algorithm randomx --pool your-pool-here --wallet your-wallet-here --disable-gpu --cpu-threads 15 --log-file Logs\log.txt --extended-log --background --start-block-height 1978433
CPU MINING SETUP
If you don’t set the –cpu-threads or –cpu-affinity parameters, miner will try to automatically find the best setup for you system.
Sometimes the auto setup won’t find the optimal settings so you should find it by experimenting.
You need to set the number of worker threads (–cpu-threads) and bind them to the appropriate PU (processing unit) with --cpu-affinity.
Examples to get you started:
4 CORE / 8 THREADS CPU :
- Use 4 threads, 1 on every core : --cpu-threads 4 --cpu-affinity 0x55
- Use 8 threads, 2 on every core : --cpu-threads 8 --cpu-affinity 0xFF
6 CORE / 12 THREADS CPU :
- Use 6 threads, 1 on every core : --cpu-threads 6 --cpu-affinity 0x555
- Use 12 threads, 2 on every core : --cpu-threads 12 --cpu-affinity 0xFFF
RANDOMX EXTRA TWEAKS
To enjoy the benefits of the increased hashrate:
1. Miner must run with administrator privileges [right click on SRBMiner-MULTI.exe->properties->compatibility-> check ‘Run this program as an administrator’ option-> click OK button
2. Make sure WinRing0x64.sys is in the same folder as SRBMiner-MULTI.exe
Enabling the extra tweaks, some register values are changed which will revert back to the original value after a computer restart.
If miner starts crashing, or creates bad results after enabling the extra tweaks, you can try the ‘–randomx-use-tweaks’ parameter, where you can define which tweak/s you want to enable.
Without this option miner enables all available tweaks (same as --randomx-use-tweaks 0123 for AMD, --randomx-use-tweaks 0 for Intel) , but with ‘–randomx-use-tweaks’ you can define which ones you want to use.
There are at the moment :
Intel - 1 tweak
Amd - 5 tweaks
Index starts from 0, so you have 0 1 2 3 4 for options on AMD.
Here’s an example :
+ Use tweaks number 0 and 3 (dont use 1 and 2)
SRBMiner-MULTI.exe --algorithm randomx --pool your-pool-here --wallet your-wallet-here --password your-password-here --randomx-use-tweaks 03
The order of tweaks doesn’t matter, so writing 03 is same as 30.
What you should do is to find which tweak, or combination of tweaks, makes the problems.
You should try enabling tweaks one by one, and test to see if miner runs stable with one tweak, and if it does, test the next one etc.
Luckily there are not so many combinations for you try (because 0123 is for example same as 3021 or 1302… )
So start with :
SRBMiner-MULTI.exe --algorithm randomx --pool your-pool-here --wallet your-wallet-here --password your-password-here --randomx-use-tweaks 0
If it runs without crashing for some time, try the next one :
SRBMiner-MULTI.exe --algorithm randomx --pool your-pool-here --wallet your-wallet-here --password your-password-here --randomx-use-tweaks 1
and so on, until you find the one that makes the trouble.
Then try combining tweaks to find a combination that works for you.
If you have older gen. Ryzens, try this combination of tweaks :
SRBMiner-Multi is supported by Awesome Miner
If you get “Insufficient system resources available to allocate X kB in large-page memory” message, that means you dont have enough FREE memory left, a computer restart should solve this.
Large-page memory regions may be difficult to obtain after the system has been running for a long time because the physical space for each large page must be contiguous, but the memory may have become fragmented.
If you still get this message even after restarting, try increasing virtual memory.
You can find all the parameters and everything you need to get going in the readme.txt file.
This thread is self moderated, because in the previous SRBMiner thread there was too much spam and garbage, which wasn’t removed by the forum moderators.
Update infos :
- added ‘ethash’ algorithm
- added ‘ubqhash’ algorithm
- added parameter ‘–esm’ to set ethash/ubqhash stratum mode for pool (0 - eth-proxy, 1 - miner-proxy)
- added parameter ‘esm’ to pools config (ethash/ubqhash stratum mode for pool (0 - eth-proxy, 1 - miner-proxy))
- added parameter ‘–enable-ethash-leak-fix’ which tries to workaround AMD driver memory leak when recreating DAG
- fixed algo switching and tweak_profiles miner freeze bug
- –gpu-tweak-profile option can now be used with auto gpu setup in cmd
- Added ‘tellor’ algorithm (trb.uupool. cn compatible stratum)
- Improvements on ‘minotaur’ algorithm
- Guided setup fixed & simplified
Added ‘tellor’ (TRB) algorithm for uupool. cn compatible stratum pools. I tried to contact hashpool. com so i could make their pool compatible too, but could not find any ways to do it (no discord, not answering on email).
Previous version introduced ‘disable_cpu’ and ‘disable_gpu’ parameters for use in configuration files.
Because of that, mixing of cmd parameters ‘–disable-cpu’, ‘–disable-gpu’ and configuration files isn’t working as before.
Now you should use only one method : either cmd parameters completely, or configuration files, don’t mix them.
- Improvements for 4G Ellesmere cards on ‘cryptonight_bbc’ algorithm
- Small improvements on ‘yespower’ CPU algorithms
- Added parameters ‘disable_cpu’ and ‘disable_gpu’ in configuration file
- Fixed ‘handshake’ algorithm mining on nicehash
- Algoswitching now reconnects back to main pool when it gets available
- Minor bug fixes
- Added CPU algorithm ‘minotaur’ (Ring coin)
- Added CPU algorithm ‘yespowerres’ (Resistance coin)
- 2-4% faster ‘cryptonight_bbc’ on some GPU’s
- Minor bug fixes
- Added algorithm ‘randomkeva’ (Keva coin upcoming algorithm)
- Added Navi support for ‘cryptonight_bbc’ algorithm
- Much better auto setup for RX 4xx/5xx 8G cards on ‘cryptonight_bbc’ algorithm
- Miner now works on older linux distributions ( Ubuntu 16.04 - glibc2.23 )
- Small hashrate increase on ‘cryptonight_bbc’ algorithm for some GPU’s
- Yescrypt algorithms hashrate increase on CPU if miner run with administrator/root privilege
- Fixed GPU sensors reading on Linux
- Fixed a bug that sometimes stopped hashing when connected to the pool after a long reconnect period
- Renamed ‘–disable-randomx-tweaks’ parameter to ‘–disable-msr-tweaks’, ‘–disable-tweaking’ to ‘–disable-gpu-tweaking’, and ‘–randomx-use-tweaks’ to ‘–msr-use-tweaks’
- Minor bug fixes
- Hashrate increased even more on ‘kadena’ algorithm, and at the same time lowered power usage
- Small hashrate increase on ‘yescrypt’ CPU algorithms
- Minor bug fixes
- ‘kadena’ algorithm huge hashrate increase
- Added ‘gpu_’ prefix to GPU related parameters in config file : gpu_intensity, gpu_raw_intensity, gpu_worksize, gpu_threads, gpu_tweak_profile, gpu_target_temperature, gpu_shutdown_temperature, gpu_off_temperature. ‘gpu_conf’ parameters naming stays the same.
- Added new parameters in config file : cpu_threads, cpu_affinity, cpu_priority
- Renamed ‘pool_use_tls’ in pool configuration file to ‘tls’
- ‘cryptonight_bbc’ algorithm updated (active from block 133060)
- Added ‘–enable-opencl-cleanup’ parameter to enable OpenCL resources freeing on miner exit/restart (default is disabled because it can hang miner if driver crashed)
- Added ‘cryptonight_bbc’ algorithm (BigBangCore) - DevFee 2.5%
- Lowered devfee for ‘bl2bsha3’ to 2%
- Some gpu’s might get a tiny hashrate increase on ‘bl2bsha3’
- Gpu sensors now available in Linux
- –adl-disable parameter now available in Linux
- –shutdown-temperature parameter (shutdown_temperature in config) now available in Linux
- –gpu-off-temperature parameter (off_temperature in config) now available in Linux
- Removed option to set difficulty for ‘bl2bsha3’ in password field
- Removed notification for 6block stratum compatibility
- Possibly fixed a bug where miner shows it’s hashing (always shows exactly same hashrate), but gpu’s are stalled
- Fixed share accepted/rejected display issue with 6block pool (HNS)
- Fixed miner auto restarting issue on Linux
- Tiny hashrate increase on ‘bl2bsha3’ with some GPU’s
- Hashrate increase on ‘bl2bsha3’ algorithm (~ 10-15% )
- Added support for Navi (gfx1010) cards for ‘bl2bsha3’ (let’s just say it works)
- Added possibility to set difficulty for ‘bl2bsha3’ . Just enter it in the pool password field.
- Miner can be now used on pools other than 6block if they implement the same stratum protocol as 6block
- Added algorithm ‘bl2bsha3’ (Handshake (HNS) coin - 6block. com pool only)
- RandomX fixes
- Added ‘–gpu-platform’ parameter to cmd config, and ‘gpu_platform’ to configuration files to force opencl platform
- Added ‘–randomx-use-1gb-pages’ parameter to config, whiche enables 1GB sized page files if supported by CPU
- When reading pools or algorithms configuration file, if invalid entry found it will be just skipped
- Fixed ERROR 0x4002
- First version for Linux (Tested on Ubuntu 18.04)
- Removed RandomV (MoneroV) algorithm
- Added ‘defyx’ algorithm (Scala)
- Added ‘yespoweritc’ algorithm (Intercoin)
- Added ‘cryptonight_catalans’ algorithm (CatalansCoin)
- Added ‘cryptonight_talleo’ algorithm (Talleo)
- –randomx-use-tweaks accepts now values from 0-4, [ 0 - Intel, 0,1,2,3,4 - AMD ]
- Added parameter ‘worker’ to pools config and cmd parameters -> defines worker name for RPC2
- a lot of bug fixes
- Added RandomV algorithm (MoneroV)
- Added --randomx-use-tweaks , to select the MSR tweak/s to apply
- Added RandomX tweak support for Intel Core2 processors
- Added RandomSFX algorithm (SafexCash)
- RandomX algorithm/s improvements
- Added RandomX algorithm/s extra tweak that can increase hashrate
- Added parameter --disable-randomx-tweaks
- Fixed K12 algorithm
- –cpu-priority now affects main process too (not just worker threads)
- default --cpu-priority is now 2 (below normal)
- bug fixes
- added license text to readme for RandomX and WinRing0
- Added Kadena algorithm (pool mining only)
- Added yespoweric algorithm (IsotopeC)
- Added yespoweriots algorithm (IOTScoin)
- Added yespower2b AVX support
- Fixed nicehash protocol compatibility with rpc2
- Miner now won’t quit if not enough free RAM for an algorithm, it will just warn the user
- Added ‘m7mv2’ algorithm (Magi coin - XMG)
- Cpu auto configuration should be better than on previous versions
- Added CPU data to API, also rearranged a few things in API
- Console window ‘Quick Edit’ mode disabled on miner start, so users won’t accidentally pause miner by clicking in the console window
- Fixed threads binding on systems with more than 1 numa node
- Better auto configuration for cpu mining
- Previous two fixes should bring higher hashrates
- Added pool controlled algorithm switching capability (Monero Ocean)
- Added pool config parameter “algo_min_time” and --algo-min-time to cli (Monero Ocean)
- Added pool config parameter “keepalive” and --keepalive to cli
- Added cpu algorithms : RandomX, RandomXL, RandomWow, RandomArq
- Added cli parameters: --cpu-priority, --disable-huge-pages, --disable-hw-aes, --disable-numa
- Added pool config parameter ‘start_block_height’ and --start-block-height cli parameter
- Fixed a bug that limited the number of threads useable for cpu mining
- Fixed cpu affinity parameter and thread/s binding to affinity mask
- Where possible, now using huge pages if available, this should bring a slight performance increase
- Removed OpenCL dependency if using only cpu miner
- Added limited Overdrive8 support (adl type 3)
- Fixed a critical bug with work creation for fast algorithms [k12, keccak]
- Keccak algorithm +15% faster on gpu
- K12 algorithm +8-12% faster on gpu [vega56 ~1700mhs]
- Guided setup got smarter a little bit
- Added support for kangaroo12 algo [k12]
- All algorithms that have a fee now have the same ~0.85% devfee
- More bug fixes
+Added yespowerltncg, yespowersugar, yespowerurx, yespower2b, eaglesong (nervos-ckb) algorithms
+Added support for SHA with yespower algorithm family