SRBMiner-MULTI AMD & NVIDIA & INTEL & CPU miner

[doktor83]

V2.4.2
+ Minor performance improvement on algorithm 'karlsenhash'
+ Added support for AMD gfx906 (VII) GPU's on Linux for drivers > 22.40
+ Added telemetry for INTEL ARC GPU's on Windows
+ Fixed power draw metric for AMD RDNA3 GPU's on Windows
+ Minor bug fixes

[Jengjeng]

Hi Doc,

Can we mine Etica ?

[Jonasuke]

V2.4.2
+ Minor performance improvement on algorithm 'karlsenhash'
+ Added support for AMD gfx906 (VII) GPU's on Linux for drivers > 22.40
+ Added telemetry for INTEL ARC GPU's on Windows
+ Fixed power draw metric for AMD RDNA3 GPU's on Windows
+ Minor bug fixes

great job!!!!

karlsenhash work dual mining?

thank you so much

[oyaebu777]

V2.4.2
+ Minor performance improvement on algorithm 'karlsenhash'
+ Added support for AMD gfx906 (VII) GPU's on Linux for drivers > 22.40
+ Added telemetry for INTEL ARC GPU's on Windows
+ Fixed power draw metric for AMD RDNA3 GPU's on Windows
+ Minor bug fixes

Hello, dear Doktor.
Can you please tell how to  set intensity for second algorithm. for example ergo+rxd: if i use --gpu-intensity it sets only for erg and rxd intensity is set to 1.
I tried also this way --gpu-id 0,1,2,3,4  --gpu-intensity 26,26,26,26,26;4,4,4,4,4 it also doesnt work for radiant. auto tune is always set rxd intensity on 2.
Also whatever --gpu-dual-max-loss  i choose its still sets second algo on 2. But i need at least 4 for second algo, so i use 2.3.9 where it sets intensity depends on overclock.
Can you please help how to set the desired  intensity for each algo?
PS: i use raveos.
PSS: sorry for my Eng.

[fenomenyaa]

V2.4.2
+ Minor performance improvement on algorithm 'karlsenhash'
+ Added support for AMD gfx906 (VII) GPU's on Linux for drivers > 22.40
+ Added telemetry for INTEL ARC GPU's on Windows
+ Fixed power draw metric for AMD RDNA3 GPU's on Windows
+ Minor bug fixes

Pls add AMD overclock and voltage settings to miner.

[Sergey2015]

who knows how to set up dual RTH+KLS tried to set up the hash on Carlson like this instead of 420 ---81((, AMD 5700XT cards

[stoniestfool]

V2.4.2
+ Minor performance improvement on algorithm 'karlsenhash'
+ Added support for AMD gfx906 (VII) GPU's on Linux for drivers > 22.40
+ Added telemetry for INTEL ARC GPU's on Windows
+ Fixed power draw metric for AMD RDNA3 GPU's on Windows
+ Minor bug fixes

Hello, dear Doktor.
Can you please tell how to  set intensity for second algorithm. for example ergo+rxd: if i use --gpu-intensity it sets only for erg and rxd intensity is set to 1.
I tried also this way --gpu-id 0,1,2,3,4  --gpu-intensity 26,26,26,26,26;4,4,4,4,4 it also doesnt work for radiant. auto tune is always set rxd intensity on 2.
Also whatever --gpu-dual-max-loss  i choose its still sets second algo on 2. But i need at least 4 for second algo, so i use 2.3.9 where it sets intensity depends on overclock.
Can you please help how to set the desired  intensity for each algo?
PS: i use raveos.
PSS: sorry for my Eng.

please try --gpu-auto-tune 0 . This allowed me to set intensity correct..

[girafa]

doktor83
We kindly ask you to mark the released map in the log not at the end of the entire log, but at the beginning of the log. Or change the timeout for the end of the log. And not like now 5 minutes.
And please add a function for changing the display time of the table.

[doktor83]

doktor83
We kindly ask you to mark the released map in the log not at the end of the entire log, but at the beginning of the log. Or change the timeout for the end of the log. And not like now 5 minutes.
And please add a function for changing the display time of the table.

I dont understand what you are trying to say. What is 'mark the released map' ?

[slappy6212]

Radeon VII refuse to start on windows 11 / SRBMiner-Multi giving me GPU0 couldn't create program for ctx->algorithm using latest AMD drivers and latest SRBMiner version.

[doktor83]

V2.4.3
+ Performance improvement on algorithm 'karlsenhash' for some AMD GPUs and INTEL ARC GPU's
+ Added support for dual mining AUTOLYKOS2/KARLSENHASH on AMD RDNA/RDNA2/RDNA3 and INTEL ARC GPU's
+ Added support for dual mining ETHASHB3/KARLSENHASH on AMD Ellesmere, RDNA/RDNA2/RDNA3 and INTEL ARC GPU's
+ DevFee for 'karlsenhash' algorithm lowered to 0.85%
+ Minor bug fixes

[slappy6212]

Can someone tell me what is this error message mean using RADEON VII ?    GPU0 couldn't create program for ctx->algorithm

[deedeeranged]

Can someone tell me what is this error message mean using RADEON VII ?    GPU0 couldn't create program for ctx->algorithm

A bit sparse with details?

OS, mining algo and maybe some more details would not go amiss.
Like what is your command line you use to start mining.

I cannot understand why people still do not provide the necessary details is beyond me.

[doktor83]

V2.4.4
+ Added algorithm 'pyrinhash' (Pyrin) for GPU mining on AMD/INTEL, fee 0.85%
+ Added algorithm 'blake3_decred' (Decred) for GPU mining on AMD/NVIDIA/INTEL, fee 1.0%
+ Added support for dual mining AUTOLYKOS2/PYRINHASH on AMD RDNA/RDNA2/RDNA3 and INTEL GPU's
+ Added support for dual mining AUTOLYKOS2/BLAKE3_DECRED on AMD/NVIDIA/INTEL GPU's
+ Added support for dual mining ETHASH/BLAKE3_DECRED on AMD/NVIDIA/INTEL GPU's
+ Added support for dual mining ETCHASH/BLAKE3_DECRED on AMD/NVIDIA/INTEL GPU's
+ Added support for dual mining ETHASHB3/BLAKE3_DECRED on AMD/NVIDIA/INTEL GPU's
+ Improved efficiency on algorithm 'blake3_alephium' for AMD RDNA2/RDNA3, NVIDIA GPU's
+ Miner will now create '--gpu-id' parameter values internally (if not already defined) when using an OC parameter that has multiple different values.
+ Minor bug fixes

[orekixu]

Couldn't connect to a devfee pool for a long time, mining will be paused until connection to the devfee pool can be established



I have checked the logs of dozens of machines and found that due to the issue with devfee, the machines have been working for almost two hours and have been stuck for more than half an hour.

Then I checked the backend links of each computer's network, with IP 193.70.81.165, port 5170, and IP 49.13.120.2, port 20033, totaling 9 links. Why does the kernel still say it can't connect to devfee? I would rather you pump 5% of the water than let the machine run empty most of the time, hoping that the developer can modify the logic of devfee. Additionally, I am digging RTH+ALPH in a dual mine

[doktor83]

Couldn't connect to a devfee pool for a long time, mining will be paused until connection to the devfee pool can be established



I have checked the logs of dozens of machines and found that due to the issue with devfee, the machines have been working for almost two hours and have been stuck for more than half an hour.

Then I checked the backend links of each computer's network, with IP 193.70.81.165, port 5170, and IP 49.13.120.2, port 20033, totaling 9 links. Why does the kernel still say it can't connect to devfee? I would rather you pump 5% of the water than let the machine run empty most of the time, hoping that the developer can modify the logic of devfee. Additionally, I am digging RTH+ALPH in a dual mine

there are 12 devfee pools total for those 2 algorithms, meaning none of them could get connected from where you are. If only one of them connects (any of the two algos) that message wont be shown, and mining wont be paused.
Fix what's blocking the connections and you wont have any issue.

[BoozyTalking]

MinotaurX is broken in last version. Dev fee links for many algos is broken (at least in old versions), dev is lazy ass =)

[doktor83]

MinotaurX is broken in last version. Dev fee links for many algos is broken (at least in old versions), dev is lazy ass =)

You are right about MinotaurX. About the broken devfee pools, which ones?

[orekixu]

Couldn't connect to a devfee pool for a long time, mining will be paused until connection to the devfee pool can be established



I have checked the logs of dozens of machines and found that due to the issue with devfee, the machines have been working for almost two hours and have been stuck for more than half an hour.

Then I checked the backend links of each computer's network, with IP 193.70.81.165, port 5170, and IP 49.13.120.2, port 20033, totaling 9 links. Why does the kernel still say it can't connect to devfee? I would rather you pump 5% of the water than let the machine run empty most of the time, hoping that the developer can modify the logic of devfee. Additionally, I am digging RTH+ALPH in a dual mine

there are 12 devfee pools total for those 2 algorithms, meaning none of them could get connected from where you are. If only one of them connects (any of the two algos) that message wont be shown, and mining wont be paused.
Fix what's blocking the connections and you wont have any issue.

As a Chinese miner, network interception is not our own behavior, but rather done by the government. We also rent overseas servers to connect to some mining pools through port forwarding. Can developers provide a user-defined devfee method? The devfee pool can be customized by a certain parameter, so that we can also achieve stable links to the devfee pool through port forwarding.

[orekixu]

Couldn't connect to a devfee pool for a long time, mining will be paused until connection to the devfee pool can be established



I have checked the logs of dozens of machines and found that due to the issue with devfee, the machines have been working for almost two hours and have been stuck for more than half an hour.

Then I checked the backend links of each computer's network, with IP 193.70.81.165, port 5170, and IP 49.13.120.2, port 20033, totaling 9 links. Why does the kernel still say it can't connect to devfee? I would rather you pump 5% of the water than let the machine run empty most of the time, hoping that the developer can modify the logic of devfee. Additionally, I am digging RTH+ALPH in a dual mine

there are 12 devfee pools total for those 2 algorithms, meaning none of them could get connected from where you are. If only one of them connects (any of the two algos) that message wont be shown, and mining wont be paused.
Fix what's blocking the connections and you wont have any issue.

SRBMiner-MULTI.exe          安全           TCP          192.168.31.97     49731        *******      55070        连接           中国 中国香港 中国香港 * *                   
SRBMiner-MULTI.exe          安全           TCP          192.168.31.97     49732        *******      51199        连接           中国 中国香港 中国香港 * *                   
SRBMiner-MULTI.exe          安全           TCP          192.168.31.97     49733        *******       51111        连接           中国 中国香港 中国香港 * *                   
SRBMiner-MULTI.exe          安全           TCP          192.168.31.97     49797        193.70.81.165     5170         连接           法国 * * * *                         
SRBMiner-MULTI.exe          安全           TCP          192.168.31.97     49817        49.13.120.2       20033        连接           德国 * * * *                         
SRBMiner-MULTI.exe          安全           TCP          192.168.31.97     50159        49.13.120.2       20033        连接           德国 * * * *                         
SRBMiner-MULTI.exe          安全           TCP          192.168.31.97     50214        49.13.120.2       20033        同步发送         德国 * * * *                         
SRBMiner-MULTI.exe          安全           TCP          192.168.31.97     50333        49.13.120.2       20033        连接           德国 * * * *                         
SRBMiner-MULTI.exe          安全           TCP          192.168.31.97     50652        49.13.120.2       20033        连接           德国 * * * *             

 
*******It is a server that I purchased myself for port forwarding. 193.70.81.165:5170 is the Devfee pool for RTH. Why does the ALPH Devfee pool for 49.13.120.2:20033 need to start 5 threads and link simultaneously. I have seen SRB start 8 threads linking the same devfee pool at most, one should be enough, or starting 8 devfee pools with different links can also be understood.