Windows or Linux? cuda 11.5 build?
Windows. Cuda 11.4 build.
Topic: Team Black Miner (ETHW ETC Vertcoin Ravencoin Zilliqa +dual +tripple mining ) - page 36. (Read 35090 times)
January 07, 2022, 11:58:45 AM
Windows. Cuda 11.4 build.
January 07, 2022, 11:56:43 AM
Windows or Linux? cuda 11.5 build? Driver?
in 1.36 you need to increase the --xintensity for the best speed.
in 1.36 you need to increase the --xintensity for the best speed.
January 07, 2022, 11:23:54 AM
Hello sp_
"Out of the box" I get more mhs with v1.28 than I do with any other version (1.29,....,1.36). I cant post pictures as I am not a jr. member yet so you´ll have to believe me.
v1.28
rx 590 31.45mhs xint 24
rx 6800 63.27mhs xint 24
1660s 32.44mhs xint 284
1660s 35.38mhs xint 284
1660s 34.93mhs xint 284
v1.36
rx 590 28.84mhs xint 24
rx 6800 62.09mhs xint 24
1660s 31.70mhs xint 192
1660s 34.58mhs xint 192
1660s 34.12mhs xint 192
I guess I could mess about with xint and kernel and all the others but it is too time consuming.
"Out of the box" I get more mhs with v1.28 than I do with any other version (1.29,....,1.36). I cant post pictures as I am not a jr. member yet so you´ll have to believe me.
v1.28
rx 590 31.45mhs xint 24
rx 6800 63.27mhs xint 24
1660s 32.44mhs xint 284
1660s 35.38mhs xint 284
1660s 34.93mhs xint 284
v1.36
rx 590 28.84mhs xint 24
rx 6800 62.09mhs xint 24
1660s 31.70mhs xint 192
1660s 34.58mhs xint 192
1660s 34.12mhs xint 192
I guess I could mess about with xint and kernel and all the others but it is too time consuming.
January 07, 2022, 09:18:42 AM
i have a 3060 running on 470 driver. i'll just keep using gminer
use the the cuda 11.3 build to get a full unlock 51MHASH with RTX 3060 LHR1 (x16 pci usb dongle 470.05 driver)
TeamBlackMiner_1_37_cuda_11_3.7z
linux build of 1.37 has been released:
This is probobly the only miner that has no hits on virus total.
https://github.com/sp-hash/TeamBlackMiner/releases/
TeamBlackMiner_1_37_Ubuntu_20_04_Cuda_11_5.tar.xz
https://www.virustotal.com/gui/file/0d16a6d23119883afc746ae748da27e677a75766079cead33477cdd82eaf7f0e?nocache=1
January 07, 2022, 09:10:11 AM
how do i start this miner. all the examples BAt in the folder do not wrok
am getting "Your system does not support cuda 11.5 api" ihave downlooad and install cuda 11.5 toolkit
am getting "Your system does not support cuda 11.5 api" ihave downlooad and install cuda 11.5 toolkit
Download and install the latest driver from NVIDIA.
Try using the 495.46 drivers
January 07, 2022, 08:52:44 AM
how do i start this miner. all the examples BAt in the folder do not wrok
am getting "Your system does not support cuda 11.5 api" ihave downlooad and install cuda 11.5 toolkit
am getting "Your system does not support cuda 11.5 api" ihave downlooad and install cuda 11.5 toolkit
Download and install the latest driver from NVIDIA.
January 07, 2022, 08:12:24 AM
how do i start this miner. all the examples BAt in the folder do not wrok
am getting "Your system does not support cuda 11.5 api" ihave downlooad and install cuda 11.5 toolkit
am getting "Your system does not support cuda 11.5 api" ihave downlooad and install cuda 11.5 toolkit
Download and install the latest driver from NVIDIA.
January 07, 2022, 06:48:15 AM
how do i start this miner. all the examples BAt in the folder do not wrok
am getting "Your system does not support cuda 11.5 api" ihave downlooad and install cuda 11.5 toolkit
am getting "Your system does not support cuda 11.5 api" ihave downlooad and install cuda 11.5 toolkit
January 06, 2022, 01:41:05 PM
Hello, three questions:
1. Is SSL implemented yet ? When I try it the miner says it is not but maybe I am doing something wrong. This is my bat file: TBMiner --algo ethash --ssl --hostname eth-de.flexpool.io --port 5555
1. Is SSL implemented yet ? When I try it the miner says it is not but maybe I am doing something wrong. This is my bat file: TBMiner --algo ethash --ssl --hostname eth-de.flexpool.io --port 5555
It's implemented, but not working 100% so we disabled it.
2. When using TBM, is it better to use the high difficulty port in flexpool 14444 (I have read in this thread that it works better in high difficulty pools)
Depends on the rig. TBM doesn't work good on low difficulty but above 1.0 should be ok. Flexpool doesn't pay for stale shares, so --xintensity above 512 (nvida) or 128 (amd) will produce alot of stales.
3. I cant use --lock-cclock with amd cards, can I ?
This option is nvidia only.
which is about 12-13% above anything else I've ever seen! (512 intensity; 1642 MHz GPU; 2122 MHz Memory {Actual runtime reported values} @ 145 Watts).
You should make a youtube video.
How to increase your ethereum mining profit +12-13% poolside with the latest AMD cards..
1. Use the new The Team Black miner.
2. Change the Mining pool (?)
3. Run on high --xintensity
January 06, 2022, 12:52:50 PM
Hello, three questions:
1. Is SSL implemented yet ? When I try it the miner says it is not but maybe I am doing something wrong. This is my bat file: TBMiner --algo ethash --ssl --hostname eth-de.flexpool.io --port 5555
2. When using TBM, is it better to use the high difficulty port in flexpool 14444 (I have read in this thread that it works better in high difficulty pools)
3. I cant use --lock-cclock with amd cards, can I ?
Thanks
1. Is SSL implemented yet ? When I try it the miner says it is not but maybe I am doing something wrong. This is my bat file: TBMiner --algo ethash --ssl --hostname eth-de.flexpool.io --port 5555
2. When using TBM, is it better to use the high difficulty port in flexpool 14444 (I have read in this thread that it works better in high difficulty pools)
3. I cant use --lock-cclock with amd cards, can I ?
Thanks
January 06, 2022, 10:42:19 AM
time: 4
Have not received response from pool
We forgot to remove the debug output. Nothing to worry about.
Also, I think this may be a windows driver issue and maybe someone has a solution for it, but I notice that my 6900XT will momentarily drop power for a second and then ramp back up to full power. I am well aware of the fun little quirky behaviour that AMD cards have for running at full power and then deciding to just go into idle and stay there, but this drop to idle and restore to full power a second later is something I've only seen with my 6900XT.. Anyone have any ideas on what's happening?
Disable the amd watchdog service.
Disable fast startup mode in windows
use AMD settings to enable fast memory timings
use MSI afterburner to set clocks.
Other than that, there were times when this wasn't occurring that I was seeing 71 MH/s as the Average (3 hour) hashrate at the 2miners pool -- which is about 12-13% above anything else I've ever seen! (512 intensity; 1642 MHz GPU; 2122 MHz Memory {Actual runtime reported values} @ 145 Watts).
Great work!
Great work!
Thanks for your support!
what does the form mean on your screen, shares 1/1/1 if these are found shares, disabled shares, obsolete shares, then there are too many obsolete shares and this is nothing good
Accepted/Rejected/ (Estimated stale shares)
The estimation code in the miner window is not good, so you should always check the pool. Some pools pay for stale shares.
edit:
cuda 11.4 and 11.2 builds added. Linux will be added in a few days.
January 06, 2022, 10:41:52 AM
v1.36 beta windows only
what does the form mean on your screen, shares 1/1/1 if these are found shares, disabled shares, obsolete shares, then there are too many obsolete shares and this is nothing good January 06, 2022, 08:49:14 AM
v1.36 beta windows only
1. Faster speed on AMD cards.
2. Fix dag validation error on startup. (ZIL dag created instead of ETH/ETC)
3. Removed buffer error/crash in buffered console output.
4. Check and correct kernel parameter on illegal user input.
5. Improved kernel autotune by letting it run longer.
6. Fixed jumping hashrate bug on amd.
7. Fix crash on ultra-low difficulty.
https://github.com/sp-hash/TeamBlackMiner/releases/
TeamBlackMiner_1_36_cuda_11_5_beta.7z
https://www.virustotal.com/gui/file/cf050d7da4c6c5ac99a1197220d4809033d9ae084c89f4ac151f7c28b9637791?nocache=1
1. Faster speed on AMD cards.
2. Fix dag validation error on startup. (ZIL dag created instead of ETH/ETC)
3. Removed buffer error/crash in buffered console output.
4. Check and correct kernel parameter on illegal user input.
5. Improved kernel autotune by letting it run longer.
6. Fixed jumping hashrate bug on amd.
7. Fix crash on ultra-low difficulty.
https://github.com/sp-hash/TeamBlackMiner/releases/
TeamBlackMiner_1_36_cuda_11_5_beta.7z
https://www.virustotal.com/gui/file/cf050d7da4c6c5ac99a1197220d4809033d9ae084c89f4ac151f7c28b9637791?nocache=1
I'm running my 6900XT on windows and the miner is doing quite well except that I am getting a large number of the following messages:
time: 4
Have not received response from pool
Other than that, there were times when this wasn't occurring that I was seeing 71 MH/s as the Average (3 hour) hashrate at the 2miners pool -- which is about 12-13% above anything else I've ever seen! (512 intensity; 1642 MHz GPU; 2122 MHz Memory {Actual runtime reported values} @ 145 Watts).
Great work!
Also, I think this may be a windows driver issue and maybe someone has a solution for it, but I notice that my 6900XT will momentarily drop power for a second and then ramp back up to full power. I am well aware of the fun little quirky behaviour that AMD cards have for running at full power and then deciding to just go into idle and stay there, but this drop to idle and restore to full power a second later is something I've only seen with my 6900XT.. Anyone have any ideas on what's happening?
January 06, 2022, 06:47:52 AM
1.27 has a higher default xintensity, that give more stale shares.
Try 1.36 with --xintensity 512 or 384
Try 1.36 with --xintensity 512 or 384
January 06, 2022, 05:27:17 AM
Hi!, i don't know why, but the 1.27 is the fastest version for me, 107mh for the 3080FTW3, all the other releases 104-105mh, with the latest version of the vga driver and older.
January 05, 2022, 03:14:37 PM
v1.36
1. Faster speed on AMD cards.
2. Fix dag validation error on startup. (ZIL dag created instead of ETH/ETC)
3. Removed buffer error/crash in buffered console output.
4. Check and correct kernel parameter on illegal user input.
5. Improved kernel autotune by letting it run longer.
6. Fixed jumping hashrate bug on amd.
7. Fix crash on ultra-low difficulty.
https://github.com/sp-hash/TeamBlackMiner/releases/
TeamBlackMiner_1_36_cuda_11_5.7z
https://www.virustotal.com/gui/file/cf050d7da4c6c5ac99a1197220d4809033d9ae084c89f4ac151f7c28b9637791?nocache=1
TeamBlackMiner_1_36_cuda_11_4.7z
https://www.virustotal.com/gui/file/5ec33b032c979eec303251fceb5825a559bce09accdd61ecdbe698f48e86ece7?nocache=1
TeamBlackMiner_1_36_cuda_11_2.7z
https://www.virustotal.com/gui/file/6f34949fa887ae014bedb661692b4bdf8cda78eca821d92441c60b34e5049c08?nocache=1
TeamBlackMiner_1_36_Ubuntu_20_04_Cuda_11_5.tar.xz
https://www.virustotal.com/gui/file/635db89ed74d94733789b6f6df46010b8346c1e0e76b6aef605cc4ba877425d2
1. Faster speed on AMD cards.
2. Fix dag validation error on startup. (ZIL dag created instead of ETH/ETC)
3. Removed buffer error/crash in buffered console output.
4. Check and correct kernel parameter on illegal user input.
5. Improved kernel autotune by letting it run longer.
6. Fixed jumping hashrate bug on amd.
7. Fix crash on ultra-low difficulty.
https://github.com/sp-hash/TeamBlackMiner/releases/
TeamBlackMiner_1_36_cuda_11_5.7z
https://www.virustotal.com/gui/file/cf050d7da4c6c5ac99a1197220d4809033d9ae084c89f4ac151f7c28b9637791?nocache=1
TeamBlackMiner_1_36_cuda_11_4.7z
https://www.virustotal.com/gui/file/5ec33b032c979eec303251fceb5825a559bce09accdd61ecdbe698f48e86ece7?nocache=1
TeamBlackMiner_1_36_cuda_11_2.7z
https://www.virustotal.com/gui/file/6f34949fa887ae014bedb661692b4bdf8cda78eca821d92441c60b34e5049c08?nocache=1
TeamBlackMiner_1_36_Ubuntu_20_04_Cuda_11_5.tar.xz
https://www.virustotal.com/gui/file/635db89ed74d94733789b6f6df46010b8346c1e0e76b6aef605cc4ba877425d2
January 05, 2022, 03:00:54 PM
I have problems since version 1.29. Speed are go lower and lower and lower every second. The miner started with very high speed - example RTX2060S with over 50mh - but the speed goes down and down.
With version 1.28 it start with low hashrate and raising to the max. I use 11.2 cuda version and no i don´t want to update my driver - it is not a solution to update the driver with every new miner version.
With version 1.28 it start with low hashrate and raising to the max. I use 11.2 cuda version and no i don´t want to update my driver - it is not a solution to update the driver with every new miner version.
We have had many bugs in the avg. hashrate. You should run the latest version 1.36.
LHR unlock not working for me, only a few seconds and the card is locked again. (tested with newest driver)
You might need manual tuning of clocks and lhr parameters. Run with --log, and --lock-cclock
example:
RTX 3070 LHR3 60% tdp 7900 memory
RX6600XT it show an inflated hashrate by around 10% (TRM = 32,7 | TBM = 34,5)
You should get much less than 10% difference, but it depends on the pool and intensity. Try the 1.36 version. We have fixed a jumping hashrate bug.
How do you set multiple intensity if those cards are on the same rig ?
First run tbminer --list-devices to get the indexes.
then for 4 cards:
--xintensity [60,24,0,256]
January 05, 2022, 11:16:54 AM
testing v1.35 now in 2miners(2 days run), get more hash average 760mh, 778reported (rtx3090x6), but get les reward vs flexpool(get more ETH), in flex average 740mh 2% stales, runing same config same oc, only run --xintensity 4096, Any recommendation?
I think 4096 is too high on the 2miners pool. Even if all shares are accepted, some can be stale and give a lower payout.
Record poolside hashrate on the 3090x with --xintensity 3108 (137MHASH) (18 hour run)
See the video at:
https://imgur.com/a/stWmdM3
bugs reported.
The bugs you reported have been added to our github and been fixed
https://github.com/sp-hash/TeamBlackMiner/issues
I have a weird bug in 1.35 with flexpool.
I am sure not a connect problem, because my other rigs are working correct in the same time, and the gminer is never happend.
I am sure not a connect problem, because my other rigs are working correct in the same time, and the gminer is never happend.
v1.35 has only 20 second default stratum timeout. This is too short for some pools. You can add -t 80 for a 80 second timout.
As a pool I use Ethermine.
Rigs in farm.
RTX 3060 Ti FHR
RTX 3060 Ti LHR
RTX 3070 FHR
RTX 3080 FHR
RX5600 XT
RX5700 XT
RX580 8GB
RX6700
RX6800
Any recommendations on specific settings would be appreciated.
Rigs in farm.
RTX 3060 Ti FHR
RTX 3060 Ti LHR
RTX 3070 FHR
RTX 3080 FHR
RX5600 XT
RX5700 XT
RX580 8GB
RX6700
RX6800
Any recommendations on specific settings would be appreciated.
ok settings on the ethermine pool:
--xintensity 284 works good on nvidia cards on ethermine. If you go higher, you get alot of stale shares.
--xintensity 60 on rx580 8GB
--xintensity 78 on RX6700/rx6800 8GB
note that you can get a few percent poolside increase by changing to the recomended pools and increasing the --xintensity.
https://github.com/sp-hash/TeamBlackMiner/blob/main/FAQ.md
How do you set multiple intensity if those cards are on the same rig ?
January 05, 2022, 08:24:36 AM
I have problems since version 1.29. Speed are go lower and lower and lower every second. The miner started with very high speed - example RTX2060S with over 50mh - but the speed goes down and down.
With version 1.28 it start with low hashrate and raising to the max. I use 11.2 cuda version and no i don´t want to update my driver - it is not a solution to update the driver with every new miner version.
LHR unlock not working for me, only a few seconds and the card is locked again. (tested with newest driver)
AMD speed on Polaris is slow, on RX6600XT it show an inflated hashrate by around 10% (TRM = 32,7 | TBM = 34,5) - i have compared it with teamredminer and let it mine for 5 days both, poolside hashrate is slightly higher on TRM
ETC+Zil mining also not working since the first release, and i asked very often - but no solution.
Verthash is also not integrated at this point.
I will give it try in the future, but at the moment i must change the miner to a more reliable one.
We hope for future improvements, keep up your work and thanks.
With version 1.28 it start with low hashrate and raising to the max. I use 11.2 cuda version and no i don´t want to update my driver - it is not a solution to update the driver with every new miner version.
LHR unlock not working for me, only a few seconds and the card is locked again. (tested with newest driver)
AMD speed on Polaris is slow, on RX6600XT it show an inflated hashrate by around 10% (TRM = 32,7 | TBM = 34,5) - i have compared it with teamredminer and let it mine for 5 days both, poolside hashrate is slightly higher on TRM
ETC+Zil mining also not working since the first release, and i asked very often - but no solution.
Verthash is also not integrated at this point.
I will give it try in the future, but at the moment i must change the miner to a more reliable one.
We hope for future improvements, keep up your work and thanks.
January 03, 2022, 04:51:05 AM
I have been using miningpoolhub with high intensity, but seems like pool hashrate has been lower with versions since the timeout bug occurred.
Try with -t 80
I have reproduced on one of my amd rigs. 1 card seem to timeout. This causes the miner to restart, and recreate dag / loss of poolside hashrate. I am trying 1.35 with -t 80 now..
The default -t 20 in v1.35 is too short.
edit: 3 hours stable with -t 80 , so this is a fix for miningpoolhub and crazypool in v1.35
Jump to: