Topic: Team Black Miner (ETHW ETC Vertcoin Ravencoin Zilliqa +dual +tripple mining ) - page 39. (Read 35053 times)

This miner is original work. The stratum code is written from scratch, and have low hits on the antivirus compared to the other miners out there.

The latest gminer:

1.35 CUDA 11.4 build hangs on start with segmentation fault. Also archive contains some footprints of upx

All that hard work stealing open source code, making some improvements, and then profiting off of it without giving back.
Must be tiring.

The 11.5 build is faster, and early x-mas holiday now. So might not be buildt in a few days. The team is Tired. 10 months of hard work completed..

Thanks, will you release cuda 11.4 build soon?

linux cuda 11.5 build released

TeamBlackMiner_1_35_Ubuntu_18_04_Cuda_11_5.tar.gz
https://www.virustotal.com/gui/file-analysis/ZjY3YWY2MTAzMzlhODRkODU1MDE4NTllZmNiODhhMjk6MTY0MDExNTE1MA==

you are killing it with new releases

Team Black Miner 1.35

1. Improved LHR unlock

https://github.com/sp-hash/TeamBlackMiner/releases/

TeamBlackMiner_1_35_cuda_11_5.7z
https://www.virustotal.com/gui/file/55d2678a7439d0f59ed78f94f9ee694e21cf9662e198ee3b0038ec737bfc6634?nocache=1

TeamBlackMiner_1_35_cuda_11_4.7z
https://www.virustotal.com/gui/file/9227f90e0e5f989a668756a1fd0464dfb0bd7f84c3870118cbc1bf47decf975e?nocache=1

TeamBlackMiner_1_35_cuda_11_2.7z
https://www.virustotal.com/gui/file/7d09ec7d76e8c691e4c6096e14bfd6b6fdd27156f76f09a9a5268d1ddbc351f5?nocache=1

TeamBlackMiner_1_35_Ubuntu_18_04_Cuda_11_5.tar.gz
https://www.virustotal.com/gui/file-analysis/ZjY3YWY2MTAzMzlhODRkODU1MDE4NTllZmNiODhhMjk6MTY0MDExNTE1MA==

v1.35 has
Stable unlock on the 3070 v4 LHR latest driver

60%tdp
7900 memory
0 core.

1.34 fix has been released for windows. We have regression testing in a windows farm environment, it we find something we delay the linux release. Linux 1.34 soon. We also found a bug with a mixed card rig AMD/NVIDIA. For mixed cards, its best to run 2 instances of the miner. One with amd cards, the other with Nvidia.

Will the latest Linux builds be released soon? Why do they always lag windows releases?

The timeout timer is set to only 5 seconds in this build. Fast reconnection seem to improve the poolspeed, but  we need to investigate why it happens.

edit: Found a fix. In testing now..

1.34

https://ibb.co/2S78sWQ

i let it run...

https://ibb.co/pQrZSfJ

Understood. I'll wait for v 1.34 to be released in SMOS.

Se in my previos post.

Regression logs showed problems with 1.32 and 1.33 . GPU timeouts, and stratum timeouts.

On some of my test rigs the rig was sleeping for 3 minutes almost every hour, loosing 5% poolside hash. For single card rigs not a problem, and might work on some pools. But still a pretty serious bug. We are working on a better fix. The 1.34 windows version still have stratum disconnects but only 5 seconds instead of 180++

Exact same settings on same 7 x 3060 TI FHR with TBM v 1.32 and v 1.33. Running at -450 Core clock, 2400 Memory clock and 130W power limit.
Getting lower hashrate on 1.33, switched back to 1.32. Running SMOS with nv470.74 drivers.

-d -K 1 --dagintensity 1 --xintensity 284 -a ethash -H eu1.ethermine.org -p 4444 -r 40 -y 20 -t 120 -W "MAIN_WALLET" -w $rigName -P x --maintenance-wallet "MAINTENANCE_WALLET" --maintenance-percent 5

RIG is online >24H.

TBM 1.32 reports 437,xx MH/s.
I keep seeing wrong reported values (pool side). Reported hashrate on MAIN_WALLET is full rigs capacity not the 95% of it (as it should) and it fully correlates with what TBM is saying (437,xx MH/s). Reported hashrate on MAINTENANCE_WALLET remains 2,5 MH/s stable.
Average hashrate in MAIN_WALLET is low. It's about 380 - 384 MH/s. Average hashrate on MAINTENANCE_WALLET is between 18-22 MH/s.

So, 2 issues here.
One is the wrong reported values in MAIN and MAINTENANCE wallets respectively.
Two is the lower overall real hashrate than the reported one.

As a side note if you have any recommendations about my config they are most than welcome.
I manage a total of 10 GH/s in mixed AMD and NVIDIA rigs and I would like to switch them all in TBM as long as it is as stable and fast as other miners. My only reason for switching is the implementation of the maintenance fee since this is exactly the feature I want to get easily paid for my maintenance work.

v1.34

1. Fixed serious pool disconnect bug / GPU timeout bug.
2. Improved the pool reconnection code.
3. Fixed bug in mixed cards rig AMD/NVIDIA

https://github.com/sp-hash/TeamBlackMiner/releases

TeamBlackMiner_1_34_cuda_11_5_fix.7z
https://www.virustotal.com/gui/file/42d8f1c8f028bc6f1fe7df57ad0eeb0c3b273f46479824eb8c88905c57c2e7b2?nocache=1

TeamBlackMiner_1_34_cuda_11_4_fix.7z
https://www.virustotal.com/gui/file/d8564bdb582cac262c8ee4b18bfe2ec04d5e5689d723116eac00321063307b7b?nocache=1

TeamBlackMiner_1_34_cuda_11_2_fix.7z
https://www.virustotal.com/gui/file/ec39886bb4b7f33b4c3d81f5857055840323007b49b762316da614e302db425e?nocache=1


linux tomorrow.

Regression logs showed problems with 1.32 and 1.33 . GPU timeouts, and stratum timeouts.

Can you run nvidia-smi please whats the output.




Then you can try to query the device for stats with:

nvidia-smi --query-gpu=gpu_bus_id,temperature.gpu,power.draw,fan.speed,clocks.sm,clocks.mem,temperature.memory -i 0000:01:00.0 --format=csv,noheader,nounits

with --templimit as it puts in the command list, what happens is that the miner does not show anything other than the hashrate, it does not show or core speed, temperatures or consumption, it only puts a 0 in all that

How do you set the limits? The correct way is:

--lock-cclock [[x,x],[x,x],[x,x],[x,x],[x,x],[x,x],[x,x]]

Why are the temperatures or consumption or anything not shown?
so it cannot establish temperature limits because the miner simply ignores the limits that I set by not showing them, the gpus are from the 5000 and 6000 series and the version of the drivers is 21,11,03