Topic: Team Black Miner (ETHW ETC Vertcoin Ravencoin Zilliqa +dual +tripple mining ) - page 112. (Read 34373 times)

Wich pool, platform version and settings?

---

v1.02 beta
1. Added stale share check on the submitted shares and display the count.
2. Added --xintensity parameter.(-E) To feed the pool(s) what they need. (default is --xintensity 32, valid values --xintensity 1 and up to 10000? --xintensity -1 For the program to calculate the best intensity.)
3. Try to fix memory problem on startup
4. Added support for more pools
5. Fixed crash in the opencl mining. Cards stopped working.

windows exe only

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


If you have time, post the best --xintensity. We need the pool and cards.  Nicehash and the other 100ms job pools probobly need a low value to avoid rejected/stale shares. Etheremine, Solominingpools, crazypool, miningpoolhub and a few others works best with --xintensity -1 or perhaps high fixed values.

Low xintensity will give poor hashrate, so try to find the sweetspots.

[moderator's note: consecutive posts merged]

I have tried the latest version with gtx 1070 x 8, got 60 shares rejected (100%) and back to t-rex.

Did you get more stales with 1.01? Stales should have been reduced with 1.01. Use the latest _fix version on git. I had to upload a patch because of the duplicate bug. They share the same version number. Stale shares are also pool depended. Some pools spam you with jobs, some are more quiet. You get payed for stale shares. Some pools pay for them as normal shares.

We will add a stale share checker in the next version and perhaps an intensity setting where you can adjust the workload sent to the gpu manually. Good for pool finetuning.

---

Here is a preview of version v1.02. The third column is the stale share amount.  The gpu is a rtx 2060 Mobile (laptop)

Accepted/Rejected/Stale

---

Added a --xintensity option. This option is similar to the one in ccminer or sgminer. It controls the amount of work the gpu does in one warp. A big workload give good hashrate, but can also give rejects and stale shares. We have also added a --xintensity=-1 option to let the program calculate the workload dynamicly based on hashrate.  Without the option the miner will run with --xintensity=32




[moderator's note: consecutive posts merged]

Try 1.01, stats fix. But got a lot stales shares.

https://i.ibb.co/DbV8RgS/vs.png

[/url]

Did you try v1.01?

The cuda 11_2 build is working on 470.05. will build v1.01 soon

---

Thanks for testing. Note that the optimal clock and power settings for gminer might not be the optimal for the Team Black miner. More power will give you more hashrate. Your requests have been put in the todo log . check for updates in the coming days. We are also working on a LHR full unlock feature...

---

- Fixed duplicate share bug on some pools
- Fixed crash in stats when using -Y or -U to manually specifying gpu list
- Added more pools

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

TeamBlackMiner_1_01_cuda_11_2_fix.7z
SHA-256   3b65b2fa7a7c1a66bc0d38778df9429066993ab2df76c3050842f715bfb045d0

TeamBlackMiner_1_01_cuda_11_4_fix.7z
SHA-256   7997f29017dbacb8c731ad7cbcd2369e23b4f209364956ca15585d936ac0683d

TeamBlackMiner_1_01_Ubuntu_18_04_Cuda_11_4.zip
SHA-256   88ad521b6171e84b91f36adc8162d4dc1830a83e9f59ad8198d0b17005f7f570

---

Try the cuda 11.2 build for windows

[moderator's note: consecutive posts merged]

little feedback, I don't know how you manage to get my rtx3070 to 63Mh? while gminer I have 61.40Mh with this same settings, after fault of pool "crazypool" not compatible I can not see if the shares are there

So I'm on hiveOs, not easy the instllation, even with the links in the read-me, I think if you had an automation screen while waiting that could help a lot of people.

I also had to put a dependency in addition I put you here


and for cuda 11.04:
https://medium.com/@anarmammadli/how-to-install-cuda-11-4-on-ubuntu-18-04-or-20-04-63f3dee2099

I did not succeed in integret in hiveos via flight-sheets

I used nvidia-smi for my OCs because in the minor one cannot put the core and memory on nvidia

on the page https://github.com/sp-hash/TeamBlackMiner

it would be nice to put more details on the command line example:
-p or --port
=> mining pool port (for example: '8888', '3333')

A function is missing to generate the dag clean and then check it if there is no error. like at gminer, phoenixminer or trex which generates in slow mode and then checks no error

now we have to fix the crash errors, so more stability and diff on pool is 8.590G

Let's see more ... when it will be more stable and easier to deploy it, I will do a test on one of my farms + 2Ghs to see the sharing rate / hour

Good continuations, with each update I will test and you will return

---

rollback driver and try it !! last version driver for mining is useless

you have to add your own tooooo





[moderator's note: consecutive posts merged]

Not working with 470.05? Are you serious??? Totally useless...

Need to debug that one. isnull indicate a zero nounce. Whats the difficulty was the pool running at?

---

Not yet. added to the todo list.. You can use the -N option (--no-cpu) to reduce cpu usage to almost 0.

---

$$$

[moderator's note: consecutive posts merged]

Is there a way of adjusting intensity? Id like to dial it down just a notch.

Ok see that



and after miner crash, juste restart and OK

We need to configure it. Use one of the pools in the 40 scripts

It's work on hiveOS 18.04 nice job

https://0bin.net/paste/DJU98qo+#RpQonM8BRaoDAwm8xMxgD30ROHbX+sYs1VT-9cn67+B


But for crazypool impossible to operate because it uses "stratum"

how to try ? what is command line ?


great job
3070 RTX: 63.40Mh
1070 | 2350 | 150

What about the gtx 750ti? The opensource code you forked and stole was optimized for cuda 5.0 by me

2014:

Thanks for testing. Good points, will add them into the todo list.

Good job, but there should also be --cuda-only option, also libapr is still required as shared library, not all distros have it installed default.
There is one more issue which was expected, RTX 3060ti/3070 if clocked a bit higher tend to corrupt DAG and produce only invalids, requires DAG checksum validation and repairing or some sort of low intensity DAG loading.

v1.01
1. Faster Nvidia kernel
2. Removed stale shares/rejects on some pools. (ethproxy/nicehash)
4. Fixed missing stats when you have unused cards in the rig.
5. Fixed crash in zil standalone mining
6. Fixed cpu rejected shares upon zil switch.
7. Fixed disconnect bug in zil dual mining mode.
8. Fixed rare sudden exit crash
9. Added amd-only option. Only use amd cl-devices
10. Added --no-cpu option for cuda to remove cpu usage. but hashrate is down -1-2%

Windows:

TeamBlackMiner_1_01_cuda_11_4.7z
SHA-256   1de3485ee14caaf5274a92079260e86bf87afe5ae2b6be2cc77b2a71fc746906
Linux: (hiveos?)

TeamBlackMiner_1_01_Ubuntu_18_04_Cuda_11_4_beta.zip
SHA-256   3805e68a981ac8ba41ddeca1aedcec531f0fd5b8e2ff21741ff8667375348941

---

Go get those coins!

[moderator's note: consecutive posts merged]

When next releases?

ubuntu 18.4 with cuda 11.4 is running. Will include it in the next release

Yes - this is what sp_ has told us:


In the meantime, I have put it behind a wrapper. Works like a charm.