Topic: [ANN]Bminer: a fast Equihash/Ethash/Cuckaroo29z miner for AMD/NVIDIA GPUs 16.4.9 - page 17. (Read 148347 times)

NONE of the Cuda/Nvidia drivers are working with bminer now.  I have tried them ALL.  Since I cannot get a response from anyone at Bminer, I will switch to another miner.  Its a shame really.   Wish everyone would just use the latest versions of drivers when offering something new to the public.  But, noooo lets use drivers from 2013.  smh.   ...Oh yeah, those dont work anymore either.  Im done.  Enjoy yourselves.  Gminer just as good  and at the moment, its far better.

...tbh, Raven is more profitable right now anyway.  Probably because everyone that cant mine Grin now have switched.  Think I'll join the masses.

Same here, GTX1070:

[INFO] [2019-07-18T10:33:36+02:00] CUDA error DRIVER: '1' in func 'bminer::cuckoo::Cuckaroo29dSolverCudaBase::CollectEdges' line 463
[FATA] [2019-07-18T10:33:36+02:00] Fatal cuda error in GPU 0. Terminate soon...

I am having exactly the same issue on an NVidia 1080Ti:

[INFO] [2019-07-18T10:19:23+03:00] Bminer: When Crypto-mining Made Fast (v15.7.1-1c421a5)
[INFO] [2019-07-18T10:19:23+03:00] Detecting platform
[INFO] [2019-07-18T10:19:28+03:00] Watchdog has started.
[INFO] [2019-07-18T10:19:28+03:00] Starting miner on 2 devices
[INFO] [2019-07-18T10:19:28+03:00] Starting the management API at 127.0.0.1:1880
[INFO] [2019-07-18T10:19:28+03:00] Free memory 9196.05 MB available on device (#0). Selecting normal solver.
[INFO] [2019-07-18T10:19:28+03:00] Operating in the thread safe mode for GPU 0
[INFO] [2019-07-18T10:19:28+03:00] Starting miner on device 0...
[INFO] [2019-07-18T10:19:28+03:00] Connected to [hidden pool]
[INFO] [2019-07-18T10:19:28+03:00] Authorized with the server
[INFO] [2019-07-18T10:19:28+03:00] Received new job 2
[INFO] [2019-07-18T10:19:29+03:00] Started miner on device 0
[INFO] [2019-07-18T10:19:29+03:00] Free memory 9326.57 MB available on device (#1). Selecting normal solver.
[INFO] [2019-07-18T10:19:29+03:00] Operating in the thread safe mode for GPU 1
[INFO] [2019-07-18T10:19:29+03:00] Starting miner on device 1...
[INFO] [2019-07-18T10:19:29+03:00] Started miner on device 1
[INFO] [2019-07-18T10:19:29+03:00] CUDA error DRIVER: '1' in func 'bminer::cuckoo::Cuckaroo29dSolverCudaBase::CollectEdges' line 463
[FATA] [2019-07-18T10:19:29+03:00] Fatal cuda error in GPU 0. Terminate soon...
[WARN] [2019-07-18T10:19:31+03:00] Miner died! It will be restarted soon...

Windows 10 x64, NVidia drivers 431.36 (26.21.14.3136 from 03-Jul-19).

Compare bminer and gminer in spakrpool. Diveces and settings equal. (5x1070 + 1x1070 ti + 2x2070)
https://drive.google.com/open?id=1bI71kVpAnTHvrM2TtgN6civGDy1aJEgo

And intresting graphics with rejected shares
https://drive.google.com/open?id=1r6WduYGiQSl-FKMuPxOL3yAfjoLhQHxr

Having same issues as well. Been using Bminer for months already and did not have any issues. For the hardfork i downloaded the latest version which was 15.7.0 at that time. This version gives a slightly different error message: CUDA error DRIVER: '500' in func 'bminer::cuckaroo29dSolverCudaBase::InitializeCudaEnvironment' line 81.

So i downloaded the one posted today which is 15.7.1 and it doesnt give any errormessages at all it just says: Miner Died! It will be restarted soon. Which keeps repeating.

Running on windows 10 with CUDA9 and 1x GTX1080-FE.

Some guidance would be appreciated. Thanks!

I too keep getting the same error as daliboard


miner is 3x ASUS Strix GTX 1070 8GB with CUDA 10.1 and Drivers 431.36 on Windows 10
bminer is 15.7.1

is there anything I can do to fix this? thanks @realbminer

I keep getting rejected shares and then it restarts and does the same. I don't know what I'm doing wrong. I have a gtx 1060 mining to f2pool. All was good till the update. Did the update but still getting rejected shares. F2pool says Bminer only works for c31 for now is that right?

[INFO] [2019-07-17T19:34:14+02:00] Bminer: When Crypto-mining Made Fast (v15.7.1-1c421a5)
[INFO] [2019-07-17T19:34:14+02:00] Detecting platform
[INFO] [2019-07-17T19:34:22+02:00] Watchdog has started.
[INFO] [2019-07-17T19:34:22+02:00] Starting miner on 1 devices
[INFO] [2019-07-17T19:34:22+02:00] Starting the management API at 127.0.0.1:1880
[INFO] [2019-07-17T19:34:22+02:00] Free memory 6721.42 MB available on device (#0). Selecting normal solver.
[INFO] [2019-07-17T19:34:22+02:00] Operating in the thread safe mode for GPU 0
[INFO] [2019-07-17T19:34:22+02:00] Starting miner on device 0...
[INFO] [2019-07-17T19:34:23+02:00] Started miner on device 0
[INFO] [2019-07-17T19:34:23+02:00] CUDA error DRIVER: '1' in func 'bminer::cuckoo::Cuckaroo29dSolverCudaBase::CollectEdges' line 463
[FATA] [2019-07-17T19:34:23+02:00] Fatal cuda error in GPU 0. Terminate soon...

[WARN] [2019-07-17T19:34:23+02:00] Miner died! It will be restarted soon...

Gtx 1070, win 10, drivers 431.36.

Here are some reality tests:
MSI Armor GTX 1070 - core +100, mem +900(+450 windows), PL: 130W
bminer: Hash - 5.52 G/s, Fidelity - 1.973
nbminer: Hash - 1: 5.23 g/s , Fidelity - 1: 2.356
gminer: Hash - GPU1 5.21 G/s , Fidelity - GPU1 1.773

I will repeat again, bminer is tolerant to memory optimizations while other miners seem more compute bound.

We're pleased to release Bminer 15.7.1.
The release provides:

* Improve the performance of the Cuckaroo29d algorithm.
* Fix the issue that the Cuckaroo29d failed to start for Turing-based cards.

Please see https://www.bminer.me/releases/ for more details.
Happy mining.

Your test is incompatible with reality, but in reality the bminer loses to the gminer and nbminer

Very strange settings. I have on my 1070 5(+-0.1)gps in sparkpool. MSI AB - 75PL +50gpu +200mem and i have cuda force p2 state = off (nvidia profile inspector)

Compare bminer vs gminer vs goldminer use msi gaming 2070 (TU106-400-A1) (P2 = off, 100PL +50gpu +0mem (default 7000/14000 mhz):
bminer: 6.4-6.5 gps
goldminer: 5.3gps
gminer: 7.3gps

Its used for encoder, not for miner. Proves that bminer is more memory intensive than others. In most cases the Pill can boost it more than with others.

downgrade 1080ti to 1070ti? nonsense

which freqs do you have in stock ? and why so low PL with default 280W ?

I compared all 4 miners currently available for cuckaroo29d using single 1080ti(while encoding at the same time).
gfxoffset: +125, memoffset: + 1000(windows +500), pwr limit: 160W
bminer: 5.44(avg)-5.78(peak)
gminer: 5.25(avg)-5.39(peak)
nbminer: 5.22(avg)-5.46(peak)
lolminer(opencl only): 4.23(avg)-4.24(peak)

Wow! Perfect speed, and "(use --algo grin29 for autoswitch after hardfork)". With telegram channel for update notification!!! Great!
It is a pity that I did not switch to it before.
Thank you so much!

Use gminer, it gives the maximum hashrate

Correct this: cuckaroo29d://
But doesn't work!!! WTF?!

Goldminer is slow but work with new grin 29d. Use this miner before bminer will fix error

Which one is correct?

• cuckaroo29://
• cuckaroo29d://
• cuckarood29://