EWBF's Cuda Equihash Miner 0.6 - page 33.

gsanjin

member

Activity: 191

Merit: 12

Quote from: xenonwon on July 05, 2018, 05:04:17 AM

suprnova.cc
20000 virtual
v0.3
miner --algo 144_5 --pers BgoldPoW --server btg.suprnova.cc --port 8866 --user user.user --pass x --pec --log 2
pause

I have tried to remove all the OC and it works stable for more than 15 hours, so I guess it was too much OC, but I do not think I had much OC.

Equihash 144,5 is demanding much more GPU memory than "old" equihash so mining it taxes more your gpu's and what was stable before, now maybe isn't. Maybe your mem was overclocked a little to high for "new" algo, so I would try with lower mem overclocks on your place.

xenonwon

newbie

Activity: 42

Merit: 0

Quote from: inudemon on July 04, 2018, 05:28:08 PM

Quote from: xenonwon on July 04, 2018, 04:14:11 PM

I have this problem all time:

04.07.2018 13:45:52 CUDA: Device: 1 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:52 CUDA: Device: 2 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:53 CUDA: Device: 3 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:53 CUDA: Device: 4 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:53 CUDA: Device: 5 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:53 WARNING: Looks like GPU0 are stopped. Restarting...
04.07.2018 13:45:53 INFO: GPU0 are restarted.
04.07.2018 13:45:53 WARNING: Looks like GPU1 are stopped. Restarting...
04.07.2018 13:45:53 INFO: GPU1 are restarted.
04.07.2018 13:45:54 WARNING: Looks like GPU2 are stopped. Restarting...
04.07.2018 13:45:54 INFO: GPU2 are restarted.
04.07.2018 13:45:54 WARNING: Looks like GPU3 are stopped. Restarting...
04.07.2018 13:45:54 INFO: GPU3 are restarted.
04.07.2018 13:45:54 WARNING: Looks like GPU4 are stopped. Restarting...
04.07.2018 13:45:54 INFO: GPU4 are restarted.
04.07.2018 13:45:54 CUDA: Device: 0 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:54 WARNING: Looks like GPU5 are stopped. Restarting...
04.07.2018 13:45:54 INFO: GPU5 are restarted.

I have 6x 1070 ti, the OC is ok i think, 65 TDP, 180 core, 550 memory, windows 10.
Thanks for helps.

Not enough info:
pool?
size of virtual memory?
v0.2 or v.03 miner?
bat file settings?

suprnova.cc
20000 virtual
v0.3
miner --algo 144_5 --pers BgoldPoW --server btg.suprnova.cc --port 8866 --user user.user --pass x --pec --log 2
pause

I have tried to remove all the OC and it works stable for more than 15 hours, so I guess it was too much OC, but I do not think I had much OC.

shmocs

newbie

Activity: 28

Merit: 0

split them up. Or, launch 2 ewbf miners (or more)

TAKCuCT

newbie

Activity: 25

Merit: 0

Still no fixes for rigs more than 12 cards.

shmocs

newbie

Activity: 28

Merit: 0

guys.. it says 144_5 takes now 1.99 but 1050 2GB cards are still not working Sad

Quote

tail: /var/run/miner.output: file truncated
+-------------------------------------------------+
| EWBF's Eqihash CUDA miner. v0.3 |
+-------------------------------------------------+
02:53:27 INFO: Used config file: /var/run/ethos/ewbf-zcash.conf
02:53:27 INFO: Current pool: eu.voidr.net:3032
02:53:27 INFO: Selected pools: 1
02:53:27 INFO: Algorithm: 144_5 "sngemPoW"
02:53:27 INFO: Devices: User defined.
02:53:27 INFO: Temperature limit: 85
02:53:27 INFO: Api: Listen on 127.0.0.1:42002
---------------------------------------------------
02:53:28 INFO: Target: 00a0000000000000...
02:53:28 INFO: Detected new work: e35d
02:53:29 CUDA Device: 0 GeForce GTX 1050, not enough memory, Available: 2012 MB, Required: 2031 MB
02:53:29 CUDA: Device: 1 GeForce GTX 1050, 1999 MB i:64
02:53:29 CUDA: Device: 2 GeForce GTX 1050, 1999 MB i:64
02:53:30 CUDA: Device: 3 GeForce GTX 1050, 1999 MB i:64
02:53:31 CUDA: Device: 4 GeForce GTX 1050, 1999 MB i:64
02:53:32 CUDA: Device: 1 Thread exited with message: "out of memory"
02:53:33 CUDA: Device: 5 GeForce GTX 1050, 1999 MB i:64
02:53:33 CUDA: Device: 2 Thread exited with message: "out of memory"
02:53:33 CUDA: Device: 3 Thread exited with message: "out of memory"
02:53:33 CUDA: Device: 4 Thread exited with message: "out of memory"
02:53:34 CUDA: Device: 5 Thread exited with message: "out of memory"
02:53:38 INFO: Detected new work: e35e
02:53:40 INFO: Detected new work: e35f
02:53:40 WARNING: Looks like GPU1 are stopped. Restarting...
02:53:40 INFO: GPU1 are restarted.
02:53:40 WARNING: Looks like GPU2 are stopped. Restarting...
02:53:40 INFO: GPU2 are restarted.
02:53:40 WARNING: Looks like GPU3 are stopped. Restarting...
02:53:40 INFO: GPU3 are restarted.
02:53:41 WARNING: Looks like GPU4 are stopped. Restarting...
02:53:41 INFO: GPU4 are restarted.
02:53:41 WARNING: Looks like GPU5 are stopped. Restarting...
02:53:41 INFO: GPU5 are restarted.
02:53:43 CUDA: Device: 1 Thread exited with message: "out of memory"
02:53:43 CUDA: Device: 2 Thread exited with message: "out of memory"
02:53:43 CUDA: Device: 3 Thread exited with message: "out of memory"
02:53:44 CUDA: Device: 4 Thread exited with message: "out of memory"
02:53:44 CUDA: Device: 5 Thread exited with message: "out of memory"

McMCrypto

newbie

Activity: 82

Merit: 0

Hi guys i am getting 20-22 sol/s , with my 1060 6GB (x6), Core +70 / Mem -350 , mining BTG (144_5 --pers BgoldPoW)

That Core/Mem setup is suitable ? thanks !

https://i.imgur.com/PYIyipW.png

daz59

newbie

Activity: 3

Merit: 0

Quote from: exquisitess on July 04, 2018, 06:10:48 PM

Using SMOS here with latest kernel 4.17

hardware:
motherboard: asrock h110 pro btc (13gpu)
memory : 8gb
cpu g4560
video cards: 12 gtx 1060 palit super jetstream

issue:

gpu gpu 11 and gpu 12 always gets stuck and stops working during gpu detection. Is there a way where you can build the miner to detect all gpus at the very start? similar to how dtsm does it? I noticed getting miners detected one by one i s very slow process and it could be that there is a specific timeout in your program that causes the gpu 11-12 to get stuck in my case.

restarting the miner will cause this issue to be intermittent. sometimes only gpu 12 gets stuck, sometimes both gpu 11 and 12 gets stuck.

please help resolve this. thanks and more power to you ewbf!

I have this same issue, running 13 cards, the last two get stuck and wont run, sometimes the 3rd from last also gets stuck and it is very very slow to detect all cards, 10+minutes.

CarlOrff

sr. member

Activity: 826

Merit: 440

@EWBF
Thank you for the v0.3 release and to maintain a Cuda 8 version (Linux).
My hashrate improvement with the 0.3 release is more than 10% :-).
(Just little tested with GTX1050Ti/Windows10/Cuda8 for the moment : will test with 1070, 1080Ti and 960/2GB next week-end).

exquisitess

newbie

Activity: 1

Merit: 0

Using SMOS here with latest kernel 4.17

hardware:
motherboard: asrock h110 pro btc (13gpu)
memory : 8gb
cpu g4560
video cards: 12 gtx 1060 palit super jetstream

issue:

gpu gpu 11 and gpu 12 always gets stuck and stops working during gpu detection. Is there a way where you can build the miner to detect all gpus at the very start? similar to how dtsm does it? I noticed getting miners detected one by one i s very slow process and it could be that there is a specific timeout in your program that causes the gpu 11-12 to get stuck in my case.

restarting the miner will cause this issue to be intermittent. sometimes only gpu 12 gets stuck, sometimes both gpu 11 and 12 gets stuck.

please help resolve this. thanks and more power to you ewbf!

inudemon

newbie

Activity: 50

Merit: 0

Quote from: xenonwon on July 04, 2018, 04:14:11 PM

I have this problem all time:

04.07.2018 13:45:52 CUDA: Device: 1 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:52 CUDA: Device: 2 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:53 CUDA: Device: 3 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:53 CUDA: Device: 4 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:53 CUDA: Device: 5 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:53 WARNING: Looks like GPU0 are stopped. Restarting...
04.07.2018 13:45:53 INFO: GPU0 are restarted.
04.07.2018 13:45:53 WARNING: Looks like GPU1 are stopped. Restarting...
04.07.2018 13:45:53 INFO: GPU1 are restarted.
04.07.2018 13:45:54 WARNING: Looks like GPU2 are stopped. Restarting...
04.07.2018 13:45:54 INFO: GPU2 are restarted.
04.07.2018 13:45:54 WARNING: Looks like GPU3 are stopped. Restarting...
04.07.2018 13:45:54 INFO: GPU3 are restarted.
04.07.2018 13:45:54 WARNING: Looks like GPU4 are stopped. Restarting...
04.07.2018 13:45:54 INFO: GPU4 are restarted.
04.07.2018 13:45:54 CUDA: Device: 0 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:54 WARNING: Looks like GPU5 are stopped. Restarting...
04.07.2018 13:45:54 INFO: GPU5 are restarted.

I have 6x 1070 ti, the OC is ok i think, 65 TDP, 180 core, 550 memory, windows 10.
Thanks for helps.

Not enough info:
pool?
size of virtual memory?
v0.2 or v.03 miner?
bat file settings?

xenonwon

newbie

Activity: 42

Merit: 0

I have this problem all time:

04.07.2018 13:45:52 CUDA: Device: 1 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:52 CUDA: Device: 2 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:53 CUDA: Device: 3 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:53 CUDA: Device: 4 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:53 CUDA: Device: 5 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:53 WARNING: Looks like GPU0 are stopped. Restarting...
04.07.2018 13:45:53 INFO: GPU0 are restarted.
04.07.2018 13:45:53 WARNING: Looks like GPU1 are stopped. Restarting...
04.07.2018 13:45:53 INFO: GPU1 are restarted.
04.07.2018 13:45:54 WARNING: Looks like GPU2 are stopped. Restarting...
04.07.2018 13:45:54 INFO: GPU2 are restarted.
04.07.2018 13:45:54 WARNING: Looks like GPU3 are stopped. Restarting...
04.07.2018 13:45:54 INFO: GPU3 are restarted.
04.07.2018 13:45:54 WARNING: Looks like GPU4 are stopped. Restarting...
04.07.2018 13:45:54 INFO: GPU4 are restarted.
04.07.2018 13:45:54 CUDA: Device: 0 Thread exited with message: "all CUDA-capable devices are busy or unavailable"
04.07.2018 13:45:54 WARNING: Looks like GPU5 are stopped. Restarting...
04.07.2018 13:45:54 INFO: GPU5 are restarted.

I have 6x 1070 ti, the OC is ok i think, 65 TDP, 180 core, 550 memory, windows 10.
Thanks for helps.

artful

newbie

Activity: 59

Merit: 0

very slow start. 1 min 30 second start on 10 GPU

http://owncloud.su/index.php/s/dCrU96ROEwQpvzL

and very big use system memory

http://owncloud.su/index.php/s/gdMo4nTspvddXyw
7 copies miner used 8Gb RAM. other miner (zm/ccminer/t-rex/bminer/claymore) 10 copies work fine on 4Gb of memory

sir_blacks

jr. member

Activity: 119

Merit: 3

Hello @EWBF_

Spec:
GPU 0: Gigabyte Windforce 2 NVIDIA GeForce GTX 1060 3GB (3072 MB)
- GPU is together with other video card what i use for my own purpose
OS: Windows 7 x64 Service Pack 1
Video Driver: 397.31

INFO: Algorithm: 144_5
CUDA: Device: 0 GeForce GTX 1060 3GB, 3072 MB i:64
ERROR: out of memory

GPU-Z Report it is 103 Mb ram use so to a simple math 3072 - 103 = 2969 MB there are left
According to https://forum.bitcoingold.org/t/our-new-equihash-equihash-btg/1512

- Quote: Equihash-BTG: "<144,5> parameters require a minimum of 700 MB to run and use about 2.5 GB to run efficiently"
So 2,5G or 2,9G 1060 with 3G Vram it suppose to work but according to EWBF it is not have inaff memory with is ?!?!?! Huh

cheers

Kr4v3n5

newbie

Activity: 1

Merit: 0

Quote from: nmcuong on June 19, 2018, 02:02:45 AM

1050 - 2GB ram not run?? help

It does not work with the gtx 1050 but if with a gtx 950. fix problem

hogwash.89m

newbie

Activity: 32

Merit: 0

Who have problem with power spikes always can try other implementation, this one is build on OpenCL (so is usabe on Nvidia and AMD).
https://bitcointalksearch.org/topic/lolminer-144-038-preview-equihash-1445-for-amd-nvidia-linux-windows-4591317

But, beware this miner also can have problems, I don't test it still. (Used version for 96.5 for mining MNX).

On my rig 1080ti got more flat power usage line, but 1080 have little jaggines (+/-5%).
Observed im MSI Afterburner, also on power meter there are no fluctuations, power consumption is constants.

kedes

newbie

Activity: 34

Merit: 0

Still no any fixes for GTX 970 cards? Sad

eliteware

newbie

Activity: 1

Merit: 0

how many sols on 192.7 algo mate

preda

sr. member

Activity: 756

Merit: 250

so now it works with 1060 3gb cards in windows 10?
* 192_7 require now 2.5GB vram, it should fit in 3Gb gpu under win 10.

dug03

newbie

Activity: 6

Merit: 0

Quote from: crypper on July 04, 2018, 07:55:02 AM

Quote from: dug03 on July 04, 2018, 07:45:53 AM

Quote from: crypper on July 04, 2018, 06:43:10 AM

Quote from: abstractHaze on July 03, 2018, 04:23:35 PM

so in the image with red font. u can see difference in power between version .2 and .3 .. .3 is a tad bit better. but in all my pc's with different mobos. different video cards different PSU's, all have now jagged lines. why is this happening.... im not gonna use another miner. but does anyone have a fix to smooth these teeth out?

can we get pics of others power lines on these algo's cause my whole entire lot, just now has this issue. and i know it's either a tweak i need to do, or miner just needs time to mature

? any ideas.

again im working on BTG new fork 144_5

I'm also concerned about erratic power and GPU load with new EWBF on new equihash (BTG 144,5).
I tried both v0.2 and 0.3 and GTX 1070 at different clocks (including stocks) - the same result; as seen on the picture (right panel on the picture):
https://imgur.com/a/Uk0ziPs

Older EWBF (0.3.4b) with older equihash is fine with the same cards (left panel on the picture).

EWBF, please comment

Actually, the miner is working fine, it's this new friggin algo.... it effin' sucks. I did same, adjusted intensity, clocks, fan speed.... I don't see the advantage and my proof is jumping on to Suprnova mining ZEC with my 1060x6's and even with the asics I'm mining much more than BTG and @ 100% efficiency. Power with 144,5 jumped 69%/86% constantly and all it did was keep pushing the gpu temp up.
- I would like to hear what folks are really getting in shares for @ least 12hrs mining and if they think the algo change was really worth it. Maybe the ZEC DEVs were right,, it isn't worth and... and now I agree.

Well mining older equihash (ZCL, ZEC, etc) is not profitable at the moment (or let's say you are missing profits by not mining something else). If you have a look at MPH estimates of profitability: BTG is 2x more profitable to mine then ZCL, and 1.5x then ether. Going for ZEC at suprnova is a really bad decision. If I cannot make EWBF work with new equihash (and I'm not accepting this power spikes and GPU load for BTG mining), i'll just mine ether.

That's why I'm gonna wait for more honest input as to actual mined BTG shares on the new algo, MPH has been great for me but the profit estimates have been way off lately. As far as ZEC on Suprnova right now I have a console full of "green" for the last 2 hours compared to the 4/5 per screen on BTG... so just curious what a day will produce with the zec since it's been a while since I mined it, using ewbf 3.3b

crypper

member

Activity: 239

Merit: 12

Quote from: dug03 on July 04, 2018, 07:45:53 AM

Quote from: crypper on July 04, 2018, 06:43:10 AM

Quote from: abstractHaze on July 03, 2018, 04:23:35 PM

so in the image with red font. u can see difference in power between version .2 and .3 .. .3 is a tad bit better. but in all my pc's with different mobos. different video cards different PSU's, all have now jagged lines. why is this happening.... im not gonna use another miner. but does anyone have a fix to smooth these teeth out?

can we get pics of others power lines on these algo's cause my whole entire lot, just now has this issue. and i know it's either a tweak i need to do, or miner just needs time to mature

? any ideas.

again im working on BTG new fork 144_5

I'm also concerned about erratic power and GPU load with new EWBF on new equihash (BTG 144,5).
I tried both v0.2 and 0.3 and GTX 1070 at different clocks (including stocks) - the same result; as seen on the picture (right panel on the picture):
https://imgur.com/a/Uk0ziPs

Older EWBF (0.3.4b) with older equihash is fine with the same cards (left panel on the picture).

EWBF, please comment

Actually, the miner is working fine, it's this new friggin algo.... it effin' sucks. I did same, adjusted intensity, clocks, fan speed.... I don't see the advantage and my proof is jumping on to Suprnova mining ZEC with my 1060x6's and even with the asics I'm mining much more than BTG and @ 100% efficiency. Power with 144,5 jumped 69%/86% constantly and all it did was keep pushing the gpu temp up.
- I would like to hear what folks are really getting in shares for @ least 12hrs mining and if they think the algo change was really worth it. Maybe the ZEC DEVs were right,, it isn't worth and... and now I agree.

Well mining older equihash (ZCL, ZEC, etc) is not profitable at the moment (or let's say you are missing profits by not mining something else). If you have a look at MPH estimates of profitability: BTG is 2x more profitable to mine than ZCL, and 1.5x than ether. Going for ZEC at suprnova is a really bad decision. If I cannot make EWBF work with new equihash (and I'm not accepting this power spikes and GPU load for BTG mining), i'll just mine ether.
In terms of hashrate it does give ~ 34 sol/s per gtx1070 as advertised, but i'm not risking to leave the miner for 24 h with the kind of issues described above.

Topic: EWBF's Cuda Equihash Miner 0.6 - page 33. (Read 85776 times)