[ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX] - page 60.

shifty252

full member

Activity: 177

Merit: 101

Quote from: djm34 on October 17, 2014, 11:17:22 AM

Quote from: shifty252 on October 17, 2014, 11:15:16 AM

Quote from: djm34 on October 16, 2014, 06:58:50 AM

Quote from: PVmining on October 16, 2014, 05:42:59 AM

spreadcoin
https://bitcointalksearch.org/topic/ann-spreadcoin-true-decentralization-no-pools-testing-new-masternodes-715435

algo: spread x11
https://github.com/spreadcoin-project/spreadcoin

1000 SPR bounty for #3 - GPU Miner
https://bitcointalksearch.org/topic/m.8953171

hmm, I don't see the difference with standard x11... did they just changed the order ? (if so you don't need to be a master gpu dev to do it yourself and cashout the 1000 shitcoins)

never heard of this coin... the thread dates back from july Shocked

edit: actually you would just need to fix cpu-miner.c ...
(can't believe nobody did that since then... they really failed in getting attention...

So? anyone willing to take the bounty?

I have other duties at the moment... and how much is it 1000SPR in BTC ?

about 0.2 BTC. 1000 is just from me, another 1000 is offered by another member

djm34

legendary

Activity: 1400

Merit: 1050

Quote from: shifty252 on October 17, 2014, 11:15:16 AM

Quote from: djm34 on October 16, 2014, 06:58:50 AM

Quote from: PVmining on October 16, 2014, 05:42:59 AM

spreadcoin
https://bitcointalksearch.org/topic/ann-spreadcoin-true-decentralization-no-pools-testing-new-masternodes-715435

algo: spread x11
https://github.com/spreadcoin-project/spreadcoin

1000 SPR bounty for #3 - GPU Miner
https://bitcointalksearch.org/topic/m.8953171

hmm, I don't see the difference with standard x11... did they just changed the order ? (if so you don't need to be a master gpu dev to do it yourself and cashout the 1000 shitcoins)

never heard of this coin... the thread dates back from july Shocked

edit: actually you would just need to fix cpu-miner.c ...
(can't believe nobody did that since then... they really failed in getting attention...

So? anyone willing to take the bounty?

I have other duties at the moment... and how much is it 1000SPR in BTC ?

(huge 2spr volume on allcrypt, and 0.03964892 on betasharex (the new exchange with the name of a scam Grin

)

shifty252

full member

Activity: 177

Merit: 101

Quote from: djm34 on October 16, 2014, 06:58:50 AM

Quote from: PVmining on October 16, 2014, 05:42:59 AM

spreadcoin
https://bitcointalksearch.org/topic/ann-spreadcoin-true-decentralization-no-pools-testing-new-masternodes-715435

algo: spread x11
https://github.com/spreadcoin-project/spreadcoin

1000 SPR bounty for #3 - GPU Miner
https://bitcointalksearch.org/topic/m.8953171

hmm, I don't see the difference with standard x11... did they just changed the order ? (if so you don't need to be a master gpu dev to do it yourself and cashout the 1000 shitcoins)

never heard of this coin... the thread dates back from july Shocked

edit: actually you would just need to fix cpu-miner.c ...
(can't believe nobody did that since then... they really failed in getting attention...

So? anyone willing to take the bounty?

AizenSou

hero member

Activity: 938

Merit: 1000

Quote from: ltcnim on October 17, 2014, 05:19:27 AM

Quote from: AizenSou on October 17, 2014, 04:42:35 AM

Quote from: ltcnim on October 16, 2014, 08:06:19 AM

well, i just put the card in the last free slot, but it happened again Sad

Code:

[ 1368.218337] NVRM: GPU at 0000:02:00.0 has fallen off the bus.

Is it possible that it has something to do with the latest nvidia drivers? I updated them twice in the last weeks/month but i'm not sure if it's connected to that.
Also, dmesg tells me the following:

Code:

[   41.839502] NVRM: Your system is not currently configured to drive a VGA console
[   41.839506] NVRM: on the primary VGA device. The NVIDIA Linux graphics driver
[   41.839507] NVRM: requires the use of a text-mode VGA console. Use of other console
[   41.839508] NVRM: drivers including, but not limited to, vesafb, may result in
[   41.839509] NVRM: corruption and stability problems, and is not supported.

I'm running ubuntu 14.04LTS server. beside the problem with the primary card, the system is running very stable.

edit: trying to downgrade the driver now...

Your GPU might be dead. Exactly what happens with my GTX780 last Tuesday. After trying so many attempts to fix, I just gave up and sent it to RMA. Let's see what they say then.

I'm afraid that might be true. However, I downgraded the driver yesterday and also set the primary display adapter to the onboard GPU. That way the error message from dmesg is gone, and the system is now hashing since 20h without a failure. But if the card fails again, I will RMA it.

Yes a best way to RMA is keep pushing the card to death and then send to RMA LOL.

ltcnim

legendary

Activity: 914

Merit: 1001

Quote from: AizenSou on October 17, 2014, 04:42:35 AM

Quote from: ltcnim on October 16, 2014, 08:06:19 AM

well, i just put the card in the last free slot, but it happened again Sad

Code:

[ 1368.218337] NVRM: GPU at 0000:02:00.0 has fallen off the bus.

Is it possible that it has something to do with the latest nvidia drivers? I updated them twice in the last weeks/month but i'm not sure if it's connected to that.
Also, dmesg tells me the following:

Code:

[   41.839502] NVRM: Your system is not currently configured to drive a VGA console
[   41.839506] NVRM: on the primary VGA device. The NVIDIA Linux graphics driver
[   41.839507] NVRM: requires the use of a text-mode VGA console. Use of other console
[   41.839508] NVRM: drivers including, but not limited to, vesafb, may result in
[   41.839509] NVRM: corruption and stability problems, and is not supported.

I'm running ubuntu 14.04LTS server. beside the problem with the primary card, the system is running very stable.

edit: trying to downgrade the driver now...

Your GPU might be dead. Exactly what happens with my GTX780 last Tuesday. After trying so many attempts to fix, I just gave up and sent it to RMA. Let's see what they say then.

I'm afraid that might be true. However, I downgraded the driver yesterday and also set the primary display adapter to the onboard GPU. That way the error message from dmesg is gone, and the system is now hashing since 20h without a failure. But if the card fails again, I will RMA it.

EbbiDos

newbie

Activity: 20

Merit: 0

Quote from: bathrobehero on October 16, 2014, 05:29:53 PM

Try K8x24 or K8x32 (cudaminer.exe -H 2 -d 0 -i 0 -C 2 -m 1 -l K8x24) and change -i 0 to 1 if you don't want your mouse to freeze.

That gives round about 278kH/s but there's only one GPU working. Using "k" instead of "K" gives 210kH/s.
Removing "-d 0" let me run both GPU with a total of 544kH/s. Working on both memory segments by removing "-m 1" gives 535kH/s. The same value as I run two seperate cudaminer processes with different "-d" parameter.
"K8x32" resulted in 1.5kH/s per GPU.

I'll test these settings at my Quad-SLI (2x GTX690) rig later.

AizenSou

hero member

Activity: 938

Merit: 1000

Quote from: ltcnim on October 16, 2014, 08:06:19 AM

well, i just put the card in the last free slot, but it happened again Sad

Code:

[ 1368.218337] NVRM: GPU at 0000:02:00.0 has fallen off the bus.

Is it possible that it has something to do with the latest nvidia drivers? I updated them twice in the last weeks/month but i'm not sure if it's connected to that.
Also, dmesg tells me the following:

Code:

[   41.839502] NVRM: Your system is not currently configured to drive a VGA console
[   41.839506] NVRM: on the primary VGA device. The NVIDIA Linux graphics driver
[   41.839507] NVRM: requires the use of a text-mode VGA console. Use of other console
[   41.839508] NVRM: drivers including, but not limited to, vesafb, may result in
[   41.839509] NVRM: corruption and stability problems, and is not supported.

I'm running ubuntu 14.04LTS server. beside the problem with the primary card, the system is running very stable.

edit: trying to downgrade the driver now...

Your GPU might be dead. Exactly what happens with my GTX780 last Tuesday. After trying so many attempts to fix, I just gave up and sent it to RMA. Let's see what they say then.

djm34

legendary

Activity: 1400

Merit: 1050

Quote from: EbbiDos on October 16, 2014, 05:20:03 PM

Quote from: djm34 on October 16, 2014, 04:20:41 PM

autotune is your friend

No it isn't. I just tried autotune again. Big mistake. Makes my computer unaccessible for minutes. Mouse and keyboard are freezed. I don't know what the autotune is trying but it doesn't work at all => in ten minutes not a single hash!!! After ten minutes I'm able to get the application closed.

try to autotune only one core at a time... (may-be using an older version of cudaminer might help, in the latest version autotune was a bit capricious although not with scrypt.. may-be "k" kernel rather than "K" kernel.

bathrobehero

legendary

Activity: 2002

Merit: 1051

ICO? Not even once.

Quote from: EbbiDos on October 16, 2014, 05:20:03 PM

Quote from: djm34 on October 16, 2014, 04:20:41 PM

autotune is your friend

No it isn't. I just tried autotune again. Big mistake. Makes my computer unaccessible for minutes. Mouse and keyboard are freezed. I don't know what the autotune is trying but it doesn't work at all => in ten minutes not a single hash!!! After ten minutes I'm able to get the application closed.

Try K8x24 or K8x32 (cudaminer.exe -H 2 -d 0 -i 0 -C 2 -m 1 -l K8x24) and change -i 0 to 1 if you don't want your mouse to freeze.

EbbiDos

newbie

Activity: 20

Merit: 0

Quote from: djm34 on October 16, 2014, 04:20:41 PM

autotune is your friend

No it isn't. I just tried autotune again. Big mistake. Makes my computer unaccessible for minutes. Mouse and keyboard are freezed. I don't know what the autotune is trying but it doesn't work at all => in ten minutes not a single hash!!! After ten minutes I'm able to get the application closed.

djm34

legendary

Activity: 1400

Merit: 1050

Quote from: EbbiDos on October 16, 2014, 03:24:36 PM

I have a terrible hashrate with my GTX690. Just 15kH/s per GPU. That's way to low!
My config for cudaminer is:

cudaminer.exe -o stratum+tcp://pool.dogepool.net:4444 -u xxxxxxx.worker1 -p xxx -d 0,1 -i 0,0 -l K32x8,K32x8

I have found this config on another forum.
By just calling cudaminer with pool parameters (cudaminer.exe -o stratum+tcp://pool.dogepool.net:4444 -u xxxxxxx.worker1 -p xxx) I get 4kH/s per GPU.
I also found this setting: cudaminer.exe -o stratum+tcp://pool.dogepool.net:4444 -u xxxxxxx.worker1 -p xxx -l K8x24 -i 0 -C 2 -m 1 -H 2
but I get only 10kH/s with this.

I use cudaMiner 2014-02-28, nVidia 340.52, Win 7x64 SP1. I have another rig with 2x GTX690. Some problem there, even worse! With auto-config I get 0.8kH/s. On that machine I run Win 8.1, nVidia 344.11.

Can anyone help? Thanks.

autotune is your friend
(for some strange reason, each time I tried setting found elsewhere for the R9-290x, it always ended by a crash... (never needed that for cudaminer because of the autotune).

EbbiDos

newbie

Activity: 20

Merit: 0

I have a terrible hashrate with my GTX690. Just 15kH/s per GPU. That's way to low!
My config for cudaminer is:

cudaminer.exe -o stratum+tcp://pool.dogepool.net:4444 -u xxxxxxx.worker1 -p xxx -d 0,1 -i 0,0 -l K32x8,K32x8

I have found this config on another forum.
By just calling cudaminer with pool parameters (cudaminer.exe -o stratum+tcp://pool.dogepool.net:4444 -u xxxxxxx.worker1 -p xxx) I get 4kH/s per GPU.
I also found this setting: cudaminer.exe -o stratum+tcp://pool.dogepool.net:4444 -u xxxxxxx.worker1 -p xxx -l K8x24 -i 0 -C 2 -m 1 -H 2
but I get only 10kH/s with this.

I use cudaMiner 2014-02-28, nVidia 340.52, Win 7x64 SP1. I have another rig with 2x GTX690. Some problem there, even worse! With auto-config I get 0.8kH/s. On that machine I run Win 8.1, nVidia 344.11.

Can anyone help? Thanks.

ltcnim

legendary

Activity: 914

Merit: 1001

well, i just put the card in the last free slot, but it happened again Sad

Code:

[ 1368.218337] NVRM: GPU at 0000:02:00.0 has fallen off the bus.

Is it possible that it has something to do with the latest nvidia drivers? I updated them twice in the last weeks/month but i'm not sure if it's connected to that.
Also, dmesg tells me the following:

Code:

[   41.839502] NVRM: Your system is not currently configured to drive a VGA console
[   41.839506] NVRM: on the primary VGA device. The NVIDIA Linux graphics driver
[   41.839507] NVRM: requires the use of a text-mode VGA console. Use of other console
[   41.839508] NVRM: drivers including, but not limited to, vesafb, may result in
[   41.839509] NVRM: corruption and stability problems, and is not supported.

I'm running ubuntu 14.04LTS server. beside the problem with the primary card, the system is running very stable.

edit: trying to downgrade the driver now...

djm34

legendary

Activity: 1400

Merit: 1050

Quote from: california_jim on October 16, 2014, 07:20:25 AM

Quote from: ltcnim on October 15, 2014, 07:22:15 PM

hmm..I currently have a problem with one (?) of my 750Ti. nvidia-smi is reporting "ERROR: GPU is lost" but it looks like it's still hashing. This happened a few times over the last weeks, sometimes it works after a reboot for another week, sometimes only a few hours. I'm also not 100% sure if it's always the same card (I will keep track of that from now on).

Does that mean one of my cards is dying, or is there another problem?

edit: from dmesg:

Code:

[14922.167340] NVRM: GPU at 0000:01:00.0 has fallen off the bus.

Quote from: ltcnim on October 16, 2014, 06:07:18 AM

Quote from: california_jim on October 16, 2014, 05:57:09 AM

Quote from: djm34 on October 15, 2014, 08:55:43 PM

Quote from: ltcnim on October 15, 2014, 07:22:15 PM

Code:

[14922.167340] NVRM: GPU at 0000:01:00.0 has fallen off the bus.

Double

Hopefully that error was coded intentionally, if not just a very funny accident.

yeah it sounds funny, but it's extremely annoying since nvidia-smi can't get any info from that card anymore, but it seems it is still working.
i will try to change the pcie slot for that card.

Yes, sorry, that wasn't fair to laugh at the message and not offer some help. I'm absolutely no expert but I'd try the usual process of elimination of swapping cards, risers, etc. about or out. What distro are you using? I've been using KopiemTu - https://litecointalk.org/index.php?topic=16800.0 - which is just magic, especially for me as I just want my stuff to work with minimal fiddling.

Before ltcnim gets back on the wagon because one of his card fell off the bus, here what I think Grin

Considering that the incident you are referring happens less than once in a week, I don't think you should really worry...

california_jim

member

Activity: 85

Merit: 10

Quote from: ltcnim on October 15, 2014, 07:22:15 PM

hmm..I currently have a problem with one (?) of my 750Ti. nvidia-smi is reporting "ERROR: GPU is lost" but it looks like it's still hashing. This happened a few times over the last weeks, sometimes it works after a reboot for another week, sometimes only a few hours. I'm also not 100% sure if it's always the same card (I will keep track of that from now on).

Does that mean one of my cards is dying, or is there another problem?

edit: from dmesg:

Code:

[14922.167340] NVRM: GPU at 0000:01:00.0 has fallen off the bus.

Quote from: ltcnim on October 16, 2014, 06:07:18 AM

Quote from: california_jim on October 16, 2014, 05:57:09 AM

Quote from: djm34 on October 15, 2014, 08:55:43 PM

Quote from: ltcnim on October 15, 2014, 07:22:15 PM

Code:

[14922.167340] NVRM: GPU at 0000:01:00.0 has fallen off the bus.

Double

Hopefully that error was coded intentionally, if not just a very funny accident.

yeah it sounds funny, but it's extremely annoying since nvidia-smi can't get any info from that card anymore, but it seems it is still working.
i will try to change the pcie slot for that card.

Yes, sorry, that wasn't fair to laugh at the message and not offer some help. I'm absolutely no expert but I'd try the usual process of elimination of swapping cards, risers, etc. about or out. What distro are you using? I've been using KopiemTu - https://litecointalk.org/index.php?topic=16800.0 - which is just magic, especially for me as I just want my stuff to work with minimal fiddling.

djm34

legendary

Activity: 1400

Merit: 1050

Quote from: PVmining on October 16, 2014, 05:42:59 AM

spreadcoin
https://bitcointalksearch.org/topic/ann-spreadcoin-true-decentralization-no-pools-testing-new-masternodes-715435

algo: spread x11
https://github.com/spreadcoin-project/spreadcoin

1000 SPR bounty for #3 - GPU Miner
https://bitcointalksearch.org/topic/m.8953171

hmm, I don't see the difference with standard x11... did they just changed the order ? (if so you don't need to be a master gpu dev to do it yourself and cashout the 1000 shitcoins)

never heard of this coin... the thread dates back from july Shocked

edit: actually you would just need to fix cpu-miner.c ...
(can't believe nobody did that since then... they really failed in getting attention...

ltcnim

legendary

Activity: 914

Merit: 1001

Quote from: california_jim on October 16, 2014, 05:57:09 AM

Quote from: djm34 on October 15, 2014, 08:55:43 PM

Quote from: ltcnim on October 15, 2014, 07:22:15 PM

Code:

[14922.167340] NVRM: GPU at 0000:01:00.0 has fallen off the bus.

Double

Hopefully that error was coded intentionally, if not just a very funny accident.

yeah it sounds funny, but it's extremely annoying since nvidia-smi can't get any info from that card anymore, but it seems it is still working.
i will try to change the pcie slot for that card.

california_jim

member

Activity: 85

Merit: 10

Quote from: djm34 on October 15, 2014, 08:55:43 PM

Quote from: ltcnim on October 15, 2014, 07:22:15 PM

Code:

[14922.167340] NVRM: GPU at 0000:01:00.0 has fallen off the bus.

Double

Hopefully that error was coded intentionally, if not just a very funny accident.

PVmining

sr. member

Activity: 330

Merit: 252

spreadcoin
https://bitcointalksearch.org/topic/ann-spreadcoin-true-decentralization-no-pools-testing-new-masternodes-715435

algo: spread x11
https://github.com/spreadcoin-project/spreadcoin

1000 SPR bounty for #3 - GPU Miner
https://bitcointalksearch.org/topic/m.8953171

sp_

legendary

Activity: 2926

Merit: 1087

Team Black developer

Performance increase is expected in x11,x13,x15,x14,x17,nist. I will work som more on it and send out exefiles(windows executables) on Saturday. If you have compute 5.0 card

Compute 5.0 or bether:

GeForce GTX 980
GeForce GTX 970
GeForce GTX 750 Ti
GeForce GTX 750
GeForce GTX 850M
GeForce 840M
GeForce 830M
GeForce GTX 980M
GeForce GTX 970M

Pm me with your email adress for more info

Topic: [ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX] - page 60. (Read 3426944 times)