Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux) - page 534.

Call_Me_Bambi

sr. member

Activity: 574

Merit: 261

Quote from: unkn0wn0x on November 17, 2017, 12:11:24 PM

Since today, I am running into problems:

I start the miner, it sets the epoch for all 6 GPUs and then instantly crashes. Does anyone have similar issues? It was working for months now.
Also, I have removed all overclocks and undervolts from the config file, tried ETH only and dual mining. It crashes all the time and I have to manually reboot the rig. The log file always ends after the json line.

Any advice?

If it has been working for months without any problems, and you have not made any changes, then it could be a hardware fault.
I would first check that the power cables into the risers have not started to burn out (I've had a few do this).
The next thing would be the usual, test each card individually, then add the cards back one-at-a-time.

Victorio

full member

Activity: 281

Merit: 100

chanyewhau,
Do you understand that 4 or 5 of your GPUs work OK with a +300 MHz memclk increase, while the 5th or the 6th does not? That one crashes the entire mining.
Set all GPUs to stock settings and raise memclk one GPU at a time. Say 1-5 hold +300 MHz, 6 does not hold even +50 MHz. You set +300 MHz for 1 to 5 and 0 MHz for 6.

unkn0wn0x,
That's a banal idea - remove the drivers via DDU and perform a fresh driver install. Perhaps some driver component is corrupt.

unkn0wn0x

newbie

Activity: 13

Merit: 0

Since today, I am running into problems:

I start the miner, it sets the epoch for all 6 GPUs and then instantly crashes. Does anyone have similar issues? It was working for months now.
Also, I have removed all overclocks and undervolts from the config file, tried ETH only and dual mining. It crashes all the time and I have to manually reboot the rig. The log file always ends after the json line.

Any advice?

chanyewhau

newbie

Activity: 7

Merit: 0

Quote from: chanyewhau on November 17, 2017, 11:15:05 AM

Quote from: BoostOrDie on November 17, 2017, 11:06:01 AM

Quote from: chanyewhau on November 17, 2017, 10:44:11 AM

The memory size is 6144 MB, memory type is Hynix.
My current OC setting is Power Limit 75%, Core Clock +50, Memory Clock +300, Fan Speed 65%.
With the current OC setting, the cards are running at GPU Clock 1566 MHz, Memory 2152 Mhz. (Based on GPU-Z)
The default clock is 1506 Mhz, Memory 2002 Mhz.

The problematic fan that runs at 100% automatically has the lowest temperature even it is running at 65% like the other cards.
So i dont think the fan switch to 100% due to overheat.

I Got 4 1060 6gb Aorus with samsungs.

I was beating about 3 weeks with strange instabilty wit h my rig. It works OK then BAM! cpu gets loaded up to 80-90% and hash rate drops. Cards seemed ok.
I reinstalled win 7 and win 10 dozens of times. tryed dozens of drves versions and settings.

The problem was EASY - too much overclock!
If you are used to artifacts and freezes due to ooverclock - so in in mining it is strange instabilty. First sign is MSI AB or any system proces start to use 60-80% of CPU even if miner is stoped! Reboot and lower OC by 5 pints backwards. watch your rig again until it is stable. then do another -5 and this is you stable OC.

Second sign of GPU error - fan speed in Claymore i 0% i logs.

And dont OC the CORE. ETH needs only Memory OC. Core can be undeclocked for coolnes..

I got 50% power limit, -200 core, +765 Memory. I get 98.2mh by x4 GTX1060 rig power consumtion it 472wh!

You're so on point !
I started to wonder, how come everytime i launch my AF, even with my claymore closed, it takes forever to launch the AF.

Let me try your OC setting. I will come back with the updates.

Thanks alot !

Hello everyone, unfortunately my cards are pretty bad.

Whenever i try to downclock the core, the miner stuck at the 6 cards detected.

So my current setting is 0 core, +300 memory. reaching to about 20.5 MHz per card. Power consumption about 622 watt for 6 cards.

The miner still crash sometimes, but will auto restart.

Is this a good OC setting ?

MikeGR7

newbie

Activity: 36

Merit: 0

Quote from: oblione on November 16, 2017, 03:32:46 PM

Hi Guys

I'm wondering if you can help me with this.

my rig keeps rebooting shortly after the miner starts on this and I cant figure out why.

I'm using 4 x vegas
1200w PSU
latest blockchain drivers
32gb virtual mem
8gb ram

I've got the fans set manually it wattman to 3000rpm min
I've disabled all other over / underclocking and am just trying to run this as a baseline test before I start playing with the other settings

The miner starts up, runs for a few seconds and then the rig just resets and wattman resets everything.

The rig works brilliantly on cryptonote but I fancied giving some eth dual mining a try for a while.

Anything that would normally cause this type of behaviour?

May i ask your PSU model?

Do not take for granted that since cryptonote runs fine so will Eth cause different algorithms have different electrical behavior and requirements.

I suggest you try to mine only Etherium without second coin.
If it's stable, please measure your power consumption in the wall and tell us how many watts it takes.
If it's not stable even with single Eth, then run Cryptonote and report the power consumption in that case.

We can draw better conclusions with the above data.

Victorio

full member

Activity: 281

Merit: 100

Quote from: hiramofTyre on November 17, 2017, 11:22:48 AM

Can anyone help me with this? the error says: NVML: cannot get current temperature, error 999 (an internal driver error occurred)

Does the miner work 100% OK at stock settings?

hiramofTyre

full member

Activity: 308

Merit: 100

Hey guys I need help.. When I first ran my gpu miner at oc settings it was running fine, until there was a power outage. When the power came back I turned the miner on and then after one hour it crashes. I tried to turn it off then remove from the riser and put it back then it would stabilize. Then the power again, turned it on then it crashes again.

Can anyone help me with this? the error says: NVML: cannot get current temperature, error 999 (an internal driver error occurred)

Victorio

full member

Activity: 281

Merit: 100

Quote from: BoostOrDie on November 17, 2017, 11:06:01 AM

The problem was EASY - too much overclock!
If you are used to artifacts and freezes due to ooverclock - so in in mining it is strange instabilty. First sign is MSI AB or any system proces start to use 60-80% of CPU even if miner is stoped! Reboot and lower OC by 5 pints backwards. watch your rig again until it is stable. then do another -5 and this is you stable OC.

Second sign of GPU error - fan speed in Claymore i 0% i logs.

Straight to the point! I had by myself such problems at the beginning, having overclocked my GTX 1060 3GB Samsung to +800 Mem Clk and spent hours parsing the logs, drawing charts, until I said - screw that 1.5 MH/s difference, my tranquility costs more than 5$/month that I'm chasing.
Working at +700 MHz like a charm (2249 MHz).

chanyewhau

newbie

Activity: 7

Merit: 0

Quote from: BoostOrDie on November 17, 2017, 11:06:01 AM

Quote from: chanyewhau on November 17, 2017, 10:44:11 AM

The memory size is 6144 MB, memory type is Hynix.
My current OC setting is Power Limit 75%, Core Clock +50, Memory Clock +300, Fan Speed 65%.
With the current OC setting, the cards are running at GPU Clock 1566 MHz, Memory 2152 Mhz. (Based on GPU-Z)
The default clock is 1506 Mhz, Memory 2002 Mhz.

The problematic fan that runs at 100% automatically has the lowest temperature even it is running at 65% like the other cards.
So i dont think the fan switch to 100% due to overheat.

I Got 4 1060 6gb Aorus with samsungs.

I was beating about 3 weeks with strange instabilty wit h my rig. It works OK then BAM! cpu gets loaded up to 80-90% and hash rate drops. Cards seemed ok.
I reinstalled win 7 and win 10 dozens of times. tryed dozens of drves versions and settings.

The problem was EASY - too much overclock!
If you are used to artifacts and freezes due to ooverclock - so in in mining it is strange instabilty. First sign is MSI AB or any system proces start to use 60-80% of CPU even if miner is stoped! Reboot and lower OC by 5 pints backwards. watch your rig again until it is stable. then do another -5 and this is you stable OC.

Second sign of GPU error - fan speed in Claymore i 0% i logs.

And dont OC the CORE. ETH needs only Memory OC. Core can be undeclocked for coolnes..

I got 50% power limit, -200 core, +765 Memory. I get 98.2mh by x4 GTX1060 rig power consumtion it 472wh!

You're so on point !
I started to wonder, how come everytime i launch my AF, even with my claymore closed, it takes forever to launch the AF.

Let me try your OC setting. I will come back with the updates.

Thanks alot !

BoostOrDie

newbie

Activity: 9

Merit: 0

Quote from: chanyewhau on November 17, 2017, 10:44:11 AM

The memory size is 6144 MB, memory type is Hynix.
My current OC setting is Power Limit 75%, Core Clock +50, Memory Clock +300, Fan Speed 65%.
With the current OC setting, the cards are running at GPU Clock 1566 MHz, Memory 2152 Mhz. (Based on GPU-Z)
The default clock is 1506 Mhz, Memory 2002 Mhz.

The problematic fan that runs at 100% automatically has the lowest temperature even it is running at 65% like the other cards.
So i dont think the fan switch to 100% due to overheat.

I Got 4 1060 6gb Aorus with samsungs.

I was beating about 3 weeks with strange instabilty wit h my rig. It works OK then BAM! cpu gets loaded up to 80-90% and hash rate drops. Cards seemed ok.
I reinstalled win 7 and win 10 dozens of times. tryed dozens of drves versions and settings.

The problem was EASY - too much overclock!
If you are used to artifacts and freezes due to ooverclock - so in in mining it is strange instabilty. First sign is MSI AB or any system proces start to use 60-80% of CPU even if miner is stoped! Reboot and lower OC by 5 pints backwards. watch your rig again until it is stable. then do another -5 and this is you stable OC.

Second sign of GPU error - fan speed in Claymore i 0% i logs.

And dont OC the CORE. ETH needs only Memory OC. Core can be undeclocked for coolnes..

I got 50% power limit, -200 core, +765 Memory. I get 98.2mh by x4 GTX1060 rig power consumtion it 472wh!

chanyewhau

newbie

Activity: 7

Merit: 0

Quote from: Victorio on November 17, 2017, 10:55:43 AM

Quote from: chanyewhau on November 17, 2017, 10:44:11 AM

The memory size is 6144 MB, memory type is Hynix.
My current OC setting is Power Limit 75%, Core Clock +50, Memory Clock +300, Fan Speed 65%.
With the current OC setting, the cards are running at GPU Clock 1566 MHz, Memory 2152 Mhz. (Based on GPU-Z)
...
The problematic fan that runs at 100% automatically has the lowest temperature even it is running at 65% like the other cards.
So i dont think the fan switch to 100% due to overheat.

No one likes Hynix memory on nVidia - it has too high latency and does not hold very well overclocking. It's a surprise to know that 6GB GTX 1060 also use Hynix chips, I thought they carry only Samsung and Micron.
Your fan goes 100% because of driver error, that is caused by unsustainable overclocking settings. Since you overclock by MSI Afterburner, check-out the Synchronize settings across similar cards and define lower clocks for the problematic GPUs.

Hello, thanks for your advise.

So i should untick the option in MSI Afterburner to sync the same setting to similar cards.
And define lower clocks to the problematic GPU?

Like probably running at core clock +30? Memory clock +200?

Victorio

full member

Activity: 281

Merit: 100

Quote from: chanyewhau on November 17, 2017, 10:44:11 AM

The memory size is 6144 MB, memory type is Hynix.
My current OC setting is Power Limit 75%, Core Clock +50, Memory Clock +300, Fan Speed 65%.
With the current OC setting, the cards are running at GPU Clock 1566 MHz, Memory 2152 Mhz. (Based on GPU-Z)
...
The problematic fan that runs at 100% automatically has the lowest temperature even it is running at 65% like the other cards.
So i dont think the fan switch to 100% due to overheat.

No one likes Hynix memory on nVidia - it has too high latency and does not hold very well overclocking. It's a surprise to know that 6GB GTX 1060 also use Hynix chips, I thought they carry only Samsung and Micron.
Your fan goes 100% because of driver error, that is caused by unsustainable overclocking settings. Since you overclock by MSI Afterburner, check-out the Synchronize settings across similar cards and define lower clocks for the problematic GPUs.

chanyewhau

newbie

Activity: 7

Merit: 0

Quote from: Victorio on November 17, 2017, 04:20:27 AM

Quote from: P00P135 on November 17, 2017, 12:15:56 AM

Looks like drivers reset because OC too high. Dial back the OC.

I second that. When overclocking settings are too high, I get "NVML error: cannot get temperature" and "NVML error: cannot get fan speed" in Claymore's log file, and the hashrate becomes null on that GPU.

chanyewhau,
Ethash algorithm is more memory-intensive than Equihash and memory instability manifests itself faster. What is the exact memory frequency - check GPU-Z or another Afterburner skin, that shows the actual frequency? Try to run the miner at stock clocks and PL 80. Sometimes it happens that cards do not hold even a 10% clock increase.
BTW, what memory size and type is installed on your GPUs?
Fans running at 100% will not shorten the life of your GPU, though the fans might fail sooner due to increased mechanical stress (mainly the bearings run dry and might jam).

The memory size is 6144 MB, memory type is Hynix.
My current OC setting is Power Limit 75%, Core Clock +50, Memory Clock +300, Fan Speed 65%.
With the current OC setting, the cards are running at GPU Clock 1566 MHz, Memory 2152 Mhz. (Based on GPU-Z)
The default clock is 1506 Mhz, Memory 2002 Mhz.

The problematic fan that runs at 100% automatically has the lowest temperature even it is running at 65% like the other cards.
So i dont think the fan switch to 100% due to overheat.

BoostOrDie

newbie

Activity: 9

Merit: 0

Quote from: Victorio on November 17, 2017, 07:10:09 AM

Quote from: NewbieMiner17 on November 17, 2017, 05:15:50 AM

But for some reason now I can't run my rig for more than 5 minutes without it getting a hard crash... I've tried running V10 but that also seems to be crashing now. Not really sure what could or might have changed to be causing it to crash now.

Post the last lines of your log files.

Quote from: BoostOrDie on November 17, 2017, 05:36:48 AM

ETH only i have 98mh\s on a rig
ETH + DCR (- decri 5 as lowest) i get 94.7 mhs + 159 h/s

Try "-ethi 16 -dcri 15" and "-ethi 16 -dcri 30". Post here the result.
If your electricity cost exceeds 0.10 USD/kWh, dual mining has a slight profitability edge over single mining on nVidia GPUs at the moment. For me, it's not worth the increased power consumption and hardware wear.

-ethi 16 -dcri 15 Gives 94.2 Eth + 471 Dcr
-ethi 16 -dcri 30 Gives 92.5 Eth + 925 Dcr

My electricity cost is $0.03 per kWh
Power Consumption is the same with ETH only and ETH + DCR = 473Wh ( On ZEC it raises up to 525Wh)

NewbieMiner17

newbie

Activity: 46

Merit: 0

Quote from: Victorio on November 17, 2017, 07:10:09 AM

Quote from: NewbieMiner17 on November 17, 2017, 05:15:50 AM

But for some reason now I can't run my rig for more than 5 minutes without it getting a hard crash... I've tried running V10 but that also seems to be crashing now. Not really sure what could or might have changed to be causing it to crash now.

Post the last lines of your log files.

I had security set to restrict write access had to remove that and let it run to get a log. Typical it has gone on to run for the last hour no problem

Victorio

full member

Activity: 281

Merit: 100

Quote from: NewbieMiner17 on November 17, 2017, 05:15:50 AM

But for some reason now I can't run my rig for more than 5 minutes without it getting a hard crash... I've tried running V10 but that also seems to be crashing now. Not really sure what could or might have changed to be causing it to crash now.

Post the last lines of your log files.

Quote from: BoostOrDie on November 17, 2017, 05:36:48 AM

ETH only i have 98mh\s on a rig
ETH + DCR (- decri 5 as lowest) i get 94.7 mhs + 159 h/s

Try "-ethi 16 -dcri 15" and "-ethi 16 -dcri 30". Post here the result.
If your electricity cost exceeds 0.10 USD/kWh, dual mining has a slight profitability edge over single mining on nVidia GPUs at the moment. For me, it's not worth the increased power consumption and hardware wear.

kkourmousis

member

Activity: 119

Merit: 10

@BoostOrDie

Mr. Claymore claims that

Quote from: Claymore on April 11, 2016, 11:30:36 AM

- Supports new "dual mining" mode: mining both Ethereum and Decred/Siacoin/Lbry/Pascal at the same time, with no impact on Ethereum mining speed. Ethereum-only mining mode is supported as well.

but I am also detecting a performance drop for ETH. Of course it is relevant to -dcri (and probably to -ethi as well). However, even with the performance drop you mentioned, dual mining should still be more profitable than solo

BoostOrDie

newbie

Activity: 9

Merit: 0

Hello everyone!
I got x4 1060 6G with pentium 4400 cpu

Im sorry if my question is dumb but i saw an update today (been using 9.8 before) and decided to update my miner.

Features states: - Supports new "dual mining" mode: mining both Ethereum and Decred/Siacoin/Lbry/Pascal at the same time, with no impact on Ethereum mining speed. Ethereum-only mining mode is supported as well.

I tryed it and when i mine ETH+DCR i get ETH hashrate drop.

ETH only i have 98mh\s on a rig
ETH + DCR (- decri 5 as lowest) i get 94.7 mhs + 159 h/s

So eth still drops as before. Is it supposed to be so and i understood a feature wrong or im doing something wrong?
Thanks in advance

NewbieMiner17

newbie

Activity: 46

Merit: 0

Hey guys,

Running a 6X480's 4G Nitro + Crimson Drivers, H81 BTC Pro, 8gb ram Win 10 Claymore V10.1 Virtual mem 32gb. Oc through Afterburner 2k 1150 (been stable for well over a month)

Have two rigs running with the same settings and they have been golden for the last 4 weeks. Around 28.5 MH while dual mining Lbry in Nicehash Pool. Then yesterday my second rig crashed and had to be manually reset (couldn't log on team viewer).

But for some reason now I can't run my rig for more than 5 minutes without it getting a hard crash... I've tried running V10 but that also seems to be crashing now. Not really sure what could or might have changed to be causing it to crash now.

Any ideas as to what could have happened? What can I check or change?

Following settings for start.bat
setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setX GPU_SINGLE_ALLOC_PERCENT 100

.exe etc....
-allpools 1 -estale 0 -esm 3 -ftime 10 -cvddc 980 -dcri 42 -ttdcr 79 -ttli 81 -tstop 85

Edit: Is there a new blockchain driver? AMD settings are saying I should update to 17.11.1 from 17.30. What is the Enable compute options under global settings? I'm not seeing that as an option anywhere

Victorio

full member

Activity: 281

Merit: 100

Quote from: P00P135 on November 17, 2017, 12:15:56 AM

Looks like drivers reset because OC too high. Dial back the OC.

I second that. When overclocking settings are too high, I get "NVML error: cannot get temperature" and "NVML error: cannot get fan speed" in Claymore's log file, and the hashrate becomes null on that GPU.

chanyewhau,
Ethash algorithm is more memory-intensive than Equihash and memory instability manifests itself faster. What is the exact memory frequency - check GPU-Z or another Afterburner skin, that shows the actual frequency? Try to run the miner at stock clocks and PL 80. Sometimes it happens that cards do not hold even a 10% clock increase.
BTW, what memory size and type is installed on your GPUs?
Fans running at 100% will not shorten the life of your GPU, though the fans might fail sooner due to increased mechanical stress (mainly the bearings run dry and might jam).

Topic: Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux) - page 534. (Read 6590565 times)