thanks I will test again.
This only speeds up gpu fan to 100% not shut down hash
you probiably mean --temp-cutoff, that doesn't work.
when temp get settings like --temp-cutoff 76 than I get
Device recovered to temperature below target, re-enabling
after that nothing happends its stuck on it.
temp is 35 oC and fan is lower than 650 RPm and nothing is going on no hashing....
Topic: OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.0 - page 330. (Read 5806004 times)
May 24, 2013, 03:38:28 PM
May 24, 2013, 03:14:40 PM
The easiest way it would be a GPU-TEMP if that rise for example over 85 oC it slows down mining....
--temp-overheat
No? Whenever it hits 85, it stops hashing till it cools down.
The manual, and stuff
Strange, works here
If my card reaches 80, the load drops to 0% till it cools down and it picks up again May 24, 2013, 02:19:15 PM
The easiest way it would be a GPU-TEMP if that rise for example over 85 oC it slows down mining....
--temp-overheat
No? Whenever it hits 85, it stops hashing till it cools down.
The manual, and stuff
May 24, 2013, 01:38:44 PM
The easiest way it would be a GPU-TEMP if that rise for example over 85 oC it slows down mining....
--temp-overheat
No? Whenever it hits 85, it stops hashing till it cools down.
The manual, and stuff
May 24, 2013, 11:26:15 AM
ati radeon 5800 and two 6800 series and one 7800 series
Uh, ok, your previous post said you had two GPU's now we're up to 4.
You should have your 7800 series in a different machine that the 5xxx and 6xxx series GPU's.
Did you read this threads top post yet? If you have multiple GPU's/types in the same machine you really need to set your clock ranges per GPU. The Top Post has easy to understand examples for that.
I have readed alot but still there is no extra flag for GPU TEMP.
I mean two GPUs in one graphic card.
I have 3 graphic cards in one PC is there not possible to have one configuration for all of them?
The safe one, for now I have regulated just a --gpu-fan 55-85 it wokrs ok for now U hope it won't turn off....
The easiest way it would be a GPU-TEMP if that rise for example over 85 oC it slows down mining....
May 24, 2013, 11:25:14 AM
I have never seen cgminer was able to restart a dead GPU. Typical situation:
[2013-05-24 19:47:14] Accepted 0314caf9 Diff 83/1 GPU 1 pool 0
[2013-05-24 19:47:15] GPU0: Idle for more than 60 seconds, declaring SICK!
[2013-05-24 19:47:15] GPU0: Attempting to restart
[2013-05-24 19:47:15] Thread 0 still exists, killing it off
[2013-05-24 19:47:16] Thread 1 still exists, killing it off
[2013-05-24 19:47:16] Error -6: Creating Command Queue. (clCreateCommandQueue)
[2013-05-24 19:47:16] Failed to reinit GPU thread 0
[2013-05-24 19:47:50] Accepted 75ce3887 Diff 2/1 GPU 1 pool 0
[2013-05-24 19:48:15] Accepted d080d191 Diff 1/1 GPU 2 pool 0
[2013-05-24 19:48:17] Thread 0 no longer exists
[2013-05-24 19:48:17] Thread 1 no longer exists
[2013-05-24 19:48:17] Error -6: Creating Command Queue. (clCreateCommandQueue)
[2013-05-24 19:48:17] Failed to reinit GPU thread 0
[2013-05-24 19:48:21] Accepted 755d2676 Diff 2/1 GPU 1 pool 0
Restarting the program solves problem.
Something is wrong here
[2013-05-24 19:47:14] Accepted 0314caf9 Diff 83/1 GPU 1 pool 0
[2013-05-24 19:47:15] GPU0: Idle for more than 60 seconds, declaring SICK!
[2013-05-24 19:47:15] GPU0: Attempting to restart
[2013-05-24 19:47:15] Thread 0 still exists, killing it off
[2013-05-24 19:47:16] Thread 1 still exists, killing it off
[2013-05-24 19:47:16] Error -6: Creating Command Queue. (clCreateCommandQueue)
[2013-05-24 19:47:16] Failed to reinit GPU thread 0
[2013-05-24 19:47:50] Accepted 75ce3887 Diff 2/1 GPU 1 pool 0
[2013-05-24 19:48:15] Accepted d080d191 Diff 1/1 GPU 2 pool 0
[2013-05-24 19:48:17] Thread 0 no longer exists
[2013-05-24 19:48:17] Thread 1 no longer exists
[2013-05-24 19:48:17] Error -6: Creating Command Queue. (clCreateCommandQueue)
[2013-05-24 19:48:17] Failed to reinit GPU thread 0
[2013-05-24 19:48:21] Accepted 755d2676 Diff 2/1 GPU 1 pool 0
Restarting the program solves problem.
Something is wrong here
May 24, 2013, 11:03:30 AM
After 13 days of continuous work cgminer 2.11.4 receive this error: ''failed to create postcalc hash thread''
help what does this mean?
help what does this mean?
May 24, 2013, 10:49:36 AM
Is there a way to disable any fan management? Just let it handle itself like it did when I was using GUIMiner?
May 24, 2013, 08:56:26 AM
I run a 5870, a 6970 and a 7950 in one machine. I was also running a 7770 as well, but it died. They worked OK together.
These threads are full of the troubles and pitfalls with mixing 7xxx with 5 and 6xxx GPU's.
This guy is having trouble regulating his temperature so I think mixing GPU types is a little above his head at the moment.
Always start with the simplest configuration and get working then and ONLY then start adding complexity.
Basic T/S techniques.
Sam
May 24, 2013, 08:04:54 AM
I run a 5870, a 6970 and a 7950 in one machine. I was also running a 7770 as well, but it died. They worked OK together.
May 24, 2013, 07:47:58 AM
ati radeon 5800 and two 6800 series and one 7800 series
Uh, ok, your previous post said you had two GPU's now we're up to 4.
You should have your 7800 series in a different machine that the 5xxx and 6xxx series GPU's.
Did you read this threads top post yet? If you have multiple GPU's/types in the same machine you really need to set your clock ranges per GPU. The Top Post has easy to understand examples for that.
May 24, 2013, 07:44:24 AM
If you're really having overheating problems, you really should set a really low lower engine limit, eg:
--gpu-engine 300-950
--gpu-engine 300-950
are not --gpu-fan regulate heating so if there is setup for --gpu-fan 25-85
it will not allow to run fan faster that 85%? so I dont ned to regulate gpu-engine?
so If I understand 25 is lower temp and 85 is the highest temp??? or why is 25-85 for example?
gpu fan is 25% to 85% fan speed. If the GPU reach's the temp overheat, 85C default, it will still send the fan to 100%.
Auto gpu with lower and upper clock limit is ALSO for regulating heat to the temp target which is 75C by default. So you NEED both to regulate temp safely.
May 24, 2013, 07:37:50 AM
If you're really having overheating problems, you really should set a really low lower engine limit, eg:
--gpu-engine 300-950
--gpu-engine 300-950
are not --gpu-fan regulate heating so if there is setup for --gpu-fan 25-85
it will not allow to run fan faster that 85%? so I dont ned to regulate gpu-engine?
so If I understand 25 is lower temp and 85 is the highest temp??? or why is 25-85 for example?
May 24, 2013, 07:36:25 AM
ati radeon 5800 and two 6800 series and one 7800 series
May 24, 2013, 07:36:09 AM
If you're really having overheating problems, you really should set a really low lower engine limit, eg:
--gpu-engine 300-950
--gpu-engine 300-950
May 24, 2013, 07:33:32 AM
thank you for advice I have looking for some help about it but no luck.
i have decreased gpu with --gpu-dyninterval 4
but it still overheat...
i will do this settings
--auto-gpu --gpu-engine 750-950 --auto-fan --gpu-fan 25-85
with a TThrootle I see I have two GPUs and cgminer is using just one is that ok?
is that normal case?
i have decreased gpu with --gpu-dyninterval 4
but it still overheat...
i will do this settings
--auto-gpu --gpu-engine 750-950 --auto-fan --gpu-fan 25-85
with a TThrootle I see I have two GPUs and cgminer is using just one is that ok?
is that normal case?
dyninterval adjusts dynamic intensity.
What GPU's do you have?
Remember replace the engine clocks in the example with clocks which make sense for your GPU.
May 24, 2013, 06:51:09 AM
thank you for advice I have looking for some help about it but no luck.
i have decreased gpu with --gpu-dyninterval 4
but it still overheat...
i will do this settings
--auto-gpu --gpu-engine 750-950 --auto-fan --gpu-fan 25-85
with a TThrootle I see I have two GPUs and cgminer is using just one is that ok?
is that normal case?
i have decreased gpu with --gpu-dyninterval 4
but it still overheat...
i will do this settings
--auto-gpu --gpu-engine 750-950 --auto-fan --gpu-fan 25-85
with a TThrootle I see I have two GPUs and cgminer is using just one is that ok?
is that normal case?
May 24, 2013, 06:14:34 AM
MY two PCs are stil shutting down because overheating, what can I do to prevent this, how to make cgminer slows down with mining when overheating!
If it is your GPU's overheating then use the auto fan and auto gpu command line arguments.
There are examples in the executive summary in the top post of this thread.
I have auto-fan flag, but no auto-gpu because it overheats more if I have auto-gpu.
That makes no sense. What's your target temp?
temp 95 oC fan 100%
Doesn't that seem a bit hot to you? It seems very hot to me.
Read the readme and/or the top post of this thread.
I would recommend using the default target temp of 75C, temp cutoff of 95 and setting auto fan with a range of 25-85.
With a temp target of 95C and 100% fan when you use auto gpu it can only raise the engine clock to get to 95C which is a bit insane if you ask me. Plus you will kill your fan pretty quick.
A command line should look something like this
--auto-gpu --gpu-engine 750-950 --auto-fan --gpu-fan 25-85
Both auto fan and auto gpu will adjust to keep the GPU at the default 75C. Set your engine clock range from a low clock range of 500 to 700 to your high overclock range, whatever your overclock is for your GPU. Try not to run your fan over 85%.
Also remove your case cover if you haven't already.
Sam
May 24, 2013, 01:07:51 AM
MY two PCs are stil shutting down because overheating, what can I do to prevent this, how to make cgminer slows down with mining when overheating!
If it is your GPU's overheating then use the auto fan and auto gpu command line arguments.
There are examples in the executive summary in the top post of this thread.
I have auto-fan flag, but no auto-gpu because it overheats more if I have auto-gpu.
That makes no sense. What's your target temp?
temp 95 oC fan 100%
May 23, 2013, 11:35:43 PM
Okay, please keep the CGMiner/BFG feud banter outside this thread. It's getting tedious seeing so many reports from here. I've split the off topic parts (and deleted a few which missed the split) to https://bitcointalksearch.org/topic/re-cgminer-asic-fpga-gpu-overc-monit-fanspd-rpc-linuxwinosxmipr-pi-311-214823
Thanks everyone.
Thanks JK. Thanks everyone.
Jump to: