[ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX]

[canigetanipa]

when ccminer go in loop,(i can't even read too fast) it mean that the gpu are crashed?

Yes, you can use a script to autokill the ccminer and then restart it every 30 mins, or reduce your overclock.

@echo off
set app=ccminer.exe
set algo=x11
set address=stratum+tcp://new.hashharder.com:9990
set user=GcJuG7oWTfGep9CEbGsurdcUH41egKQzME
set password=x
:start
  start "" /min %app% -a %algo% -s 1 -d 0,1,2,3,4,5 -o %address% -u %user% -p %password%
  timeout 1800   
  taskkill /im %app% /f
goto start

[1715468287]

1715468287

[Amph]

the problem is that one of the card appears to be in sick mode or so(the hash is very low) and i must restart the pc to fix it

oc is very limited +75/+200 on msi

on msi afterburner i can't rise the voltage above 31

[myagui]

By the time i got back from work i was so tired i just got some. Obviously didn't pay full wack for them, i did save some. But i didnt pay as low as you suggested because i never would have got any xD

Ah craps... 
I got some more XMR at 0.0043 overnight (at Bittrex), so, you could have certainly bought some at the range I mentioned. You could even buy some now at that price on Polo. No trading while sleepy/tired, you'll be missing a key trading quality: patience...

On the positive side: I'm confident you'll see gains from that purchase, if not any obscene figures. With XMR, I'm betting things will evolve slow and steady, so I'd expect just small spikes and dips along the way. Anyways, I'm just pulling guesses out of my ass now. Good luck mate!

~ Myagui

[bigjme]

Well my webserver is set to check the rates on the 3 coins, every so many minutesm calculate profit/loss and tell me if profit goes up over 15%

[S_tring]

the problem is that one of the card appears to be in sick mode or so(the hash is very low) and i must restart the pc to fix it

oc is very limited +75/+200 on msi

on msi afterburner i can't rise the voltage above 31

That's what happens to me... gpu4 every time.

[S_tring]

when ccminer go in loop,(i can't even read too fast) it mean that the gpu are crashed?

Yes, you can use a script to autokill the ccminer and then restart it every 30 mins, or reduce your overclock.

@echo off
set app=ccminer.exe
set algo=x11
set address=stratum+tcp://new.hashharder.com:9990
set user=GcJuG7oWTfGep9CEbGsurdcUH41egKQzME
set password=x
:start
  start "" /min %app% -a %algo% -s 1 -d 0,1,2,3,4,5 -o %address% -u %user% -p %password%
  timeout 1800   
  taskkill /im %app% /f
goto start

Just run this as a bat file?

[Amph]

the problem is that one of the card appears to be in sick mode or so(the hash is very low) and i must restart the pc to fix it

oc is very limited +75/+200 on msi

on msi afterburner i can't rise the voltage above 31

That's what happens to me... gpu4 every time.

mine is the one on the pcie x16, i know is the oc, but +75 is a very poor oc, it shouldn't crash...

[zelante]

https://github.com/zelante/ccminer/releases
v1.2.3 "Split Screen ccMiner" (2014-06-18) Source + Windows Binary release

add physical number of card in system to gpu info string
fixed order of gpu info after upgrade to nvidia driver 340.43

Physical (bus) order number to info string in [ ] must be equal to gpu number in system (ex. in msi afterburner monitor)

In my case i have 3 card on this rig. After upgrade to nvidia driver 340.43 my order after launch ccminer wthout key -d look like this


i want to reorder my stats string to equal system order (like in msi afterburner monitor) i setup key -d 1,2,0 and got this

[S_tring]

I'm getting :

Code:

Error: The process ccminer35-50.exe not found

When trying to run this batch file from the same directory as ccMiner 1.2

Code:

@echo off
set app=ccminer35-50.exe
set algo=jackpot
set address=stratum+tcp://erebor.dwarfpool.com:3352
set user=JPC_ADDRESS
set password=x
:start
  start "" /min %app% -a %algo% -s 1 -d 0,1,2,3,4,5 -o %address% -u %user% -p %password%
  timeout 1800   
  taskkill /im %app% /f
goto start

 

[myagui]

Code:

set app=ccminer35-50.exe

It's now ccminer.exe (dropped the compute level variations)
~ Myagui

[S_tring]

Code:

set app=ccminer35-50.exe

It's now ccminer.exe (dropped the compute level variations)
~ Myagui

Not in the version I have? I'm running the 35-50 in my normal batch file, but it gives that error in the auto restart batch file.

[S_tring]

Strange, I just downloaded from the github repo again and it only has the ccminer.exe file there now.

[myagui]

Oddball suggestion: rename the exe file and change the batch to make it just ccminer.exe, or figure out if it needs some bracketing to get properly called as a variable? Does your other batch file also invoke the executable as a variable, or is it a full path statement?

So, suggesting something like:

Code:

set app='ccminer35-50.exe'

Or maybe:

Code:

set app="ccminer35-50.exe"

Out of ideas here!  
~ Myagui

[Amph]

how to set core clock via command line on desktop shortcut, what the parameters?

[zelante]

how to set core clock via command line on desktop shortcut, what the parameters?

NVIDIA Inspector can generate shortcuts with clocks, voltages, fan speed...

[S_tring]

Oddball suggestion: rename the exe file and change the batch to make it just ccminer.exe, or figure out if it needs some bracketing to get properly called as a variable? Does your other batch file also invoke the executable as a variable, or is it a full path statement?

So, suggesting something like:

Code:

set app='ccminer35-50.exe'

Or maybe:

Code:

set app="ccminer35-50.exe"

Out of ideas here!  
~ Myagui

Quotation marks aren't working either. Hmm. Do I need to set a path to the file rather than just the filename?

[Lucie]

Christian a question.
I was testing around with keccak algo and I enabled checkCudaErrors macro in the nvkernel.cu.
Using K1024x32 with a lookup gap of 1024 (I have a GTX 660).
Everything seems to work but I do get a runtime error at: (nvkernel.cu:99 in repository)

Code:

NVKernel::set_scratchbuf_constants(int MAXWARPS, uint32_t** h_V)

at the call:

Code:

cudaMemcpyToSymbol(c_V, h_V, MAXWARPS*sizeof(uint32_t*), 0, cudaMemcpyHostToDevice)

The c_V array on the device only supports up to TOTAL_WARP_LIMIT elements but since MAXWARPS under these settings
is higher the memcpy fails.
What I don't understand is, the pool is still accepting shares. Shouldn't this break the algo?
Is this a bug or can it be ignored?

[cbuchner1]

Is this a bug or can it be ignored?

well Keccak doesn't actually use the scratchpad. Keccak and Blake were quickly hacked add-ons to cudaminer
and the memory handling was never fully adjusted for these algos. The algos should probably moved to ccMiner.

It's a bug, but this one is quite safe to ignore.

[S_tring]

Can anyone get this to work on Windows? I really like the idea of having the miner restart every half hour as it could solve my driver crash problems.

Code:

@echo off
set app=ccminer.exe
set algo=jackpot
set address=stratum+tcp://erebor.dwarfpool.com:3352
set user=JPC_ADDRESS
set password=x
:start
  start "" /min %app% -a %algo% -s 1 -d 0,1,2,3,4,5 -o %address% -u %user% -p %password%
  timeout 1800   
  taskkill /im %app% /f
goto start

[Amph]

how to set core clock via command line on desktop shortcut, what the parameters?

NVIDIA Inspector can generate shortcuts with clocks, voltages, fan speed...

i need to OC only 3 cards rather than four, it is possible?

i need something like --gpu-engine xxxx