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

[atc1]

ok guys so to elaborate on my problem, so i've clocked all the 750Ti's so they run the same clock and memory.
The one that plays up is running at 33mv higher then the other 2, i am unable to under voltage the card in msi afterburner
when the miner crashes, if i restart it the card pretty much stays idle and mines at around 4Mh/s
Doesn't clock itself back up, which seems like maybe a driver crash. So i un-installed all my drivers, and re-installed them

even with new drivers the same card is still crashing the same, so the loop trick wont work as the card never clocks itself back up. Seems a weird one to me, and i dont have a clue why afterburner wont let me undervolt the card

who ever helps me get this sorted (besides suggesting returning the card, i will send you every heavycoin i have mined, at the moment its only 40, but i will send through more as i get them)

EDIT
Ooo and the card will happily mine for ages when its sitting at 4MH/s

I dunno,but have you tried to use any other OC software other than Afterburner?  And when the miner crashes,close it,go to device manager,disable and reenable the specific card. It'll reset the drivers,maybe that'll help?

[1714330099]

1714330099

[1714330099]

1714330099

[1714330099]

1714330099

[1714330099]

1714330099

[bigjme]

i've tried evga precision aswell, that will let me over-volt but not under
i haven't tried disabling the card yet no, i will try that if it crashes soon

EDIT

ok so disabling and enabling the card does reset it back to full mining speed

FIXED THE ISSUE
may help others, so the card that was playing up, turned out to be plugged into a spare PCI 3.0 x16 slot.
The temps on that card were higher then the others aswell

I had a spare PCI-E 2.0 x1 slot, moved the card there and its mining like a dream!
so be warned anyone buying a rig, and planning to use the PCI-E 3.0 slots, they may not work!

[christian1980]

I'm trying to make a batch file that restarts ccminer when it crashes, cause I cant figure out why it crashes. But cant get this command to work. Now I'm stuck :

I posted a solution at page 509.

[nomad1109]

i've tried evga precision aswell, that will let me over-volt but not under
i haven't tried disabling the card yet no, i will try that if it crashes soon

EDIT

ok so disabling and enabling the card does reset it back to full mining speed

FIXED THE ISSUE
may help others, so the card that was playing up, turned out to be plugged into a spare PCI 3.0 x16 slot.
The temps on that card were higher then the others aswell

I had a spare PCI-E 2.0 x1 slot, moved the card there and its mining like a dream!
so be warned anyone buying a rig, and planning to use the PCI-E 3.0 slots, they may not work!

There was a mention a while back of making sure your 16x slot was set to rev 2 in bios or something like that. Might check that out and see if it helps with that slot.

[bigjme]

I only have the 3 cards so i dont need to use that slot, unsure why i used it in the first place if im honest.

Got all my clocks sitting at 1.2ghz now, mining at 11.5MH/s

i dont want to increase power usage too much, as the idea was to replace my 780 with 3 of these to save on power

[brainape]

Novice miner (and computers in general) installs 4 - EVGA 750ti and finds the following...

Waited on these for almost a month.  Jumped from 1 - GTX660 to the 4 - 750ti. 

Running Windows 8x64, CUDAminer 2014-02-28 x64 and Nvidia driver 335.23

All 4 cards are on powered risers.

CUDAminer set to auto (I think) -o stratum yadayada -0 (GPU clock +30 Mem clock +530)

Each card averages 260 - 265K with the total average at 1050K.

Temps lowest at 51 and highest at 61.

Watts at the wall showing through my UPS are between 390 and 450 and bouncy.

Overall I'm pleased with my first real project.  Thanks CUDAminer.   

Thought I would share.  Searching around to get a bit more hashing without crashing.


 

[bigjme]

Power wise, my main system running 3 instances of cccminer is sitting at 120W, not sure why, ccminer must be keeping my cpu running at full speed, even with cpu usage of 7%?
My 3 750Ti's are pulling 150W from the wall, on a 80% efficiency psu, at best! Their clocked to 1.2GHz and mining HVC

giving a total load of 270W! With my 780 by itself i had a total power usage of 350W
that's a power saving of 80W, or £110/$190 a year, one card will pay for itself just in power saving! not to mentioned the extra power i get from the 3 cards
To pay power i would need to mine 74HVC a day, and anything extra is profit. All in all, their beast cards, and im looking to order a load more soon

[cbuchner1]

Power wise, my main system running 3 instances of cccminer is sitting at 120W, not sure why, ccminer must be keeping my cpu running at full speed, even with cpu usage of 7%?

in one of the algos (fugue?) we forgot to call MyStreamSynchronize after the kernel call. I just noticed today.

But you're saying your CPU isn't clocking down even with CPU use at 7%? That's hard to believe.

Christian

[bigjme]

in one of the algos (fugue?) we forgot to call MyStreamSynchronize after the kernel call. I just noticed today.

Ahh i didnt notice that

But you're saying your CPU isn't clocking down even with CPU use at 7%? That's hard to believe.

PIC

i don't have anything else running, thats my xeon which normally sits at 1.4GHz and only goes up to 3.5GHz maximum, (3.7GHz never gets hit without 100% load)
power settings are set to cpu lowest power state is 5%

your computer might be part of some botnet network  
You might be mining something you don't know

haha i hope not

[djm34]

 your computer might be part of some botnet network  
You might be mining something you don't know

[sin242]

Some big news here if its real:



https://bitcointalk.org/index.php?topic=549572.0;all

[cbuchner1]

Some big news here if its real:
https://bitcointalk.org/index.php?topic=549572.0;all

won't work. LTC is too big of a network, and I seriously doubt the majority is supportive of a hardfork.

[djm34]

in one of the algos (fugue?) we forgot to call MyStreamSynchronize after the kernel call. I just noticed today.

Ahh i didnt notice that

But you're saying your CPU isn't clocking down even with CPU use at 7%? That's hard to believe.

PIC

i don't have anything else running, thats my xeon which normally sits at 1.4GHz and only goes up to 3.5GHz maximum, (3.7GHz never gets hit without 100% load)
power settings are set to cpu lowest power state is 5%

your computer might be part of some botnet network 
You might be mining something you don't know

haha i hope not

may-be you have a wallet opened with "setgenerate true" ?

[bigjme]

may-be you have a wallet opened with "setgenerate true" ?

no wallets on this machine, here's it with all my ccminers in one launch file
PIC

so my cpu usage drops by about 3%
and my cpu clock went down, even to the point of idling down to the 1.6MHz region for a few seconds every so often, i am still on it so it will idle down properly when im not. Yes i lost a small amount of hashing power doing this, but it seems to not keep my system fully clocked all the time. just thought i would throw it out there :-)

[sin242]

in one of the algos (fugue?) we forgot to call MyStreamSynchronize after the kernel call. I just noticed today.

Ahh i didnt notice that

But you're saying your CPU isn't clocking down even with CPU use at 7%? That's hard to believe.

PIC

i don't have anything else running, thats my xeon which normally sits at 1.4GHz and only goes up to 3.5GHz maximum, (3.7GHz never gets hit without 100% load)
power settings are set to cpu lowest power state is 5%

your computer might be part of some botnet network 
You might be mining something you don't know

haha i hope not

may-be you have a wallet opened with "setgenerate true" ?

Or "gen=1" in the conf

[bigjme]

ok, so the power usage drove me to believe that my cpu was using the power, monitoring it now this is what i get

No ccminer, my 780 sits at 5% tdp, total power usage 60W
launch ccminer, my 780 goes to 27% tdp, total power usage 130W

This is my ccminer launch config; (my 780 is device 0)

Code:

ccminer-v0.5\x86\ccminer35.exe -a heavy -q -d 1,2,3 -o stratum+tcp://hvcpool.1gh.com:5333 -u wallet -p x -v 512

now the gpu load of my 780 is 0% but it is clocking itself up the second i start ccminer
is it perhaps still telling my 780 it may be needed? causing the clock to go up? as the clock stays down when im using chrome, and doing basic desktop stuff

Ooo and the power usage from my 780 is the same with 1 or 3 instances of ccminer at the same time

[djm34]

 you know that with my system, I use less power than you to hash hvc   
(something like 80W per equivalent 750ti hashing power)

[bigjme]

you know that with my system, I use less power than you to hash hvc  
(something like 80W per equivalent 750ti hashing power)

how much do you hash at? and whats your total power?
if i count just my cards, i get 145W giving my around 35MH/s

once this thing with the 780 is sorted my total system will be 205W for 35MH/s
Don't forget my system is water cooled so i expect around 25W extra for the pump etc.

[djm34]

you know that with my system, I use less power than you to hash hvc  
(something like 80W per equivalent 750ti hashing power)

how much do you hash at? and whats your total power?
if i count just my cards, i get 145W giving my around 35MH/s

once this thing with the 780 is sorted my total system with me
205W for 35MH/s
Don't forget my system is water cooled so i expect around 25W extra for the pump etc.

The gtx780ti does 28-30MHash/s for 207W (83% tdp)
the R9290x 13MHash for around 120W
So this is roughly equivalent to for 750ti mildly overclocked

[bigjme]

so your 780TI is around 5MH/s slower, and is that gpu power or total system?
right now i can watch 1080p movies and use no extra power because of my gpu clocking up
lmao