Author

Topic: [ANN] cudaMiner & ccMiner CUDA based mining applications [Windows/Linux/MacOSX] - page 1056. (Read 3426947 times)

newbie
Activity: 8
Merit: 0
Maybe use one of the old cudaminer versions that dont seem to hammer the card so much?
sr. member
Activity: 462
Merit: 250
Does anyone have settings for a 690 that doesn't absolutely melt the device? How can I throttle this thing to make it less intense?

Try running with -i -- it'll reduce the speed a little bit.

I _just_ got my 690 up and running and am still having driver issues with it (I can't run on both devices at the same time, sigh).

But with a single device running with -d1 -m1 -lK8x16
I'm seeing about 270-275 kh/s
and after a few minutes my card is at 78C.  It's freestanding (motherboard-on-a-table kind of thing).  78 is not something you want to stick your tongue on, but it shouldn't hurt the card.

What kind of temperatures are you seeing from nvidia-smi?  What hash rates and what config?

(And - for my own use - if you're running Linux, which driver are you using that works?  *grins*)

I tried running "-i --" but that just put interactive mode to 0 and made my pc unresponsive.

I'm seeing 400 KH/s in total with default settings on my 690 on Windows 8.1

Is there really no way for to have my 690 only mine at 60 % for example?
newbie
Activity: 9
Merit: 0
My cudaminer for gigabyte gtx 760 cudaminer.exe -H 1 -i 0 -t 1 -C 2 -l K12x16  - got from 180kh/s to 280!! and then i havent even OC yet. This made my day!
newbie
Activity: 6
Merit: 0
Loving the new cuda, ~325 kh coming from ~200 with a slight OC on my GTX680.

-d 0, -l K16x16  is my current setup. Not sure if i can tweak it a little more.
dga
hero member
Activity: 737
Merit: 511
Does anyone have settings for a 690 that doesn't absolutely melt the device? How can I throttle this thing to make it less intense?

Try running with -i -- it'll reduce the speed a little bit.

I _just_ got my 690 up and running and am still having driver issues with it (I can't run on both devices at the same time, sigh).

But with a single device running with -d1 -m1 -lK8x16
I'm seeing about 270-275 kh/s
and after a few minutes my card is at 78C.  It's freestanding (motherboard-on-a-table kind of thing).  78 is not something you want to stick your tongue on, but it shouldn't hurt the card.

What kind of temperatures are you seeing from nvidia-smi?  What hash rates and what config?

(And - for my own use - if you're running Linux, which driver are you using that works?  *grins*)
hero member
Activity: 821
Merit: 503
Not sure if it matters but when trying to run default settings the newest version of cudaminer on my nvidia 650ti boost. it loads and when it tries to run the auto tune it crashes .. I have tried both 32/64 bit versions, but if i re-add all my config file. it runs and even runs faster Smiley

Just thought i would let you know, running latest nvidia drivers and card is not o/c


Icon

sr. member
Activity: 462
Merit: 250
Does anyone have settings for a 690 that doesn't absolutely melt the device? How can I throttle this thing to make it less intense?
member
Activity: 84
Merit: 10
SizzleBits
That's a very old version, go get the latest one and see if it works better: http://www.guru3d.com/files_details/msi_afterburner_beta_download.html

I suppose I'll give the update a try with the voltage config hack. Maybe the usage of Cuda 5.5 has broken the stats in the older version. Back in a few if that helps.

EDIT: Nope same story even with the update and the voltage hack. Something with the update to the miner is causing it. I think it thinks my voltages are too high and is throttling the core clock since if I drop my core clock down to +0 my core clock shoots back up to almost normal OC settings before cudaminer 12/18. But i know from knowing this card that it will crash at those lower voltages in  ~15 mins at 100% load. And I tested it with kombuster and furmark just now and both of those stress tests bring me back up to my normal OC core clock and with normal overvolting. So definitely something to do with todays update.
hero member
Activity: 840
Merit: 1000
x64 version gives cannot validate on CPU errors? x86 works fine though, getting 500kh/s per 780, impressive!
newbie
Activity: 43
Merit: 0
That's a very old version, go get the latest one and see if it works better: http://www.guru3d.com/files_details/msi_afterburner_beta_download.html
member
Activity: 84
Merit: 10
SizzleBits
see the issue with 2.3.1 is i cant overvolt the card without changing some config files :/ So i stick to 2.2.3 where the overvolting controls havent been disabled in the background of the application.
full member
Activity: 173
Merit: 100
MSI N660 TF 2GD5/OC x 2

Bumped up a bit with Afterburner 2.3.1
Core Voltage +25
Power Limit 119
Core Clock +65
Memory Clock +162

-d 0,1 -i 0,0 -l K5x12,K5x12 -H 1

214kh/s per card

The -H 1 yields a +10kh/s increase over not including it.

Christian, can you explain why general computer use seems to be unaffected despite the -i 0 flag on both cards?

newbie
Activity: 14
Merit: 0
GTX 570 phantom - same as previous two versions... as expected!

EDIT: BUT...  temps with x86 are ~2C lower (same settings and kH/s results)! .... so this is THE FIRST win for x86 @ my place Smiley
member
Activity: 84
Merit: 10
SizzleBits
Lol I got my EVGA 780 Classified up to 655KH/s Smiley
I have the mem overclocked +700.

Its running at 150KH/s more than my Titans... so sad for Titans...

I thought core clock was more important than mem clock for scrypt mining...?
Thats crazy though. Ive gotten my 660 ti up to +525 and +110 stable in mining and games
Vbs
hero member
Activity: 504
Merit: 500
On GTX 780 (Ti) and Titan cards, one of the best launch configurartions seems to be

T number of SMX x 32

i.e. T15x32 for 780 Ti
T14x32 for Titan
T12x32 for GTX 780

Upped from 550 to 570 kHash simply by going from x24 to x32. autotune will currently not find x32 configurations.

Christian


Yep, thanks for that! Grin 675kH/s now! Grin

(Gigabyte 780 Ti OC edition)
sr. member
Activity: 305
Merit: 250
Lol I got my EVGA 780 Classified up to 655KH/s Smiley
I have the mem overclocked +700.

Its running at 150KH/s more than my Titans... so sad for Titans...
full member
Activity: 142
Merit: 100
Awesome update. Got an appox 33% increase in hashing power. From 38 to 51 khash/s on my GT 650M. Slow, I know, but I get free electricity and only have this laptop.

Temps have gone up about 4 oC though (to 72 oC). Slight concern on my laptop. Don't wanna blow this thing Smiley

mine hits 95 deg before starting its jet engines... 750M GT


My GT 540M always runs at 95°C, at 96°C it halves the GPU and shader clocks.

Is anybody else running cudaminer on this card? My current settings are -i 0 -H1 -m1 -l F64x2. Autotune also has selected F96x2, F80x2, F32x2 and F108x2 so far with nearly the same results. Any better settings?
hero member
Activity: 756
Merit: 502
Awesome update. Got an appox 33% increase in hashing power. From 38 to 51 khash/s on my GT 650M. Slow, I know, but I get free electricity and only have this laptop.

Temps have gone up about 4 oC though (to 72 oC). Slight concern on my laptop. Don't wanna blow this thing Smiley

mine hits 95 deg before starting its jet engines... 750M GT
full member
Activity: 140
Merit: 100
Awesome update. Got an appox 33% increase in hashing power. From 38 to 51 khash/s on my GT 650M. Slow, I know, but I get free electricity and only have this laptop.

Temps have gone up about 4 oC though (to 72 oC). Slight concern on my laptop. Don't wanna blow this thing Smiley
hero member
Activity: 756
Merit: 502
On GTX 780 (Ti) and Titan cards, one of the best launch configurartions seems to be

T number of SMX x 32

i.e. T15x32 for 780 Ti
T14x32 for Titan
T12x32 for GTX 780

Upped from 550 to 570 kHash simply by going from x24 to x32. autotune will currently not find x32 configurations.

Christian
Jump to: