I've used EVGA Precision but I liked NvidiaInspector more.  It seemed to have more/better control for my cards.  I have 2 gtx560se's and Precision would only adjust one of them and leave the other at stock.  NvidiaInspector lets you set each cards speeds, fan, and voltage separately.

Try them both. 

Google a tool called NvidiaInspector (I think it's from TechPowerup).  It will let you adjust the fan speeds, voltage, and clock speeds of the core/mem/shader of most all nvidia cards.

I use it to set the clocks on my gt430 card to 882Mhz core, 810Mhz mem, 1760Mhz shader, .990v.  That combo yields between 36kh/s-38kh/s.  If I go any higher than that I get the driver crash.

Your mileage may vary though on the clock speeds you can attain.  I have an EVGA GT430 so I'm not sure how it compares to other flavors.

If you overclock that gt430 a bit you can get in the 30kh/s range. 
I currently get 36kh/s on my gt430 with configuration 20x8. 

1LjKLZ1i6HJ5JopHXFc6wq1ADZfzreAKsU

check here
https://docs.google.com/spreadsheet/ccc?key=0AjMqJzI7_dCvdG9fZFN1Vjd0WkFOZmtlejltd0JXbmc#gid=1


that should give you an idea

6rZ1ojHEw4uEEd66KgoF4Py8k2v66HH7Y6

Thanks!

I have 2 gtx560se and 1 gt430.  I got 1 560se for free (I asked for a 560 but they weren't that computer literate so whatever lol ) and figured I may as well sli it with another for $120 (still not sure why...other than I always thought it would be cool to have sli.  I'm not a big gamer).  The 430 was a leftover from an htpc build.

The 560se I must say I'm impressed with.  It overclocks like a dream.  Stock clocks are 776/1915mem/1552shader.  I have it running at 971core/2121mem/1938shader at 1.062v.  Stock cooling.  Never gets above 80c when cranking away mining.

I tried my gtx560se's with -C 1 and got a very slight drop in performance.  Hardly measurable in my opinion.

I used to average 276kh/s with the previous version, with this new one and the -C 2 variable I'm averaging 278.5kh/s.

edit:  sorry...the 278 is a total combined hashrate of all three cards

Tried autotune with -C option set to anything other than 0 and got a program close and the gpu clocks went into the recovery state (half speed).  Needed to reset to fix it. 

Setting -C to 0 and letting autotune run came up with nearly the same results as the 4/14 build.  Adding the -C 1 or -C 2 variable had almost no effect (maybe .5kh/s).

2 gtx560se (gpu's 0 and 2) and a gt430 (gpu 1) with the following commandline generate about 277kh/s


cudaminer.exe -i 0,0,0 -d 0,1,2 -C 2,2,2 -l 12x8,20x8,12x8

Will let it run for a while and check back.

Looks like you have a huge number of HW errors.  Try messing with worksize or threadconcurrency to get those down.  I was runninng into the same problem and once I got those numbers in the HW column under control my numbers on the pool started to reflect what I was seeing locally.

Thanks - amazingly that did it.  I also lowered the Intensity to 17 and it seems better now.  Now I'm getting 0 HW errors and very few rejects. 

Trying to setup a 4870x2 for some mining and have run into some weird issues.

The system is as follows:
Asus P5N7A-VM motherboard with onboard nvidia gpu (enabled and used as primary display).
4GB ram
40gb hdd.
4870x2

Current cgminer config:

{
"pools" : [
   {
      "url" : "http://poolname",
      "user" : "username",
      "pass" : "password"
   }
]
,
"intensity" : "18,18",
"vectors" : "1,1",
"worksize" : "128,128",
"kernel" : "scrypt,scrypt",
"lookup-gap" : "2,2",
"thread-concurrency" : "3200,3200",
"shaders" : "800,800",
"gpu-engine" : "0-800,0-800",
"gpu-fan" : "60-100,60-100",
"gpu-memclock" : "920,920",
"gpu-memdiff" : "0,0",
"gpu-powertune" : "0,0",
"gpu-vddc" : "1.250,1.250",
"temp-cutoff" : "95,95",
"temp-overheat" : "85,85",
"temp-target" : "75,75",
"api-port" : "4028",
"expiry" : "120",
"gpu-dyninterval" : "7",
"gpu-platform" : "1",
"gpu-threads" : "2",
"hotplug" : "5",
"log" : "5",
"no-pool-disable" : true,
"queue" : "1",
"scan-time" : "60",
"scrypt" : true,
"temp-hysteresis" : "3",
"shares" : "0",
"kernel-path" : "/usr/local/bin"
}

1st issue:  when using the nvidia as primary display and dedicating the card to mining, ATI CCC complains that my primary display device is not compatible.  Understandably since it's nvidia, but shouldn't I be able to adjust the ATI card settings stil?  In this configuration I get the message from cgminer that ADL found less devices than opencl and that hardware monitoring may not match up to devices.
  When I get that error, it is in fact correct...there is no hardware monitoring nor does it control the fan...but I get equally decent hashrates on both gpu's (~150Kh/s).  The problem here is that I also get a high number of rejects and hw errors.  So high in fact that for every 4 accepts, I get 15 hw errors.

2nd issue:  When I do a fresh install of the ATI drivers I don't get any of the messages from the 1st issue.  It goes smoothly, cgminer sees the card correctly, adjusts the fat, etc.  However gpu0 will get a hashrate of say 80kh/s while gpu1 will get a hashrate of near 160kh/s.  I still run into the hw errors though.

Any ideas?  I've tried to play with the thread concurrency but haven't really found a magic number yet.

Added my gtx560se and gt430 oc to the list.

Go to techpowerup.com and grab nvidiainspector.  It will let you adjust the fan speeds on the cards manually (I just set mine at 100).  It will also let you set clockspeeds and voltages through command line (I made a batch file to do all of the above on my 3 nvidia cards).

I definately like the speed improvement on the last update.  I went from a total hashrate of 230kh/s (2 gtx560se's and a gt430) to 264kh/s.

Unfortunately the hashrate of my gtx260 went down by 10kh/s.  I plan on replacing it soon anyway.

Downloaded and testing now.  

First thing I noticed with my gtx560se's is a higher hashrate than before using the same config.  
4/13/2013
gtx560se
config: 24x4
105-107kh/s

4/14/2013
gtx560se
config: 24x4
109kh/s

EDIT:

Seems that now when I try to exit it, it reboots the computer (not a huge deal really).

I'm getting a similar result as that.

Local hashrate is awesome...but the pool isn't seeing it.  I disabled the texture cache and now it's back to the pool reporting shares - but at a MUCH lower hashrate...lower than the 4/10 build.  I tried to go back to the 4/10 build and was still getting a low hashrate (gtx260 only, and hashrate droped to the 40's) so I'm not sure what's going on.

Edit
I figured out what's going on.  I've lost my mind.  seems the only explanation.  I could have swore that stupid gtx260 got a higher hashrate than 44kh/s on the 4/10 build. 

4/12 build still doesn't seem to send shares when the texture cache is enabled.  I'm not sure what's going on...but cudaminer says it sends them and they're accepted...I can see the same activity going on in the proxy window, yet the shares never show up on the pool.  Tried 2 different pools with the same result, and 2 different machines with the same result. 

GTX260 went from 41kh/s > 80kh/s
9300 (nforce 730i) remained at 4.5kh/s whether the texture cache was enabled or not


Here is some interesting results with my fermi cards:
gtx560se autotune + textcache = 77kh/s
gtx560se autotune - textcache = 102kh/s   (46x2 config)
enabling the textcache AFTER performing the autotune = 107kh/s (46x2 config)

gt430 went from 33-35kh/s down to 31kh/s no matter what.  (48x4 config)

Nice!  Cant wait to get home and try it.
Maybe you already found the magic combo for fermi...althought its hard to believe that a gtx260 is almost at the performance of my gtx560se's. 

What happens if you run only using only 1 gpu?  My gtx 260 gets 40kh/s itself, I would expect at least double that with a 295.