After another 24000 shares I got 2 blocks, 1 in 12000 shares at a difficulty between 2000 and 5000.
CPU load was 2%, GPU load was 99%.

We really need a pool, solo mining sucks.

Thanks, I meant GPU usage from Afterburner.
But you are correct, it seems like the reaper eats up 38% of my CPU, which may affect the wallet as mentioned previously, thanks, will try the priority thing.

Wow, I really hate you now, I got 4 blocks at 30 000 shares, running at 7GH/s and still getting next to nothing.

I am at aggression level: Very angry.
The reaper was set at agression 26, think I ran the default agg value.

Anyways, my new favorite settings for the XFX 79xx cards (that some claim are bad for mining)

worksize 256
aggression 25
threads_per_gpu 2
cpu_mining_threads 0
vectors 1
gpu_thread_concurrency 8192
sharethreads 32

At 1000Mhz it gives me 2.34Mh/s with instant 99% GPU usage, unlike other settings that slowly ramp up.
Still sucks at delivering blocks though.

/EDIT: Lowered aggression to 25 for a 2% CPU usage, that hopefully does not lock the wallet...

Any recommended settings for 7950 and 7970 cards?

GPU usage at 99% but temps are lower than scrypt/scrypt-jane/Sha256, though my settings are just start values...
Measured with MSI Afterbruner.

Got 4 blocks from 24000 shares, one in 6000 for me, far from 1/2000.

10 days have passed, erundook needs to hire an extra support technician...

Cannot login to coinex.pw anymore, says "User not found".
Emailed the support twice, no response.
PMed captainfuture, no response.

Is my account lost forever?
Had a little BTC there, guess that's gone. 

Are NVIDIA cards completely unsupported with no chance of even working?

Runs fine and fast for 15 minutes on 7970, but 0 blocks for me and 0 payout for the author.
Please add pool support.

Would prefer a finished Quark miner with pool support, before getting another new miner...
At 12.5% fee, you should get the cash to buy a mid-range NVIDIA card in no time at all.

Answering myself:
Looks like it may be the old nvidia opencl driver that is the problem:
http://stackoverflow.com/questions/15456987/opencl-could-not-find-clretaindevice-in-dll

It happens when you use the C++ bindings header file with OpenCL 1.2 header. For instance, when you run an application compiled with AMD SDK (OpenCL 1.2) on NVIDIA platform (OpenCL 1.1 only).
As fast and dirty work around, you can just edit the AMD SDK cl.h header and undef "CL_VERSION_1_2" preprocessor symbol. If you are not interested to 1.2 features, it should fix your problem.

Any chance of a quick and dirty fix?

Wanted to check if NVIDIA cards may be of any use, but I get:
"The procedure entry point clRetinDevice could not be located in the dynamic link library."
Updating the driver...

Edit, updated from 320.00 to 320.86, no change, driver is supposed to support OpenCL 1.1, is it too old?

Will you guys add the new DEM info at the old DEM wikipedia page?  
http://en.wikipedia.org/wiki/Deutsche_Mark

Awesome pool, thanks guys!
Love the "Disable autotrade" function, which means that I mine the favorite coins at low diff automatically, nice.

JarzcoxK7DExJapRzbTGXZ95Nzv8H5HCjm

Thanks!

Tried 1.2 with a little solo mining, cgminer found a block, the wallet still says 0 coins.
WTF? 

So it is a miner and also a sex toy?
Those crazy italians... 

QfYb6D8MBrahFtvTXMFKVxUXAV9WoKQByd

Thanks!

A67Wenvs8WUuUHoTwKzbW8VcVYduEGa4Zj

thnx

e8Sw3sasvW2NWApnveiwZFSJnXTrP3Acbg

Thanks!