Topic: [ANN] Yet another GPU miner release! [YAC] - page 24. (Read 53371 times)

Is this really finding blocks? i get msg "block found for pool 0" every 30 seconds but theres nothing happening in the wallet...

Has anyone managed to get I greater than 12 without it crashing?

yacminer compiles and runs under x86 on Windows just fine, but I had to rename the .cl file to scrypt130511.cl *shrug*

7850 - 140 kh/sec
5770 - 60 kh/sec

EDIT:was underclocking memory for BTC mining initially... fixed that!

I don't think it's a real hardware problem, had it many times when settings are off.

thank you the updated version works

but i think my graphics card is not ok, if i set I bigger then 9 i get

 [2013-06-12 08:18:46] GPU0: invalid nonce - HW error - d17e2089
 [2013-06-12 08:18:47] GPU0: invalid nonce - HW error - cdb3c1f4
 [2013-06-12 08:18:47] GPU0: invalid nonce - HW error - 2f63b98a
 [2013-06-12 08:18:47] GPU0: invalid nonce - HW error - b428d90b
 [2013-06-12 08:18:46] GPU0: invalid nonce - HW error - 71a9f146
 [2013-06-12 08:18:47] GPU0: invalid nonce - HW error - 36910f4a

guess its running too hot on intensity higher then 9 

( I renamed the newly built cgminer to yacminer)
./yacminer -D --scrypt -o (URL) -p X -I 12 --worksize 256 --thread-concurrency 32768 --lookup-gap 2

[?1049h(B[?7h [2013-06-12 13:32:04] Started cgminer 3.2.0
 [2013-06-12 13:32:06] No GBT coinbase + append support found, using getwork protocol
 [2013-06-12 13:32:07] X-Roll-Ntime expiry set to 100
 [2013-06-12 13:32:07] Successfully retrieved and deciphered work from pool 0 http://yacpool.cloudapp.net:8336
 [2013-06-12 13:32:07] Pushing pooltest work to base pool
 [2013-06-12 13:32:07] Network diff set to 48.8K
 [2013-06-12 13:32:07] New block: 14fd2024b14d5b12... diff 48.8K
 [2013-06-12 13:32:07] Pushing work to getwork queue
 [2013-06-12 13:32:07] Long-polling activated for http://yacpool.cloudapp.net:8336/long-polling8] List of devices:
 [2013-06-12 13:32:08]  0Tahiti8]  1       Tahiti8] Selected 1: Tahiti
 [2013-06-12 13:32:08] Popping ping in miner thread8] Popping work from get queue to get work8] Got work from get queue to get work for thread 08] Generated target 00000000000000000000000000000000000000000000000000000000ffff0000
8] Timestamp: 1371007927, Nfactor: 1024, Target: ffff
 [2013-06-12 13:32:08] Error: clSetKernelArg of all params failed.
 [2013-06-12 13:32:08] GPU 0 failure, disabling!
 [2013-06-12 13:32:08] Thread 0 being disabled [2013-06-12 13:32:08] Creating scrypt buffer sized 2147483648
 [2013-06-12 13:32:08] initCl() finished. Found Tahiti
 [2013-06-12 13:32:08] 2 gpu miner threads started
 [2013-06-12 13:32:08] API not running - API will not be available
 [2013-06-12 13:32:08] Popping ping in miner thread
 [2013-06-12 13:32:08] Popping work from get queue to get work
 [2013-06-12 13:32:08] Got work from get queue to get work for thread 1
 [2013-06-12 13:32:08] Generated target 00000000000000000000000000000000000000000000000000000000ffff0000
 [2013-06-12 13:32:08] Timestamp: 1371007928, Nfactor: 1024, Target: ffff
 [2013-06-12 13:32:08] Error: clSetKernelArg of all params failed.
 [2013-06-12 13:32:08] GPU 1 failure, disabling!
 [2013-06-12 13:32:08] Thread 1 being disabled(B cgminer version 3.2.0 - Started: [2013-06-12 13:32:08]

What's your command line?
Can you add -D to your command line and get some debug info?

ok  

Ah, it's not needed. I satisfy myself already. I've mined with GPU for two weeks. Thank you anyway. 

Oaky I have it built now, but now I also get...

 [2013-06-12 13:18:25] Long-polling activated for http://yac.ltcoin.net:5555/LP
 [2013-06-12 13:18:25] Error: clSetKernelArg of all params failed.
 [2013-06-12 13:18:25] GPU 0 failure, disabling!

hanzac, you should post your YAC addy

This might be it's based on the latest cgminer code. The network threads seems react slow at the beginning. But it will become normal after a while.

I'm also getting the "Found Block for Pool 0!" message a lot, with no accompanying Accepted Share in cgminer.  Just want to make sure that this is normal.

My screen looks a lot like Aggrophobia's above, without the crazy blue color.

So, IS this normal behaviour or am I mis-configured?

chmod +x ./configure

Because I packaged it with zip, the exe permission is lost.

The cl file is not compatible with cgminer.

You need to build again, rename the cgminer to yacminer etc. and copy yacminer & scrypt-chacha.cl to /usr/local/bin then it will be OK.

I do already have it installed.  When I run ./configure I get "command not found"??
Do I just need to copy scryot-cacha.cl to my previosuly installed cgminer location and rebuild that (the one I use for LTC etc)

Ha, it's OK.

I do wonder what we could do, my 7970s are doing about 140s @ 1050/1500  --worksize 256 --lookup-gap 2 -I 11 --thread-concurrency 8192

1090/1500 gets 150

I just realized how low the GPU temps are

to tired i think it's 4:45 in the morning time to go to bed thanks for sharing the gpu miner