AMD OpenCL Momentum miner

[NewMine]

Is failover the only new update to the miner? Because my col/min dropped about 30% with this new one and share/min was even lower.

[Mario24]

ok

but for 290X is better Options a 1 (3250c/m) as a 0 (3050c/m)

[NaN]

ok its too fast

but with v0.2.2 it was not so

It is the same "issue" with the old miner, but it did not report any reconnects because it is based on the jhProtominer code.

[wsnnwa]

I will be giving this a try within the next couple of days. It looks like I will see a hefty speed increase on all my cards (280x/290x).

[Mario24]

ok its too fast

but with v0.2.2 it was not so

[NaN]

That's not an issue of the miner because the connection is closed by ypool.net.

[Mario24]

I'm working on stratum support. Which pool do you use because the miner should reconnect by itself only every 30min but then twice within 36s if you do not use ypool? If ypool is used then the miner does not reconnect by itself unless the server closes the connection or an error occurs. As I mentioned earlier some servers close the connection occasionally.

i'm using ypool.net too. is about separately every 15 minutes, I think

clpts_x86-64.exe -o 6 -u workername.1 -p x -t 0.0 -a 1


and 15min late

[Bigboppa]

To give some feedback:
Running fine, but only at catalyst 14.3beta (as told before).
I got 7150 cpm with 2x 280x @ default clock, win7-64 at ypool (-t 0,0,1,1 -a 2)
Thank you for the good work! 

[NaN]

I'm working on stratum support. Which pool do you use because the miner should reconnect by itself only every 30min but then twice within 36s if you do not use ypool? If ypool is used then the miner does not reconnect by itself unless the server closes the connection or an error occurs. As I mentioned earlier some servers close the connection occasionally.

[Mario24]

the miner reconect the pool every 15 minutes? Thats right?

after 8 hours 2421 c/m v1.0a


nrs or stratum support would be great

[NaN]

Ok, I will install Catalyst 13.12 and create a version for older Catalyst drivers.
EDIT: done

[Mario24]

great stuff v1.0a

i have hd7970 3gb on ypool.net win7 64x 1000 mhz core and 1325 mhz memory and newest beta driver from amd

with v0.2.2 was the sharerate 175-180 share/h option a1 and 1 thread

with v1.0a now i get 250-275 share/h preference is the same

2400 col before 1600 col

[stanmin]

No stratum/getwork support? I was hoping for those so I could use the this miner for NoirShares as well. Would you be interested in supporting NoirShares along with PTS? It's the same PoW algo, but it's lacking xpt pools.

[Ykw]

The new version v1.0 with multi-pool support and pool fallback. I'm sorry that the miner is late but I had not much time in the last weeks.

Thanks man...

Does it need the new beta AMD SDK?

I am getting this now.. (with previous miner I had to do nothing):
*****************************************************************************
** clpts - OpenCL PTS Pool Miner v1.0 by NaN
**
** If you like this software, please consider sending tips to:
         opencl.miner@gmail.com
** PTS:  PajuVUWrcPn5ksCFfrBiWQXvpwtE29dcN2
** BTC:  1AumJ5uzz1nuER7pBA6Bh4gNusaxhN85rc
** Your donations will encourage further optimization and development
**
** press CTRL+C to exit
*****************************************************************************
Vendor of used platform (#1 / 1): Advanced Micro Devices, Inc.
        Name of device #1 / 2 (deviceID 0): Hawaii
        Name of device #2 / 2 (deviceID 1): Pitcairn


[MASTER] | Spawning 2 worker threads
[WORKER[WORKER1] Hello, World!
0] Hello, World!


Error during compilation of program for device 'Pitcairn':

Internal error: Input OpenCL binary is not for the target!


EDIT: It only give me that for: device #2 / 2 (deviceID 1): Pitcairn

Win 7k SP1 + latest non-beta AMD drivers

Oh, I published the binaries instead of the LLVMIR code... Perhaps you can just use the old gpuhash_* files. I'm working on a fix. edit: fixed

Using the version 1.0a I just get the same error as user justmwa for both cards, using AMD 13.12 drivers with win7 SP1 x64:

Hi,
I also have

Code: [Select]

Error during compilation of program for device 'Tahiti':

Frontend phase failed compilation.
Error: Compilation from LLVMIR binary to IL text failed!
with 13.12 on Ubuntu.
Maybe adding a note to use beta catalysts with v1.0 is in order? 
I will wait before switching 'til you get your hands on a Tahiti card and fix perf if needed.

If I switch the "gpuhash_*" files for the "clpts-v0.1.4_win_x86-64" version... they work fine. Same performance.. my best run over the 290X is using -a 1 (not 0)

[femida]

When you plan to support Nvidia GPU?