[ANN][BLC] Blakecoin Blake-256 for GPU/FPGA With Merged Mined Pools Stable Net

[kramble]

It finally finished compiling (I did say raspi was slow) seems to be OK

pi@tvpi ~/.blakecoin $ ./blakecoind getinfo
{
    "version" : 80900,
    "protocolversion" : 70001,
    "walletversion" : 60000,
    "balance" : 0.04261342,
    "blocks" : 39494,
    "timeoffset" : 0,
    "connections" : 3,
    "proxy" : "",
    "difficulty" : 3816.54040249,
    "testnet" : false,
    "keypoololdest" : 1386504023,
    "keypoolsize" : 101,
    "paytxfee" : 0.00000000,
    "errors" : ""
}

[BlueDragon747]

It finally finished compiling (I did say raspi was slow) seems to be OK

cool 

I did do some testing at least on the windows build last night and it does seem stable, no sync issues or dropping connections

but if anyone is getting an issue with 0.8.9 please report

[mogrith]

Just started it up. I'll keep an eye on it.
I'm also pool mining now. I've seen a few high Hash lines but only a few then back to normal.

[BlueDragon747]

Just started it up. I'll keep an eye on it.
I'm also pool mining now. I've seen a few high Hash lines but only a few then back to normal.

feedback on the wallet is always welcome, if you get a chance to try solo over the weekend let me know if it is any more stable

I am sure kr105 will fix the high hash issue on the pool soon   

[kr105]

Just started it up. I'll keep an eye on it.
I'm also pool mining now. I've seen a few high Hash lines but only a few then back to normal.

feedback on the wallet is always welcome, if you get a chance to try solo over the weekend let me know if it is any more stable

I am sure kr105 will fix the high hash issue on the pool soon   

I did some tweaks, I have not got a high-hash error since a lot of time, please let me know if you do.

[BlueDragon747]

Just started it up. I'll keep an eye on it.
I'm also pool mining now. I've seen a few high Hash lines but only a few then back to normal.

feedback on the wallet is always welcome, if you get a chance to try solo over the weekend let me know if it is any more stable

I am sure kr105 will fix the high hash issue on the pool soon   

I did some tweaks, I have not got a high-hash error since a lot of time, please let me know if you do.

sounds good I will give it some testing 

[kr105]

@ALL:
Please download the new version of cgminer from the pool's website: http://blakecoinpool.org/cgminer-blake256.rar and let me know how it goes the H-not-zero issue.

[kramble]

@ALL:
Please download the new version of cgminer from the pool's website: http://blakecoinpool.org/cgminer-blake256.rar and let me know how it goes the H-not-zero issue.

The rar just contains the windows executable. Could you update the github with the changes too please

PS The (high-hash) errors stopped at 3am (GMT) but I've been getting (H-not-zero) errors (1714 rejected in around 8.5 hours, compare 4215 accepted) using the fpga cgminer 3.1.1 (hence the request for the github update).

[boymok]

@ALL:
Please download the new version of cgminer from the pool's website: http://blakecoinpool.org/cgminer-blake256.rar and let me know how it goes the H-not-zero issue.

Much better here!

 (5s):2.030G (avg):2.039Gh/s | A:3375  R:8  HW:0  WU:55.6/m

[mogrith]

I was getting H not Zero errors fairly often. Switched to solo mining.

Still got disconnected on new wallet.

[kr105]

I was getting H not Zero errors fairly often. Switched to solo mining.

Still got disconnected on new wallet.

What miner are you using?

[BlueDragon747]

I was getting H not Zero errors fairly often. Switched to solo mining.

Still got disconnected on new wallet.

What miner are you using?

no source code update for cgminer 

no pool support for fpga 

[kr105]

New version of cgminer, this one is based on cgminer 3.1.1 so everyone with FPGA should be happy now. I updated the code of the 3.9 version just in case someone finds it useful.

Download:

https://github.com/kR105/cgminer-alt
https://www.blakecoinpool.org/cgminer-alt.rar

@BlueDragon747: Please use this links on the OP
@Everyone: Stop using the old versions of cgminer, use this one instead

[BlueDragon747]

New version of cgminer, this one is based on cgminer 3.1.1 so everyone with FPGA should be happy now. I updated the code of the 3.9 version just in case someone finds it useful.

Download:

https://github.com/kR105/cgminer-alt
https://www.blakecoinpool.org/cgminer-alt.rar

@BlueDragon747: Please use this links on the OP
@Everyone: Stop using the old versions of cgminer, use this one instead

the clock switches for ztex/cm1 are not working yet and some of the checks also fail on the fpga maybe kramble can help

much better to see binaries and source code

[mogrith]

yeah get this on starting

Code:

 [2013-12-20 14:29:40] Icarus Detect: Test failed at \\.\COM26: get 00000000, shou
ld: 000187a2
 [2013-12-20 14:29:40] Icarus Detect: Test failed at \\.\COM27: get 00000000, shou
ld: 000187a2
 [2013-12-20 14:29:41] Icarus Detect: Test failed at \\.\COM28: get 00000000, shou
ld: 000187a2
 [2013-12-20 14:29:41] Icarus Detect: Test failed at \\.\COM29: get 00000000, shou
ld: 000187a2
 [2013-12-20 14:29:43] No devices detected!

Correct USB ports for CM1

[kr105]

yeah get this on starting

Code:

 [2013-12-20 14:29:40] Icarus Detect: Test failed at \\.\COM26: get 00000000, shou
ld: 000187a2
 [2013-12-20 14:29:40] Icarus Detect: Test failed at \\.\COM27: get 00000000, shou
ld: 000187a2
 [2013-12-20 14:29:41] Icarus Detect: Test failed at \\.\COM28: get 00000000, shou
ld: 000187a2
 [2013-12-20 14:29:41] Icarus Detect: Test failed at \\.\COM29: get 00000000, shou
ld: 000187a2
 [2013-12-20 14:29:43] No devices detected!

Correct USB ports for CM1

Patches accepted Or point me in the right direction to fix it

[BlueDragon747]

yeah get this on starting

Code:

 [2013-12-20 14:29:40] Icarus Detect: Test failed at \\.\COM26: get 00000000, shou
ld: 000187a2
 [2013-12-20 14:29:40] Icarus Detect: Test failed at \\.\COM27: get 00000000, shou
ld: 000187a2
 [2013-12-20 14:29:41] Icarus Detect: Test failed at \\.\COM28: get 00000000, shou
ld: 000187a2
 [2013-12-20 14:29:41] Icarus Detect: Test failed at \\.\COM29: get 00000000, shou
ld: 000187a2
 [2013-12-20 14:29:43] No devices detected!

Correct USB ports for CM1

Patches accepted Or point me in the right direction to fix it

some kramble magic is needed 

[kramble]

some kramble magic is needed  

I'm a bit short of time right now, it probably won't be until Sunday before I can devote significant time to this.

The CM1 is fairly easy to fix, just need to use the correct golden_ob/nonce values, see (link) lines 618-625 of driver-icarus.c

The clock switches are a bit more involved. It should work OK without, though CM1 will be a little slow and the ztex may hunt around the optimal clock. Unfortunately it will take more than just copying the driver code into kr105's 3.1.1 as the switches are coded into miner.h and cgminer.c

EDIT. Just did a quick scan through https://github.com/kR105/cgminer-alt and concluded that this will need a fair bit of work to get the ztex to run as there are changes to libztex.c (just the clock defaults, so it may be OK as it is) and more importantly driver-ztex.c where the ztex_checkNonce() function needs to be replaced with the blake hash (just commenting it out may also work, but will probably break the automatic frequency management which relies on HW error counts). It may be worth just copying my driver code over the clean code in kr105's 3.1.1 and fixing up the ztex_checkNonce() which I'd done a quick'n'dirty hack on by moving it into cgminer.c (should no longer be needed if the makefile has been updated so blake is available globally rather than my hack of #including it into cgminer.c).

[mogrith]

No Rush on my part.

If I see too many errors on pool I'll go back solo. But so far it's not too bad on pool with the kramble miner

[kramble]

No Rush on my part.

If I see too many errors on pool I'll go back solo. But so far it's not too bad on pool with the kramble miner

I'm losing perhaps 25% of shares (from the stats I checked earlier today, not sure how its going right now). I'll take a look at what's been changed later (Sunday most likely). It may be best if I fork kr105's code, add in my driver changes and git-pull them back to kr105 so as to keep a single consistent code base. Anyway I'll look into it as soon as I find time. Gotta go now. Bye.