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

[mogrith]

I have 3 X6500 FPGAs I am willing to part with for cheap.

@kramble: did you get these to work? At what speed?

They are a pain. you have to load the bitstreams all the time.

[BlueDragon747]

just finished
http://eu2.blakecoin.com
http://sg1.blakecoin.com

you need to re register on these pools as it is using the new MPOS and has email based security

[Ignatius]

I have 3 X6500 FPGAs I am willing to part with for cheap.

@kramble: did you get these to work? At what speed?

They are a pain. you have to load the bitstreams all the time.

I wonder why others are experiencing issues with the x6500 boards and blake? I have only loaded the bitstream once and have not experienced any issues with them. Speed ~400MHs.

Edit: Using krambles miner.py, his X6500-Robust-v04-2core-fmax-100MHz.bit, against eloipool-blake.

[mogrith]

I have 3 X6500 FPGAs I am willing to part with for cheap.

@kramble: did you get these to work? At what speed?

They are a pain. you have to load the bitstreams all the time.

I wonder why others are experiencing issues with the x6500 boards and blake? I have only loaded the bitstream once and have not experienced any issues with them. Speed ~400MHs.

Edit: Using krambles miner.py, his X6500-Robust-v04-2core-fmax-100MHz.bit, against eloipool-blake.

I'll try again What driver do I need to do to allow ubuntu to see the x6500 I do not see an entry in the /dev folder.

[Ignatius]

I'll try again What driver do I need to do to allow ubuntu to see the x6500 I do not see an entry in the /dev folder.

I failed to get an x6500 running with a Win 7 64b host. I have yet to try it under linux. What has been working fine for me is a 32b Windows XP virtual machine, python 2.6, and PyUSB 1.6 for python 2.6.

I just tried to get it going on Ubuntu 12.04.3 and failed. "No module named d2xx".

 

[kramble]

I wonder why others are experiencing issues with the x6500 boards and blake? I have only loaded the bitstream once and have not experienced any issues with them. Speed ~400MHs.

Edit: Using krambles miner.py, his X6500-Robust-v04-2core-fmax-100MHz.bit, against eloipool-blake.

Thanks for the report. It would be useful to put together a guide to how to set these up.

I assume you are using my modified X6500 python miner (link). Does this work against the standard pools? I did get reports that it was not working (the script expects getwork to supply midstate, even though it's not actually needed, and it gives an error). I think I've got a fix for this (see post), but I'd want to get confirmation before updating the github.

[Ignatius]

Thanks for the report. It would be useful to put together a guide to how to set these up.

I assume you are using my modified X6500 python miner (link). Does this work against the standard pools? I did get reports that it was not working (the script expects getwork to supply midstate, even though it's not actually needed, and it gives an error). I think I've got a fix for this (see post), but I'd want to get confirmation before updating the github.

Yes I am using your mine.py. I did make one edit to it per what you mentioned in this thread. I did not notice any issue, blocks were being found and submitted, but since you said it couldn't hurt to uncomment the line, I did so.

I did solo-mine, would have found two blocks.

But the base difficulty and golden nonce are off in mine.py for the coin so it counted the solved blocks as rejected.

There is an easy work-around, just uncomment line 104 of mine.py. This will submit all shares regardless of the difficulty target, which is harmless when solo mining, and since (AFAIK, Blue may want to chip in here) the blakecoin pools are currently fixed diff=1 (even though the dashboard may report higher), it should be OK there too.

A slightly more sophisticated fix would retain the subsequent CheckTarget() and logger calls, but always return True.

I have not tried to mine against any of the public pools, I will go try now and report back. I find it odd it wouldn't work, since I run against a private eloipool-blakecoin build, which afaik is the same software the pools use.

EDIT: Tested, I was unable to mine against ny1.blakecoin.com. "Error getting work! Retrying"

[Jude Austin]

Yes, a write up would be great.

I would much rather mine BLC than sell them.

On my Windows 7 Laptop, I can't even get the program.py to recognize the X6500.

Keeps saying no FT232 device found.

[mogrith]

Yes, a write up would be great.

I would much rather mine BLC than sell them.

On my Windows 7 Laptop, I can't even get the program.py to recognize the X6500.

Keeps saying no FT232 device found.

Same issue I had on win7 64.

[Ignatius]

Yes, a write up would be great.

I would much rather mine BLC than sell them.

On my Windows 7 Laptop, I can't even get the program.py to recognize the X6500.

Keeps saying no FT232 device found.

Nor could I, so I tried a 32b Windows XP virtual machine, and it worked. What mining software did you previously use with x6500's? I used MPBM, and could even write the blake bitstream to the x6500 via MPBM...it just wouldn't mine after(obviously, mpbm being unmodified for blake).

[Jude Austin]

Yes, a write up would be great.

I would much rather mine BLC than sell them.

On my Windows 7 Laptop, I can't even get the program.py to recognize the X6500.

Keeps saying no FT232 device found.

Nor could I, so I tried a 32b Windows XP virtual machine, and it worked. What mining software did you previously use with x6500's? I used MPBM, and could even write the blake bitstream to the x6500 via MPBM...it just wouldn't mine after(obviously, mpbm being unmodified for blake).

I used MPBM.

Perhaps I will load the bitstream with MPBM like you did and try running miner.py.

I had it running at one point on my Windows 8.1 machine but every time it would submit a block it would get rejected, and it would not get shares from a pool either.

[kramble]

Yes, a write up would be great.

I would much rather mine BLC than sell them.

I think we almost got there a couple of weeks back ...

I will try shortly.

No problem, its still early evening here so I've got a while.

I think I may have found the problem (other than the http:// prefix). The x6500 miner is expecting the pool to send midstate, which is missing. I think we can get away with commenting out line 436 of fpga.py vis:
      #job.midstate = work['midstate']

Or if that doesn't work, set it to something benign, eg
      job.midstate = work['target']

The driver problem needs to be sorted out first though. It's getting a little late here though, so I'm not going to be online very much longer tonight.

[kramble]

I used MPBM.

Perhaps I will load the bitstream with MPBM like you did and try running miner.py.

I had it running at one point on my Windows 8.1 machine but every time it would submit a block it would get rejected, and it would not get shares from a pool either.

I did a quick hack of MPBM to support Blakecoin way back (link), NB it's the master branch, not the default testing. I don't know if it works though (it probably needs some tweaking to work on pool rather than solo). I couldn't even get the original bitcoin version to work nicely with my browsers (firefox, IE).

[Ignatius]

No problem, its still early evening here so I've got a while.

I think I may have found the problem (other than the http:// prefix). The x6500 miner is expecting the pool to send midstate, which is missing. I think we can get away with commenting out line 436 of fpga.py vis:
      #job.midstate = work['midstate']

Or if that doesn't work, set it to something benign, eg
      job.midstate = work['target']

The driver problem needs to be sorted out first though. It's getting a little late here though, so I'm not going to be online very much longer tonight.

NOTE: This works!

Commenting out line 436 of fpga.py allows me to connect and mine against ny1.blakecoin.com, whereas I previously could not do so.

[Ignatius]

I did a quick hack of MPBM to support Blakecoin way back (link), NB it's the master branch, not the default testing. I don't know if it works though (it probably needs some tweaking to work on pool rather than solo). I couldn't even get the original bitcoin version to work nicely with my browsers (firefox, IE).

I have always had to use Chrome to get MPBM to display properly. If I get time I will try the modified mpbm.

[kramble]

No problem, its still early evening here so I've got a while.

I think I may have found the problem (other than the http:// prefix). The x6500 miner is expecting the pool to send midstate, which is missing. I think we can get away with commenting out line 436 of fpga.py vis:
      #job.midstate = work['midstate']

Or if that doesn't work, set it to something benign, eg
      job.midstate = work['target']

The driver problem needs to be sorted out first though. It's getting a little late here though, so I'm not going to be online very much longer tonight.

NOTE: This works!

Commenting out line 436 of fpga.py allows me to connect and mine against ny1.blakecoin.com, whereas I previously could not do so.

Thanks! I'll update the github version with those two changes (it'll be tomorrow as I'm a bit tired tonight so I don't want to risk messing it up).

[Jude Austin]

Guess who's mining BLC?

ME!

Haha, thanks Ignatius, Mogrith, Kramble, and BlueDragon for being very supportive!

[mogrith]

Guess who's mining BLC?

ME!

Haha, thanks Ignatius, Mogrith, Kramble, and BlueDragon for being very supportive!

Win32? linux? or what?

I guess that means your not selling x6500's now? 

[Jude Austin]

Guess who's mining BLC?

ME!

Haha, thanks Ignatius, Mogrith, Kramble, and BlueDragon for being very supportive!

Win32? linux? or what?

I guess that means your not selling x6500's now? 

Windows 8.1

Krambles X6500 stuff, with the line 436 commented out in FPGA.py.

Also had to remove http:// from the address.

And I also had to pull the usb on the x6500 after programming to get miner to connect.

[Ignatius]

I used MPBM.

Perhaps I will load the bitstream with MPBM like you did and try running miner.py.

I had it running at one point on my Windows 8.1 machine but every time it would submit a block it would get rejected, and it would not get shares from a pool either.

I did a quick hack of MPBM to support Blakecoin way back (link), NB it's the master branch, not the default testing. I don't know if it works though (it probably needs some tweaking to work on pool rather than solo). I couldn't even get the original bitcoin version to work nicely with my browsers (firefox, IE).

Your modified MPBM worked for me with a couple changes.

I overwrote my last working copy of mpbm with your updated files.
I added a blakecoin getwork source. (stratum doesn't work, old divide by zero bug between mpbm <-> eloi)
I had to copy the blake8.py file found in the /mpbm/core/ directory to the /mpbm/ directory. (The directory which contains mpbm.exe, run-mpbm.py) The need for this may be a simple path typo, I haven't checked the code yet.
I changed line 431 of /mpbm/modules/fpgamining/x6500/x6500worker.py from:
self.checksuccess = False
to:
self.checksuccess = True

With just these modifications the x6500 successfully submits shares to a pool. The hashrate is reported incorrectly. Just a quick and dirty copy pasta to get mpbm going in windows with blakecoin firmware from kramble and an x6500. Thanks for all your efforts kramble.