logically based on what you said at the beginning, it is time to open-source (and I paid a licence for your soft)
Do you realise that once we open source it, there won't be any more updates?
Actually I didn't realise that. Open Source in my opinion allows to bring new ideas.
But in view how it works with xpm (
the so called open source has 1 undisclosed binary, only after a few days... then some other part are used
for the other fully closed...), that may not be a great idea after all and not really what I was expecting lets be honnest...
Which one was that?
The last one... something working with a bin file in a patch isn't something I call open source.
If you are referring to:
https://bitcointalksearch.org/topic/m.6710729and
https://bitcointalksearch.org/topic/m.6714779You got it wrong. Do you not need the bin files, what are necessary are the .cl files; the bin files were provided for convenience .. as the communication wasn't that clear, it was not indicated what card the bin were compiled for. I suspected it could be for 290. Anyway you just need to put the .cl files (gpu folder) and run the client, it will automatically compile the .cl into .bin (though you need Catalyst 14 files even if you are using Catalyst 13.12 as the compiler produces error).
TL;DR version: The instruction might not have been provided by the dev (though I did in one of the post) but source codes were there.
The corresponding cl to that binary hasn't been released (it is a mix of the first open source which had a faster fermat code and the second open source if I understood well). Also there wouldn't be any profit if it was just a bin compiled for a different card...
Anyway, this would be my last post on other miners in primeGPU's thread. Further qns can be posted to eXtremal's thread.
So far, eXtremal has been opening up his sources so I doubt it is the case. The first file he released has the cl files for 10ch version. The patch_11_v2 has the cl files for 11ch version.
eXtremal said earlier:
Mining algorithm consists of two parts - sieve and Fermat test. madPrimeMiner have very fast sieve and slow non-optimized Fermat test, xpmclminer have too poor sieve, but fast Fermat test. I will merge two miners code for create fastest XPM miner.
Basically he just needed to improve madmax's Fermat test code, which was what he did, and it would be the better version compared to madmax and eXtremal's original codes. (xpmclminer was eXtremal open-source code, the first one before madMax released his).
