Aaand it's out:
https://github.com/tsiv/ccminerInsert the standard "if your house burns down, totally not my fault" disclaimer here.
Windows build should be fine as well, at least I managed to compile it myself. I won't be releasing any binaries, feel free to do so yourselves though.
Christ gonna release killer goretl version probably before or during weekend. The new version will upgrade x11 speed gtx 750 ti to 2.8 - 3. Maybe u should wait for the new release and implement your code which will increase x13 speed significantly. Or maybe both of u can combine code and do a combine release ? So the new version will have killer goretl + x13. That is pretty exciting.
Should be an easy modification, right now my X13 is using the same code for the first 11 algos as the original X11. At the simplest it just takes a recompile after throwing in the new groestl version.
no reason to wait
Actually I just curious to know what I did wrong
I have a very good idea about that
I did end up messing with the original echo hash, it seemed to have an extraneous xor that made the full 512 bit hash come out wrong. Then using that result for the additional hamsi+fugue hashes kinda ends badly. The funny thing is that it worked fine for X11 because it was the last hashing algo in the chain and the X11 algo only takes the first 256 bits of the hash, discarding the rest and the flipped bit ended up in the discarded part. Was a bitch to track down for sure.
ok, it's working.
Actually, I had a couple of other mistakes regarding bit swapping (I am rather bit illiterate...), which could have been solved by random trials
but for the echo, it was clearly outside my level of competence (however I was suspecting something funny, as it wasn't possible to add new algo on top of this one).
In terms of speed I am getting between 1700khah/s and 1800khash/s with the 750ti.
I found something else interesting, it is possible to run on X13 p2pool with X11 algo
and to get validated share (not sure if it works for the "big" share, but for the rest it works)
(it's compute_35 but with the constant at 1)
