Quote from: jonnybravo0311 on November 24, 2015, 02:34:53 PM

Quote from: sawa on November 24, 2015, 02:28:14 PM

Quote from: -ck on November 24, 2015, 01:35:21 PM

Quote from: wariner on November 24, 2015, 01:19:47 PM

Quote from: -ck on November 24, 2015, 11:44:38 AM

Runs fine yes, but the bitmain driver throws away what could be potential block solves on p2pool...

the problem is only for s7? or even s5 - s3....?

There is a discussion that contains all modified antminer drivers for p2pool?

Except for Kano's firmware and my drivers for S1-3, all antminer S* cgminer forks from bitmain have the same driver design flaw. We mention it at regular intervals but it keeps getting forgotten or lost in the discussion by p2pool miners...

Maybe does not make sense to do firmware forks for S *, so we are constantly catching up and more and more losing hashrate.
Now it is so small that it can be said that p2pool is dead.
And many of those who get ASICs, can never replace firmware and getting out on p2pool they switch to a another type of pools.

Maybe it's time for develpers to join and modify and adapt it for the new ASICs? Perhaps it's needed to rewrite it on C, as in Python it loads only one core and often on 100%
We need creative solutions, or this topic can be closed and we should stop our nodes and move to antpool, f2pool and similar.

We should absolutely not, in any way, shape or form, EVER move to AntPool or f2pool. Not only are those pools way too big as it is, but they are also poorly written and take shortcuts that are bad for Bitcoin. In case you don't remember, they caused a fork in BTC earlier this year precisely because of their shoddy software and mining practices.

I agree with you, but most people do not understand this.

Quote from: jonnybravo0311 on November 24, 2015, 02:34:53 PM

As to your other points, we've discussed numerous times rewriting p2pool in a language that can handle multiple threads. Thus far, nobody's stepped up to take on that challenge. Even so, rewriting the pool code in C or some other language will not handle the underlying problem of crappy cgminer forks not submitting shares which could be solving blocks.

I'm not sure, but why don't we raise funds for the development team that can do this (maybe it's a bad idea).
I think we should act more active rather than passively wait for volunteers.