Well thanks to difficulties I'm finally in on the beginning of a launch, all set up and ready to fly. So, am I right in assuming that we're waiting on the new wallet before firing up the miners? (poor dev, let's all buy him a brain-pillow).
Yes, a new Wallet version 0.3.6 with yet another hard fork is coming later today. Please, hold your miners till this version is released.
Saw this discussion by chance and thought to give you some input from discussions on another coin.
Main point is, how much is the hashrate changing?
If the changes are rather decent, everything from DS, KWG, DGW to DGW3 should do.
DGW3 i.e. uses the last 24 blocks (can be easily modified) to recalculate the diff every block (like most modern algos do to react quicker).
It has a minor 'one-off' bug, but nothing serious and easy to fix. Even the btc algo had a bug like this at one point.
If you see big jumps in hashrate, none of the actual algos are helping you here.
There is a dev building a hashrate simulator to develop and test new diff algos for this problem.
If you need more info, you can PM me because I don't want to take over your thread and I think I saw some familiar names from the other thread here too.
Thanks for the info.
But, I think I found why the PID was acting up. I wasn't averaging Block time using the correct set of TX times.
Like what my teacher of Control Theory course in my University used to say about controlling systems using feedback loops .... If your sensors aren't reporting correctly, you won't be doing squat.