Cryptonote & Zawy difficulty algorithms

[zawy12]

I see that Forknote was trying to include my old simple moving average difficulty algorithm with N=17 like Sumokoin and Karbowanec used to fix their Cryptonote problem. The default Cryptonote difficulty algorithm with N=720 simple moving average reliably "breaks" new small coins by having unacceptable delays (a day or more to find a block). The N=17 of Zawy v1xx fixed the problem for Sumo and Karbo, but it does not perform well, and might be terrible if solvetimes are less than 240 seconds. A higher N like 30 to 60 would have been a lot better for all and not cause the delays like N=720.

I see there were a lot problems with the old Zawy algorithm attempts here, but I can't see if that was from the algorithm being too fast or from the rest of the code not meshing correctly with it.

5 Monero clones are having excellent success with my new algorithm, LWMA. Stellite reported it handled hash attacks with 52x the baseline hash rate without a problem. Masari has been using it since November.

I posted an issue to Cryptonote that more thoroughly discusses this, and gives another option for fixing Cryptonote more simply by lowering the N based on the target solvetime.

[pmitchev]

I saw your "issue" on Forknote too. We will move to your new algorithm (LWMA) as default when it's implemented and tested. Thank you for your amazing work

[zawy12]

Thanks for giving my new one a chance despite it appearing like my old one was causing problems.

I have a question. Qwertycoin and Bitsum seem to have used forknote, and they say they are using my algorithm. But in looking at their difficulty charts, their difficulty is changing very slowly as if they are using the default Cryptonote algorithm (neither the old or new zawy algos). Do you know what that's all about? Is there something (like the code or old documentation) that makes them think they are using mine?

[SoraKohaku]

thank now used LWMA