Beta vs latest protein speed question

[Chinasaur]

All you testers.

I haven't been testing the Beta..and can't find what I'm looking for in the DF forum...I know the DF engine is changing which changes the crunching methodology.

Question: Will protein size matter "speed wise" with the new engine? Like it does with the current engine?

TIA :)


:cheers:

Have a good Friday night and a great weekend!!!

[bwkaz]

I don't think anyone knows. :p

All of the betas so far have been on the same protein (the 96AA one that the normal client was using when the beta split off). I would assume there would be some impact (if there are more amino acids, there are more calculations to do, regardless of algorithm -- you still have to place each of them, so the algorithm CAN'T be better than O(n)), but I don't know if there's more or less of an impact compared to the normal client.

[Paratima]

Yeah. What he said.

You do 5,000 iterations just like the old client. It somehow picks one it likes and you process or "drill down" that one to death, each "generation" after the first being 50 folds. Even on the 96AA protein we were working on, a generation could take 10 minutes or a day and a half! :eek:

Of course the size and complexity of the target will have an effect, but as kaz says, it's not a sure thing either way.