Hello,

after a few months break I am back and running boinc/rosetta a much as possible. My cpu is really not thaaaaaat super powerful anymore, but as I am not getting a new cpu soon, this is currently all I can do.

I want to help more and I don't know how.
In other projects I would start to hack the code... but there is no code... why?

I know I can code, and I know there are LOTS of people who can code even more and better, why can't they help you? As I think this is a very important project you could only benefit from more help, or? This would not only be beneficial for you, but maybe also for the whole mankind as this could lead to faster problem solving and maybe faster healing diseases.

I just can not understand why you demand on your closedsource client. Maybe you should rethink opening your sourcecode. I just want to help you...

Hello,

after a few months break I am back and running boinc/rosetta a much as possible. My cpu is really not thaaaaaat super powerful anymore, but as I am not getting a new cpu soon, this is currently all I can do.

I want to help more and I don't know how.
In other projects I would start to hack the code... but there is no code... why?

I know I can code, and I know there are LOTS of people who can code even more and better, why can't they help you? As I think this is a very important project you could only benefit from more help, or? This would not only be beneficial for you, but maybe also for the whole mankind as this could lead to faster problem solving and maybe faster healing diseases.

I just can not understand why you demand on your closedsource client. Maybe you should rethink opening your sourcecode. I just want to help you...

It's not closed source - it is worked on by labs around the world. The code base has to be tightly controlled so that it remains managable. I presume you would need to understand computational biochemistry and be in contact with other developers to be able to usefully contribute though, so it wouldn't be a casual investment.

If you're serious and could spare the time then there might be things that they could give you (and others) as a project such as trying to improve the efficiency of a module, by porting it to GPU for example. Whether that's viable, or even desireable I don't know. My understanding is that the focus is mainly on improving the accuracy of the code rather than optimising its speed.

---

Then let me rephrase, it is open to me or any other developer who wants to help?
-> Nope.

I am not even talking about that everyone can use its own client to push data to rosetta, I would still prefer to get the official binarys from rosetta directly.

I am also not talking about making super huge changes here. But how about just investigating? like a crashing testdata, or maybe something that makes the client "stuck"?

I can't do that without the code. Or rather is super painful without the code.

So I just want make the life a tiny bit more easier for them. But a performance optimization may also be possible, but as you said too, thats not the main focus for now.
I am also kinda worried that rosetta currently misses the arm cpu train, as those devices get more and more powerful and are widely spreed. Well thats another topic... but "someone" could do the porting, which in best case is rather minimal and would almost change nothing. But without code....