I've been using threads for years now in various languages (mainly Java (painless) and C++ (slightly less so)).
There is a serious learning curve on threads (some of my team still stuggle depsite working on a very multithreaded application for a couple of years now) but I would suggest that once you get over the curve threads are no more difficult or unreliable than something like pointers.
The comparision with pointers is a good one actually - remember when you first started using them - all those mysterious GP faults, odd behaviour from reading and writing the wrong bit of memory etc. You soon get to the point where that kind of thing is rare - I think the same is true of threads once you get your head around them they are just as straight forward.
cheers
mark
-----Original Message----- From: Olaf Matthes [mailto:olaf.matthes@gmx.de] Sent: 24 June 2002 12:40 To: pd-list Subject: Re: [PD] we want threads !!!!
Miller Puckette schrieb:
Yes, I'm scared to death of threads. The readsf~/writesf~ objects do use them (there's no alternative) but my experience in general is that
it's very hard to maintain threaded code if you want a high level of reliability (which I do in Pd)...
The threaded versions of my ogg vorbis streaming externals are largely based on the readsf~/writesf~ code. It took me some time to fix all the bugs in my code but after that the externals now work more reliable than the unthreaded ones.
There's also a (partly) threaded netsend available at http://www.akustische-kunst.de/puredata/winport.html for those who haven't noticed it. Maxlib's netdist uses multithreading as well...
Olaf