cpu load will not make the thing freeze, it will just make bad audio , and potentially slow to react)
generally as i stated in the top post, I need much more detail than this…
what were you doing at the time it froze?, what had you being do before? can you reproduce it?
if you can’t reproduce ,then neither can it, and that means we don’t get it fixed…
(in your case, I tried your chain, and it worked fine for me)
let me give some background, as I think some thing developer should ‘find their own bugs’, and i can understand that viewpoint
but I cannot sit here for hours, trying every possible combination, that the user might have done, to see if I can get it to crash
lets remember too, whilst I developed Orac, I tested orac thoroughly as i went along, I also tested every module… so I know for the most part its stable. as you can be sure, I tested many obvious, and less obvious scenarios.
(thats not to say there aren’t any bugs , just that its likely bugs are caused by more complex combinations)
why I can’t I test every combination?
you think 1008 has a lot of combination? - theres over 970,000,000,000,000,000 module combinations (50^10) in orac, and that doesn’t include the parameters combinations ( i dont even know how to say that number )
ok, thats a silly number,
but, the point is with the complexity of Orac, if I don’t get bug reports which are clear and accurate about what was being done, how things were setup, and how you can repeat a problem - Ive very little chance (on my own!) of reproducing, let alone fix a bug.
let me give you a concrete example… the S-czz bug I found, i discovered because I had a crash/freeze, like your reporting. (id been testing lmnts at the time, due to bugs reported)
So, then I went back and tried the same combo, it didnt crash,
so I tried to remember what Id done when it crash, initially I couldn’t,
so I kept trying different combinations, rarely after a while it might crash.
then one time, I realised that it had crashed twice, but Id been loading two different modules.
so i now suspected, it was highly unlikely, it was the module being loaded… rather it must be the module I was switching out/unloading.
viola, now i had a hypothesis, and it started to become clear it was S-czz.
the thing to note here is… this was not looking at any code… or patch, this was something anyone here could have found.
… of course, it then still took me about 30 minutes, to look at the code and detemine the cause,
so even if you submit the perfect bug report, be under no illusion - Ive still go my work cut out!
BUT if this had just reported that it crashed when loading LMNTS (and I suspect this has happened) , this would lead me down the garden path, as the bug had nothing to do with LMNTS… so Id waste hours trying to crash LMNTS and might never even load S-czz and get the real bug.
this is why developers need users to help, to report clearly, to really think about what might have caused the issues - this kind of community effort can really help to get rid of pesky bugs.