[Csnd] --realtime flag
Date | 2018-03-02 16:22 |
From | Rene |
Subject | [Csnd] --realtime flag |
Attachments | reinit.csd |
Hello, Instruments using reinit opcode and --realtime flag seems to have a faulty behavior.The reinit is done for the complete instrument regardless of the reinit label and rireturn locations. |
Date | 2018-03-02 17:07 |
From | Victor Lazzarini |
Subject | Re: [Csnd] --realtime flag |
It could be a bug. Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland
|
Date | 2018-03-02 17:48 |
From | Victor Lazzarini |
Subject | Re: [Csnd] --realtime flag |
Actually, can you try the latest (6.11)? This version has a completely overhauled realtime mode, and I remembered testing reinit and it worked.
Victor Lazzarini
Dean of Arts, Celtic Studies, and Philosophy
Maynooth University
Ireland
|
Date | 2018-03-02 19:26 |
From | Rene |
Subject | Re: [Csnd] --realtime flag |
Hello, Tanks for your answer. I have tested from git (today download) csound-master : give same bug as 6.10.0 csound-develop and csound-feature-realtime : (may be wrong version, i can't find 6.11) WARNING: instr 0 uses 3 p-fields but is given 0 then csound hang. Best, René 2018-03-02 18:48 GMT+01:00 Victor Lazzarini <Victor.Lazzarini@mu.ie>:
|
Date | 2018-03-02 20:03 |
From | Victor Lazzarini |
Subject | Re: [Csnd] --realtime flag |
The master branch is 6.10, old code. The develop branch is 6.11, e.g. Csound version 6.11 beta (double samples) Mar 1 2018 Feature-realtime has been merged into develop, so that would have similar code. I tested your code and I am getting the WARNING, which I think it is a bug in the latest checked in code. That problem is there with or without —reatlime Once that is fixed, you can test it again. ======================== Prof. Victor Lazzarini Dean of Arts, Celtic Studies, and Philosophy, Maynooth University, Maynooth, Co Kildare, Ireland Tel: 00 353 7086936 Fax: 00 353 1 7086952 > On 2 Mar 2018, at 19:26, Rene |
Date | 2018-03-02 20:09 |
From | Victor Lazzarini |
Subject | Re: [Csnd] --realtime flag |
actually, it’s a problem with realtime. I’ll fix it. I think it is to do with timout, not reinit. ======================== Prof. Victor Lazzarini Dean of Arts, Celtic Studies, and Philosophy, Maynooth University, Maynooth, Co Kildare, Ireland Tel: 00 353 7086936 Fax: 00 353 1 7086952 > On 2 Mar 2018, at 19:26, Rene |
Date | 2018-03-02 21:29 |
From | Victor Lazzarini |
Subject | Re: [Csnd] --realtime flag |
I have fixed this now (in git develop), but there are a couple of caveats: 1) reinit should be avoided in --realtime mode: it stops performance to run a reinit pass, and that is not a good idea. Even though it runs the reinit on a separate thread, performance has to pause for reinitialisation. 2) opcodes like timeout would potentially have gaps in their time counting due to the reinit pause. In other words: although the instrument will work (with maybe some timout time counting issues), it’s not a good realtime instrument. ======================== Prof. Victor Lazzarini Dean of Arts, Celtic Studies, and Philosophy, Maynooth University, Maynooth, Co Kildare, Ireland Tel: 00 353 7086936 Fax: 00 353 1 7086952 > On 2 Mar 2018, at 19:26, Rene |
Date | 2018-03-03 18:51 |
From | Rene |
Subject | Re: [Csnd] --realtime flag |
Tested git develop, it is fine, and thanks for the advise :)2018-03-02 22:29 GMT+01:00 Victor Lazzarini <Victor.Lazzarini@mu.ie>: I have fixed this now (in git develop), but there are a couple of caveats: |