Hello,
I couldn't have bonk working with last release of pd-extended under windows vista, the object is created but it doesn't seem to receive audio signal, it's happening with any release of bonk~ external.
Colet Patrice
Hello Patrice,
Can you check if [sqrt~] works in the releases where [bonk~] doesn't work? [bonk~] uses the same optimized sqrt code, and there has been something wrong with that recently.
Katja
On Sun, Dec 25, 2011 at 10:26 PM, Patrice Colet colet.patrice@free.fr wrote:
Hello,
I couldn't have bonk working with last release of pd-extended under windows vista, the object is created but it doesn't seem to receive audio signal, it's happening with any release of bonk~ external.
Colet Patrice
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
[sqrt~] have same behavior, nothing flows out from outlet.
Colet Patrice
----- Mail original -----
De: "katja" katjavetter@gmail.com À: pd-list@iem.at Envoyé: Lundi 26 Décembre 2011 00:09:07 Objet: Re: [PD] pdx+bonk~
Hello Patrice,
Can you check if [sqrt~] works in the releases where [bonk~] doesn't work? [bonk~] uses the same optimized sqrt code, and there has been something wrong with that recently.
Katja
On Sun, Dec 25, 2011 at 10:26 PM, Patrice Colet colet.patrice@free.fr wrote:
Hello,
I couldn't have bonk working with last release of pd-extended under windows vista, the object is created but it doesn't seem to receive audio signal, it's happening with any release of bonk~ external.
Colet Patrice
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
On Tue, Dec 27, 2011 at 4:07 PM, Patrice Colet colet.patrice@free.fr wrote:
[sqrt~] have same behavior, nothing flows out from outlet.
These problems may well be related. I submitted a bug report for [sqrt~] some time ago (http://sourceforge.net/tracker/?func=detail&aid=3442082&group_id=557...). It is fixed, at least for OSX10.5 i386, where both [sqrt~] and [bonk~] work in the latest autobuild (20111221). Which builds did you check?
Katja
----- Mail original -----
De: "katja" katjavetter@gmail.com À: pd-list@iem.at Envoyé: Mardi 27 Décembre 2011 16:24:56 Objet: Re: [PD] pdx+bonk~
On Tue, Dec 27, 2011 at 4:07 PM, Patrice Colet colet.patrice@free.fr wrote:
[sqrt~] have same behavior, nothing flows out from outlet.
These problems may well be related. I submitted a bug report for [sqrt~] some time ago (http://sourceforge.net/tracker/?func=detail&aid=3442082&group_id=557...). It is fixed, at least for OSX10.5 i386, where both [sqrt~] and [bonk~] work in the latest autobuild (20111221). Which builds did you check?
I've been using an older release, with the very last one, bonk~ is okay, thank you.
Katja
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Le 2011-12-27 à 16:07:00, Patrice Colet a écrit :
[sqrt~] have same behavior, nothing flows out from outlet.
You could modify init_rsqrt in d_math.c to make sure that it does write stuff in the table. If the tables are not blank, then the problem is in the two functions after it.
BTW, in those functions, «long» may be 32 or 64 bits, whereas «float» is just 32 bits, and this looks like the kind of thing that would fail on a big-endian 64-bit machine, no ? But that's a separate problem : win64 has a 32-bit long.
| Mathieu BOUCHARD ----- téléphone : +1.514.383.3801 ----- Montréal, QC
----- Mail original -----
De: "Mathieu Bouchard" matju@artengine.ca À: "Patrice Colet" colet.patrice@free.fr Cc: "pd-list" pd-list@iem.at Envoyé: Mardi 27 Décembre 2011 18:13:02 Objet: Re: [PD] pdx+bonk~
Le 2011-12-27 à 16:07:00, Patrice Colet a écrit :
[sqrt~] have same behavior, nothing flows out from outlet.
You could modify init_rsqrt in d_math.c to make sure that it does write stuff in the table. If the tables are not blank, then the problem is in the two functions after it.
d_math.c doesn't seem to be used anymore in pdx, the code is scattered into extra/vanilla and half is missing into sqrt~.c, so it's quite hard to see what's happening, but I'm giving a try, thanks for the tip.
| Mathieu BOUCHARD ----- téléphone : +1.514.383.3801 ----- Montréal, | QC
Le 2011-12-28 à 03:02:00, Patrice Colet a écrit :
d_math.c doesn't seem to be used anymore in pdx, the code is scattered into extra/vanilla and half is missing into sqrt~.c, so it's quite hard to see what's happening, but I'm giving a try, thanks for the tip.
Well, is d_math.c still compiled in pd-extended ? I bet that it is. (I haven't checked out the git version yet)
I bet that there are two copies of the same table, and at least one is not initialised, if not both, and [bonk~] calls the wrong functions with the wrong tables (if any table even happens to be right).
Miller's pd 43 has a [rsqrt~] that works fine, right ?
extra/vanilla exists so that you can run into that kind of trouble and spend time debugging it. I haven't seen any benefit yet.
| Mathieu BOUCHARD ----- téléphone : +1.514.383.3801 ----- Montréal, QC
On Dec 27, 2011, at 7:22 PM, Mathieu Bouchard wrote:
Le 2011-12-28 à 03:02:00, Patrice Colet a écrit :
d_math.c doesn't seem to be used anymore in pdx, the code is scattered into extra/vanilla and half is missing into sqrt~.c, so it's quite hard to see what's happening, but I'm giving a try, thanks for the tip.
Well, is d_math.c still compiled in pd-extended ? I bet that it is. (I haven't checked out the git version yet)
I bet that there are two copies of the same table, and at least one is not initialised, if not both, and [bonk~] calls the wrong functions with the wrong tables (if any table even happens to be right).
Miller's pd 43 has a [rsqrt~] that works fine, right ?
extra/vanilla exists so that you can run into that kind of trouble and spend time debugging it. I haven't seen any benefit yet.
Yeah, that splitting out things into vanilla/ is pretty messy at the moment. Perhaps its not even the best approach. But its an experiment.
.hc
If you are not part of the solution, you are part of the problem.
Le 2011-12-27 à 22:09:00, Hans-Christoph Steiner a écrit :
On Dec 27, 2011, at 7:22 PM, Mathieu Bouchard wrote:
I bet that there are two copies of the same table, and at least one is not initialised, if not both, and [bonk~] calls the wrong functions with the wrong tables (if any table even happens to be right). [...]
Yeah, that splitting out things into vanilla/ is pretty messy at the moment. Perhaps its not even the best approach. But its an experiment.
How about an experiment to make sqrt work ?
You need to delete one of the two copies of Miller's fast sqrt functions, or at least call both inits, but having two copies is redundant.
| Mathieu BOUCHARD ----- téléphone : +1.514.383.3801 ----- Montréal, QC
On Jan 6, 2012, at 11:03 AM, Mathieu Bouchard wrote:
Le 2011-12-27 à 22:09:00, Hans-Christoph Steiner a écrit :
On Dec 27, 2011, at 7:22 PM, Mathieu Bouchard wrote:
I bet that there are two copies of the same table, and at least one is not initialised, if not both, and [bonk~] calls the wrong functions with the wrong tables (if any table even happens to be right). [...]
Yeah, that splitting out things into vanilla/ is pretty messy at the moment. Perhaps its not even the best approach. But its an experiment.
How about an experiment to make sqrt work ?
You need to delete one of the two copies of Miller's fast sqrt functions, or at least call both inits, but having two copies is redundant.
I do accept patches.
.hc
¡El pueblo unido jamás será vencido!
-
Hans-Christoph Steiner -
katja -
Mathieu Bouchard -
Patrice Colet