Quantcast

Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

classic Classic list List threaded Threaded
20 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

Hanns Holger Rutz
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi there,

has anyone seen this strange kind of behaviour:

    File '<some-path>.aif' could not be opened: Error. Bad format
field in SF_INFO struct when openning a RAW file for read.

It goes without saying that the particular AIFF files are all fine and
normal. They are _not_ raw files.

When I re-ran the same synth / b_read / DiskIn, it was working all
fine. This is with SC 3.6.3 on Debian. Any idea what could have made
libsndfile / DiskIn choke here?

Best, .h.h.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
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=vXfw
-----END PGP SIGNATURE-----

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

Hanns Holger Rutz
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 10/29/2014 03:03 PM, Hanns Holger Rutz wrote:
> Bad format field in SF_INFO struct when openning a RAW file for
> read

D'oh:

http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/Buffer-cueSoundFile-trouble-with-more-than-one-channel-td7611357.html

Jost writes:

> if the cued file is not immediately played, then cueSoundFile
> throws
an error

which could explain why I could not reproduce the problem easily. It
seemed to have occurred when the system load was maybe particularly
high so that there was a short delay between bundle completion from
b_read and actual s_new?

I'm preparing a show so I will _not_ compile libsndfile or anything
from source any more, this is too late. I'm hoping that if I warm up
the system, this doesn't happen any longer. I have used the same SC
version since ages, have never encountered this, so it must be some
rare condition.

Best, .h.h.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
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=C4G7
-----END PGP SIGNATURE-----

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

Hanns Holger Rutz
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

@Jost (not sure you are on the list, so I'll also PM you) - do you
remember if this problem goes away if one doesn't use DiskIn but
PlayBuf etc. with individual b_read commands? I have a mechanism in
place that can swap out DiskIn/VDiskIn for an alternative streaming
facility. I have to do that anyway occasionally because of all the
num-channels woes with these UGens.

Best, .h.h.


On 10/29/2014 03:10 PM, Hanns Holger Rutz wrote:

> On 10/29/2014 03:03 PM, Hanns Holger Rutz wrote:
>> Bad format field in SF_INFO struct when openning a RAW file for
>> read
>
> D'oh:
>
> http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/Buffer-cueSoundFile-trouble-with-more-than-one-channel-td7611357.html
>
>  Jost writes:
>
>> if the cued file is not immediately played, then cueSoundFile
>> throws
> an error
>
> which could explain why I could not reproduce the problem easily.
> It seemed to have occurred when the system load was maybe
> particularly high so that there was a short delay between bundle
> completion from b_read and actual s_new?
>
> I'm preparing a show so I will _not_ compile libsndfile or
> anything from source any more, this is too late. I'm hoping that if
> I warm up the system, this doesn't happen any longer. I have used
> the same SC version since ages, have never encountered this, so it
> must be some rare condition.
>
> Best, .h.h.
>
>
> _______________________________________________ sc-users mailing
> list
>
> info (subscription, etc.):
> http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml 
> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/ search:
> http://www.listarc.bham.ac.uk/lists/sc-users/search/
>

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
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=1kCo
-----END PGP SIGNATURE-----

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

carltesta
Hello Everyone,

I'm getting this same error using VDiskIn.ar in 3.7.1 but not in 3.6.6. Anyone know how I can fix it?

Carl
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

Constantin Popp
hi all,

i also get the same problem same as carl and the others. i am on OSX 10.9.5. (i've tried to replace the libsndfile with a recent pre-built libsndfile but without success. file came from here: https://github.com/bastibe/libsndfile-binaries).

does anybody know how to fix this?

many thanks,
constantin

On 16 April 2016 at 01:08, carltesta <[hidden email]> wrote:
Hello Everyone,

I'm getting this same error using VDiskIn.ar in 3.7.1 but not in 3.6.6.
Anyone know how I can fix it?

Carl



--
View this message in context: http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/Strange-sporadic-libsndfile-DiskIn-problem-Bad-format-field-in-SF-INFO-struct-tp7614418p7624770.html
Sent from the SuperCollider Users New (Use this!!!!) mailing list archive at Nabble.com.

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx



--
::::::::::::::::::::::::::::::::::::::::::::::::::::

[21/9/2016, Warsaw Autumn, live with Antoni Baksiak]
[7/11/2016, points lines planes @ Liverpool Hope Uni]
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

testing a local quark

Iain Mott-2

Hello,

I'm trying to learn how to make quarks. I have a basic class prepared and I'm able to use it by placing it in the ~/.local/share/SuperCollider/Extensions directory in Linux.

I've read the quark guide here:

https://github.com/supercollider-quarks/quarks

and have created the *.quarks file and a help file in the same directory as my class file. I've also placed it all up on the GitHub.

My class is not ready to be distributed publicly as a quark, however I'd like to be able to load it and test it locally as such. I can't work out how to do this. For example I tried the following, as suggested in the guide, to add my "personally developed quarks":

Quarks.addFolder("~/path/to/my/quarks/directory");

The addFolder method seems not to be found.

How can I test a local quark?

Thanks!

Iain

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: testing a local quark

Chris Sattinger

It should be just:

Quarks.install("~/path/to/my/quarks/your-quark");

Quarks.addFolder is for adding an entire folder that would have many quarks inside it.
Then you can add and remove from them using the gui.

so 

Quarks.addFolder("~/path/to/my/quarks");

and there you would have

quarks
   my-first-quark
   my-second-quark

etc.

Obviously that documentation should be rewritten a bit to make it clear.



On Mon, Sep 19, 2016 at 4:06 PM Iain Mott <[hidden email]> wrote:

Hello,

I'm trying to learn how to make quarks. I have a basic class prepared and I'm able to use it by placing it in the ~/.local/share/SuperCollider/Extensions directory in Linux.

I've read the quark guide here:

https://github.com/supercollider-quarks/quarks

and have created the *.quarks file and a help file in the same directory as my class file. I've also placed it all up on the GitHub.

My class is not ready to be distributed publicly as a quark, however I'd like to be able to load it and test it locally as such. I can't work out how to do this. For example I tried the following, as suggested in the guide, to add my "personally developed quarks":

Quarks.addFolder("~/path/to/my/quarks/directory");

The addFolder method seems not to be found.

How can I test a local quark?

Thanks!

Iain

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: testing a local quark

Iain Mott-2

Thanks, that was it. Built 3.7 and it all works.

All the best,

Iain


Em 19/09/2016 18:03, Chris Sattinger escreveu:

Ah yeah those are all 3.7 features.  It's a much impr9ved Quarks system.


On Mon, Sep 19, 2016, 22:53 Iain Mott <[hidden email]> wrote:

Yes, 3.6.6 on Linux.


Em 19/09/2016 17:44, Chris Sattinger escreveu:

What sc version are you using ?  It sounds like you are on 3.6



Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

carltesta
In reply to this post by Constantin Popp
Hi All,

Just downloaded the 3.8 build to test this "Bad format field in SF_INFO struct when openning a RAW file for read" issue and it's still present. Anything I can do to try and sort this out? Should I ask on the sc-Dev list?

Carl
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

Nathan Ho
carltesta wrote
Hi All,

Just downloaded the 3.8 build to test this "Bad format field in SF_INFO struct when openning a RAW file for read" issue and it's still present. Anything I can do to try and sort this out? Should I ask on the sc-Dev list?

Carl
Hi Carl,

Sorry about that. I couldn't find an issue in our tracker, so here it is: https://github.com/supercollider/supercollider/issues/2474


Nathan
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

Robert van Heumen
Just reporting that I have the same problem. I made a DiskPlayer class that plays files from disk using DiskIn to play mono and stereo files (and VDiskIn for 4 channel files, but haven't tried that recently). Pretty sure it worked well for a while (can't remember when I built it), but now quite randomly throws the above mentioned error. It happens most often when I try to loop shorter sounds. Also when I close and free the buffer and then start to instantiate a new DiskPlayer. It's quite weird.

If I can assist anyone working on this, let me know!

Best,
RobertDiskPlayer.sc
*
Robert van Heumen
Composer and improvising musician
http://west28.nl/ 
http://shackle.eu/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

Rainer Schuetz
Dear Robert,

there has been quite some change regarding libsndfile during the period the various error reports span, it is quite possible that something is going awry here with SC. On the other hand it is also possible that the problem is caused by your system, in your code, or in the sound file you use, and that older reports refer to a different SC than the one you are using. SC used to use a quite old version of libsndfile (1.0.24) statically linked, and then switched to dynamically linking any libsndfile version that is provided by homebrew (a OSX package manager) at the time of the release. The libraries are provided inside the mac application bundle, so an interference with other libraries installed in your system is unlikely. This change was made to allow for flac-, and I think ogg support, and was introduced in April 2016 with Version 3.7.1. I must say that not much testing was done when the change was introduced, but at the time of the introduction of the feature we got no error-reports at all that could be substantiated. I think it is fair to say we didn't get any serious ones at all, and that reports concerning sound-file read/write problems are quite rare (although the implementation is not really in a well cleaned up state). The ones assembled in the issue created in reaction to your report are not very useful, as they are not reproducible. Also starting with 3.7 a bug in multi-channel recording was fixed which was under suspicion of playing into this problem as some reports claim the problem occurs with stereo files only.

I tell this for two reasons: we are seeing a lot of issues with similar symptoms, but it is very unlikely (though not impossible) that they have the same underlying reason. So while it appears to be, it is quite unlikely that you have *the same* issue.

In order to do anything, we need to be able to reliably reproduce your problem in an as simple as possible setting. Please provide the shortest code possible that exposes your problem and make the soundfile you use accessible. Then I will try to reproduce it on my system. You could also try to temporarily use a pre 3.7.1 version and see if that makes a difference.

Thanks and all the best
Rainer

> On 2 Mar 2017, at 20:47, Robert van Heumen <[hidden email]> wrote:
>
> Just reporting that I have the same problem. I made a DiskPlayer class that
> plays files from disk using DiskIn to play mono and stereo files (and
> VDiskIn for 4 channel files, but haven't tried that recently). Pretty sure
> it worked well for a while (can't remember when I built it), but now quite
> randomly throws the above mentioned error. It happens most often when I try
> to loop shorter sounds. Also when I close and free the buffer and then start
> to instantiate a new DiskPlayer. It's quite weird.
>
> If I can assist anyone working on this, let me know!
>
> Best,
> Robert DiskPlayer.sc
> <http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/file/n7630974/DiskPlayer.sc>  
>
>
>
> -----
> *
> Robert van Heumen
> Composer and improvising musician
> http://west28.nl/ 
> http://shackle.eu/
> --
> View this message in context: http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/Strange-sporadic-libsndfile-DiskIn-problem-Bad-format-field-in-SF-INFO-struct-tp7614418p7630974.html
> Sent from the SuperCollider Users New (Use this!!!!) mailing list archive at Nabble.com.
>
> _______________________________________________
> sc-users mailing list
>
> info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/


_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

Robert van Heumen
Dear Rainer,
Thanks for your response!
So I simplified the problematic code, and pinpointed the issue on my system. It is a weird combination of things. The simplified code is below. 
It seems it has to do with alternatingly sending SynthDefs and playing them. Of course that is not necessary (just the way I set up the class), so I can solve the problem on my end. 
But it would be great of course if this could be resolved. Let me know how/if I can help!
Best,
Robert


////////////// most simple example /////////////////

// Using OS X 10.9 and SC 3.8

( // 1: send SynthDefs
SynthDef(\play2, { arg buf; Out.ar(0, VDiskIn.ar(2, buf)) }).add;
SynthDef(\play4, { arg buf; Out.ar(0, VDiskIn.ar(4, buf)) }).add;
)

( // 2: start playback
~path = "/Users/Robert/---data---/Audio/random samples/oguz004a.wav";
~buffer = Buffer.alloc(s, 2 * 32768, 2);
~buffer.cueSoundFile(~path);
~synth = Synth(\play2, [\buf, ~buffer]);
)
( // 3: stop playback and clean up // you can repeat 2 and 3 a couple of times, no problem
~synth.free;
~buffer.close; ~buffer.free;
)
( // 4: send SynthDefs again (not necessary of course, but this is the way it works in my class, and it turns out to be the problem)
SynthDef(\play2, { arg buf; Out.ar(0, VDiskIn.ar(2, buf)) }).add;
SynthDef(\play4, { arg buf; Out.ar(0, VDiskIn.ar(4, buf)) }).add;
)

( // 5: start playback again; breaks on the cueSoundFile
~path = "/Users/Robert/---data---/Audio/random samples/oguz004a.wav";
~buffer = Buffer.alloc(s, 2 * 32768, 2);
~buffer.cueSoundFile(~path);
~synth = Synth(\play2, [\buf, ~buffer]);
)

/*
Note: the SynthDef with VDiskIn and 4 channels is essential here, if you remove that and only keep the 2 channel SynthDef, there is no problem.
Note: a buffer size of 32768 (not 2*32768 as in this example) breaks the cueSoundFile already in step 2.
Note: in SC3.7 the example does NOT break in step 5.
*/






On 02 Mar 2017, at 22:50, Rainer Schuetz [via New SuperCollider Mailing Lists Forums (Use These!!!)] <[hidden email]> wrote:

Dear Robert,

there has been quite some change regarding libsndfile during the period the various error reports span, it is quite possible that something is going awry here with SC. On the other hand it is also possible that the problem is caused by your system, in your code, or in the sound file you use, and that older reports refer to a different SC than the one you are using. SC used to use a quite old version of libsndfile (1.0.24) statically linked, and then switched to dynamically linking any libsndfile version that is provided by homebrew (a OSX package manager) at the time of the release. The libraries are provided inside the mac application bundle, so an interference with other libraries installed in your system is unlikely. This change was made to allow for flac-, and I think ogg support, and was introduced in April 2016 with Version 3.7.1. I must say that not much testing was done when the change was introduced, but at the time of the introduction of the feature we got no error-reports at all that could be substantiated. I think it is fair to say we didn't get any serious ones at all, and that reports concerning sound-file read/write problems are quite rare (although the implementation is not really in a well cleaned up state). The ones assembled in the issue created in reaction to your report are not very useful, as they are not reproducible. Also starting with 3.7 a bug in multi-channel recording was fixed which was under suspicion of playing into this problem as some reports claim the problem occurs with stereo files only.

I tell this for two reasons: we are seeing a lot of issues with similar symptoms, but it is very unlikely (though not impossible) that they have the same underlying reason. So while it appears to be, it is quite unlikely that you have *the same* issue.

In order to do anything, we need to be able to reliably reproduce your problem in an as simple as possible setting. Please provide the shortest code possible that exposes your problem and make the soundfile you use accessible. Then I will try to reproduce it on my system. You could also try to temporarily use a pre 3.7.1 version and see if that makes a difference.

Thanks and all the best
Rainer

> On 2 Mar 2017, at 20:47, Robert van Heumen <<a href="x-msg://53/user/SendEmail.jtp?type=node&amp;node=7630976&amp;i=0" target="_top" rel="nofollow" link="external">[hidden email]> wrote:
>
> Just reporting that I have the same problem. I made a DiskPlayer class that
> plays files from disk using DiskIn to play mono and stereo files (and
> VDiskIn for 4 channel files, but haven't tried that recently). Pretty sure
> it worked well for a while (can't remember when I built it), but now quite
> randomly throws the above mentioned error. It happens most often when I try
> to loop shorter sounds. Also when I close and free the buffer and then start
> to instantiate a new DiskPlayer. It's quite weird.
>
> If I can assist anyone working on this, let me know!
>
> Best,
> Robert DiskPlayer.sc
> <http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/file/n7630974/DiskPlayer.sc>  
>
>
>
> -----
> *
> Robert van Heumen
> Composer and improvising musician
> http://west28.nl/ 
> http://shackle.eu/
> --
> View this message in context: http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/Strange-sporadic-libsndfile-DiskIn-problem-Bad-format-field-in-SF-INFO-struct-tp7614418p7630974.html
> Sent from the SuperCollider Users New (Use this!!!!) mailing list archive at Nabble.com.
>
> _______________________________________________
> sc-users mailing list
>
> info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/



To unsubscribe from Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct", click here.
NAML

*
Robert van Heumen
Composer and improvising musician
http://west28.nl/ 
http://shackle.eu/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

Rainer Schuetz
Dear Robert,

yes, I can reproduce something in 4-Channel-mode that has nothing to do with the audio-file per-se, and still produces an error-message that comes from libsndfile:

-> Synth('play4' : 1028)
File '/Users/rainer/test/a11wlk01-4Channel.wav' could not be opened: Error. Bad format field in SF_INFO struct when openning a RAW file for read.

Same code with 1 or 2 channel doesn't produce the error for me, but others got the error-message for 2-channel too. Something weird is going on,
but I am not sure yet how to trigger the error best.

Do you use 4-channel files? Could you maybe try triggering the error with even simpler code, using only a 4 channel file?

I'll inspect a bit further ;)
.r.

On 5 Mar 2017, at 14:11, Robert van Heumen <[hidden email]> wrote:

Dear Rainer,
Thanks for your response!
So I simplified the problematic code, and pinpointed the issue on my system. It is a weird combination of things. The simplified code is below. 
It seems it has to do with alternatingly sending SynthDefs and playing them. Of course that is not necessary (just the way I set up the class), so I can solve the problem on my end. 
But it would be great of course if this could be resolved. Let me know how/if I can help!
Best,
Robert


////////////// most simple example /////////////////

// Using OS X 10.9 and SC 3.8

( // 1: send SynthDefs
SynthDef(\play2, { arg buf; Out.ar(0, VDiskIn.ar(2, buf)) }).add;
SynthDef(\play4, { arg buf; Out.ar(0, VDiskIn.ar(4, buf)) }).add;
)

( // 2: start playback
~path = "/Users/Robert/---data---/Audio/random samples/oguz004a.wav";
~buffer = Buffer.alloc(s, 2 * 32768, 2);
~buffer.cueSoundFile(~path);
~synth = Synth(\play2, [\buf, ~buffer]);
)
( // 3: stop playback and clean up // you can repeat 2 and 3 a couple of times, no problem
~synth.free;
~buffer.close; ~buffer.free;
)
( // 4: send SynthDefs again (not necessary of course, but this is the way it works in my class, and it turns out to be the problem)
SynthDef(\play2, { arg buf; Out.ar(0, VDiskIn.ar(2, buf)) }).add;
SynthDef(\play4, { arg buf; Out.ar(0, VDiskIn.ar(4, buf)) }).add;
)

( // 5: start playback again; breaks on the cueSoundFile
~path = "/Users/Robert/---data---/Audio/random samples/oguz004a.wav";
~buffer = Buffer.alloc(s, 2 * 32768, 2);
~buffer.cueSoundFile(~path);
~synth = Synth(\play2, [\buf, ~buffer]);
)

/*
Note: the SynthDef with VDiskIn and 4 channels is essential here, if you remove that and only keep the 2 channel SynthDef, there is no problem.
Note: a buffer size of 32768 (not 2*32768 as in this example) breaks the cueSoundFile already in step 2.
Note: in SC3.7 the example does NOT break in step 5.
*/






On 02 Mar 2017, at 22:50, Rainer Schuetz [via New SuperCollider Mailing Lists Forums (Use These!!!)] <[hidden email]> wrote:

Dear Robert, 

there has been quite some change regarding libsndfile during the period the various error reports span, it is quite possible that something is going awry here with SC. On the other hand it is also possible that the problem is caused by your system, in your code, or in the sound file you use, and that older reports refer to a different SC than the one you are using. SC used to use a quite old version of libsndfile (1.0.24) statically linked, and then switched to dynamically linking any libsndfile version that is provided by homebrew (a OSX package manager) at the time of the release. The libraries are provided inside the mac application bundle, so an interference with other libraries installed in your system is unlikely. This change was made to allow for flac-, and I think ogg support, and was introduced in April 2016 with Version 3.7.1. I must say that not much testing was done when the change was introduced, but at the time of the introduction of the feature we got no error-reports at all that could be substantiated. I think it is fair to say we didn't get any serious ones at all, and that reports concerning sound-file read/write problems are quite rare (although the implementation is not really in a well cleaned up state). The ones assembled in the issue created in reaction to your report are not very useful, as they are not reproducible. Also starting with 3.7 a bug in multi-channel recording was fixed which was under suspicion of playing into this problem as some reports claim the problem occurs with stereo files only. 

I tell this for two reasons: we are seeing a lot of issues with similar symptoms, but it is very unlikely (though not impossible) that they have the same underlying reason. So while it appears to be, it is quite unlikely that you have *the same* issue. 

In order to do anything, we need to be able to reliably reproduce your problem in an as simple as possible setting. Please provide the shortest code possible that exposes your problem and make the soundfile you use accessible. Then I will try to reproduce it on my system. You could also try to temporarily use a pre 3.7.1 version and see if that makes a difference. 

Thanks and all the best 
Rainer 

On 2 Mar 2017, at 20:47, Robert van Heumen <[hidden email]> wrote: 

Just reporting that I have the same problem. I made a DiskPlayer class that 
plays files from disk using DiskIn to play mono and stereo files (and 
VDiskIn for 4 channel files, but haven't tried that recently). Pretty sure 
it worked well for a while (can't remember when I built it), but now quite 
randomly throws the above mentioned error. It happens most often when I try 
to loop shorter sounds. Also when I close and free the buffer and then start 
to instantiate a new DiskPlayer. It's quite weird. 

If I can assist anyone working on this, let me know! 

Best, 
Robert DiskPlayer.sc 
<http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/file/n7630974/DiskPlayer.sc>   



----- 
* 
Robert van Heumen 
Composer and improvising musician 
http://west28.nl/  
http://shackle.eu/
-- 
View this message in context: http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/Strange-sporadic-libsndfile-DiskIn-problem-Bad-format-field-in-SF-INFO-struct-tp7614418p7630974.html
Sent from the SuperCollider Users New (Use this!!!!) mailing list archive at Nabble.com. 

_______________________________________________ 
sc-users mailing list 

info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/


_______________________________________________ 
sc-users mailing list 

info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/


If you reply to this email, your message will be added to the discussion below:
http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/Strange-sporadic-libsndfile-DiskIn-problem-Bad-format-field-in-SF-INFO-struct-tp7614418p7630976.html
To unsubscribe from Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct", click here.
NAML





-----
*
Robert van Heumen 
Composer and improvising musician 
http://west28.nl/ 
http://shackle.eu/
--
View this message in context: http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/Strange-sporadic-libsndfile-DiskIn-problem-Bad-format-field-in-SF-INFO-struct-tp7614418p7631011.html
Sent from the SuperCollider Users New (Use this!!!!) mailing list archive at Nabble.com.

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

Robert van Heumen
Hello Rainer,
I can try that tomorrow.
Best,
Robert

On 05 Mar 2017, at 18:28, Rainer Schuetz [via New SuperCollider Mailing Lists Forums (Use These!!!)] <[hidden email]> wrote:

Dear Robert,

yes, I can reproduce something in 4-Channel-mode that has nothing to do with the audio-file per-se, and still produces an error-message that comes from libsndfile:

-> Synth('play4' : 1028)
File '/Users/rainer/test/a11wlk01-4Channel.wav' could not be opened: Error. Bad format field in SF_INFO struct when openning a RAW file for read.

Same code with 1 or 2 channel doesn't produce the error for me, but others got the error-message for 2-channel too. Something weird is going on,
but I am not sure yet how to trigger the error best.

Do you use 4-channel files? Could you maybe try triggering the error with even simpler code, using only a 4 channel file?

I'll inspect a bit further ;)
.r.

On 5 Mar 2017, at 14:11, Robert van Heumen <<a href="x-msg://55/user/SendEmail.jtp?type=node&amp;node=7631012&amp;i=0" target="_top" rel="nofollow" link="external">[hidden email]> wrote:

Dear Rainer,
Thanks for your response!
So I simplified the problematic code, and pinpointed the issue on my system. It is a weird combination of things. The simplified code is below. 
It seems it has to do with alternatingly sending SynthDefs and playing them. Of course that is not necessary (just the way I set up the class), so I can solve the problem on my end. 
But it would be great of course if this could be resolved. Let me know how/if I can help!
Best,
Robert


////////////// most simple example /////////////////

// Using OS X 10.9 and SC 3.8

( // 1: send SynthDefs
SynthDef(\play2, { arg buf; Out.ar(0, VDiskIn.ar(2, buf)) }).add;
SynthDef(\play4, { arg buf; Out.ar(0, VDiskIn.ar(4, buf)) }).add;
)

( // 2: start playback
~path = "/Users/Robert/---data---/Audio/random samples/oguz004a.wav";
~buffer = Buffer.alloc(s, 2 * 32768, 2);
~buffer.cueSoundFile(~path);
~synth = Synth(\play2, [\buf, ~buffer]);
)
( // 3: stop playback and clean up // you can repeat 2 and 3 a couple of times, no problem
~synth.free;
~buffer.close; ~buffer.free;
)
( // 4: send SynthDefs again (not necessary of course, but this is the way it works in my class, and it turns out to be the problem)
SynthDef(\play2, { arg buf; Out.ar(0, VDiskIn.ar(2, buf)) }).add;
SynthDef(\play4, { arg buf; Out.ar(0, VDiskIn.ar(4, buf)) }).add;
)

( // 5: start playback again; breaks on the cueSoundFile
~path = "/Users/Robert/---data---/Audio/random samples/oguz004a.wav";
~buffer = Buffer.alloc(s, 2 * 32768, 2);
~buffer.cueSoundFile(~path);
~synth = Synth(\play2, [\buf, ~buffer]);
)

/*
Note: the SynthDef with VDiskIn and 4 channels is essential here, if you remove that and only keep the 2 channel SynthDef, there is no problem.
Note: a buffer size of 32768 (not 2*32768 as in this example) breaks the cueSoundFile already in step 2.
Note: in SC3.7 the example does NOT break in step 5.
*/






On 02 Mar 2017, at 22:50, Rainer Schuetz [via New SuperCollider Mailing Lists Forums (Use These!!!)] <<a href="x-msg://55/user/SendEmail.jtp?type=node&amp;node=7631012&amp;i=1" target="_top" rel="nofollow" link="external">[hidden email]> wrote:

Dear Robert, 

there has been quite some change regarding libsndfile during the period the various error reports span, it is quite possible that something is going awry here with SC. On the other hand it is also possible that the problem is caused by your system, in your code, or in the sound file you use, and that older reports refer to a different SC than the one you are using. SC used to use a quite old version of libsndfile (1.0.24) statically linked, and then switched to dynamically linking any libsndfile version that is provided by homebrew (a OSX package manager) at the time of the release. The libraries are provided inside the mac application bundle, so an interference with other libraries installed in your system is unlikely. This change was made to allow for flac-, and I think ogg support, and was introduced in April 2016 with Version 3.7.1. I must say that not much testing was done when the change was introduced, but at the time of the introduction of the feature we got no error-reports at all that could be substantiated. I think it is fair to say we didn't get any serious ones at all, and that reports concerning sound-file read/write problems are quite rare (although the implementation is not really in a well cleaned up state). The ones assembled in the issue created in reaction to your report are not very useful, as they are not reproducible. Also starting with 3.7 a bug in multi-channel recording was fixed which was under suspicion of playing into this problem as some reports claim the problem occurs with stereo files only. 

I tell this for two reasons: we are seeing a lot of issues with similar symptoms, but it is very unlikely (though not impossible) that they have the same underlying reason. So while it appears to be, it is quite unlikely that you have *the same* issue. 

In order to do anything, we need to be able to reliably reproduce your problem in an as simple as possible setting. Please provide the shortest code possible that exposes your problem and make the soundfile you use accessible. Then I will try to reproduce it on my system. You could also try to temporarily use a pre 3.7.1 version and see if that makes a difference. 

Thanks and all the best 
Rainer 

On 2 Mar 2017, at 20:47, Robert van Heumen <[hidden email]> wrote: 

Just reporting that I have the same problem. I made a DiskPlayer class that 
plays files from disk using DiskIn to play mono and stereo files (and 
VDiskIn for 4 channel files, but haven't tried that recently). Pretty sure 
it worked well for a while (can't remember when I built it), but now quite 
randomly throws the above mentioned error. It happens most often when I try 
to loop shorter sounds. Also when I close and free the buffer and then start 
to instantiate a new DiskPlayer. It's quite weird. 

If I can assist anyone working on this, let me know! 

Best, 
Robert DiskPlayer.sc 
<http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/file/n7630974/DiskPlayer.sc>   



----- 
* 
Robert van Heumen 
Composer and improvising musician 
http://west28.nl/  
http://shackle.eu/
-- 
View this message in context: http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/Strange-sporadic-libsndfile-DiskIn-problem-Bad-format-field-in-SF-INFO-struct-tp7614418p7630974.html
Sent from the SuperCollider Users New (Use this!!!!) mailing list archive at Nabble.com. 

_______________________________________________ 
sc-users mailing list 

info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/


_______________________________________________ 
sc-users mailing list 

info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/


If you reply to this email, your message will be added to the discussion below:
http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/Strange-sporadic-libsndfile-DiskIn-problem-Bad-format-field-in-SF-INFO-struct-tp7614418p7630976.html
To unsubscribe from Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct", click here.
NAML





-----
*
Robert van Heumen 
Composer and improvising musician 
http://west28.nl/ 
http://shackle.eu/
--
View this message in context: http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/Strange-sporadic-libsndfile-DiskIn-problem-Bad-format-field-in-SF-INFO-struct-tp7614418p7631011.html
Sent from the SuperCollider Users New (Use this!!!!) mailing list archive at Nabble.com.

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/




To unsubscribe from Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct", click here.
NAML

*
Robert van Heumen
Composer and improvising musician
http://west28.nl/ 
http://shackle.eu/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

Robert van Heumen
In reply to this post by Rainer Schuetz
Dear Rainer,

Below is the example with only 4 channels. If I change it to 2 channels there is no problem. The issue seems to be the 4 channel VDiskIn, even if you don’t use it but only send a SynthDef with a 4 channel VDiskIn to the server.

Best,
Robert


// 

////////////// even more simple example /////////////////

// Using OS X 10.9 and SC 3.8

( // 1: send SynthDefs
SynthDef(\play4, { arg buf; Out.ar(0, VDiskIn.ar(4, buf)) }).add;
)
( // 2: start playback
~path = "/Users/Robert/---data---/Audio/random samples/oguz004a-quadro.wav";
~buffer = Buffer.alloc(s, 2 * 32768, 4);
~buffer.cueSoundFile(~path);
~synth = Synth(\play4, [\buf, ~buffer]);
)
( // 3: stop playback and clean up // you can repeat 2 and 3 a couple of times, no problem
~synth.free;
~buffer.close; ~buffer.free;
)
( // 4: send SynthDefs again (not necessary of course, but this is the way it works in my class, and it turns out to be the problem)
SynthDef(\play4, { arg buf; Out.ar(0, VDiskIn.ar(4, buf)) }).add;
)

( // 5: start playback again; breaks on the cueSoundFile
~path = "/Users/Robert/---data---/Audio/random samples/oguz004a-quadro.wav";
~buffer = Buffer.alloc(s, 2 * 32768, 4);
~buffer.cueSoundFile(~path);
~synth = Synth(\play4, [\buf, ~buffer]);
)

/*
Note: when changing 4 channels in to 2 channels in this example then there is no problem. The issue seems to be the 4 channel VDiskIn, even if you don't use it but only send a SynthDef with a 4 channel VDiskIn to the server.
Note: in SC3.7 the example does NOT break in step 5.
*/



On 05 Mar 2017, at 18:28, Rainer Schuetz [via New SuperCollider Mailing Lists Forums (Use These!!!)] <[hidden email]> wrote:

Dear Robert,

yes, I can reproduce something in 4-Channel-mode that has nothing to do with the audio-file per-se, and still produces an error-message that comes from libsndfile:

-> Synth('play4' : 1028)
File '/Users/rainer/test/a11wlk01-4Channel.wav' could not be opened: Error. Bad format field in SF_INFO struct when openning a RAW file for read.

Same code with 1 or 2 channel doesn't produce the error for me, but others got the error-message for 2-channel too. Something weird is going on,
but I am not sure yet how to trigger the error best.

Do you use 4-channel files? Could you maybe try triggering the error with even simpler code, using only a 4 channel file?

I'll inspect a bit further ;)
.r.

On 5 Mar 2017, at 14:11, Robert van Heumen <<a href="x-msg://65/user/SendEmail.jtp?type=node&amp;node=7631012&amp;i=0" target="_top" rel="nofollow" link="external">[hidden email]> wrote:

Dear Rainer,
Thanks for your response!
So I simplified the problematic code, and pinpointed the issue on my system. It is a weird combination of things. The simplified code is below. 
It seems it has to do with alternatingly sending SynthDefs and playing them. Of course that is not necessary (just the way I set up the class), so I can solve the problem on my end. 
But it would be great of course if this could be resolved. Let me know how/if I can help!
Best,
Robert


////////////// most simple example /////////////////

// Using OS X 10.9 and SC 3.8

( // 1: send SynthDefs
SynthDef(\play2, { arg buf; Out.ar(0, VDiskIn.ar(2, buf)) }).add;
SynthDef(\play4, { arg buf; Out.ar(0, VDiskIn.ar(4, buf)) }).add;
)

( // 2: start playback
~path = "/Users/Robert/---data---/Audio/random samples/oguz004a.wav";
~buffer = Buffer.alloc(s, 2 * 32768, 2);
~buffer.cueSoundFile(~path);
~synth = Synth(\play2, [\buf, ~buffer]);
)
( // 3: stop playback and clean up // you can repeat 2 and 3 a couple of times, no problem
~synth.free;
~buffer.close; ~buffer.free;
)
( // 4: send SynthDefs again (not necessary of course, but this is the way it works in my class, and it turns out to be the problem)
SynthDef(\play2, { arg buf; Out.ar(0, VDiskIn.ar(2, buf)) }).add;
SynthDef(\play4, { arg buf; Out.ar(0, VDiskIn.ar(4, buf)) }).add;
)

( // 5: start playback again; breaks on the cueSoundFile
~path = "/Users/Robert/---data---/Audio/random samples/oguz004a.wav";
~buffer = Buffer.alloc(s, 2 * 32768, 2);
~buffer.cueSoundFile(~path);
~synth = Synth(\play2, [\buf, ~buffer]);
)

/*
Note: the SynthDef with VDiskIn and 4 channels is essential here, if you remove that and only keep the 2 channel SynthDef, there is no problem.
Note: a buffer size of 32768 (not 2*32768 as in this example) breaks the cueSoundFile already in step 2.
Note: in SC3.7 the example does NOT break in step 5.
*/






On 02 Mar 2017, at 22:50, Rainer Schuetz [via New SuperCollider Mailing Lists Forums (Use These!!!)] <<a href="x-msg://65/user/SendEmail.jtp?type=node&amp;node=7631012&amp;i=1" target="_top" rel="nofollow" link="external">[hidden email]> wrote:

Dear Robert, 

there has been quite some change regarding libsndfile during the period the various error reports span, it is quite possible that something is going awry here with SC. On the other hand it is also possible that the problem is caused by your system, in your code, or in the sound file you use, and that older reports refer to a different SC than the one you are using. SC used to use a quite old version of libsndfile (1.0.24) statically linked, and then switched to dynamically linking any libsndfile version that is provided by homebrew (a OSX package manager) at the time of the release. The libraries are provided inside the mac application bundle, so an interference with other libraries installed in your system is unlikely. This change was made to allow for flac-, and I think ogg support, and was introduced in April 2016 with Version 3.7.1. I must say that not much testing was done when the change was introduced, but at the time of the introduction of the feature we got no error-reports at all that could be substantiated. I think it is fair to say we didn't get any serious ones at all, and that reports concerning sound-file read/write problems are quite rare (although the implementation is not really in a well cleaned up state). The ones assembled in the issue created in reaction to your report are not very useful, as they are not reproducible. Also starting with 3.7 a bug in multi-channel recording was fixed which was under suspicion of playing into this problem as some reports claim the problem occurs with stereo files only. 

I tell this for two reasons: we are seeing a lot of issues with similar symptoms, but it is very unlikely (though not impossible) that they have the same underlying reason. So while it appears to be, it is quite unlikely that you have *the same* issue. 

In order to do anything, we need to be able to reliably reproduce your problem in an as simple as possible setting. Please provide the shortest code possible that exposes your problem and make the soundfile you use accessible. Then I will try to reproduce it on my system. You could also try to temporarily use a pre 3.7.1 version and see if that makes a difference. 

Thanks and all the best 
Rainer 

On 2 Mar 2017, at 20:47, Robert van Heumen <[hidden email]> wrote: 

Just reporting that I have the same problem. I made a DiskPlayer class that 
plays files from disk using DiskIn to play mono and stereo files (and 
VDiskIn for 4 channel files, but haven't tried that recently). Pretty sure 
it worked well for a while (can't remember when I built it), but now quite 
randomly throws the above mentioned error. It happens most often when I try 
to loop shorter sounds. Also when I close and free the buffer and then start 
to instantiate a new DiskPlayer. It's quite weird. 

If I can assist anyone working on this, let me know! 

Best, 
Robert DiskPlayer.sc 
<http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/file/n7630974/DiskPlayer.sc>   



----- 
* 
Robert van Heumen 
Composer and improvising musician 
http://west28.nl/  
http://shackle.eu/
-- 
View this message in context: http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/Strange-sporadic-libsndfile-DiskIn-problem-Bad-format-field-in-SF-INFO-struct-tp7614418p7630974.html
Sent from the SuperCollider Users New (Use this!!!!) mailing list archive at Nabble.com. 

_______________________________________________ 
sc-users mailing list 

info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/


_______________________________________________ 
sc-users mailing list 

info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/


If you reply to this email, your message will be added to the discussion below:
http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/Strange-sporadic-libsndfile-DiskIn-problem-Bad-format-field-in-SF-INFO-struct-tp7614418p7630976.html
To unsubscribe from Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct", click here.
NAML





-----
*
Robert van Heumen 
Composer and improvising musician 
http://west28.nl/ 
http://shackle.eu/
--
View this message in context: http://new-supercollider-mailing-lists-forums-use-these.2681727.n2.nabble.com/Strange-sporadic-libsndfile-DiskIn-problem-Bad-format-field-in-SF-INFO-struct-tp7614418p7631011.html
Sent from the SuperCollider Users New (Use this!!!!) mailing list archive at Nabble.com.

_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/




To unsubscribe from Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct", click here.
NAML

*
Robert van Heumen
Composer and improvising musician
http://west28.nl/ 
http://shackle.eu/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

Rainer Schuetz

On 6 Mar 2017, at 09:35, Robert van Heumen <[hidden email]> wrote:

Dear Rainer,

Below is the example with only 4 channels. If I change it to 2 channels there is no problem. The issue seems to be the 4 channel VDiskIn, even if you don’t use it but only send a SynthDef with a 4 channel VDiskIn to the server.

Best,
Robert

Yea, it's the ~buffer.free that causes the subsequent error. I can't look at it/think about it right now, hopefully this weekend, sorry.
Best
.r.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

Nathan Ho
In reply to this post by Rainer Schuetz
Rainer Schuetz wrote
Dear Robert,

there has been quite some change regarding libsndfile during the period the various error reports span, it is quite possible that something is going awry here with SC.

[...]
Hi all,

Good news: we've fixed this bug in the upcoming 3.9 release.


Nathan
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

Rainer Schuetz

> On 14 Mar 2017, at 17:30, Nathan Ho <[hidden email]> wrote:
>
> Rainer Schuetz wrote
>> Dear Robert,
>>
>> there has been quite some change regarding libsndfile during the period
>> the various error reports span, it is quite possible that something is
>> going awry here with SC.
>>
>> [...]
>
> Hi all,
>
> Good news: we've fixed this bug in the upcoming 3.9 release.
>
> Nathan

It was a short but dramatic hunt in which Nathan proved Holmes like qualities! :D

Thanks, man
.r.


_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Strange sporadic libsndfile / DiskIn problem "Bad format field in SF_INFO struct"

Robert van Heumen
Great! Thanks for the work.


On 14 Mar 2017, at 17:39, Rainer Schuetz <[hidden email]> wrote:

>
>> On 14 Mar 2017, at 17:30, Nathan Ho <[hidden email]> wrote:
>>
>> Rainer Schuetz wrote
>>> Dear Robert,
>>>
>>> there has been quite some change regarding libsndfile during the period
>>> the various error reports span, it is quite possible that something is
>>> going awry here with SC.
>>>
>>> [...]
>>
>> Hi all,
>>
>> Good news: we've fixed this bug in the upcoming 3.9 release.
>>
>> Nathan
>
> It was a short but dramatic hunt in which Nathan proved Holmes like qualities! :D
>
> Thanks, man
> .r.
>
>
> _______________________________________________
> sc-users mailing list
>
> info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/


_______________________________________________
sc-users mailing list

info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
*
Robert van Heumen
Composer and improvising musician
http://west28.nl/ 
http://shackle.eu/
Loading...