headers
Kai Vehmanen | 2 Mar 2008 21:33

Re: "short-length-file-with-effects bug" now unavoidable

Hi,

finally responding to this, 1+ year old, bug report.

On Sat, 23 Dec 2006, linuxmedia4 <at> netscape.net wrote:

> If I run the following ECS file during the running time of "Short.wav"
> (obviously, "Short.wav" is shorter than the other two files), then
> sometime after "Short.wav" stops (file ends, but other files are
> playing) there's extreme "stuttering and shuffling" sounds. On one
> occasion, it locked up my computer and I had to shut it off without a
> proper shutdown. But if I do one of two things, the bug doesn't show up...

Hmm, I've tried out various things to reproduce this. I've tried both with 
ecasound 2.4.4 and the most recent devel version (from the public git 
tree), but I've couldn't reproduce the "stuttering and shuffling".

Has anyone else had similar experiences (or in general weird mixed 
output after one of the inputs has reached end-of-file)?

--

-- 
  links, my public keys, etc at http://eca.cx/kv

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Ecasound-list mailing list
Ecasound-list <at> lists.sourceforge.net
(Continue reading)

Permalink | Reply | 1 comment |
headers
Kai Vehmanen | 2 Mar 2008 21:54

Re: Ecasound and LV2 plugins

Hi,

On Sat, 12 May 2007, Julien Claassen wrote:

>  I just wonder, did you think about integrating LV2 soon? Or what is the
> state of consideration (problem solving :-)) anyway?

now that LV2 [1] is out, adding support for it to Ecasound has definitely 
moved up a notch on my todo-list. I'm waiting with great interest how 
support for LV2 develops in general.

Of course, you can already use LV2 plugins with Ecasound through the 
combination of JACK and Zynjacku [1].

[1] http://lv2plug.in/
[2] http://home.gna.org/zynjacku/

--

-- 
  links, my public keys, etc at http://eca.cx/kv

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Ecasound-list mailing list
Ecasound-list <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ecasound-list
(Continue reading)

Permalink | Reply | 0 comments |
headers
Kai Vehmanen | 4 Mar 2008 18:52

Re: cs-option -B:*

Hi,

don't ask how/why I decided to handle this now :) (after 5 years), but 
nevertheless, read on...

On Fri, 12 Dec 2003, Jan Weil wrote:
> The question remains: If I set the buffering mode explicitly (using
> cs-option) why isn't the status of the chainsetup affected immediately?

Yes, this indeed was a bug. If you explicitly set a buffering mode (e.g.
'cs-option -B:rt'), but did not connect the chainsetup (which would cause
ecasound to calculate the final buffering params), the output of 
"cs-status" was essentially stale.

I now fixed this in the development git tree, and the bugfix will be
part of 2.4.7. Thanks for the bugreport! :)

--

-- 
  links, my public keys, etc at http://eca.cx/kv

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Ecasound-list mailing list
Ecasound-list <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ecasound-list
(Continue reading)

Permalink | Reply | 1 comment |
headers
Jan Weil | 4 Mar 2008 20:36
Picon

Re: cs-option -B:*

Hi Kai,

Am Dienstag, den 04.03.2008, 19:52 +0200 schrieb Kai Vehmanen:
> don't ask how/why I decided to handle this now :) (after 5 years), but 
> nevertheless, read on...

well, although I have to admit that I moved on since then, it's good to
see that all those bugs are tracked. :-)

All the best,

Jan

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Ecasound-list mailing list
Ecasound-list <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ecasound-list
Permalink | Reply | 0 comments |
headers
Kai Vehmanen | 4 Mar 2008 22:41

EWF refactoring, testers needed

Hi all,

I've just pushed a big rewrite of Ecasound's EWF (Ecasound Wave Files, see 
the user's guide for more info) support to the public git repo.

The EWF related code hasn't been touched in a long, long time, and it was 
supposed to be replaced many times by something completely different 
(EWF-v2, audio locators, etc). As a result, the code has rotted quite 
badly, and many severe bugs were never fixed. Oh well, now I decided it's 
time to either polish it up or drop it from the codebase altogether. I 
ended up choosing the former path...

Also noteworthy is that the ability to write to EWF files has been now 
deprecated. Let me know if someone will miss this feature.

So here's my first attempt at a better EWF implementation which should 
gracefully handle all the corner cases as well, and in general, be more 
robust.

Testers are now welcome! I listed a few people on the cc-list who have 
reported EWF related bugs during the last few years. This most recent 
snapshot just might fix the problems you reported. :)

Get the git repo with:
   - sh> git clone http://ecasound.seul.org/ecasound.git

Or download a snapshot from:
   - http://ecasound.seul.org/download/snapshots/ecasound-2.4.6.1dev-20080304.tar.gz

--

--
(Continue reading)

Permalink | Reply | 1 comment |
headers
Eric Dantan Rzewnicki | 5 Mar 2008 06:53

Re: EWF refactoring, testers needed

On Tue, Mar 04, 2008 at 11:41:47PM +0200, Kai Vehmanen wrote:
> Hi all,
> 
> I've just pushed a big rewrite of Ecasound's EWF (Ecasound Wave Files, see 
> the user's guide for more info) support to the public git repo.
> 
> The EWF related code hasn't been touched in a long, long time, and it was 
> supposed to be replaced many times by something completely different 
> (EWF-v2, audio locators, etc). As a result, the code has rotted quite 
> badly, and many severe bugs were never fixed. Oh well, now I decided it's 
> time to either polish it up or drop it from the codebase altogether. I 
> ended up choosing the former path...
> 
> Also noteworthy is that the ability to write to EWF files has been now 
> deprecated. Let me know if someone will miss this feature.
> 
> So here's my first attempt at a better EWF implementation which should 
> gracefully handle all the corner cases as well, and in general, be more 
> robust.
> 
> Testers are now welcome! I listed a few people on the cc-list who have 
> reported EWF related bugs during the last few years. This most recent 
> snapshot just might fix the problems you reported. :)
> 
> Get the git repo with:
>   - sh> git clone http://ecasound.seul.org/ecasound.git
> 
> Or download a snapshot from:
>   - 
>   http://ecasound.seul.org/download/snapshots/ecasound-2.4.6.1dev-20080304.tar.gz
(Continue reading)

Permalink | Reply | 0 comments |
headers
Kai Vehmanen | 9 Mar 2008 10:08

Re: Question about loops

Hi,

On Tue, 20 Nov 2007, Rémi R wrote:

> I'm building a Multiband Limitter for FM transmitter with ecasound.
[...]
> With ctest2.ecs :
> -z:mixmode,sum -f:24,2,96000 \
> -a:IN -f:24,2,96000 -i alsahw,0,0 -o loop,1 -o loop,2 -o loop,3 \
> -a:LOW -f:24,2,96000 -i loop,1 -el:bandpass_iir,1500,3000,2
> -el:fastLookaheadLimiter,0,-0.5,0.060 \
> -a:MED -f:24,2,96000 -i loop,2 -el:bandpass_iir,7500,7000,2
> -el:fastLookaheadLimiter,0,-4,0.050 \
> -a:HIG -f:24,2,96000 -i loop,3 -el:bandpass_iir,14000,6000,2
> -el:fastLookaheadLimiter,0,-12,0.030 \
> -a:LOW,MED,HIG -f:16,2,96000 -o stdout
>
> What I want to do is to spleet the output of chain 'IN' to the 3 chains
> 'LOW' 'MED' 'HIG'.

You can achieve this without using loop objects:

  -a:LOW,MED,HIG -f:24,2,96000 -i alsahw,0,0

And then modify the LOW/MED/HIG chains:

  -a:LOW -f:24,2,96000 -el:bandpass_iir,1500,3000,2

.. and so forth. So what happens is that the audio captured from the 
soundcard routed to three chains (LOW+MED+HIG). All of these chains have
(Continue reading)

Permalink | Reply | 0 comments |
headers
Kai Vehmanen | 9 Mar 2008 12:31

Re: erroneous error condition?

Hi,

On Sat, 13 Oct 2007, Noah Mercer wrote:

> When I issue the following command:
[ writing to an mp3 file ]
> I get the following error:
[...]
> (eca-engine) WARNING: An output object has raised an error! Possible 
> causes: Out ... of disk space, permission denied, unable to launch 
> external applications needed in ... processing, etc.
[...]
> Yet the mp3 file is written out correctly and seems fine.

hmm, I've been trying to reproduce this today but so far without success.

> If this were just a one-time problem I wouldn't worry about it. But (a) 
> it happens with multiple files (although not all files) and (b) this is 
> happening in a scripted environment where I depend on the return code 
> from ecasound to let me know whether things were successful or not.

Agreed, this should definitely be tracked down and fixed.

> Does anyone have any suggestions about how to track this down? Does 
> ecasound attempt to do any file permission operations as part of 
> finalizing the output write? (I have group write perms on the directory 
> the file is being written to, but not ownership of it.)

Hmm, the only way for an error to be raised is if ecasound will fail to 
write bytes to the pipe it uses to communicate with lame. Based on your
(Continue reading)

Permalink | Reply | 501 comments |
headers
Kai Vehmanen | 9 Mar 2008 12:34

Re: libecasoundc errors

Hi,

On Wed, 26 Sep 2007, sara lidgey wrote:

> I'm getting this error from libecasoundc:
> (ecasoundc_sa) Error='read() error', cmd='c-select mychain' last_error='' cmd_cnt=142 last_cnt=141.

this usually means that...

  a) ecasound has crashed on the background
  b) the ecasound binary could not be started at all

One thing to try is to start ecasound manually and feed the commands 
(you program will send) one by one and see if it works this way. If the 
cause for error was (a), you should get the same crash this way.

Br, Kai

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Ecasound-list mailing list
Ecasound-list <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ecasound-list
Permalink | Reply | 501 comments |
headers
Kai Vehmanen | 9 Mar 2008 12:37

Re: convert sample rate

Hi,

On Fri, 25 Jan 2008, Julien Claassen wrote:

>  I know about the high-quality resampling and I have it. But alas, it still
> gives a kind of low-frequency crackling. Don't ask me where that comes from.

ugh, oh -- that sounds bad! :( That sounds like the one-off buffering 
error we had a long time ago with libsamplerate (fixed in 2.4.0).

With which version of libsamplerate you get the low-freq crackle?

Br, Kai

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Ecasound-list mailing list
Ecasound-list <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ecasound-list
Permalink | Reply | 1 comment |

Gmane