Updated build of TiMidity
Alex Taylor
based on Darwin O'Conner's original port.
I made a few minor modifications, mostly to preprocessor stuff to get it
compiling on my system. But I did actually update the program code in
order to fix the annoying bug where it can't find the configuration file
unless you specify it on the command-line every time.
I also created a WarpIN installer package for it, which not only installs
the files and creates a program object, but also adds the program
directory to the configuration file. And I created a companion WPI
containing the EAWPATS collection, which will also update the configuration
file during installation if installed at the same time.
Both packages are currently sitting in Hobbes incoming:
* timidity_player_2104_a1.wpi
* timidity_eawpats_r12.wpi
This was the first step in my quest to improve the MIDI playing experience
on OS/2. What I hope to try doing next is update the OS/2 port to the
latest TiMidity++ release level (2.13.3). I actually have it building
already, but unfortunately it won't play anything; all it does is output
the message "Couldn't open output device". I still have to try and trace
through the code to see why.
I'm even thinking about taking a poke at updating Darwin's ancient
TimidityPM GUI program for the latest code level. But that's getting
really ambitious.
--
Alex Taylor
Fukushima, Japan
http://www.socis.ca/~ataylo00
Please take off hat when replying.
Dave Yeo
> I'm even thinking about taking a poke at updating Darwin's ancient
> TimidityPM GUI program for the latest code level. But that's getting
> really ambitious.
>
http://hobbes.nmsu.edu/download/pub/os2/apps/mmedia/jukebox/audc100.zip
makes a nice PM frontend to Timidity (and others like Z!).
Anyways be nice if you can get Timidity++ working.
Dave
Leonard N Small
>I've just rebuilt TiMidity++ v2.10.4 (text/standalone version) for OS/2,
>based on Darwin O'Conner's original port.
>
<Snip>
>--
>Alex Taylor
>Fukushima, Japan
>http://www.socis.ca/~ataylo00
>
>Please take off hat when replying.
Not sure if it's just me, but when I try to install
'timidity_player_2104_a1.wpi', I get an error from WarpIn about the file
having an invalid header. I re-downloaded the file just to make sure. The
'timidity_eawpats_r12.wpi' file doesn't throw the error. I'm running WarpIn
v1.0.18. Let me know if there's anything else I can provide.
Thanks for all your work for us!
--
Len Small
lns...@verizon.net
Alex Taylor
> Not sure if it's just me, but when I try to install
> 'timidity_player_2104_a1.wpi', I get an error from WarpIn about the file
> having an invalid header. I re-downloaded the file just to make sure. The
> 'timidity_eawpats_r12.wpi' file doesn't throw the error.
Yeah, I know. It looks like I accidentally uploaded the first file in
ASCII mode. (As you say, the eawpats package is OK.) I didn't catch it
because the Hobbes FTP server software very regrettably doesn't report the
transfer mode when uploading.
Unfortunately, the documented process for correcting a bad upload (i.e.
renaming the file) no longer seems to work on the new Hobbes server. I'm
not sure if I have to wait for some period of time before it'll allow me
to rename it.
If I still can't fix it by the end of today, I guess I'll have to email
the administrator...
At any rate, you can get the (correct) file from my website.
http://users.socis.ca/~ataylo00/programming/ports/
Alex Taylor
wrote:
> This was the first step in my quest to improve the MIDI playing experience
> on OS/2. What I hope to try doing next is update the OS/2 port to the
> latest TiMidity++ release level (2.13.3). I actually have it building
> already, but unfortunately it won't play anything; all it does is output
> the message "Couldn't open output device". I still have to try and trace
> through the code to see why.
Incidentally, I now have this version (TiMidity++ 2.13.3) working here.
It looks like not all of Darwin's code actually made it into the mainline
source tree (oddly enough, since the major part of it did).
I'm going to test a while more before I release it, though. There may be
other parts of the OS/2 code missing.
Paul Ratcliffe
> Unfortunately, the documented process for correcting a bad upload (i.e.
> renaming the file) no longer seems to work on the new Hobbes server. I'm
> not sure if I have to wait for some period of time before it'll allow me
> to rename it.
You only get a small window to delete/rename immediately after you uploaded
it. Once that has expired you will never be able to do so.
> If I still can't fix it by the end of today, I guess I'll have to email
> the administrator...
That is now your only option.
Lars Erdmann
be fixed is that Timidity can handle only one simultaneous instance.
That leads to the famous hang on Warp 4 when you open the Midi player.
Reason: the Midi player starts 2 instances of itself (don't ask me why) and
then, it closes one instance (or rather: it would).
But opening leads to Timidity hanging.
Another one: Chris multimedia classes don't get along with Timidity. I would
not be surprised if it is for the same reason.
Lars
"Alex Taylor" <mai...@reply.to.address> schrieb im Newsbeitrag
news:mdq090pMZSKk-pn2-ql0qmdCD1Lgu@mahoro...
Lars Erdmann
I was referring to the MCD version of Timidity and not to the standalone
program. However I would still favour the MCD version over the standalone
version.
It just better fits into the OS/2 multimedia system and it allows to use
whatever player you want as long as it uses the standard Multimedia
interface.
Lars
"Lars Erdmann" <lars.e...@arcor.de> schrieb im Newsbeitrag
news:4a7e9d89$0$31337$9b4e...@newsspool4.arcor-online.net...
Alex Taylor
> I was referring to the MCD version of Timidity and not to the standalone
> program. However I would still favour the MCD version over the standalone
> version.
> It just better fits into the OS/2 multimedia system and it allows to use
> whatever player you want as long as it uses the standard Multimedia
> interface.
Yeah, I'm inclined to think so too.
Unfortunately, I know nothing about MMPM programming, or the WPS, and I'm
pretty whoever debugs the MCD is going to have to.
I have no plans to touch TiMidity MCD, I'm afraid.
Ilya Zakharevich
> Oops,
>
> I was referring to the MCD version of Timidity and not to the standalone
> program. However I would still favour the MCD version over the standalone
> version.
Judging by what I saw on Usenet (+ my limited experience): for
majority (or at least a significant minority) of people MCD locks the
computer.
Hardly a "favourable" situation...
Yours,
Ilya
Leonard N Small
>On Sat, 8 Aug 2009 23:17:42 UTC, "Leonard N Small" <lns...@verizon.net> wrote:
>
>> Not sure if it's just me, but when I try to install
>> 'timidity_player_2104_a1.wpi', I get an error from WarpIn about the file
>> having an invalid header. I re-downloaded the file just to make sure. The
>> 'timidity_eawpats_r12.wpi' file doesn't throw the error.
>
>Yeah, I know. It looks like I accidentally uploaded the first file in
>ASCII mode. (As you say, the eawpats package is OK.) I didn't catch it
>because the Hobbes FTP server software very regrettably doesn't report the
>transfer mode when uploading.
<Snip>
I hate it when that happens ;) Got the good version from your site. Thanks
again!
--
Len Small
lns...@verizon.net
Ilya Zakharevich
> On Sat, 8 Aug 2009 23:17:42 UTC, "Leonard N Small" <lns...@verizon.net> wrote:
> Yeah, I know. It looks like I accidentally uploaded the first file in
> ASCII mode. (As you say, the eawpats package is OK.) I didn't catch it
> because the Hobbes FTP server software very regrettably doesn't report the
> transfer mode when uploading.
>
> Unfortunately, the documented process for correcting a bad upload (i.e.
> renaming the file) no longer seems to work on the new Hobbes server. I'm
> not sure if I have to wait for some period of time before it'll allow me
> to rename it.
Meanwhile, you can upload it as
OLDNAME-corrected.zip
;-)
Yours,
Ilya
Marty
I had it working reliably here, but I needed to modify PLAY.CMD to do so
(when playing using this method). Other methods worked well for me, but
I understand the ALSA drivers had an issue with trying to play multiple
audio streams, with Timidity as one of them, which caused some failures.
--
Reverse the parts of the e-mail address to reply by mail.
Alex Taylor
wrote:
> > I was referring to the MCD version of Timidity and not to the standalone
> > program. However I would still favour the MCD version over the
> > standalone version.
>
> Judging by what I saw on Usenet (+ my limited experience): for
> majority (or at least a significant minority) of people MCD locks the
> computer.
Crashes it (causing it to restart) in my experience, actually. This was
remarked on in a recent thread in the eCS mailing list. This post from Andy
Willis:
The bug is actually documented in Timidity, it is a WPS bug. The new
eCS classes just make it difficult to apply the fix/work around because
they read the files as a folder is opened. There is a way to disable
that, then you could appy the fix/work around but I don't recall now
where I found that setting to turn it off.
On <http://doconnor.homeip.net/timidity.html> you'll see:
*Bug:* Be aware that there is if you play a MIDI using the Player menu
item of a MIDI file or the MIDI app from the multimedia folder it will
cause the Workplace Shell to crash. It you play a MIDI by double
clicking it directly it will work. After you've played it once, the
other methods of playing MIDI files will then work.
The eCS classes load the midi file causing the bug to be triggered and
the desktop restarts so you never get to the point of double clicking it
directly.
Paul Ratcliffe
>> Judging by what I saw on Usenet (+ my limited experience): for
>> majority (or at least a significant minority) of people MCD locks the
>> computer.
>
> Crashes it (causing it to restart) in my experience, actually. This was
> remarked on in a recent thread in the eCS mailing list. This post from Andy
> Willis:
>
> The bug is actually documented in Timidity, it is a WPS bug. The new
> eCS classes just make it difficult to apply the fix/work around because
> they read the files as a folder is opened.
Ah, the CWMM problem again. Mr. CW will never admit that there are bugs or
design flaws in his crappy code, but there patently are. He will never
publish his source code either for peer review.
Alex Taylor
<ab...@orac12.clara34.co56.uk78> wrote:
> > The bug is actually documented in Timidity, it is a WPS bug. The new
> > eCS classes just make it difficult to apply the fix/work around
> > because they read the files as a folder is opened.
>
> Ah, the CWMM problem again. Mr. CW will never admit that there are bugs or
> design flaws in his crappy code, but there patently are. He will never
> publish his source code either for peer review.
Well, to be fair I think reading the files on folder open qualifies as a
conscious design decision, not a bug. It just happens to trigger a
pre-existing bug that lies elsewhere, in this particular case.
I think in this case the responsibility has to lie with TiMidityMCD,
because I'm not aware of any other MCD causing this problem.
Alex Taylor
but never mind.)
I've re-uploaded both TiMidity packages and checked them. Everything
should be OK now.
Lars Erdmann
see my inital postings. The problem is that the MIDI player opens a second
instance of itself. And Timidity MCD cannot handle that. If you double
click, the MIDI player is not opened and only one instance of Timidity MCD
is necessary
I think, once Timidity supports multiple instances instead of just one,
those probs go away.
Lars
"Alex Taylor" <mai...@reply.to.address> schrieb im Newsbeitrag
news:mdq090pMZSKk-pn2-iFoH6b20lhqZ@localhost...
Alex Taylor
> No,
> see my inital postings. The problem is that the MIDI player opens a second
> instance of itself. And Timidity MCD cannot handle that. If you double
> click, the MIDI player is not opened and only one instance of Timidity MCD
> is necessary
> I think, once Timidity supports multiple instances instead of just one,
> those probs go away.
Interesting, thanks.
Do you have any idea what kind of changes would need to be made to the
MCD code? Is it fairly standard logic, or would it depend on the code?
Mentore
> On Sat, 8 Aug 2009 14:19:02 UTC, "Alex Taylor" <mail...@reply.to.address>
> wrote:
>
> > This was the first step in my quest to improve the MIDI playing experience
> > on OS/2. What I hope to try doing next is update the OS/2 port to the
> > latest TiMidity++ release level (2.13.3). I actually have it building
> > already, but unfortunately it won't play anything; all it does is output
> > the message "Couldn't open output device". I still have to try and trace
> > through the code to see why.
>
> Incidentally, I now have this version (TiMidity++ 2.13.3) working here.
> It looks like not all of Darwin's code actually made it into the mainline
> source tree (oddly enough, since the major part of it did).
>
> I'm going to test a while more before I release it, though. There may be
> other parts of the OS/2 code missing.
Nice work Alex. I was trying to complete a diff between the two source
archives (2.10.4 for OS/2 and 2.13.3 sources) before my main machine
broke, so I was ought to stop.
Now I see someone else has my very same interests :D Have luck and
publish it as soon as possible! I wonder if there's a way to build
another GUI over it.
Mentore
Alex Taylor
> > Incidentally, I now have this version (TiMidity++ 2.13.3) working here.
> > It looks like not all of Darwin's code actually made it into the mainline
> > source tree (oddly enough, since the major part of it did).
> >
> > I'm going to test a while more before I release it, though. �There may be
> > other parts of the OS/2 code missing.
>
> Nice work Alex. I was trying to complete a diff between the two source
> archives (2.10.4 for OS/2 and 2.13.3 sources) before my main machine
> broke, so I was ought to stop.
>
> Now I see someone else has my very same interests :D Have luck and
> publish it as soon as possible! I wonder if there's a way to build
> another GUI over it.
Funny you should mention that... see my other message. :)
Paul Ratcliffe
>> > The bug is actually documented in Timidity, it is a WPS bug. The new
>> > eCS classes just make it difficult to apply the fix/work around
>> > because they read the files as a folder is opened.
>>
>> Ah, the CWMM problem again. Mr. CW will never admit that there are bugs or
>> design flaws in his crappy code, but there patently are. He will never
>> publish his source code either for peer review.
>
> Well, to be fair I think reading the files on folder open qualifies as a
> conscious design decision, not a bug.
That's why I said "or design flaws".
Mentore
> On Thu, 13 Aug 2009 08:56:23 UTC, Mentore <mentore.sie...@libero.it> wrote:
> > > Incidentally, I now have this version (TiMidity++ 2.13.3) working here.
> > > It looks like not all of Darwin's code actually made it into the mainline
> > > source tree (oddly enough, since the major part of it did).
>
> > > I'm going to test a while more before I release it, though. There may be
> > > other parts of the OS/2 code missing.
>
> > Nice work Alex. I was trying to complete a diff between the two source
> > archives (2.10.4 for OS/2 and 2.13.3 sources) before my main machine
> > broke, so I was ought to stop.
>
> > Now I see someone else has my very same interests :D Have luck and
> > publish it as soon as possible! I wonder if there's a way to build
> > another GUI over it.
>
> Funny you should mention that... see my other message. :)
Saw it just some minutes ago. I'll surely take a look at it :D
Mentore
Alex Taylor
> > > Nice work Alex. I was trying to complete a diff between the two source
> > > archives (2.10.4 for OS/2 and 2.13.3 sources) before my main machine
> > > broke, so I was ought to stop.
> >
> > > Now I see someone else has my very same interests :D Have luck and
> > > publish it as soon as possible! I wonder if there's a way to build
> > > another GUI over it.
> >
> > Funny you should mention that... see my other message. :)
>
> Saw it just some minutes ago. I'll surely take a look at it :D
I've fixed several problems since I uploaded that build... I'll try and
put another one up on the weekend.