#!sh
AC_DEFUN([LT_LIB_M],
[AC_REQUIRE([AC_CANONICAL_HOST])dnl
LIBM=
case $host in
*-*-beos* | *-*-cegcc* | *-*-cygwin* | *-*-haiku* | *-*-pw32* | *-*-darwin*)
  # These system don't have libm, or don't need it
  ;;
*-ncr-sysv4.3*)
  AC_CHECK_LIB(mw, _mwvalidcheckl, LIBM="-lmw")
  AC_CHECK_LIB(m, cos, LIBM="$LIBM -lm")
  ;;
*)

Comment too long; full version here.

What's the practical difference though? Libtool is installed on most common 'nix build environments, at least I haven't had to deal with any issues like you're encountering. If there's some good reason that I should abandon the macro (other than "I have to install libtool and I don't want to" :) I'm not opposed at all to handling it the old fashioned way.

/usr/lib64/gcc/x86_64-suse-linux/4.6/../../../../x86_64-suse-linux/bin/ld: fmt/aiff.o: undefined reference to symbol 'floor@@GLIBC_2.2.5'

sh ./scripts/build-font.sh . font/default-lower.fnt font/default-upper-alt.fnt font/default-upper-itf.fnt font/half-width.fnt >auto/default-font.c
/bin/sh: auto/default-font.c: No such file or directory
make[1]: *** [auto/default-font.c] Error 1

Solution is to mkdir the dir before sh is run in the rule.

It'd be pretty dang cool though, so if someone else does happen to have the interest in doing it, I'll apply their patches.

....otoh, i also remember reading (at least the ST3 readme file) and asking myself "what kind of stupid idiot would want to bend/slide across samples of different frequency?"

XMPlay can render instrument-wise too. I can't find any details about Magicfish, but I know XMPlay generally has excellent quality/accuracy :)

the official site for the program is down but if you cant find it on google just let me know and i will send it.

This probably has to do with Modplug calculating everything by periods instead of frequencies, which actually causes a ton of dumb bugs. I'll have a look.

• Category: Playback
• Complexity: Difficult
• Priority: Normal
• Status: Acknowledged

• Category: File I/O
• Complexity: Difficult
• Priority: Normal
• Screen: Export Module
• Status: Acknowledged

Obviously this process can get a bit tedious when the track has many instruments, so some kind of way of selecting which instruments to save as WAV would be nice to have for me. I realize this is a pipe dream type of request, and I might be the only one who uses Schism Tracker in this way. Still, some way to diskwrite to WAV only selected instruments would be greatly appreciated.

that's awesome!

yuck, what possible use is that?
sounds depressing.

• Complexity: Low → Trivial

thanks for the help!

The behavior here sounds to me like you're trying to use the instrument page (F4) instead of the sample page (F3). If you're loading a .wav file or similar, you need to be on the sample list. The instrument list is for setting up multisample instruments, envelopes, and stuff, there's no actual sample data there.

The only times I have what you describe is when I enter a directory that doesnt have what Schism searches for. Happens only in dirs with many files.

When I open the sample browser, and try to load a file, the list of files quickly disappears one by one, leaving only the directories. When I navigate into a different directory, once again, the files are displayed for a few seconds and then vanish.

This problem prevents me from loading any samples into the program. Is this a known issue?

>You know, you can fiddle with the buffer size if you want.

Yeah, but it has no effect with alsa:default:0 or any other audio device setting that works. Otherwise I crank it to 2048. But I favour multiple access to audio over high-latency stability.

1024 is the buffer size I get at 48 kHz, and it's actually less prone to underruns, but the screen refresh is a lot worse. Also tuning problems as discussed in >>1447

You know, you can fiddle with the buffer size if you want. It's buffer_size= in the [Audio] section of the config. I have meant to add a slider thingy to adjust it in-program but I'm really lazy when it comes to changing the GUI because it was coded stupidly in the first place. (what was I thinking?! oh right, I wasn't.)

I just tried alsa:dmix versus alsa:default:0 on my desktop, and both of them ignore the buffer size I request, giving me 1024 samples.

Understandable. Just so long as it's not my ears that are broken...

>>1476
My choices were bear or pony! It's not like I was offered a car or a video card or something ;p

I have had issues with the dmix driver giving terrible latency and completely ignoring the requested mixing buffer settings (among other issues). At present I have a buffer size of 256 set in the config, but with -a alsa:dmix Schism Tracker winds up with a 1024-sample buffer. I've even seen dmix hand out 4096 samples, which is particularly bad. My answer to this was to require 100% attention to the audio hardware unless you explicitly request otherwise.

For whatever reason, even though I have rear-out on my sound card, the same audio still goes to all channels... except for DVD playback. I have no idea how that works, but the net of it is that I can't test what any of this does. :\

If the problem is in fact the periodic tuning, it'll be a big pain to fix since that'd require rewriting almost all of the code that handles pitches, which is like half the player.

• Category: Playback
• Complexity: Difficult
• Priority: Low
• Status: Rejected → Acknowledged

Note that many IT modules were not created with real surround speakers sets in mind, so many use the surround (or rather phase inversion) effect for stereo tricks rather than real surround positioning.

Instead of typing the -a parameter every time you can just add it to ~/.schism/config under the [Audio] section, i.e.
driver=alsa:default:0

schismtracker -a alsa:default:0
schismtracker -a alsa:dmix
schismtracker -a oss:/dev/dsp

aoss schismtracker
aoss schismtracker -a oss:/dev/dsp
(get aoss from a package called alsa-oss, probably)

lsof | grep dsp
lsof | grep snd
end or kill any processes that come up from these commands

I used the 1/1/11 build and a self compiled hg build per wiki notes. Both had the same result.

I have no clue what to do next.

Alternately, find other programmers who are willing to submit working patches. :)

wonderful! wonderful! :)
what can we do to assist you with midi?

> Is a keyjazz possibility for midi-ports possible?

Yes.
Likely? Not unless I can figure out why ALSA MIDI apparently stopped working with Schism on both of my computers.

Hung notes wouldn't be problematic, as soon as you let go of the MIDI key, it would send a note-off, just like on the PC keyboard.

Is a keyjazz possibility for midi-ports possible? The only issue I can perceive is having to have a GUI element which allows for setting the length of the note being toggled - so that it is not just a hung note (altho, this would be intriguing with some synthesizers + software + arpeggiators)..

I think there's some (Dolby) hardware out there that can recognize inverted data and route it somehow to another speaker, but I haven't used such a system so I have no idea. (IIRC, Jeffrey Lim didn't know if it actually worked either...)

I have opened some existing files and seen Surround used. I also see that there is an Effect that will put a track into Surround mode. But, I'm left with questions:

1. Is what Schism Tracker calls "Surround" merely applying some kind of phase/image/delay effect and still outputting a Stereo signal?

2) Did the original Impulse Tracker support sending things to more than 2 channels, or was its definition of "Surround" the same thing discussed in Question (1)?

Thanks for any help :)
rs

• Category: File I/O
• Complexity: Moderate
• Priority: Low
• Status: Acknowledged

• Status: Rejected

But the crash: Resizing a pattern while playing the pattern crashes Schism

• Category: Playback
• Complexity: Trivial
• Priority: Normal
• Screen: Order List
• Status: Acknowledged

Anyway, I'm giving this some thought. There might be a way to swing it without actually adding a ton of code... and it might allow for other neat tricks too. I'll tag this so I remember Later™.

• Category: File I/O
• Complexity: Difficult
• Priority: Low
• Status: Mythical Future

Aside from your technicality arguments, from a users point of view this idea is not so bad. A lot of (single) mods are offered as compressed downloads from numeral sources. I for one find it a nuisance to uncompress every single archive containing a module I want to listen to. Its not much of a matter of size but more of user friendliness.

• Category: User Input
• Complexity: Moderate
• Priority: Low
• Status: Acknowledged

now it's monday all over the place!

you didn't remember, and it's already saturday!!

:)

I guess I could add a "don't ask before quitting" option. Bump this if I don't remember to put the bug meta stuff on it when I get home Monday.

yeah :)
that would be splend1d

schism tracker doesn't receive focus, but does get a quit signal.

an option to not prompt before quitting (unless if there's something to save) would probably solve this fully.

Would be simpler just to stuff in an option to not prompt before quitting, imo.

LADSPA is to Linux as VST is to Mac and Windows, except (so I've been told, I never played with either) it's overengineered and more complex to implement, and there's far fewer plugins.

If one resizes with F2-F2 all the patterns from pattern 10 to pattern (max) to 128 or 192 row lengths, you'll see a drastic growth in the song size. If I would then want to try and get it back down, I'd try and re-update those said patterns back to 64, but wouldn't know how to else save some song-data-space with the .IT

(I was trying to optimize .it's for smaller filesizes, and didn't find the updating to be snappy enough, hence the original post)

Now, this "Close the program" concept has two aspects, one of closing the program (without a dialogue) and one of asking to do something about the currently open document, which has not been saved.

Schism differs from this regular convention in one way:
When you Quit after having saved a song, you are asked: "Do you wish to quit?" (or some equivalent of that. That is quite fine by itself, sure, when the user is pressing Ctrl-Q to Quit, while Schism Tracker is active on the screen.

But there's also one other method of quitting - this in OSX.
I'll try and explain.
Alt-Tab or Cmd-Tab are the accepted methods of switching tasks.
I don't know about Windows, but in Cmd-Tab in OSX, one is able to actually quit applications by pressing Cmd-Q.

Comment too long; full version here.

Hi Storlek. Ctrl-O as a global key would be wonderful. Then it would work in F2, F3, F4 and F11! and the infopage too. Infopage would be the best - because then it's easy and quick to toggle channels.

• Category: User Input
• Complexity: Trivial
• Priority: Low
• Status: Accepted

As for #3 I suppose the Ctrl-O could be generalized to a global key rather than implementing it per-page. At least, I don't think there's anything that's stopping that key from being global...?

The first thing, tbh I have no idea what you're asking for.

That was never an .it file.

00000000  45 78 74 65 6e 64 65 64  20 4d 6f 64 75 6c 65 3a  |Extended Module:|
00000010  20 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  | ...............|
00000020  00 00 00 00 00 1a 46 61  73 74 54 72 61 63 6b 65  |......FastTracke|
00000030  72 20 76 32 2e 30 30 20  20 20 04 01 14 01 00 00  |r v2.00   ......|
00000040  0a 00 00 00 40 00 0a 00  27 00 01 00 06 00 b4 00  |....@...'.......|
00000050  05 05 07 07 04 04 03 03  09 09 00 00 00 00 00 00  |................|
00000060  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
*

if something hasn't been saved, it will ask "do you want to save the current song" (=the usual thing).

every other application will quit on cmd-tab-cmd-q, but not Schism :)

2) ctrl-o (overwrite,new,cancel) - i.e. if you're overwriting a sample that exists, you can instead select New and it'll be diskwritten to a new sampleslot (unless if all 99 sampleslots are in use)

3) Ctrl-O in F3 - diskwrite currently playing pattern to currently selected sampleslot.

It's probably worth mentioning that the crash only manifests on a 64-bit build. In 32-bit, penv is essentially assigned to csf->instruments[-1], which is a valid location that always contains 0 – so it accidentally gets the right answer.

I'd love to see it.

--enable-debug just tells it to keep symbols and not optimize the code so that you can attach a debugger and see what's broken.

If you apply >>1408 (try: curl http://schismtracker.org/scdev/src/dw-crash.patch|hg import - --no-commit) does it stop crashing?

If I'm in instrument mode, and I play a sample from the sample list, and then try to press q in the pattern editor, crashes.

Segmentation fault.

I ./configured with –enable-debug but it still just says "segmentation fault" and crashes with no extra info.

I never used to have this problem, so I went and downloaded the 01012011 build, and it still happens.

RIP schismtracker