Ok, looks like the load/save routines are broken because of a change in the compiler at some point that makes it to assume there should be  padding. The worst part is that I think some songs have been saving with the padding (and until now they have loaded OK), so I need to include code that handles that padding. What a clusterfuck.

IceWolf wrote:

I got an error trying to compile it on a 64bit debian based distro.

You should have both the klystron and the klystrack sources like so:

~/code/klystron
~/code/klystrack

Then simply go to ~/code/klystrack and make. That will run make also im the klystron directory. There is now bin.debug/klystrack, which is the binary. You need to copy the "res" directory (~/code/klystrack/res) with the binary, it contains the themes and is required. That should be all that is needed.

Now, the instruction set problem is probably because there is a "-mpentium2" flag in the Makefiles. I think it might work if you just remove that in both Makefiles (klystron and klystrack).

Ah noticed some songs indeed crash everything. I'm on it. At least castlevania.kt loads properly.

Edit: Ok the load routine is broken, dammit.

Edit 2: Songs saved with the current version seem to load A-OK so this affects only for older songs that have effects, need to look at this bug tomorrow. Dammit, dammit, dammit.

I just compiled it on Ubuntu, works perfectly.

Ok, time for another preview of 1.6... new stuff include the above command and also a song timer (attn n00bstar) and some other fixes.

http://code.google.com/p/klystrack/down … n=2&q=

OK, added command 2D00 which skips to the end of the current pattern. Basically, if you have a 64-step pattern that has 2D00 on row 15, it will make the whole song jump to the row right after the last row of the pattern.

Hmm I think I'll add something for doing that. It would be silly to have to put the rest of the patterns on weird rows because of a shorter pattern!

IceWolf wrote:

Hope it's ok if I ask a klystrack question in this thread. Does klystrack's skip pattern feature work? I tried command 0D (I'd used milkytracker before and have seen this command in action) and it didn't skip to the next pattern at all like it should. I tried 0D00 as well as 0D09 (9 was the next pattern I had).

In klystrack, Dxx actually jumps to the position inside the current pattern. For example, if you have a 16 step pattern and have 0D00 on the very last row, it will loop back to the beginning and the pattern will play until another pattern comes up in the sequence. All other channels remain unaffected.

I'm not sure if the behavior you were looking for is useful given you can put patterns on any row in the sequence (not just multiples of, say, 16).

Settings needed are just enable/disable MIDI and which channel to listen to. If there's a start signal klystrack will start playing the song and stop signal will stop etc. If there's no such item in the settings menu it might be disabled in the build.

The sync tries to sync klystrack from an external clock signal but it falls out of sync quite fast (I have no idea how to do this). You can also use the MIDI keyboard to play notes but I don't know if it works as intended at the moment (enabling MIDI devices on my computer simply freezes klystrack so it's kinda hard to debug...).

The listening party was way past my bedtime! Thanks mom.

Also, I ordered the physical CD while waiting for that vinyl, which I would prefer.

P.S. Vinyl.

Excellent!

I'll add some extra debug messages that probably show where it hangs.

Thanks, that should help if it has anything to do with the changes between those two.

That font.bmp issue is irrelevant since it basically means "didn't find this file in the current theme, so I will load it from Default instead".

until 1232

Do the r1230 or r1230M nightlies work?

n00bstar wrote:

Ah yeah. Dass wut I'm talkin'bout.

I still crash with the process being stuck in the list when I try to quit though. Anything I can do to help you with that? There doesn't seem to be any kind of log anywhere hmm I'm about to reinstall Windows though, I'll let you know if that fixed anything.

Maybe it's related to the multiple soundcard setup. Does the stderr.txt say "klystrack has left the building"?

Not yet.

EDIT: Uploading right now.

Alright, added to the trunk. Turns out the oversampling is not needed to eliminate the distortion, I just had a bug in the interpolation code which caused similar noise. The oversampling code exists in the source, it can be turned on in case there still is distortion in some cases.