@HS: I downloaded the AXS demo for DOS from Maz soundtools-page and it won't run unless I disable audio, complaining on the lack of 16 bit soundcard (which is probably more a software than a hardware problem imho). I tried some tricks but nothing worked. The free Windows-version worked nicely on the laptop though.

@ijsf: Nice to read about the MIDI-support and the googleproject. :) Will update the links here. I also mainly work with Renoise (everything at http://soundcloud.com/platimatic) but I would probably have a hard time making the fm-sounds fit within my music (and I've gone through a couple of Yamaha FM-synths over the years). :)

@HS: Sure, if you can provide with me a link to the DOS-version (all I could find online was the win-version), I'll give it a shot.

@marm: I would keep the 'LCD Display stretch' on and instead play with screen_mode (and perhaps comp_text_mode) in the adtrack2.ini-file. See what works for ya. :)

@ encore -> regarding your computer, can you try and run AXS for DOS and see if it runs? If so, I'm going to hunt down this computer of yours.

@ijsf: I have no words, hats off and all my thumbs up! :-) As soon as I catch Encore online, we will update the page news / links.

http://at2.googlecode.com/

I just spent a few hours porting the code over to FPC. It's perfectly possible, if you enable Turbo Pascal compatility, swap out the TMT-specific stuff (DPMI unit, a few types/calls such as TRealRegs, etc.) with their Free Pascal counterparts. AFAIK inline assembly is in Intel-style syntax with TASM/MASM dialect, in theory you can all enable that in FP.

Unfortunately.. and extremely frustating.. I basically ended up giving up because of bugs.. in Free Pascal, fairly ridiculous. They claim to be a "professional" compiler, but the bottom line is that the DOS support just sucks balls and elementary stuff such as external assembler support (TASM/MASM) is completely broken.

I guess I'll open up a GoogleCode project for this stuff.

@ijsf: That was quick... you're just great, man :) I hardly believed somebody would make a crap in that undocumented stuff! As your question regarding Google Code and GitHub - please feel free to submit the sourcecode to any code-hosting-alike site you find interesting, I currently don't have time to make such things in 'good quality', it means I would just register with mandatory minimum to make a new project, which would result in bad starting point. Please take this as you are doing me a favor :-)

@GhstWlf: Honestly, can't really remember what exactly was the point, but I'm sure I tried to use some of the latest version of FPC back then (around 2005 or so), but there were issues which were not trivial to resolve (too much inline asm to rewrite, or something like that). But feel free to try this... or anybody. I would be grateful, and so will be AT2 community, maybe. As long as I can remember, FPC-compiled code was way much better and faster than the TMTPC stuff.

@ijsf (and probably subz3ro): Have you tried free pascal to compile AT2?

@Newk: No, basically because this software is solely written for MS-DOS. It'd pretty much require a complete rewrite.

@subz3ro: BTW, have you thought about hosting your code on Google Code (or perhaps GitHub) and adding some other contributors? IMHO it's a neater place to put some source+docs, etc. in comparison to SF.

Also, I stumbled upon a bug with the latest AT2 release .exe crashing on the newer (pentium 3) slave system I was testing on. Compiling the sourcecode with TMT Pascal 3.90 seemed to fix the problem.

ijsf: awesome!

would it be possible to compile it to run on linux machines?