[linux-audio-user] [ANN] Audio* Tekno Composer

Hi all, I have got the second alpha release coming on Sunday. Audio* is a project to create real-time music software as Open Source. I announced the project a week ago. http://audiostar.sourceforge.net There is plans to make a Reason-like program next for Linux. I have moved to Linux for good. I don't even notice I am using it anymore, there's no differece between using Linux or Windows to me anymore, except one is free and get's better latency. Tekno Composer is a copy of Rebirth without copying it. (A project named Reborn tried and was pulled because of copyright problems). Tekno Composer features a Silver Mojo synthesizer, a 16-voice polyphonic drum and sample stack wavetable synthesizer, and a demented Loop machine I am still having problems getting to sync properly but should be fixed by release tommorow. It is just as good as Rebirth almost (it needs features still not implemented). I have plans to make a Rebirth file loader so you can listen to Rebirth files on Linux since Propellerheads never supported Linux in any way, although it will not be able to play them exactly the same as Rebirth (remember it's not a clone of Rebirth, it is just an alternative that sounds pretty close to it *snicker*, I think it's better than a 303... more toys to play with). Silver Mojo is a monophonic, 3 oscillator synthesizer with plenty of features and toys to make cool sounds. It was more or less designed to copy a SH-9, an M-100, Korg MS-20, and Mini Moog, but with a step sequencer like a 303. That is why there are 3, and one is label SUB but you can make it up to 2' pitch if you want. It has two LFOs which can be synced to the tempo. One goes to the oscillators (PWM and Vibrato) and one goes to the filter. It also has Tremolo and Ping/Pong panning which will be syncable to the tempo soon; I am real busy on other stuff right now. It has all the basic F/X pedals you could want, and one custom Quad Chorus I created which is a multi-tap chorus box and sounds very fat. It has a 64-step sequencer, like a 303, but a little different. It is not an emulator, and does not intend to be one, because most projects that copy Roland gear or Rebirth get into trouble with them. The drum machine is a 16-voice 16-stack polyphonic wavetable synthesizer. With some more work it will be a synthesizer in it's own, much like the stacking synths of the 80's, except you can use any wave form, not just a period of a wave in it. You do not really need to use it as drum machine if you load other samples in it. It is very much like an 808/909 programming wise, 16-step sequencer. Right now it uses a very primitive text file to load the samples... but in the future it will be an XML file format. The Loop machine (will) allow you to sync samples/loops to the drum machine beat. You can loop or do one-shot synced to a step on the drum machine. It also has a barbaric text file to load the samples. That too will be XML in the future. I can only do so much in a weekend. It should run on a P2-450 but I have only tested it on a P4 2.0 and a P3 550 so far. The goal is for it to run on a Pentium Pro 200 Dual system as the base line. It is not designed for speed but for the highest quality sound possible that I can figure out. I do not have oversampling enabled yet it is not fast enough for it. You will need FLTK 1.1.X and Port Audio v18. I have an ALSA driver in there, but I have not confirmed it works yet it is a rip from my other project. There will be Jack drivers as soon as I can get Jack to startup as a daemon on my system. It is still Alpha and not all of the program has been implemented yet. The project could use one dedicated programmer to help implement features. I can only do so much at a time by myself. Although Rebirth isn't any big deal now, I wouldn't feel complete if I didn't build a program that could mimic it without infriniging on copyrights in any way :) I'd like to thank Julius Orion Smith from CCRMA for all of his papers, and Tom Stilson for analyzing the Moog VCF transfer function, and Perry Cook for STK. It took about 6 months of intense reading of them ( I read the BLIT paper 100 times), but I learned alot and because of that I was able to make a program like this. I thought it was fun, but frustrating at times, getting to the point I could make something cool with all that technical information. Even though I've been programming for 10 years I didn't know alot about DSP and DSP is not something you learn in an afternoon (unless you are a super-genius type). So many thanks to the awesome people who took the time to explain the math and formulas, and also thanks for Music DSP for the many code snippets. Also I'd like to thank Alex (something) who made the DigiFX, where I learned how to code many of the sound F/X. Big thanks and much gratitude to you all. I hope I can return some cool stuff in exchange for that knowledge. That's the reason I am making it Open Source and free to say thanks for teaching me. Thanks list! -- Nick <nicktsocanos(a)charter.net>