nitro pulse wrote:Get lsdj get a super gameboy get a snes put lsdj in the super gameboy turn on the snes . And there you go a seqencer on your snes
You could do that. :-P (I'm assuming that's a joke!! although fair play... ) But then you aren't taking advantage of the whole APU in the SNES.
So I'm still here with Ferris testing / building / coding (reverse order) the SNES music tool. Ferris is kickin' goals with some mad devs on the play-routine code and sample basics.
Updates:
- Looping samples.
Kudos to Ferris for finally getting this ticking. It's quantised to a 16 sample loop so you gotta be pretty mathematical on your note frequency vs. samplerate if you want a correct zero crossing on loop. I'm in the process of making a library of single cycle looping wave forms including sampling from MC202, DX7, Prophet 600. When this is released, it's going to be dope ++ on the sample department. Trying to find the right combos to get cleanly converted waveforms ;-) Triangle and square / pulse cycles sound mad. Sine sounds.... well, sound SNES "dirty". The BRR audio compression system "vectorises" (for the sake of a better term!) the waveform. So everything you import has this "SNES" grime about it. It's damn cool - but not in a subtle way!
^^^^ Yes.... waveforms get deeeeestroyed! Part of the fun of the SNES ;-)
- Envelopes
Yup - envelopes are working. We want to eventually add some graphical representation of what the envelopes actually are doing. As far as functionality (along with loops) this is a big step to shaping the sound in that SNES way.
- RLE Compression of Patterns
Finally!!! We are getting some optimisation into our play-routine :-) Where, a week ago, every pattern was eating 5kb of APU RAM... now it's a fraction of that (depending on complexity). We also have 16 note (or "1 bar" of 16ths in 4/4) detection of repetition. For those who want to do odd time signatures... we'll do pattern jump in time - or maybe offer variable windowing.
Ferris also fixed a huge frikkin bug on the ROM image exporter. In fact re-coded it. The cool thing about the two of us workin' on this is that Ferris does some insane piece of code... then I break it. From the outset I'm applying every command-combo trick I know (from the Amiga + DMG) to push this baby to the limits. From shuffles to echos to stupid tempo shifted arps to a ridiculous number of samples... it's all being tested as we go, and for the most part is very stable.
So yup (pun intended) : just wanted to let you all know that it's still happening. Step by step. Just like Bobby Brown said in the 90's... ("every little step you take"). Ehhh?! Hey, it's a sat night and I've been drinkin' beers and tracking all night ok?!
Still can't give an ETA on a public beta at the mo. Tool is very early stages and as long as we are having fun and it's coming along basic functionality wise, we will keep it alpha. Until the day one of us says "you know what? I think we could public beta this shiz..." And that day will come!
Also - TUNE: http://weeklybeats.com/#/ctrix/music/su
ld-of-tool (scroll down past description and hit play)
^^^ that's what I'm making so far. (and this was pre the looping samples build with proper envelopes!)
Huuuuugeeee big-ups to Ferris. He really is the driving force doing all the assembly coding on this one. I'm just the testing / bug reporting / complaining / nodding at the code / sample lib making / hyping-up guy. Yes, I look at the DSP / ASM code (and countless block diagrams) and get what he's achieving, but it's a true joy to work with a fellow chip artist who methodically makes efficient assembly / C++ routines happen. In that logical "oh, cool!" way :-) Right down to the SNES BRR audio compression level (which shouldn't even work!)... and writing an assembler from scratch for the SPC700. The work on the true guts of this project has been massive.
We're a lucky bunch to have this guy on board! Which means I owe him beer... damn.
Last edited by cTrix (Apr 20, 2014 3:56 am)