I think the prop chip might make a nice synthesiser using an algorithm like the Karplus-Strong plucked string algorithm (like a feedback filter) , rather than the sample-based methods. If that could be combined with a couple of cogs running some nice phoneme stuff that is found on this forum, with user input via about four rotary encoders, I think it would be very interesting, and probably quite economical since the prop chip would do it all. Does that sound feasible?
yerpa58 said...
I think the prop chip might make a nice synthesiser using an algorithm like the Karplus-Strong plucked string algorithm (like a feedback filter) , rather than the sample-based methods. If that could be combined with a couple of cogs running some nice phoneme stuff that is found on this forum, with user input via about four rotary encoders, I think it would be very interesting, and probably quite economical since the prop chip would do it all. Does that sound feasible?
I have never worked with the Karplus-Strong algorithm before, But I have heard it in action.
I don't see why something like this couldn't be done in the Propeller. Just look at what Chip and Phil have done with the speech systhesis!
I wrote (Hss) as the most processing and Hub-RAM minimalistic approach.
Basically to get the most channels out of the least CPU and RAM.
7 channels at 32khz sampling rate from 1 COG.
Another COG provides realtime controll, playback and the high level interface.
Looking forward to the release of the HSS.· I have a quick question - will the lowpass settings be·configurable so that the object(s) would work with either a Hydra or Demo Board?
Thanks for working on this, it sounds exciting.· Can't wait for the code!
Filip S said...
What's the url for the webpage, I'm very interessed in this project.
The site still needs to be uploaded to a host. But I'll PM you as soon as I do that.
I'm aiming to get this all up and running in time for next year. Which as we all know means within a week or so.
But I'll keep you posted. Right now I'm looking into adding a few more commands to the music tracker engine and writing a simple demo.
Thank you all who have been waiting for your great patients.
Your welcome Cat. Maybe this code could help excite your other propeller systhesis development.
Having a big old DRAM for a wavetable would sound very nice. ^^
I'm going to write up the rest of the docs when I get the time.
I just added another song to the archive (Both a high quality version and a low quality version) along with 3 sound FX files for people to play with.
I really like your sound engine. I got it up a and running in about 5 minutes. The sound is impressive and memory requirements are great. Do you have any information on how to create the HMUS file? Just listening to your sample files has inspired me. But now I can't wait to get started on my own!
From there you can convert the songs into .hmus format with a tool I developed.
Remember to keep your music in 4 channels and use the S3M format as a starting point.
That's a greate MOD site.· I was introduced to the Mod format some years ago by some Vietnamese friends of mine.· Do you have the converter ready for yet?· Want a beta tester?
mcstar said...
That's a greate MOD site. I was introduced to the Mod format some years ago by some Vietnamese friends of mine. Do you have the converter ready for yet? Want a beta tester?
That's great! I've been looking around for more composers to jump onto this platform!
MCstar if you want. I can send you some information on how to write music for this.
The (hss) tracker tools are still in beta though so they aren't the easiest to use.
But it's very rewarding for any chiptune/mod composer to write something for a physical chip that is supported todate and has growing support.
This is my first time using ModPlug.· I'm considering creating some accompaniment tracks for a performer I know that only plays the guitar.· This may be a way to use the propeller in a live performance for them.· I'm planning on using an SD card for storage of samples if possible to get the best quality.·· Any information you can send on writing the music or converting to the HMUS file format would be appreciated.· I'm not afraid of beta code or command line utilities and the like.· I'm a programmer by trade.· I understand the basics of what Mod is. Basically you take a single sample of an instrument/soundfx whatever, and compose an entire melody/hamony from it.· The sound engine follows the instructions in the composition file to determine what freq, how long, what volumne, to play and any effects that are specified.· Do this in multiple channels to get the appearance of a realistic composition.·· I was considering doing this accompaniment music in midi, but using the propeller just sounds way more interesting.
You can send any files, instructions etc to my email address mcstarATgmailDOTcom.· BTW, you've done a great thing putting this music engine together.· I think it will be used more and more as others find out how awesome it is! ·
I'm very new to Parallax Forums, and I'm not sure if this will help, but there is a great sound recording/editing suite that can be downloaded for free at http://audacity.sourceforge.net/.
If you choose to download it, I would also recommend downloading the Lame MP3 Encoder.
If this comment is not applicable to the subject, I am very sorry.
Ductshui
Yeah, that would be really cool. I play keyboard and thought the Prop would make a really cool sequencer of some sort.
I personally use Goldwave for audio editing. Though at the production studio, I use Adobe Audition.
Goldwave is really cool because it allows more low-level editing of waveforms then other programs.
It displays number of samples instead of number of seconds. (lol)
asterick said...
Splendid audio quality. Much kudos. I would expect nothing less from someone named after one of my favorite FM chips. [noparse];)[/noparse]
Thanks buddy! I always loved the old Yamaha FM chips.
They had a very cool and ringy sound to them.
Anyway in the end, it's all just a lot of fun doing this kind of stuff with the Prop.
Right now I'm writing a quick datasheet on how to use the FXsynth to generate sounds.
Good documention is very important.
I've not had a chance to look at it deeply... but I'm assuming the two cogs work in a pair, one does generation, where as the other generates the PWM signals?
asterick said...
It's an ADSR based system, correct?
I've not had a chance to look at it deeply... but I'm assuming the two cogs work in a pair, one does generation, where as the other generates the PWM signals?
Either way, it's darn impressive.
Yeah, that's pretty close to the truth.
The way it works is one COG does everything a sound chip would do... DAC output, Wavetable Synthesis, Mixing, ADPCM decoding, etc.
The other COG does everything a sound coprocessor would do!
Realtime control of parameters, format decoding, music playback, channel mapping, etc.
It works much like the z80+ym2612 layout of the Sega Genesis.
One COG plays the role of the Z80 coprocessor while another COG plays the role of the ym2612 sound chip!
My design can also work using only one COG. Only your game or application engine controlling it would have to directly update the sound parameters its self and not just issue high level commands to the sound co-processor.
So basically...
1 COG is a Sound Chip. (with 6 channels of wavetable and 1 channel ADPCM)
1 COG is a Sound co-processor running a sound API engine. (That does music playback, sound FXs and decompression)
Send commands to the co-processor and it does the rest while your program is free to do other things. (No CPU overhead from your program is needed)
Thanks man!!
Yeah, I know more and more people will start using it as the Hydra system takes off. ^^
I wanted to write something that is both powerful enough and simple enough for most everyone, while saving the most space one could.
Ym2413a said...
Thanks man!!
Yeah, I know more and more people will start using it as the Hydra system takes off. ^^
I wanted to write something that is both powerful enough and simple enough for most everyone, while saving the most space one could.
It's starting to look like I'm going to have to make my own board. I tried ordering a demo board, but it keeps freaking out and saying that shipping is going to cost $100 to get it to me. (???) The hydra is cool and all, but it's a little high cost for my tastes.
Good thing I have a pile of ICs I can use in my desk.
EDIT: Thank goodness for digikey, and the fact that the people I work for order components though them. Piggybacking orders rocks.
Comments
I have never worked with the Karplus-Strong algorithm before, But I have heard it in action.
I don't see why something like this couldn't be done in the Propeller. Just look at what Chip and Phil have done with the speech systhesis!
I wrote (Hss) as the most processing and Hub-RAM minimalistic approach.
Basically to get the most channels out of the least CPU and RAM.
7 channels at 32khz sampling rate from 1 COG.
Another COG provides realtime controll, playback and the high level interface.
Looking forward to the release of the HSS.· I have a quick question - will the lowpass settings be·configurable so that the object(s) would work with either a Hydra or Demo Board?
Thanks for working on this, it sounds exciting.· Can't wait for the code!
Chat with you soon,
- Earl
The (HSS) website is almost done. The files will be hosted there.
Thanks Earl.
Is HSS a musicplayer or can it be used as a software "soundchip" ?
Sorry for the delay. I was out on vacation.
Cool! I would be interested in helping out to make an editor for the player if anyone else is interested.
It will be interesting to see the code for this... if it could be made to some type of sound "standard" for the propeller.
Will HSS be released any time soon? or are you fully occupided mr. Ym2413a? [noparse]:)[/noparse]
Yeah, The project's website is almost done.
I'm occupided, But I do get some free time every weekend to work on the project when I'm not backed up.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Check out my game page: http://maze3d.no-ip.org (Swedish only, sorry.)
I'm aiming to get this all up and running in time for next year. Which as we all know means within a week or so.
But I'll keep you posted. Right now I'm looking into adding a few more commands to the music tracker engine and writing a simple demo.
Thank you all who have been waiting for your great patients.
www.andrewarsenault.com/hss/
Download and enjoy.
I have a version for the DemoBoard uploaded.
Also two song files are available in the media section. More stuff is on the way.
Post Edited (Ym2413a) : 12/27/2006 1:41:35 AM GMT
Thanks!
THX!
Having a big old DRAM for a wavetable would sound very nice. ^^
Awesome!· Thanks a million Andrew!
Looking forward to info on the file formats (particularly .hmus) as this moves forward.
Happy New Year to everybody!
- Earl
I'm going to write up the rest of the docs when I get the time.
I just added another song to the archive (Both a high quality version and a low quality version) along with 3 sound FX files for people to play with.
^^
--Andrew Arsenault
It's a free windows based tracker program that can write module music.
www.modplug.com/trackerinfo.html
From there you can convert the songs into .hmus format with a tool I developed.
Remember to keep your music in 4 channels and use the S3M format as a starting point.
Many Thanks!
--Andrew Arsenault
·
That's great! I've been looking around for more composers to jump onto this platform!
MCstar if you want. I can send you some information on how to write music for this.
The (hss) tracker tools are still in beta though so they aren't the easiest to use.
But it's very rewarding for any chiptune/mod composer to write something for a physical chip that is supported todate and has growing support.
Have you written anything in MODplug before?
You can send any files, instructions etc to my email address mcstarATgmailDOTcom.· BTW, you've done a great thing putting this music engine together.· I think it will be used more and more as others find out how awesome it is!
·
If you choose to download it, I would also recommend downloading the Lame MP3 Encoder.
If this comment is not applicable to the subject, I am very sorry.
Ductshui
I personally use Goldwave for audio editing. Though at the production studio, I use Adobe Audition.
Goldwave is really cool because it allows more low-level editing of waveforms then other programs.
It displays number of samples instead of number of seconds. (lol)
Thanks buddy! I always loved the old Yamaha FM chips.
They had a very cool and ringy sound to them.
Anyway in the end, it's all just a lot of fun doing this kind of stuff with the Prop.
Right now I'm writing a quick datasheet on how to use the FXsynth to generate sounds.
Good documention is very important.
I've not had a chance to look at it deeply... but I'm assuming the two cogs work in a pair, one does generation, where as the other generates the PWM signals?
Either way, it's darn impressive.
Yeah, that's pretty close to the truth.
The way it works is one COG does everything a sound chip would do... DAC output, Wavetable Synthesis, Mixing, ADPCM decoding, etc.
The other COG does everything a sound coprocessor would do!
Realtime control of parameters, format decoding, music playback, channel mapping, etc.
It works much like the z80+ym2612 layout of the Sega Genesis.
One COG plays the role of the Z80 coprocessor while another COG plays the role of the ym2612 sound chip!
My design can also work using only one COG. Only your game or application engine controlling it would have to directly update the sound parameters its self and not just issue high level commands to the sound co-processor.
So basically...
1 COG is a Sound Chip. (with 6 channels of wavetable and 1 channel ADPCM)
1 COG is a Sound co-processor running a sound API engine. (That does music playback, sound FXs and decompression)
Send commands to the co-processor and it does the rest while your program is free to do other things. (No CPU overhead from your program is needed)
Yeah, I know more and more people will start using it as the Hydra system takes off. ^^
I wanted to write something that is both powerful enough and simple enough for most everyone, while saving the most space one could.
It's starting to look like I'm going to have to make my own board. I tried ordering a demo board, but it keeps freaking out and saying that shipping is going to cost $100 to get it to me. (???) The hydra is cool and all, but it's a little high cost for my tastes.
Good thing I have a pile of ICs I can use in my desk.
EDIT: Thank goodness for digikey, and the fact that the people I work for order components though them. Piggybacking orders rocks.
Post Edited (asterick) : 1/6/2007 3:25:01 AM GMT