Blitter sound routine
Moderators: simonsunnyboy, Mug UK, Zorro 2, Moderator Team
Re: Blitter sound routine
can you help me with how the timer and setting he tempo is accomplished?
The radioactive half-life : )
Atari is a lifestyle,not a hobby.
HOLD ON ! ! ! Im printing unreadable characters ...!
Atari is a lifestyle,not a hobby.
HOLD ON ! ! ! Im printing unreadable characters ...!
Re: Blitter sound routine
The score is interpreted at each frame (50hz).
The $F (15 in decimal) effect allow to specify how many frames a line into the score will last.
For lowest tempos there is another calculation method as described in the mod format
http://www.textfiles.com/programming/FO ... d-form.txt
(look at 15 Set speed effect)
The $F (15 in decimal) effect allow to specify how many frames a line into the score will last.
For lowest tempos there is another calculation method as described in the mod format
http://www.textfiles.com/programming/FO ... d-form.txt
(look at 15 Set speed effect)
Re: Blitter sound routine
thanks ....ive been trying for years to implement a tempo adjust for a midi playback routine I wrote by adjusting the timers resolution...maybe this will help .....
brace yourself
ill prob have more questions
brace yourself
ill prob have more questions
The radioactive half-life : )
Atari is a lifestyle,not a hobby.
HOLD ON ! ! ! Im printing unreadable characters ...!
Atari is a lifestyle,not a hobby.
HOLD ON ! ! ! Im printing unreadable characters ...!
Re: Blitter sound routine
I don't have anything to contribute but I really like this discussion, great job 

Re: Blitter sound routine
Thanks 

Re: Blitter sound routine
Major update of the sound routine available here :
https://github.com/gibs75/demOS
https://github.com/gibs75/demOS/tree/ma ... CS/BLITSnd
https://github.com/gibs75/MilkyTracker-fork
(this version used in this prod https://www.twitch.tv/videos/1553977161?t=8h17m22s)
- 4 + 3 voices with added support for standard YM (no timer) voices
- conversion from .xm files allows variable length patterns
- more tools
https://github.com/gibs75/demOS
https://github.com/gibs75/demOS/tree/ma ... CS/BLITSnd
https://github.com/gibs75/MilkyTracker-fork
(this version used in this prod https://www.twitch.tv/videos/1553977161?t=8h17m22s)
- 4 + 3 voices with added support for standard YM (no timer) voices
- conversion from .xm files allows variable length patterns
- more tools
Re: Blitter sound routine
Really great work, metalages!
>> > raZen/Paradox < <<
Atari 1040STE, TOS 2.06, 4MB, MC68010, IDE 8GB SSD, Gigafile
Atari 1040STE, TOS 2.06, 4MB, MC68010, IDE 8GB SSD, Gigafile
Re: Blitter sound routine
Thank you
Also the result of the great tunes by Dok

Also the result of the great tunes by Dok

- viking272
- Atari Super Hero
- Posts: 705
- Joined: Mon Oct 13, 2008 12:50 pm
- Location: west of London, UK
Re: Blitter sound routine
This is really cool, great work! I played the Twitch stream and cranked up the volume.
Soon I had all the kids in the study jumping and dancing away! Great memories
Soon I had all the kids in the study jumping and dancing away! Great memories

Re: Blitter sound routine
I think about adding a new feature.
Currently we have an hybrid configuration with voices blittered PCM / YM with their own pro / cons :
- YM :
* portamento (variable freq)
* squares (except if the STe has LCM -12db mix hardware fix) because of mixing level. only 10/11 volume levels (else it is too loud)
* no YM interrupt effects (the goal is to fit the upper border, also interupts would collide with blitter when running) => no pulse width
- blittered PCM :
* no portamento (pre determined frequencies)
* steep volume effects (7 levels / -6 db steps)
* samples (looping or not)
* dynamic bit decimation
The new feature would provide something in between what YM and current PCM routine offers.
This new feature will be a blittered PCM square.
The square wave is specifically convenient because of its constant values, so you can loop the blitter where you want
Because of this, with only one source square sample in memory and setting up the loop accordingly it would give :
* portamento (variable freq)
* square on 127 volumes step (with input value of 0 or 127 constant, volume can be done throught half tone masking)
* variable pulse width
Currently we have an hybrid configuration with voices blittered PCM / YM with their own pro / cons :
- YM :
* portamento (variable freq)
* squares (except if the STe has LCM -12db mix hardware fix) because of mixing level. only 10/11 volume levels (else it is too loud)
* no YM interrupt effects (the goal is to fit the upper border, also interupts would collide with blitter when running) => no pulse width
- blittered PCM :
* no portamento (pre determined frequencies)
* steep volume effects (7 levels / -6 db steps)
* samples (looping or not)
* dynamic bit decimation
The new feature would provide something in between what YM and current PCM routine offers.
This new feature will be a blittered PCM square.
The square wave is specifically convenient because of its constant values, so you can loop the blitter where you want

* portamento (variable freq)
* square on 127 volumes step (with input value of 0 or 127 constant, volume can be done throught half tone masking)
* variable pulse width

Re: Blitter sound routine
Quick & dirty test
https://youtu.be/tvO1BLg9iSk
https://youtu.be/tvO1BLg9iSk