Open303 - open source 303 emulation project - collaborators wanted

[brambos]

It would be a little insane, yet cool, if you could somehow flip the lid of the software clone and adjust those trimpots yourself .. so several instances of the instrument would sound different (add user-triggerable random function for the positions)

Even better: upon first use the plugin should determine it's pot-characteristics so indeed, like a hardware unit, each unit would sound a little different yet consistently so!

[antto]

rv0: i'm still thinking alot on your opinion about the sequencer pattern editor
i pretty much understand what you mean..
you got two (as far as i know) TB-303s, you know a lot of people who own one (or more) too
you probably share your experience, patterns, and so on
you probably have sheets of paper, where you've written some great patterns of yours, and reprograming one in your TB-303 is easy, because you are used to that, and it's written in a simple format on the paper

i'm not sure if i'll be able to make my pattern editor to work like a real one
but i think i can make some additional display on my panel, where the current pattern can be displayed in this format, i just have to take some closer looks and think more
what do you think?
this way also, while you program the pattern in the "modern" way, you will see in realtime how the pattern looks on that display
well, you still won't be able to do some of the tricks you mentioned, that are related to the way the TB-303 memory works.. i'm not sure i fully understand these..

[rv0]

@rv0 Are you going to add shuffle to your sync unit plug ? That would be something!

yes, multiple types of shuffle will be implemented... uhm.. how did you know I'm working on that?

It would be a little insane, yet cool, if you could somehow flip the lid of the software clone and adjust those trimpots yourself .. so several instances of the instrument would sound different (add user-triggerable random function for the positions)

Even better: upon first use the plugin should determine it's pot-characteristics so indeed, like a hardware unit, each unit would sound a little different yet consistently so!

corrected your quote..

well, with the real thing you can tweak those settings as you wish, better not make it a final constant...
i have to admit i always open up a new 303 and tune it my way..
last yearr i did a live set and my 303s were in for repair.. I got a spare 303 from the "303doc" and I opened it just to change the pitch response (I always tune it so that the maximum tune setting gives a steady tuned note)

rv0: i'm still thinking alot on your opinion about the sequencer pattern editor
i pretty much understand what you mean..
you got two (as far as i know) TB-303s, you know a lot of people who own one (or more) too
you probably share your experience, patterns, and so on
you probably have sheets of paper, where you've written some great patterns of yours, and reprograming one in your TB-303 is easy, because you are used to that, and it's written in a simple format on the paper

yes, but not only because i'm used to that method,
that method just requires less button presses and offers some flexibility while composing.

i'm not sure if i'll be able to make my pattern editor to work like a real one
but i think i can make some additional display on my panel, where the current pattern can be displayed in this format, i just have to take some closer looks and think more
what do you think?
this way also, while you program the pattern in the "modern" way, you will see in realtime how the pattern looks on that display
well, you still won't be able to do some of the tricks you mentioned, that are related to the way the TB-303 memory works.. i'm not sure i fully understand these..

Don't worry about those special tricks, I don't fully understand them too, it would be insane to implement them as it's based on flaws in the memory adressing.
I will one day find it out exactly.. but it's not that important, just a geeky point of interest.

Yes, you can easily implement that interface or display I guess.. I'm not going to use it though, mac only setup here, so don't do it just to do me a favor


BTW:
Has any of you guys done OSX Audio Unit development?
I've been thinking of starting one up.. Although not right away as I'm still working on my own app. I'm still learning the API.

[antto]

i'm not sure if i'll be able to make my pattern editor to work like a real one
but i think i can make some additional display on my panel, where the current pattern can be displayed in this format, i just have to take some closer looks and think more
what do you think?
this way also, while you program the pattern in the "modern" way, you will see in realtime how the pattern looks on that display
well, you still won't be able to do some of the tricks you mentioned, that are related to the way the TB-303 memory works.. i'm not sure i fully understand these..

Don't worry about those special tricks, I don't fully understand them too, it would be insane to implement them as it's based on flaws in the memory adressing.
I will one day find it out exactly.. but it's not that important, just a geeky point of interest.

Yes, you can easily implement that interface or display I guess.. I'm not going to use it though, mac only setup here, so don't do it just to do me a favor

well, you won't use it, but others that share your opinion might just do
also, if i got a nice pattern on my synth, i could just make a screenshot and you'll be able to program the pattern in your TB-303 just by looking at the screenshot ;]
now that's nice, and doesn't matter what OS you got ;]
you'll just have to help me a little, so this "display" displays the patterns in the correct way (someone will have to verify it)

[rv0]

you'll just have to help me a little, so this "display" displays the patterns in the correct way (someone will have to verify it)

well i linked 2 empty pattern sheets in this topic 2 days ago.
but as soon as i get home i will see if i can scan a filled in sheet for you..

[Kriminal]

you'll just have to help me a little, so this "display" displays the patterns in the correct way (someone will have to verify it)

well i linked 2 empty pattern sheets in this topic 2 days ago.
but as soon as i get home i will see if i can scan a filled in sheet for you..

and i found these patch sheets on my hard-drive, the first one contains some explanation ( >5mb word .doc)
http://users.pandora.be/darffader/303patterbackup.doc
http://users.pandora.be/darffader/303%2 ... 0sheet.pdf

[Kritical Audio]

@rv0 I read I have made the same thing as your sync unit with reaktor, but I'm too lazy implementing shuffle atm.

[antto]

yeah, i got these, there are some aspects that are unclear to me
like, what is the difference between "C(-)" and "c(U)" ?
also "c(-)" and "C(D)"
..
and, for an accents ..as far as i understand, accents are added only in PitchMode, right?
i guess the only way for an accent to work is if there is "gate" on that step (erm, if there is a filled circle, note-on..)
i guess an accent won't work on a note after a slide if there is no slide before that grrr.. hard to explain what i mean

if i understand correctly, if you "tie" a note, it is extended, but the pitch is the same for the whole note
now to actually "tie" and "slide" a note, there must be a "slide" programmed too, and then the sequencer will take a "note" .. blah

so, these are "tied" and "slided" notes, right?
if i understand correctly, you can't have an accent on the "second part" of a tied note, but you CAN have an accent on the second part of a slided note?

would a graphic help? i can draw something that might explain what i mean better than my words ;]

EDIT: i actually forgot to report what i found out (was distracted)
there must be another (3rd) HP filter there, using plain-old saw, i can't get the exact shape of the TB-303 sawtooth
there is either a third HP filter, or one of the two others is a two pole?!
gonna try that.. it maybe just the damn A/D converter

[rv0]

yeah, i got these, there are some aspects that are unclear to me
like, what is the difference between "C(-)" and "c(U)" ?
also "c(-)" and "C(D)"

no difference at all, it is just used to extend the octave range (trivia: additional extending +1 octave can be done with track mode, using a programmed transpose to C)

and, for an accents ..as far as i understand, accents are added only in PitchMode, right?

yes. accent, up, down and slide are parameters tied to notes only.

i guess the only way for an accent to work is if there is "gate" on that step (erm, if there is a filled circle, note-on..)

yes.

i guess an accent won't work on a note after a slide if there is no slide before that grrr.. hard to explain what i mean
if i understand correctly, if you "tie" a note, it is extended, but the pitch is the same for the whole note
now to actually "tie" and "slide" a note, there must be a "slide" programmed too, and then the sequencer will take a "note" .. blah
so, these are "tied" and "slided" notes, right?
if i understand correctly, you can't have an accent on the "second part" of a tied note, but you CAN have an accent on the second part of a slided note?

yes i think you started of wrong in this paragraph and ended right.

2 things i might wanna clear up:
- yes you can accent notes following on a slide
- when sliding to a note of the same pitch value it will delete that note from the pitchmode and automaticly make it a 'G O' in trackmode.. i have to check out what this does with accents on the second note.. or up/down, as i'm not sure.

[antto]

then i think, i pretty much understand it, i just had to be concentrated for a while, now i won't forget it ever ;]

BTW: about my last report: looks correct
readjusted the 3 filters, sawtooth looks 97% perfect (at least at C-2 and C-3)

[antto]

looking at the "303patterbackup.doc" i really think i'm gonna rewrite the sequencer (erm, the part that reads the patterns and transforms them to CV/GATE)
it really looks simple that way there

[rv0]

looking at the "303patterbackup.doc" i really think i'm gonna rewrite the sequencer (erm, the part that reads the patterns and transforms them to CV/GATE)
it really looks simple that way there

Great! that document is great indeed.. i dont follow the 'LC' notation though, on my sheets i fill in C' for the high C and just C for the normal C.

i recorded some things for you, here's the file:
http://users.pandora.be/darffader/rv0te ... dtests.ogg
and here's the explanation:

1)

Code: Select all

pm: c, C(D), C, c(U)
tm: G G G G (4)

internal tempo from max to min.

so you can hear you get the same note.

2)

Code: Select all

pm: c(S), C(D), C(S), C(U)
tm: G G G G (4) 
sounds exactly the same as:
pm: c, C
tm: G O G O (4)

i alter between the above patterns all the time in the recording: you wont hear any change or difference


3)

Code: Select all

pm: c(S), c(S), c(S), c(S)
tm: G G G G (4)

although programmed like this, the 303 will automaticly save it as:

Code: Select all

pm: c(S)
tm: G O O O (4)

4) same as previous but without infinite slide:

Code: Select all

pm: c(S), c(S), c(S), c
tm: G G G G (4)

although programmed like this, the 303 will automaticly save it as:

Code: Select all

pm: c
tm: G O O O (4)

5) 'full octave range' in A
T: 11 (tuned in A with guitar tuner), C: max, RDA: min, E: center
i used the square wave.
made with two 12step patterns, one with 'down' and other with 'up'.. programmed with transposes in track mode to get full range.

6) same as previous, tuned in G#
T: 10

7) same as previous, tuned in G
T: 9,5

same as previous, tuned in F#
T:9 (yeah almost same reading as previous, just a lil bit less.. the tune control is not that accurate, luckily a tuner is)

9) same as previous, tuned in F
T: 8,5

10) same as previous, tuned in E
T: 8

11) same as previous, tuned in D
T: 7

12) same as previous, tuned in C#
T: 6,5

..... Inserted a beep tone in case you lost count by now ......

13) same as previous, tuned in C
T: 6 (= center)

14) same as previous; tuned in B
T: 5

15) same as previous, tuned in A (ooh thats the same note i started with, only one octave lower)
T: 4,5

16) same as previous, tuned in G (i'm getting lazy)
T: 3,5

17) same as previous, tuned in F
T: 3

18) same as previous, tuned in E (getting really low and dirty here)
T: 2,5

19) same as previous, tuned in D
T: 2

20) same as previous, lowest setting (detuned D)
T: 1




Have fun.;
Next up: the accent behaviour in relation to slides.

[antto]

wow, thanks!
first impression: the square wave is all about HP filters doing bad stuff to the phase! (and there is some fundamental assymetry)
after the second beep, it sounds so nasty, like the "bad wolf" tune ;]

this will keep me really busy (and productive) for a while now ;]

[antto]

i just tested generating the Square by waveshaping the sawtooth with a tanh() (and adding some DC offset inside, to get the PW-like effect)

well, that must be it, it really explains (in practice) why the filter doesn't get too excited on one of the edges
and um, it's computationaly cheaper (when i find a faster tanh() approximitation)
i'm still tweaking the whole thing, but the square is really getting there, really easy ;]

[rv0]

wow, thanks!
first impression: the square wave is all about HP filters doing bad stuff to the phase! (and there is some fundamental assymetry)
after the second beep, it sounds so nasty, like the "bad wolf" tune ;]

this will keep me really busy (and productive) for a while now ;]

i must add, i just checked my 2nd 303 and the tuning is slightly different there. i can also get the full range (from high to low: AGFEDCBAGFED), but i can go a bit higher and lower. on the 303 from the it has been tuned to be a straight 'A' on its highest setting, but you could also go a bit lower than the D.
I think it has been changed by the guy who installed the midi interface.

i can find out how it is supposed to be but its gonna take a while.