The haskell source for my vocable synthesis system used in my previous screencasts is now available. I’ve been having fun rewriting this over the last couple of days, and would appreciate any criticism of my code.
Making music with text[ure]
Fist, I have to comment line
> import Sound.SC3.UGen.Dot
in KarplusStrong.lhs because there is no such file in hsc3.
I install SuperCollider server, and all examples from “* Haskell SuperCollider, a Tutorial.” work fine, and I can hear sound 🙂
When i start ./Main i see strings:
initialised 5
…
initialised 0
~>
Then I enter different strings and nothing 🙁
No errors, and no sound:(
Hi Cyril,
Thanks for pointing out the spurious Dot dependency, this version isn’t using that so commenting it out is fine.
Did you compile it with the -threaded option? You need to manually delete all the .hi and .o files then recompile if not. Otherwise I’m not sure what’s wrong – let me know what o/s and library you’re using.
(BTW there are many more fine hsc demos in the hsc3 help files.)
Yes, I compile with -threaded option. And I can see 3 treads of ./Main in # ps -eLf output.
About my system:
Kubuntu 7.10, ghc-6.6.1, darcs versions of hsc3 and hosc fom 15 of November. supercollider-server-20060416.
Hm, that’s very similar to my system. Could you try running from ghci?
ghci Main.lhs
then
main
Hi, I have the same problem as Cyril. The SuperCollider tutorial works but when I start Main, it just kills my speaker, and I don’t get any sound back until I kill scsynth… I also see 3 threads.
This is on Mac OSX Leopard. ghc-6.8.1, hsc3 and hosc from darcs and SuperCollider 3.1.1
I have tried to run from ghci. Same thing, no sound.
Sorry about this, I’ll try to find time to try on some different installations soon. If anyone else comes across this problem please comment, especially if you managed to fix it!
alex,
this is rather distracting! thanks kindly for posting it.
the problem noted above is due to un-initialized control buses, leading to the resonating filter ‘exploding’. adding something to the effect of:
mapM ((flip playWord) “test”) [0..voices-1]
to initSynth, before initVoices, is a ‘fix’.
i found it helpful when finding my way about to isolate the voices,
ie.
a8 = pan2 a7 (Constant ([0, -1, 1, -0.5, 0.5, -0.25, 0.25] !! i)) 1
regards,
rohan
Thanks a lot Rohan!
I’ll upload a fixed version soon.
alex