From: Javier Serrano Polo <jasp00@te...> - 2006-06-26 13:29:36
Sorry for double-mailing you, Tobey.
-------- Missatge original --------
Assumpte: Re: [LMMS-devel] 3 ideas
Data: Mon, 26 Jun 2006 15:30:26 +0200
De: Javier Serrano Polo
A: Tobias Doerffel
En/na Tobias Doerffel ha escrit:
> Am Montag, 19. Juni 2006 16:22 schrieb Javier Serrano Polo:
>> My problem is I'm in the middle of the automation stuff and won't be
>> able to help. If you decide to go ahead let me upgrade those organic
>> presets before.
> what exactly are you doing at the moment? it would be fine if you would inform
> us from time to time about the state of your work, especially for avoiding
> conflicts with other developers, because they might to the same thing...
I was going to commit today. The most important change: automation, many
knobs already use it.
> There're also some other things to change: currently you pass pointers to
> floats (or whatever) to oscillator-class, but in my opinion it should be
> pointers to automatableObject<TYPE> and use their value()-method. This is
> much more generic.
I think float is more generic than automatableObject. You could use the
oscillator within other projects and dependencies are reduced. Besides,
you need the transformed value rather than the object's value.
> The only thing is, that some code in tripleosc etc. has to
> be rewritten, as they calculate the final value from the knobs-value. As
> knob, all button-classes etc. are derived from automatableObject<...> it'd be
> more logical to use pointers on them instead of "3rd-party-data" ;-)
> Otherwise things get too complex very soon...
The problem is triple osc isn't the only one using oscillators. Organic
does too and it uses different algorithms.
>> Then there's libsamplerate, it should be a requirement. Fixes weren't
>> applied to non-libsamplerate code.
> is it a lot of work to fix them? then it'd be better to fix them... otherwise
> we could add a serious warning to configure informing about the issue..
I choose the serious warning. There was a lot of changes and I don't
even know if that code worked in the first place.
> beside that I would prepare a release within this week, even if it won't be
> perfect and complete, but it's better and more stable than 0.1.4 for sure!
> and that's important...
If you don't want the automation stuff, please tell me. I was going to
upload it this evening. I think it's worth a couple of days delay.