Re: [Jsynthlib-devel] 10th anniversary

[Frankie F. <jsy...@te...>]

On 23/09/2011 17:04, Joe Emenaker wrote:
> On 9/23/2011 6:21 AM, frankster wrote:
>> From the docs (which have fallen off the internet in the last couple 
>> of days!): "The advantage of a scene is that it contains the locations 
>> for the patches in the synth's memories in addition to the patches 
>> themselves" So a scene has more to do with a particular synth's setup. 
>> It does sound useful - though I have to confess I've only really used 
>> libraries so far.
> Ah! I thought scenes were something else. Now that I look at the code, 
> it is becoming clear to me (and it's also becoming clear that it's coded 
> improperly... not just in how the GUI's for them look, but in how 
> they're coded on the back-end).
>
> Here are my general thoughts on this:
>   - The "Library" frame is a spreadsheet with the patches and comments, 
> without any location info. The "Scene" frame is a spreadsheet with the 
> patches and the bank/patch location on the target synth. But the 
> spreadsheets look too similar. I'd prefer that the "Scene" spreadsheet 
> have its rows/columns match the banks/patches on the synth, so we can 
> drag/drop them around to arrange the patches however we want.

I think there is scope for a synth-level scene and then something to
configure ALL synths in one go. Basically a group of scenes (performance?).


>   - Both the Library and Scene frames allow us to have patches for 
> different synths in a single window. This strikes me as more confusing 
> than helpful. I'd prefer to have each Library be limited to just one synth.

Yep I agree that we should restrict libraries to one synth. as I comment
above, I think we need a mechanism to control the setup of one synth,
and a mechanism to control the setup of all synths. This could be
achieved either by allowing scenes to hold setups for several synths, or
have a new concept of a grouping of scenes. Probably the latter I think,
so that would give us 3 concepts. Library, SynthSetup, Performance.

>   - If we were to have Library frames limited to one synth, then I think 
> a Library frame should have *all* of the patches for that synth. So, the 
> DX7 Library frame would have *all* of the patches for a DX7 that JSL 
> knows of, regardless of where they came from. Because of *that*, you 
> could make the argument that there's no need to ever have more than one 
> Library frame for a synth open at one time. You could have multiple 
> *Scenes* open, and you could be dragging patches from the one Library 
> window into several different Scenes... but you wouldn't need to have 
> multiple Library frames for a single synth.

Disagree. Firstly libraries with thousands of patches in might be
unwieldy. Secondly, it could make it harder to organise sounds. For
example, last night I was separating some TX81z sounds so that I could
use the "cross breed" functionality on them. I wanted to generate some
bass synth sounds and some drum-like sounds. Because of the way cross
breed works (randomly off any patch in the library), I found it useful
to have a group of drum-like patches in one library, and a group of bas
synthy sounds in another library. This way the cross-breed wouldn't
attempt to use the shitty "choir" patches in the main library that
haven't really stood the test of time, and when I bred drummy sounds
with other drummy sounds, the output was more likely to be a drummy sound.

That's not to say that we can't have a *view* of all patches for a
device. Just as itunes has albums, but might display the individual
tracks in a big list.

>   - The more I think about it, I think of JSL's relationship to synths 
> as similar to iTunes' relationship to iPhone/iPod, in that iTunes holds 
> all of your known media, and then some subset of that is synced to your 
> iPhone or iPod. So, I'm trying to work out some kind of "sync" paradigm 
I agree with this abstraction/workflow, I think it would be useful and
intuitive.

> for JSL. For starters, since the Library is holding the computer-side 
> representation of the patches, it's not limited by the synth's name 
> limitations, so I think JSL should let us give long names to the patches 
> in the library. When they're placed into a Scene, then their name either 
> gets truncated to what the synth's limitation is, or we could maintain a 
> LongName and a ShortName... who knows?
There's no reason we can't add all sorts of tags to patches, such as
author, comments, notes, ratings etc.

>   - Going further with the "sync" notion, we could have JSL deal, in a 
> smart way, with changed patches it sees from the synth. For example, 
> suppose I've got a patch called "Swoopy" in my library, and I send it to 
> my synth. Then, I modify the patch on the synth (during a performance, 
> say). And then, afterward, I sync a Scene with my synth and JSL would 
> notice that there's a patch called "Swoopy" on the synth which doesn't 
> match the "Swoopy" in the Library. I could be given a choice of: 1) 
> Overwrite the one in the Library, 2) Overwrite the one on the synth, or 
> 3) Make a new "Swoopy-2" in the library to hold the new one found on the 
> synth.

yep - although in my experience scanning synths entire preset libraries
can take a long time, but that doesn't mean it wouldn't be a useful
feature albeit with some caveats.

frankie