setronic
Synthedit regular
Posts: 293
Posts: 293
|
Post by setronic on Feb 4, 2014 10:41:36 GMT 1
build code skeleton i don't have any experience of this bcs i still use 1.017, but: a/ build GUIBools/GUIList things in eg: 1.017 - just all the multiples, that get the module count up, involve hundreds of cables, etc. b/ load the se1 file into v1.1914, see what it looks like there, figure out how 'build code skeleton' works, then do it, and get resultant file. this is C++, right? ( i should just try it i know, i will) what next? i mean, could i then just send this file to a C++ savvy person willing to SDK3 it up? does that involve a lot more work? ; i know i need to spend some time on this, learning how it all wraps up into a sem and any bit of basic C++ i need - i'm still reluctant to stick my head in there, another thing to take on  eeek/shame) (or even sdk2? i'm not sure i understand the distinction, i just knw that sdk3 won't show in SE 1.017) nb: just really talking about making some specific multi-components/arrays to use in projects that are starting to work at this end, re-usables. dang, i'mma gonna have to learn some c++... |
|
|
|
Post by Rob on Feb 4, 2014 12:57:09 GMT 1
Dom,
The Build Code Skeleton option just creates a empty module with some pins already in place.
It is handy for developers wanting to convert their modules to SDK3, normal users don't have any use for it.
You can't just do containerise on your modulestructure and "Build Code Skeleton".
|
|
|
|
Post by sepoweruser on Feb 5, 2014 14:05:21 GMT 1
Yes, the code skeletton doesn't provide the main processing code of the module, it just provides the surroundings (pin definitions, empty basic code blocks). You only can complete the module if you know the main processing code.
If you want to learn coding modules then have a look at the module examples provided in the SDK and read the SDK documentation on the synthedit website.
|
|
setronic
Synthedit regular
Posts: 293
Posts: 293
|
Post by setronic on Feb 5, 2014 20:44:16 GMT 1
yeah spoke too soon, not something i'd thought
about or looked into.
my interest is stuff like: just made a boolsplitter
with 16x12x3 pins, so that was an exciting way to
spend the time..it's a tiny part of the whole thing
and in the end it's going to be a container within
a project.
when you're in C++, are there shortcuts that save time
will all this kind of stuff? i mean, i'm curious how
people can visualize bools when they are looking at C++..
but of course if you save time with a line to make it
repeat etc., makes sense.
i've had quite a satisfying time making little flashing
LED prefabs in GUI sems in v 1.0 - you have to take time
out of anything else, to make apparently pointless
constructions. i just wanted to see how set/reset flipflop
types could be built from gates, stuff like that. ha, with
the 2way thing, it gets complicated,inserting extra gates
as 'yes'/1way blockers. of course Rob's LogicGates2 is a
very versatile module - does it use much cpu when it operates?
or maintaining the extra pins? it is an interchange for all
the GUI pin types, as well as all the basic gates > so i've
just also made a container with 16x 12x RHLogic2.. this has
its lists, values, patchmems and a container of DW NoteQuantize.
erm, yeah, so that's where i'm up to with that. i have to keep
extracting parts to work on the speed up connections, and then
re-import the sub-prefab(getting some good 'spares'), so you
come to see that your multiples and cables are costing you time
and maybe performance, when you're dealing with 0/1 switching and list
selection. this doesn't have to produce audio, it is just a stepped
control prefab with patch memory, octave & semitone allocation.
i'm thinking i will be needing XOR and OR in some arrangement next,
haven't worked that out, need to gen up on scales. (suggestions?)
and yeah of course, if this could then result in a sem, in some
way using the work done to build this, drafting VST something and
reverse engineering the *.dll to get the code..is this possible?
which then lol...eek copy+paste somewhere, with nice tidy SDK
around it to minimize the damage ? also: would i be correct in
thinking that, if you could run both sdks simultaneously -on
separate machines if necessary- the same core code would be good
in both SDKs? or with minor/major adjustment? just asking...
i find the 'blue background' pins in v1.0x nice to deal with
(don't seem to be blue in 1.1), and of course, for checking
out all 3rd party free sems(haven't seen CK collection),you
see them as they were intended to be viewed > what happens
in 1.1 with blue backgrounds? i've got both versions installed
but don't really run 1.1 to avoid confusion with files(for time being).
back to my crochet...
|
|
eeek/shame)
