Thanks for sharing those observations about VCV Rack and the internal
software design in particular.
On Mon, 3 Dec 2018, 13:56 Paul Davis, <paul(a)linuxaudiosystems.com> wrote:
On Mon, Dec 3, 2018 at 7:42 AM Dave Phillips <dlphillips(a)woh.rr.com>
wrote:
[ ... ]
Following on from Louigi's question, I want to say on this mailing list a
couple of things about VCV Rack that i've said often on IRC but not
elsewhere. Really just two.
1) I want to give HUGE thanks and congratulations for Andrew Belt and what
he has done with VCV Rack. Not so much the program itself, but the
ecosystem he either deliberately or accidentally created around it. Some of
you (*cough* Dave *cough*) might remember that before I wrote Ardour and
JACK, my first foray into Linux audio was something called Quasimodo.
Powered by a reimplementation of CSound, the GUI was a lot like VCV Rack
(and other similar software). Quasimodo never succeeded, for a variety of
reasons, and it is so good to see a GPL'ed modular synth now finally really
finding and creating a community and success. The visual appeal, ease of
use, and relatively simple module API of VCV Rack have all been critical in
its success, and Andrew deserves many kudos for this. It really is amazing
to see the set of available modules (and their quality) and the dual
business model (no-cost vs. paid) for modules. I'm actually jealous.
2) All that being said, as a programmer, the internals of VCV Rack's
engine are deeply disturbing. It is really amazing that VCV Rack works as
well as it does. It isn't properly coupled to the audio hardware at all (it
uses a timer to drive the graph), and it can't be trivially modified.
Thankfully, someone has done the modification for the VST version of
VCVRack (because in a plugin, you have no choice), and so perhaps the
redesign might make it back into the mainline code. Given the lovely to use
GUI and the fantastic ecosystem, it's a little sad to see the internal code
suggest almost no understanding of how to write a realtime audio
application. Maybe Andrew had his reasons for the internal design - I
don't know. But it certainly makes it much more likely to have audio
glitches and to be incapable of operating at the lowest latencies. It works
well enough for me, however (and is a huge risk because I could spend hours
playing with it).
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
https://lists.linuxaudio.org/listinfo/linux-audio-user