Saturday, November 4, 2017

The Potential of Open Source Audio Software

Although there is lots of good music software out there, lots of hope and advances, audio software these days could be much better. Most software is proprietary which most musicians pirate anyways, exposing their machines to cracked binaries and thus being hacked. Even than, the inputs to the software are limited, using very ancient protocols such as MIDI, VST, etc. Inputs to hardware are also limited, with drivers being of varying quality between hardware vendors. You have to blindly use your intuition and create a "stack" which includes your computer hardware, operating system, and audio interface. There’s lots of misinformation in these areas, including audiophile BS (we've all been there) which makes it hard to make an informed choice.

The challenger is linux audio, which is unfortunately in an equally sorry state. But the open source (or open $X) model seems to be the way forward. Currently, it's honestly like stepping back 10+ years in the evolution of audio. One major sore spot is hardware support. This will probably never be fully supported like as proprietary hardware manufacturers produce almost all audio hardware. Maybe some startup will create an interesting hybrid hardware/software product like the AKAI and EMU samplers of old? I hope so, but again doubt it, as there’s a huge barrier to entry to the audio hardware industry, and not much money to be made.

Some possible reasons for this barrier to entry are: Musicians are brand-loyal, and while not technologically illiterate, hold onto alot of superstitions about technology. (ie. know just enough to be dangerous) There's a preference to old technologies, and a hierarchy built based on cost, image, as well as anecdotes about tech they or others have had experience with.

My vision is to have visibility and access to all parts of a digital audio stack. This will probably never really happen on Windows, or even OSX. It seems the the potential is there on Linux, with a wide array of open source projects, and a few commercial programs (Renoise, Bitwig) supported. But these programs are extremely limited on the linux platform, and scoped down to basically only editing and playing back files. The proprietary software has very limited integration with the open source components. Sure you can listen on your headphones, using your motherboard’s default sound card.  But outputting line-level audio to common studio hardware is difficult due to lack of widespread hardware support. Forget about using all your outboard gear, routing audio in and out of software freely.

The worst part of this is not just missing out on commercial plugins, but interfacing with the physical world outside computers, which is where music originally comes from. For a musician that may want to record their acoustic guitar while the idea is fresh, messing around at the layer of the code and it’s execution environment is not an acceptable alternative. While the development of software runs smoother on open source platforms, this has yet to translate into musician-accessible programs for composition and recording.
Probably no musician is going to learn code and start cloning git repositories and compiling binaries on their machine just to be able to record. Maybe they shouldn’t have to, but surely music software can benefit from an open source approach.

No comments:

Post a Comment