Re: [Audacity-devel] OpenGL for Rendering Waveforms, etc?

[Campbell B. <ide...@gm...>]

On Sat, Sep 28, 2013 at 1:18 AM, Thales <tha...@ya...> wrote:
> Hello Campbell,
>
> I don't know off hand, but judging by the interfaces for packages like Adobe
> Audition and Pro Tools, I'm guessing they are using either DirectX or
> OpenGL.
>
>
> Here are several screenshots from both Audio Edition and Pro Tools:
> http://images.amazon.com/images/G/01/software/detail-page/B004TCG89C-1.jpg
> http://www.renderosity.com/adobe-audition-3-the-best-just-got-better-cms-13972
> http://www.heise.de/software/screenshots/315.jpg
> http://fire-software.blogspot.com/2011/02/avid-pro-tools-9-windows.html
> http://goprotools.com/blog/wp-content/uploads/2010/10/m_audio_pro_tools_recording_studio_5a.png
>
> The rich colors, sharp details, transparent images, free form layouts, etc.,
> make me think they're using OpenGL or DX!  :-)
>
>
> Thanks,
> ...John

Hi John, note that I'm not speaking as part of the Audacity team (I
only did minor patches so far),
but I think you approach it backwards.

Rather then say "theres a shiny technology - lets use it", Id approach
from other direction.

- Audacity is limited because "..."
- It can be solved by "..." (improved drawing I assume?)
- Improved/Advanced drawing can be achieved by "..."

I'd actually be surprised if all the applications you linked to are
using OpenGL/DirectX,
Its quite feasible to have smooth lines without hardware acceleration.

Also don't under estimate the potential of cheap laptop hardware and
buggy drivers to perform baddly with OpenGL,
especially 2D functions which are sometimes done in software anyway
(cheap cards tend to do this since games don't use them).

It depends on the scope of the task but you might end up having to
check for different opengl hardware and implement workarounds for
known bugs. thats in fact very common in projects using OpenGL.
Such bugs are incredible annoying since often the developers can't
redo them and users complain that their interface is broken.

Long term - I wouldn't assume this to be a straightforward task or
even a net gain.

Suggest you first define the problem - then solutions can be considered,
though unless a developer plans to pick up this project, it probably
ends up being a lot of noise with no outcome.