Re: [Rosegarden-user] Encouragement

[Holger M. <ho...@ma...>]

On Sat, 26 Jan 2013, Abrolag wrote:

> For Michael and the other long-suffering devs.
> 
> I've just completed a project that uses Rosegarden to drive two instances of
> Yoshimi at a total of 32 different instruments for 8 minutes. It is intended to
> be a demo of Yoshimi's capabilities, but I thought you'd like to know about it
> as it's also a demo of Rosegarden at the same time.
> 
> The entire project was built up relying mostly on melodic parts played in real
> time with some editing later. Arps and chord stabs were also recorded live, but
> unfortunately I had to quantise some of these as my playing wasn't regular
> enough - particularly on the longer sections where these old fingers are
> getting a little stiff :(

I like it an think of a children's dream :)

But there's something I must say because I hear some latency issues.
Most music I encounter that's made with Rosegarden or other Open Source
Software avoids instruments that need exact timings. I don't know if
it's caused by lacking/faulty latency compensation. But it's an
important thing, at least for me. It's easy to imagine that many
musicians tried Open Source software and ran away when they noticed that
it doesn't sound as exact and doesn't get the groove as in commercial
sowftare.

That's bad.

Unfortunately I can't do C++ and doesn't understand the internal
concepts of Rosegarden. But I think it's damn important that

- we get the off-by-one latency compensation bug in subgroups fixed
- we get the missing "audio tracks don't get latency compensated at all" 
  feature added
- we get a positive/negative time offset in milliseconds per track
  added or at least a milliseconds entry field, so I could delay every
  track by 100 ms and use e.g. 90 ms as a substitue for -10 ms.

That would Rosegarden make an outstanding DAW. Its track concept and its
matrix and drum editors are already very good. It's (for me) much more
important to get these bugs fixed than to add LV2 plugin capability or
audio automation.