[linux-audio-user] Re: Call for collaboration

R Parker rtp405 at yahoo.com
Wed Jul 13 20:59:05 EDT 2005


> what about a small shell script (or two) which takes
> an
> ardour-session-dir, compresses all the audiofiles
> with flac (its
> lossless isn't it?) and then tars the dir. On the
> other side it untars
> the dir, decompresses all the flacs to waves and
> your session is ready
> to be loaded in ardour. Additionally the peak-files
> could be removed
> (they will be regenerated cause the wav-file is
> newer than the peak
> file).

The session average between 1.5 and 2.5gig. My
experience with flac produces about a 30% reduction.
If it's only the bass track that has changed A. I
don't want to transer the entire session and B. I
don't want to manually keep track of changes.

> What SEX is missing is the ability to join two
> branches of music.
> Example: A starts the song with some tracks, B takes
> it and adds the
> bass in an own snapshot.

In this case B wouldn't create a new snapshot but for
the sake of argument let's follow that course.

 But while B is doing this,
> A also adds
> tracks.

Now we've got snapshotA-v01, snapshotA-v02 and
snapshotB-v01.

 After A loaded the snapshot from B, he can
> listen and think
> about both versions but the info when B's tracks and
> recordings start
> are not usable in A's snapshot.

Before A can listen to B's snapshot he'd have to pull
that entire snapshot to his local station.

 Only the pure files
> can be used but
> what if the bass-track of B is in fact several takes
> which don't clue
> on beats?

I think all the problems are prevented by never
creating a new shapshot. The solution is to use Ardour
Playlists.

If I understand B and A are modifying snapshotA-v01 at
the same time. A completes her work which creates what
we'll call snapshotA-v02 and then pulls B's bass track
 into that version which produces snashotA-v03.

B has never heard snapshotA-v02. The only manual
intervention in this exchange is A needs to consider
the consequences of modifying the song by asking
himself will whaatever B produces again v01 still work
with v02.

Playlists have a further advantage over snapshots
because Ardour is nondestructive. If you edit the
region view of Playlist "bass" the bass.wav file
remains unmodified and so the only change is to the
snapshot session file "snapshot?-v??". If SEX is
designed properly and I suspect it is only the session
file is exchanged.

IMO the bandwidth that's available today dooms any
scheme that doesn't account for versioning in one way
or another.

Anyway, my description of how SEX works might be
pretty poor but I think the principle idea is sound.
Do not create new snapshots. A snapshot is the
equivalent of a new song. Use playlists and bear in
mind that everyone participating needs to think about
the consequences of their actions.

When we all agree that the basic tracks are poored
into concrete then we begin adding our solos, vocals,
etc. In order to modify the basic tracks everyone
involved needs to complete their current work, update
to the current version and then all Playlists are fair
game again. If we have a descension amongst the
participants then the song can be forked. heh

Maybe I make sense, maybe I'm wrong about SEX.
Reguardless, I maintain the positon that I like SEX

ron

> BTW: I wouldn't use CVS for ardour-sessions. Every
> VCS (version
> control system) sucks when it comes to big files.
> And CVS is so
> oldschool...
> 
> Arnold
> 
> -- 
> Wenn man mit Raubkopien Bands wie Brosis oder
> Britney Spears wirklich
> verhindern könnte, würde ich mir noch heute einen
> Stapel Brenner und
> einen Sack Rohlinge kaufen.
> 
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 



More information about the Linux-audio-user mailing list