-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Thorsten Wilms schrieb: >>> Well, it would make perfect sense to add per region effects to >>> Ardour ;) .... > You could just have automation tracks that are not tied to other > tracks. That should allow controlling several things with one > automation track and processing stages between automation and > destinations.
Hi Thorsten, your response is somewhat confusing for me -- basically I don't know if you are a user or a developer. If you are mainly a user, you should consider that for implementing things, we need to be 100% precise, up to the last tiny and boring detail you never even wanted to know it existed. "..should allow controlling several things with one automation track" is by far to fuzzy! There seems to be a compelling idea buried within this approach, but we'd need to nail down how each of this "things" can be attached to this common automation. And, moreover, what to do if your "things" are moved and rearranged? Leave the automation where it is? Extract a part of the automation and move it alongside with the "thing". And is this really helpful for the user? In Ardour, indeed, automation stays fixed to the track and this is a mayor obstacle in the production workflow. Personally, I'm using a bunch of python scripts to be at least able to move automation data (I've spent weeks to create) in case the customer wants some correction afterwards. Cinelerra at least allows to copy/past keyframes, which is vastly superior, but far from perfect. It dodges the problem of how to combine pasted automation data with the data already there. Ideally, I'd want the automation data to follow the objects automatically. Now, you could easily jump into inventing clever mechanisms for this and that, but -- hold on! We were talking about design. This mandates us to restrain ourselves from "I can do this and that" and step back. Rather, the question is, to find a general approach such that the greatest number of issues and problems just "falls into place", rather then necessitating the invention of clever solutions for this and that. To summarize my argument: Ardour's main strength is the real-time capable wiring of a virtual sound studio environment. This is achieved by building on JACK. Thus, Ardour is necessary track/bus-centric. The editing and arranging of source material is subordinated to this principle and solved by employing a track-wide entity called "playlist". For Lumiera, we take a much more object-centric aproach, which decreases the importance of the tracks, but brings us to have to deal with the wiring in a more active manner, and also to organize the actual calculations differently (we employ pull-processing). > > I agree that video leads to requirements that reach into the audio > realm. Without a doubt making an app that is good at both video and > audio is extra hard. It's just that you would need an insane amount > of work to get even close to Ardour's audio editing and routing > capabilities. Starting any larger software development project is an act near to insanity. Somewhat similar to wanting to climb a mountain. You need to ignore the hugeness of the task and keep going. > Given a special interest in music videos, I see a hard separation of > editing audio and editing video as a problem. I agree fully with you regarding the importance of this point. To my understanding, you should be able to get to the level of a complete draft of the soundtrack within the editing/compositing App. On the other hand, a dedicated DAW has its own strengths we shouldn't try to compete with. Think for example at overdubbing. So, probably we'll have to care for a viable session exchange mechanism, so you could take your draft tracks over to the DAW for the sound specific work and bring some finished chunks back into the video session. Cheers, Hermann -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFILwt8ZbZrB6HelLIRAmhrAKDpUeNT/Cdj/gZu6QAt3n/qzlLvngCfVPkw 4Jof42h6nBjr4LxHiTSYClk= =pOqX -----END PGP SIGNATURE----- _______________________________________________ Cinelerra mailing list [email protected] https://init.linpro.no/mailman/skolelinux.no/listinfo/cinelerra
