Hi Thomas,
the list has been pretty quiet after the workshop, although me and Nils and Jasch have had some exchange off-list.
Anyway, first of all thanks to everyone at T-labs for hosting the Berlin meeting, and likewise to everyone that participated. To me the meeting was productive, and raised several new questions and thoughts.
One of the main impressions that I came away from the meeting with, is that it might be a good idea to enforce a clearer separation of the various layers of spatialisation than we have done so far. I'm here referring to the stratified model for spatialisation proposed in [1].
The list of topics that there were an expressed interest in discussing, dealt with at least the top four layers (spatial authoring, scene description, encoding and decoding). At the same time it was clear that some of us had a stronger interest and need for getting things sorted out at one of the layers, while others were more concerned about other layers. Currently I'm thinking that the best probably will be to develop separate descriptive formats for (or some other way split how we address each of the following):
- Layer 3-4: Encoding-decoding. I doubt that these can be separated, as several spatialisation techniques don't separate encoding and decoding, but deal with them in one go (e.g. VBAP)
- Layer 5: Scene description
- Layer 6: Authoring
Maybe later on we also want to be able to address the bottom two layers, but that don't seem urgent at the moment.
I also got the impression that the fifth layer (scene description) seems to be the one that we want to address first, as there will be possibilities for interoperability between Jamoma, ICST-tools for MaxMST and IOSONO products. It also lays an important foundation for further work on descriptors of the layers below and above.
Based on the discussions of spline functions that Matthias raised, I have afterwards thought that the 6th layer will be the most challenging to address, but also possibly the most interesting one, if we manage to find a good approach to it. One of the challenges we'll face here, is that authoring approaches are not just a question of moving stuff around manually, but might draw on various mathematical functions and physical simulation algorithms. Some examples are spline functions and flock/boids algorithms. A first step will probably be to collect a set of algorithms that can be "use cases" for further studies, helping us to get a more clear idea of what kind of authoring strategies we'll need to be able to cater for, and how (and if) we manage to find a unified way of describing them. It might be that rather than adding support for algorithms through extensions, we'll need to develop a system where the author can describe the functions required for the algorithm so that the
y can be interpreted and processed in real time. Else way I'm afraid that we might end up with a potential infinite numb of extensions, in order to cater for one for each and every algorithm that someone came up with at a whim at some point in the past.
I've planned to go through the 0.2 version of SpatDIF, and make an overview of what layer each of the descriptors discussed belong at, in order to see if that might help further structuring the specification and possibly separate how we deal with each of the layers. It might take a few weeks before I get to this though.
Cheers,
Trond
References:
[1] Peters N., Lossius T., Schacher J., Baltazar P., Bascou C. & Place T. (2009): A stratified approach for sound spatialization. Proceedings of The 6th Sound and Music Computing Conference, 23-25 July 2009, Porto.
http://spatdif.org/papers/Spatialization-SMC09.pdfOn Oct 12, 2011, at 8:08 AM, Thomas Sporer wrote:
> Dear all,
>
> the attached email is the first email i've seen on this reflector since subscribing Sept 6th 2011. I assume that there was some exchange after the Berlin meeting. Can you please tell me whether this is a problem on my side, or whether there had not been any emails on the reflector?
>
> best regards,
> Thomas
_______________________________________________
Spatdif mailing list
[hidden email]
https://mail.bek.no/mailman/listinfo/spatdif
Locked
