[z-machine] 1.1 proposal

[David Kinder]

> This may seem restrictive, but the only alternative is to have the number
> of available channels change depending on what samples are currently
> playing, which is the sort of complexity I'm trying to avoid.

To me, that's the beauty of Kevin's system in the proposed spec: it copes
with as many channels as the interpreter can provide, depending on what
sounds are currently playing.

Your system, then, imposes a different set of restrictions from those
imposed by the proposed spec. As I see it, those restrictions are:

1) There is a maximum number of possible sound channels that any
   interpreter could ever offer (you've put it at 16, I'd say it should
   be higher, say 256). The draft 7 spec imposes no limit.

2) The game may think that no more sounds could play when, in fact,
   the system really could play them. As an example, on a system where
   stereo samples take up twice as many channels as mono samples, if a
   game uses only mono samples it can only make use of half of the
   available system resources.

I have to say that these restrictions are worse than those in the draft
7 spec, especially 2), IMHO.

Of course, from the traffic on this list, it appears only David and I
care :)

David