VSD Use cases (was two more questions)
[email protected] Fri, 16 Nov 2012 22:12:30 -0500
| Newsgroups | gmane.comp.lib.openal |
|---|---|
| Message-ID | <[email protected]> |
All, Thank you for the discussion in response to my question(s) about contexts and listeners and devices. There is a LOT there, and I'll probably have to read your responses 3 or 4 times to soak it all in. I'll share a bit more of what I'm thinking at a high level, in hopes that you might see any major "holes" or perhaps have some ideas to help make things easier or better. I see two modes of operation, basically, which are scaled up or down to match the size of the layout (anywhere from a single operator on a 1x3' shelf layout to a several thousand square foot club setup with dozens of operators and trains). What follows is a mix of what I was already thinking, with the feedback you've already given. ROOM AMBIENT: This is sound-out-in-the-air. A stereo speaker pair for a tiny shelf layout, up to a 5.1 or 7.1 home theater surround setup for a room sized layout. One Device, one Context, one Listener. The Listener is in a fixed position at the center of the "room" (or, for table or shelf sized layouts, at the operator's position). The error in the apparent sound locations from operators being away from the sweet spot is taken as an acceptable compromise in exchange for (a) not having everyone wear headphones, and (b) not having the cacophony of ten different sound images overlapping each other in a spare bedroom. For larger, multi-room home layouts or large club layouts, the above scenario might be expanded to multiple Zones. Each Zone would have its own speaker system, Device, Context and Listener. The Zones are set not to overlap any more than absolutely necessary. The case of a single, stationary operator wearing headphones instead of using speakers, but remaining basically stationary around a small layout is considered a ROOM AMBIENT scenario in this case. HEADPHONE: In this case, one or (more likely) several operators are around the layout. Each operator has his own pair of headphones, either a bluetooth/wireless set, or attached to his smart phone (with the audio streamed to the phone). The easy method is that each operator hears the "cab view" of the locomotive he is controlling... the sound as though his listener position was inside the cab. Somewhat harder is "room view", where the listener hears what he would hear from where he is physically standing in the room, only it's his own personal version, due to the headphones. Clear? The hard part here is tracking and recognizing the listener's location in the room, but I'm working on that. This case would require one Context and Listener for each operator. Whether multiple Devices are required seems a bit fuzzier, but it also sounds like with the wireless headset case, we get that for "free". A couple of points. Even in the room ambient case, volumes are generally kept quite low, for two reasons. One is that the sound works better if it is kept "to scale" with the models. It doesn't sound right to have this tiny locomotive you can hold in your hand have a diesel rumble so loud it rattles the windows. The other is that often there is a good bit of conversation when a group is running trains, and you don't want the train sounds to interfere with the talk. So when thinking about the multiple zone case, keep the low volumes in mind. Another thing is that actually tracking the location of the trains (or the listeners!) rather difficult and expensive, much more so than keeping track of characters in a video game or sound effects in a movie. So while spatially locating the sounds precisely is a laudable goal, the reality for most people is that at best they will get a rough approximation. So, in summary, I haven't heard anything that says I'm busted, and I've heard several things (like the Bluetooth headsets having their own devices) that will actually make things easier, when I get to that. Well, I've filled your inboxes enough for now. I'm sure I'll have some more specific questions soon. Again, thanks a bunch for your help! - Mark (TD) _______________________________________________ Openal mailing list [email protected] http://opensource.creative.com/mailman/listinfo/openal