Hi,
I'm not sure that an orientation application would need HrtF as buildings and such are big enough to not require pinpoint accuracy. Also, typical mobile computing ear buds are probably not accurate enough (unless you go for something like those from Shure at about $250) to deliver the audio information in enough detail to really use the HrtF capabilities. I think it would be dangerous to wear a good enough headset while walking places where we might encounter vehicles or other dangerous obstacles - dangerous, of course, until we develop virtual landscapes based on laser scans of our surroundings. The latter idea is, as of right now, purely science fiction but I can imagine how it would work and can probably talk some of my electronic hardware hackers to use laser range finders used by hunters to build a real time wireframe models of our surroundings which, of course, would "see" cars coming almost instantly and then we need to figure out how to tell the user all of the information our vision synthesizer (audio has had synthesizers for a long time, why not vision?) has collected and then couple that with location services and points of interest and the like. Happy Hacking, cdh On Jul 31, 2009, at 10:19 AM, Chris Blouch wrote: > Apple's audio APIs do support HRTF on OSX: > > http://developer.apple.com/audio/overview.html > > but apparently HRTF has not made it to the iPhone: > > http://www.steamboatmountaindesigns.com/blog/2009/02/openal-for-iphone-does-not-support-hrtf.html > > Would be nice if these guys ported the Mac fmod sound system to > iPhone: > > http://www.garagegames.com/community/blogs/view/16489 > > CB > > Chris Hofstader wrote: >> >> The most "accurate" spatial audio information uses a technique >> called Head Relative Transform Functions (HRTF) which simulates >> relatively precise tonal qualities of sounds that one perceives >> from a specific direction. >> >> Some of the research into what became these functions had humans >> sitting with dozens of very tiny microphones set up in an array in >> their ears. there was more than enough similarity for the >> physicists to resolve the raw data into a number of mathematical >> equations and the output of said functions is amazing. >> >> the problem with HRTF is that it is very compute intensive. Of >> course, when I first started diddling about with it, I was on what >> is now a six year old Dell desktop. A dual core, 64 bit laptop >> should be able to handle these equations in real time much more >> easily. >> >> Microsoft has for a number of years now included HRtF in >> DirectSound (the audio component of Drect X) and my experiments >> with them showed very promising results. I don't know the >> Macintosh API too well so I don't know if they have something >> similar, identical or not. The Microsoft documentation does warn >> about relying too much on these functions as they can bog down the >> processor and cause other threads to behave kind of badly (again, >> all of the faster new computers should be able to calculate these >> sounds in real time. >> >> I've loads of ideas for a combination of really accurate sound >> combined with a haptic interface using the Falcon to simulate a >> whole lot of 3 dimensional real world objects. >> >> cdh >> >> >> >> >> On Jul 30, 2009, at 5:06 PM, Chris Blouch wrote: >> >>> I've heard there is a lot more to spacial placement than just >>> fiddling with the pan to adjust volume levels. At least this seems >>> obvious to me when a little pan to the left makes the sound fly >>> way off to the left of the sound field. Probably a lot more going >>> on in the ear that isn't fooled by that trick. So, that said, do >>> you have any idea if the VoiceOver positional audio is doing real >>> sound field magic or is it just a subtle pan? >>> >>> One underrated feature on the new iPhone is the compass. With this >>> an application can not only know where you are but also which way >>> you (or at least your phone) is pointed. This makes possible real >>> walking directions from where you're at using left and right >>> rather than assuming the user has their cardinal orientation >>> correct. Even feedback to say what is in front of you in the >>> direction you are pointing. Should be good things coming. >>> >>> CB >>> >>> Chris Hofstader wrote: >>>> >>>> Including the two textbook chapters (one entirely on games and >>>> another >>>> with a portion on audio games), a couple of peer reviewed >>>> articles on >>>> the matter, a pile of blog articles >>>> (http://www.blindconfidential.blogspot.com >>>> ) and the odd item here and there, I have studied a handful of >>>> audio >>>> games in depth, fooled around with a few others and talked to a >>>> lot of >>>> their authors but the breadth of my knowledge is far more narrow >>>> than >>>> a dozen hardcore gamers I know and rely on for brief descriptions >>>> for >>>> new entries so I can quickly ascertain how they may move the >>>> science >>>> in a new and/or different and interesting direction. >>>> >>>> As I said this morning, I am far too boring to actually build a >>>> real >>>> game and, when you scratch beyond the surface, you will find that >>>> first and for mostly, I think about putting technology in the >>>> hands of >>>> blinks that will provide them with the tools they may choose to >>>> employ >>>> in a job or school situation. >>>> >>>> >>>> If you are interested in mixing sound and tactile feedback, get a >>>> look >>>> at a Falcon 3D tactile controller. I think they have an SDK for >>>> Mac >>>> and they do .5 mm tactile resolution (most humans can feel no >>>> better >>>> than .1 mm) so you can do extraordinary things with this device and >>>> they only cost about $ 100 (really). >>>> >>>> I like the idea of using the iPhone for orientation and to provide >>>> information about one's surroundings. I'm not especially >>>> impressed by >>>> the iPhone's Maps application but its location services are pretty >>>> excellent. I thought of this while walking through Harvard Yard >>>> after >>>> lunch today and wondered how a student new to the university would >>>> find the library as opposed to the historical center next door as >>>> none >>>> of these buildings have traditional street addresses and are >>>> arrived >>>> at by foot paths. Your idea could give a student a lot of good >>>> information and can be used as a fallback if they get confused. >>>> >>>> Cool idea. >>>> cdh >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> On Jul 30, 2009, at 12:29 PM, Mark Baxter wrote: >>>> >>>> >>>>> I really wonder why no one's heard of AudioQuake with MindGrid. >>>>> The >>>>> project has somewhat stalled out over the past two years, but the >>>>> "Jedi Quake," version, written by Cara Quinn and others, is the >>>>> best >>>>> 3D audio simulation I've seen. I can't speak for the parallel- >>>>> processing audio capacity of the brain, but immediately this >>>>> game and >>>>> its maps made me start thinking about "mobility orientation," >>>>> applications. Having only ever encountered a mobility orienting >>>>> instructor for the first time in memory last year, I don't know >>>>> much >>>>> about how most blind people do it, but I'm sure my skills aren't >>>>> that >>>>> different from the norm, and I saw implications for map-making and >>>>> environment-learning, which is where I wanted to go with apps >>>>> for the >>>>> Iphone. I still have a Windows laptop running XP; I'll check out >>>>> shades of doom. >>>>> >>>>> >>>>> Mark BurningHawk >>>>> >>>>> Skype and Twitter: BurningHawk1969 >>>>> MSN: burninghawk1...@hotmail.com >>>>> My home page: >>>>> http://MarkBurningHawk.net/ >>>>> >>>>> >>>>> >>>> >>>> >>>> >>>> >>> >>> >>> >> >> >> > > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "MacVisionaries" group. To post to this group, send email to macvisionaries@googlegroups.com To unsubscribe from this group, send email to macvisionaries+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/macvisionaries?hl=en -~----------~----~----~----~------~----~------~--~---
