Re: [Bug 20681] spatial coords should just be in metres

Den 2013-04-27 10:40:47 skrev Frederick Umminger  
<frederick.umminger@gmail.com>:

> I don't think I have a bugzilla login, so I can't post to the bugzilla.
>
> I don't think it is a good idea to go with undefined units.
> Just because OpenAL and some other APIs did does not mean that it is the  
> right thing to do.
> We should aspire to something higher than just ripping off OpenAL.

Absolutely.

>
> Many sound algorithms, and other physical modeling algorithms, have to  
> touch base with physics.
> As pointed out above, this API is already running into this with the  
> speed of sound.
> You are going to run into this again if you do virtual analog modeling,  
> or physical modeling synthesis, and so on.

While I agree that SI is the only sane system to use, there's really no  
part of the Web Audio API that mandates any unit. For instance, if there  
was some hard-coded constant under the hood, there would be a real  
dependency, but [I really hope that] this is not the case.

So really, if the Web Audio API should require SI units, it's still only  
informative text. A user could still use other units and the API would  
just do the right thing. That given, I think that any mention of SI units  
other than seconds should only be in informative parts of the spec - not  
in normative parts of the spec.

/Marcus


>
> Physics is done in real-world units, and worldwide those units are  
> standardized to SI units.
> Using anything else is going to create a morass of unit conversion and  
> bugs.
> This is a hard lesson learned from experience.
>
> Furthermore there is really no point in going with undefined units.
> If a game wants to map its world in some strange game-specific unit  
> system,
> then the game can take responsibility for converting to SI units at the  
> point of API calls.
>
> Sincerely,
>    Frederick
>
>
>
>
>
> On Fri, Apr 26, 2013 at 12:11 AM, <bugzilla@jessica.w3.org> wrote:
>> https://www.w3.org/Bugs/Public/show_bug.cgi?id=20681
>>
>> --- Comment #4 from Marcus Geelnard (Opera) <mage@opera.com> ---
>> ...also just found that the PannerNode and AudioListener specifies the  
>> velocity
>> in meters / second, which would have to change.
>>
>> --
>> You are receiving this mail because:
>> You are the QA Contact for the bug.
>>
>



-- 
Marcus Geelnard
Android Browser Developer
Opera Software ASA