W3C home > Mailing lists > Public > public-audio@w3.org > January to March 2012

Re: Adding MIDI APIs to Audio WG Charter (was: MIDI enumeration (was: Re: getUserMedia use cases))

From: Joseph Berkovitz <joe@noteflight.com>
Date: Mon, 6 Feb 2012 09:49:30 -0500
Cc: public-audio@w3.org
Message-Id: <1E66FF90-D10E-49A0-9E77-26A85807969E@noteflight.com>
To: James Ingram <j.ingram@netcologne.de>
There is no single answer to this question. I don't disagree with the fact that *some* applications will not be sensitive to synth consistency, but it is absolutely the case that others will require it.

At the same time, any prospective MIDI API need not provide a guaranteed-consistent synthesizer.  It's fair to ask any application that wants this level of consistency to provide its own JS-based synth (and the viability of doing so, on top of the Web Audio API at least, is proven).

By the way, it is not the case that any JS-based wavetable synth is fundamentally flawed idea as per Marat's previous post.  It is not a good decision for every application but if one is willing to live within certain constraints and also make use of local storage, it can definitely be done.


On Feb 6, 2012, at 4:51 AM, James Ingram wrote:

> Robert O'Callahan said
>> Other people want to be able to manipulate real-time MIDI streams and synthesize output from them. Where do those applications come down on system synthesizer vs consistent synthesis?
> I don't think its important to guarantee that exactly the same audio results from playing a particular MIDI stream on different systems.
> Many more application scenarios are possible if browsers allow their users to decide which of their installed MIDI output devices they want to use, by listing them in a preferences dialog. On Microsoft OSs the "Microsoft GS Wavetable Synth" would appear there, and would be the default synthesizer. But if I've got a better piano installed, then I want to be allowed to use it. If I want to link up with a lighting system, then I want to be able to do that. If I've got the Vienna Philharmonic installed, I may want to use that.
> So even if there was a common default synthesizer across all browsers, there's no guarantee that users would actually be using it.

> Marat Tanalin said
>> It makes sense to take into account that GM standard itself defines just set of general timbres and does not regulate their exact sounding, so different sounding depending on specific GM device is perfectly acceptable for GM. If _some_ web-applications need 100% consistency across browsers and platforms, authors of that applications are free to implement their own pure-script synths _simultaneously_ with having ability to use universal system GM synth in all other, more general usecases.
> Exactly.
> I don't think its too much to ask of an operating system, that it should supply a default software synthesizer in the way that Microsoft does. Perhaps Microsoft could  be persuaded to make their's open source... :-)
> James

... .  .    .       Joe

Joe Berkovitz

Noteflight LLC
84 Hamilton St, Cambridge, MA 02139
phone: +1 978 314 6271
Received on Monday, 6 February 2012 15:50:05 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:49:57 UTC