W3C home > Mailing lists > Public > public-audio-comgp@w3.org > August 2020

Re: [Agenda] WG/CG Telecon Thurs Aug 20, 9:00 am PDT

From: Raymond Toy <rtoy@google.com>
Date: Thu, 20 Aug 2020 10:15:33 -0700
Message-ID: <CAE3TgXGFiVc_v6n-9wC4_iC2vma-3+q3ukEV8k=uTjt+=Tq57Q@mail.gmail.com>
To: "public-audio@w3.org Group" <public-audio@w3.org>, public-audio-comgp@w3.org
Attendees:  Jeff Switzer, Ruth John, Christoph Guttandin, Chris Lilley,
Raymond Toy, Jack Schaedler

Minutes:

   - V1
      - Next steps for V1
         - Chris Lilley: Show two implementations pass tests, create report
         on test passes, and see if director agrees.  wpt.fyi can be
good enough for
         the report.
         - Christoph:  Would nodejs count?
         - Chris L: Yes!
         - Raymond: no obvious gaps in testing based on wpt.fyi.
         - Chris L: Looks good.
         - Raymond and Chris to work on the report.
      - New editorial issues
         - Raymond: Summarizes issue
      - V2
      - Pulse width oscillator
      <https://github.com/WebAudio/web-audio-api-v2/issues/7>.
         - Raymond:  Should it be a type of oscillator or a separate node?
         - Jack: Yes. a separate node.
         - Raymond: links to andy.harmon demo
         <https://github.com/pendragon-andyh/WebAudio-PulseOscillator>.
         - Chris: Only works if wave shaper is bandlimited
         - Raymond: I don't think it is, in general.
         - Conclusion:  Use a different node, called PulseWidthOscillator.
         - Attributes:  frequency, detune, pulseWidth (range 0 to 1). 0 is
         silent, 1 is 1.
         - Chris: Should be bipolar instead of unipolar.
         - Raymond:  will update issue with proposal for node.
      - Bring your own buffer
      <https://github.com/WebAudio/web-audio-api-v2/issues/4>
         - Jack: From last time, a profile would be very useful to see what
         the cost would be.
         - Raymond: V8 has good profiling tools but I need to look it up.
         - Jack: Will try to profile using Chrome's tracer.
         - Raymond: Will find the V8 profiling tools
      - Selectable render size
      <https://github.com/WebAudio/web-audio-api-v2/issues/13>
         - Jack: This is important on Android.
         - Jeff: another fingerprinting issue
         - Jack: 2 primary cases: optimize for HW (Android) for better CPU
         usage or large sizes to process large chunks at once where
latency isn't an
         issue.
         - Conclusion:  Work on API proposal, allowing user to specify that
         the HW size should be used, but also allow setting larger sizes.  Some
         constraints on values need to be specified too.
      - AOB
      - Virtual F2F meeting to be arranged sometime around TPAC 2020 in
      October.  Dates/times TBD, but similar to virtual F2F in June.


On Wed, Aug 19, 2020 at 2:40 PM Raymond Toy <rtoy@google.com> wrote:

> Here's the agenda for our WG/CG meeting.
>
> Meeting venue https://meet.google.com/wgg-wjtd-kdu
>
> Agenda:
>
>    - V1, depending on attendees available
>       - Next steps since privacy issues are closed
>       - A couple of new editorial issues.  May skip these.
>    - V2
>       - Continue with pulse width oscillator
>       <https://github.com/WebAudio/web-audio-api-v2/issues/7>
>       - Bring your own buffer
>       <https://github.com/WebAudio/web-audio-api-v2/issues/4>
>       - ?
>    - AOB
>
>
Received on Thursday, 20 August 2020 17:15:59 UTC

This archive was generated by hypermail 2.4.0 : Thursday, 20 August 2020 17:16:00 UTC