- From: Chris Wilson <cwilso@google.com>
- Date: Wed, 28 Aug 2013 12:58:20 -0700
- To: s p <sebpiq@gmail.com>
- Cc: Paul Adenot <paul@paul.cx>, "public-audio@w3.org" <public-audio@w3.org>
- Message-ID: <CAJK2wqXwVinkzfTX3GB7gRu-TnnBiZCmRDm46LPn4oyYJrGsSA@mail.gmail.com>
Well, ScriptProcessor is kinda heavyweight - if you're just looking for general feedback (i.e. not precise "OMG clipping happened!"), you don't need to process EVERY sample; you'd be more efficient by sampling using an AnalyserNode's getByteTimeDomainData() giving you less data you have to process in response to a requestAnimationFrame, which is likely how you're showing the data. Of course, if you're really doing a fancy monitor, you'd want RMS as well as instantaneous clip indicators, etc., and yeah, you'd need a ScriptProcessor. On Wed, Aug 28, 2013 at 12:46 PM, s p <sebpiq@gmail.com> wrote: > > it's relatively non-obvious how to show a meter of reduction > > It is equally non-obvious to show a level-meter for example. But rather > easy in fact with a ScriptProcessorNode, ... I would say it is even less > obvious how to use a read-only monitor value as the input for a param or > node. > From a dsp perspective, I certainly prefer having a control signal as > output, that I can plug-in to another node. > > As there is no good solution, probably the one that is the most consistent > with the spec is the nicest! I cannot think of a place where a "dsp" > read-only value is provided (except in AudioParam) ... but I cannot think > of a node that outputs a control signal either. > > > 2013/8/28 Chris Wilson <cwilso@google.com> > >> On Wed, Aug 28, 2013 at 12:20 PM, s p <sebpiq@gmail.com> wrote: >> >>> > We would still make the reduction a read only float, because it would >>> really be a monitoring value. >>> >>> If you want to make an envelope follower, an output would probably be >>> more useful, as you could plug-it directly to a gain for example. >>> >> >> Although that's true, one problem is that it's relatively non-obvious >> how to show a meter of reduction (or overall volume, for that matter). The >> reduction output as a read only floating-point value is valuable for that >> purpose. >> > >
Received on Wednesday, 28 August 2013 19:58:47 UTC