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

RE: audio delay node max delay time

From: Liu, Raymond <raymond.liu@intel.com>
Date: Fri, 17 Feb 2012 03:13:41 +0000
To: "Liu, Raymond" <raymond.liu@intel.com>, Yotam Mann <yotammann@gmail.com>, Chris Rogers <crogers@google.com>
CC: "public-audio@w3.org" <public-audio@w3.org>
Message-ID: <391D65D0EBFC9B4B95E117F72A360F1A03B3D5@SHSMSX101.ccr.corp.intel.com>
Oops. Why it cost my mail almost 12 hours to reach this mail list.....

Best Regards,
Raymond Liu

> -----Original Message-----
> From: Liu, Raymond [mailto:raymond.liu@intel.com]
> Sent: Thursday, February 16, 2012 10:42 AM
> To: Yotam Mann; Chris Rogers
> Cc: public-audio@w3.org
> Subject: RE: audio delay node max delay time
> 
> >Seeing as how the delayTime is an AudioParam node, why not just expose
> some of the attributes of that node to allow developers to tweak the maxValue,
> minValue, defaultValue, etc?
> 
> I guess the problem here is not to expose the interface, but that what kind of
> interface should be exposed. You definitely hope it can be done as
> xx.setMaxDelay() than xx.getdelayParam().setmaxValue(), (and what's
> meaning of setMinValue() here ;)  Also for later case, you need to make it
> available for every node to make it an universal approaching.
> 
> >The AudioParam is turning out to be one of the most crucial nodes for my work
> with web audio, so I would personally like to see more functionality come out of
> that.
> >As far as an upper limit on delays, I don't see a need given that you have some
> performance monitoring written into the spec and i could see potential uses for
> really long delays.
> 
> About the upper limit, I guess there had to be some limits, since buffer say
> more than 1000s data for a bit-rate around 100-200KB/s might be a good way
> to kill your memory quickly.
> I also see some potential uses for maybe tens of seconds delay to give a chance
> for real time analysis and prepare some visual effect which need an animation
> and need to be sync upon the rhythm. While I am curious on what kind of usage
> might need more than hundreds or thousands of seconds delay? ;)
> 
> Raymond
> 
> 
Received on Friday, 17 February 2012 03:15:50 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 17 February 2012 03:15:52 GMT