- From: Cullen Jennings <fluffy@iii.ca>
- Date: Tue, 24 Jan 2012 06:57:26 -0800
- To: Dan Burnett <dburnett@voxeo.com>
- Cc: public-webrtc@w3.org
On Jan 24, 2012, at 6:07 , Dan Burnett wrote: > At first glance I really like the simplicity here. I am quite concerned by the "by any camera" aspect, thought, since there is no relationship given between width and height available *together*, something which I personally really care about when taking photos! Imagine how hard it would be to take a picture if you couldn't know how to frame it :) Agree - and I had the same concern but could not figure out how to fix it. I think Aleksandr had a good suggestion of make the max and min both be a 2d vector with the size. The comparison of the vector to order them for min and max would be based on number of pixels. So for example, if you could do 176 by 144 bigger and 160 by 160. 176 * 144 = 25344 which is smaller than 160 * 160 = 25600. If the number of pixels is the same then we pick the one that is wider is considered bigger. So the object would return that the min = (176,144) This way the min and max are both and actually size you could get. > > -- dan > > On Jan 23, 2012, at 11:32 AM, Cullen Jennings wrote: > >> >> For Video: >> >> haveCamera - boolean true indicates have one or more cameras >> >> maxWidth - the max width supported by any camera >> >> maxHeight - the max height supported by any camera >> >> maxFps - the max frame rate supported by any camera >> >> >> For Audio >> >> haveMicrophone - boolean >> >> maxInputChannels - max number of input channels or any audio input source >> >> haveHeadset - boolean. I really don't know that we need this and it does increase the fingerprint information. We should talk about if really needed or not >> >> >> >> >> > >
Received on Tuesday, 24 January 2012 14:57:56 UTC