Re: [rtcweb] Video resolution SHOULDs (Re: resolutions in draft-cbran-rtcweb-codec-01)

The highlighted part of Harald's suggestion below

>    SHOULD support resolutions of 1920x1080, 1280x720, 720x480, 1024x768,
> 
>       800x600, 640x480, 640 x 360 , 320x240 *if these resolutions can be
> displayed on the target device*
> 
seems sensible.  However, I have another issue.  I don't believe that naming
specific resolutions is advisable.  In a system where the rendering side has
almost undefined picture properties because they are typically rendered in
"windows", it IMO does not make much sense to limit yourself to certain
aspect ratios, picture sizes etc., even if those correspond to today's
preferred sensor resolutions (or integer fractions thereof).  Resampling an
input signal is cheap nowadays (at least compared to the encoding itself),
and cropping is even cheaper.
I would instead recommend naming one single MUST resolution (as the draft
does) to ensure minimum interoperability, and otherwise negotiate a
processing rate in macro blocks per second (or equivalent), with fixed
constraints around the aspect ratio.  Just as what H.264 does in its level
specification.  In fact, if H>264 turns out to be our mandatory codec, then
I would adopt the level spec wholesale.
As for said mandatory resolution, I believe that 320x240 is sooo yesterday.
Even on something like an iPhone, you can do better today.  Who uses 4:3
ratio nowadays?  My vote would go towards something like WQVGA or even WVGA.
Stephan
From:  Harald Alvestrand <harald@alvestrand.no>
Date:  Thu, 17 Nov 2011 16:50:54 +0800
To:  <rtcweb@ietf.org>
Subject:  [rtcweb] Video resolution SHOULDs (Re: resolutions in
draft-cbran-rtcweb-codec-01)

 On 11/17/2011 11:19 AM, Roni Even wrote:
>     
>  
> 
> Hi,
>  
> The text in the draft on video codec is
>  
>  
>  
> o  MUST support a minimum resolution of 320X240
>  
>  
>  
> o  SHOULD support resolutions of 1280x720, 720x480, 1024x768,
>  
>       800x600, 640x480, 640 x 360 , 320x240
>  
>  
>  
> I propose adding 1920 x 1080 to the SHOULD.  This resolution are used by video
> conferencing systems
>  
>  

 I think I want to object to making this a SHOULD, but it's possible we just
don't understand the requirement the same way....

 The RFC 2119 SHOULD definition says:

 3. SHOULD   This word, or the adjective "RECOMMENDED", mean that there
    may exist valid reasons in particular circumstances to ignore a
    particular item, but the full implications must be understood and
    carefully weighed before choosing a different course.

 When designing an RTCWEB implementation for my mobile phone, with its
(comparatively) microscopic screen and anemic processor and battery, it
takes approximately zero seconds to decide that I won't waste energy
(literally) on decoding HD; I'll make sure I negotiate down to something
more sane for me if someone's unkind enough to suggest it to me.

 I would be happy to write the requirement as

    SHOULD support resolutions of 1920x1080, 1280x720, 720x480, 1024x768,
      800x600, 640x480, 640 x 360 , 320x240 if these resolutions can be
displayed on the target device

 but I would not be happy to just extend the requirement as it is proposed
today.

(Yes, I know, this means that I've turned down-negotiation of video
resolution into a MUST. I think it already is, but I think we haven't talked
explicitly about it.)

>  
>  
> 
>  
>  
>  
>  
>  
> I also assume that there will be a way to negotiate the video parameters (no
> issue if we use offer/ answer RFC 3264)
>  
>  
>  
> Roni Even
>  
>  
>  
>  
>  
>  
>  
>  
> 
> _______________________________________________
> rtcweb mailing list
> rtcweb@ietf.orghttps://www.ietf.org/mailman/listinfo/rtcweb
>  

_______________________________________________ rtcweb mailing list
rtcweb@ietf.org https://www.ietf.org/mailman/listinfo/rtcweb