Re: Document Coordinate System intepretation

I think I'm agreeing with both of you except to say that I do not think that storage pixels have any spatial dimensions at all – they are just colour tuples that can be addressed in a two dimensional array. Calling them "square" is useful only in the sense that, if they exist at all*, then when you multiply the number of them in each dimension by PAR you get the DAR.

* The pixels might not exist, if for example a transformation processor is performing logical operations without mapping to an output raster of any kind. This is entirely feasible without a root container extent, if only percentage of root container units rh and rw are used, say.

Not sure how much that helps, it's another mental view anyhow.

Nigel


From: Glenn Adams <glenn@skynav.com<mailto:glenn@skynav.com>>
Date: Wednesday, 24 May 2017 at 06:46
To: Pierre-Anthony Lemieux <pal@sandflow.com<mailto:pal@sandflow.com>>
Cc: "public-tt@w3.org<mailto:public-tt@w3.org>" <public-tt@w3.org<mailto:public-tt@w3.org>>
Subject: Re: Document Coordinate System intepretation
Resent-From: <public-tt@w3.org<mailto:public-tt@w3.org>>
Resent-Date: Wednesday, 24 May 2017 at 06:47



On Tue, May 23, 2017 at 11:19 PM, Pierre-Anthony Lemieux <pal@sandflow.com<mailto:pal@sandflow.com>> wrote:
Hi Glenn,

> hmm, it is the aspect ratio of the physical pixels that PAR applies to

What do you mean by "physical pixels"? If the DAR and SAR are those of
the root container, then the PAR is also that of the root container,
not the physical display. In other words, both SAR and PAR apply to
the logical pixels of the coordinate system, completely independently
of the physical pixels.

I mean quasi-physical pixels, i.e., PAR pixels in the DAR space.

Consider that all SAR pixels (pixels in the sample space) are square, then PAR scales such square (logical) pixel to a possibly non-square DAR pixel (pixels in the output space). Since we use "physical" in PAR, then when I refer to physical pixels here I am referring to whatever output pixels are produced by TTML rendering. These are quasi-physical in the sense that the still need to be mapped (possibly multiple times) to an eventual dot pattern on a physical display.


> what I mean is that width divided by height selected by the document processing
> context must equal SAR, and that SAR * PAR must still equal DAR;

The only thing the document processing context can control by setting
the extent in pixel is the SAR. The DAR and the PAR are computed
according to the algorithm specified in Annex H, so no need to repeat
it.

Agreed. It is redundant to include "and PAR" in that cited text.


Best,

-- Pierre

On Tue, May 23, 2017 at 10:09 PM, Glenn Adams <glenn@skynav.com<mailto:glenn@skynav.com>> wrote:
>
>
> On Tue, May 23, 2017 at 10:50 PM, Pierre-Anthony Lemieux <pal@sandflow.com<mailto:pal@sandflow.com>>
> wrote:
>>
>> Hi Glenn et al.,
>>
>> Before I continue filing issues on PR #321, I thought I would confirm
>> (or not) my summary understanding of the document coordinate system as
>> currently proposed.
>>
>> - the coordinate system is in logical pixels
>
>
> yes, effectively in the storage (sample) coordinate space (as governed by
> SAR)
>
>>
>> - the aspect ratio of each of the logical pixels is set by the value
>> of Pixel Aspect Ratio (PAR) of the root container
>
>
> hmm, it is the aspect ratio of the physical pixels that PAR applies to, or,
> to be more precise, pixels in the display coordinate space (as governed by
> DAR); logical pixels in the storage coordinate space (as governed by SAR)
> have no aspect ratio per se;
>
>>
>> - the origin of the coordinate system is the top left corner of the
>> root container
>
>
> yes
>
>>
>> - the positive x-axis points to the right edge of the root container,
>
>
> yes
>
>>
>> while the positive y-axis points to the bottom edge of the root
>
>
> yes
>
>>
>> contain
>> - the extent of the root container in this coordinate system is set
>> either by (a) tts:extent if specified in "px" or (b) arbitrarily by
>> the document processing context such that the ratio of the width
>> divided by the height is equal to the computed Storage Aspect Ratio
>> (SAR) of the root container
>
>
> yes for (a), for (b), this is not exactly what I wrote, which is
>
> otherwise, the resolution of the root container region is determined
> (arbitrarily) by the document processing context in a manner that respects
> the resolved values of SAR and PAR as determined by H.1 Aspect Ratios above
>
> in saying that the resolved value of SAR be respected, what I mean is that
> width divided by height selected by the document processing context must
> equal SAR, and that SAR * PAR must still equal DAR;
>
>>
>>
>> Thanks,
>>
>> -- Pierre
>>
>




----------------------------

http://www.bbc.co.uk

This e-mail (and any attachments) is confidential and may contain personal views which are not the views of the BBC unless specifically stated.
If you have received it in error, please delete it from your system.
Do not use, copy or disclose the information in any way nor act in reliance on it and notify the sender immediately.
Please note that the BBC monitors e-mails sent or received.
Further communication will signify your consent to this.

---------------------