Any way to tell what --sar to use (ITU or non-ITU) for a DVD encode?

[Ghitulescu]

People keep telling me BBC had shot on film.
However, all BBC movies and serials I have bought here in Germany, all look like they have been if not shot at least edited, and if not edited then at least one intermediary copy was on tape. Maybe UK-market is fed with better editions, who knows....?

That is, at least once IMHO they had to work with non-square pixels. And this has been done according to their own rules. However, BBC does not manufacture any equipment. So they rely on foreign manufacturers, Sony, JVC, Panasonic. The way they implement what BBC requires may vary.

This Babel thing happens everywhere in the world where such methods are used. I was therefore one of the most early supporters of the BD, from the consumer point of view, for it eradicated (in the sense of WHO :) ) the rectangular PARs - and reverted to the natural way, of identical units on each scale. Yes, for compatibility sake BD kept some very used PARs, but it provided the technical/formal provision for an undisturbed view.

The correct way of resizing an image is to find a circle or a square or anything that can give (disclose :) ) the real ratio.

[Logan9778]

Yeah, agreed. Cricles and squares seem to be the only way with these old movies. A I can tell some of these were "filmed" off of a TV screen, because you can tell how circles seem to change near the edges. Some early WHO video looks like it was shot with a fish eye camera.

[Logan9778]

Good that you mentioned BBC.

BBC used a different approach.

The digital frame sizes are 788 and 1050 respectively (PAR=1), for they include also what BBC calls blanking. The extra pixels are usually painted black.

Resize then to 788/1050 then cut the black borders and you'll end with some strange number of pixels but quadratic.

Thanks! After a 2 month break to clear my head of all this, I realize what you're talking about now. So I assume it looks like ITU, but 788 pixels wide with 10 pixel wide black ( or some say it could be white) columns on each side? And I assume the black or white columns would be overscan on an old 4:3 TV, same as ITU?

The DVDs I have (Old Doctor Who from the 60's) seem to just be the whole film frame ( you can actually see the slight rounded corner at the bottom left sometimes ), so I'm assuming non-ITU and 768 pixels, and the 3 or 4 pixels of black you sometimes see are just sloppy capture as they went beyond the frame itself on to the actual frame holder. I now understand that the digital width is what "fits" into 720x576 non-square pixels, and the "digital pixel size" varies according to different companies. i.e. BBC's digital "pixel width" is different from most other standard "digital pixel widths", so 788 "BBC pixels" fit into 720x576 the same as the "standard 768 pixels" fit into 720x576.

Found a good post on this here:

http://www.lafcpug.org/phorum/read.php?1,263245,263324

@mods - Sorry for the double post, but it's been two months since my last post.

[hello_hello]

If you read some of what I quoted from the ABC pdf earlier (and as far as I know the ABC do everything the same as the BBC):

2.2.3 All SDTV programs produced using modern digital equipment shall have narrow horizontal blanking, ie. active vision shall be 720 horizontal pixels wide. Legacy programs with 702 pixel wide vision (wide horizontal blanking) will also be acceptable. A single program shall have consistent blanking throughout.

And combine it with the 788 square pixel width Ghitulescu's mentioned, to me it still adds up to an mpeg4/ITU pixel aspect ratio but there's very little to no black at the sides.

I did a little more searching and I'm still a bit confused, but some anecdotal (almost) evidence the picture is supposed to be 720 pixels wide, although wider than 4:3.

EBU Technical Recommendation R92-1999 (http://tech.ebu.ch/docs/r/r092.pdf) (pdf)
Active picture area and picture centring in analogue and digital 625/50 television systems
The EBU is aware of a certain amount of confusion about the active picture area in the implementation and use of digital signals conforming to the ITU-R Recommendation BT.601 [1].
Recommendation BT.601 specifies a line length of 720 luminance pixels (13.5 MHz sampling). ITU-R Recommendation BT.470[2], specifies a line length of 52 µs for 625 line analogue signals. This corresponds to 702 luminance pixels.
This apparent difference can lead to difficulties in conversion, especially if users wish to maintain the correct aspect ratio of the pictures.
Recommendation BT.601 accommodates modest variations in the position and length of analogue blanking which arise before a signal is digitised or when digital signals pass through any subsequent analogue process.
The EBU recommends that:
In 625-line television systems sampled to ITU-R Rec. BT.601 part A, only the central 702 luminance samples of the digital active line (samples 9-710 inclusive) and their associated chrominance samples should be used to carry the active picture. The remaining 18 luminance samples and their associated chrominance samples may be used to carry picture information only but for no other purpose. It cannot be guaranteed that picture information in these samples will be displayed in either 4:3 or 16:9 aspect ratio images.

BBC Technical Standards for Network Television Delivery (https://web.archive.org/web/20100331085936if_/http://www.bbc.co.uk:80/guidelines/dq/pdf/tv/tv_standards_london.pdf)
4.1.4 Video Signal Timings.
Digitally delivered pictures are considered to have a nominal active width of 702 pixels (52us) starting on the 10th pixel and ending on the 711th pixel in a standard REC 601(720 sample) width. A minimum width of 699 pixels (51.75us) within these limits must be achieved. Additional active pixels outside the above limits must be an extension of the main picture.

There's information in this ITU PDF explaining the relationship between digital and analogue.
Studio encoding parameters of digital television for standard 4:3 and wide-screen 16:9 aspect ratios (http://www.itu.int/dms_pubrec/itu-r/rec/bt/R-REC-BT.601-7-201103-I!!PDF-E.pdf)
Encoding parameter values for the 4:2:2, 13.5 MHz member of the family
The specification (see Table 3) applies to the 4:2:2 member of the family, to be used for the standard digital interface between main digital studio equipment and for international programme exchange of 4:3 aspect ratio digital television or wide-screen 16:9 aspect ratio digital television when it is necessary to keep the same analogue signal bandwidth and digital rates.

It refers to a chart on page 7 that lists the total number of samples per line for PAL as 864, with the number of samples per digital active line being 720.
On page 10 there's a diagram showing how the "digital blanking" interval is from pixel 720 to pixel 864.

My take on all that, and hopefully someone will correct me if I'm wrong, is the original BT.470 analogue spec is equivalent to 702 pixels wide in digital-speak, which is exactly 4:3, and BT.601 doesn't change that, only adding some bonus picture on each side.
If that's correct, it makes sense that a 4:3 DVD containing a video capture would have lashings of black down the sides, while a digital source (and those old Doctor Who episodes have been digitally process every which way), would be 720 pixels wide, although I assume still only 704 of the width would be 4:3. I'm still reasonably sure most 4:3 DVDs I've come across have an ITU/mpeg4 pixel aspect ratio, be they of BBC origins or otherwise..

I haven't seen enough of them to be certain, but I strongly suspect the few 16:9 BBC DVDs I've seen are ITU, despite the 720 width being almost entirely picture, but as a rule the rest of the world seems to be sticking to the 720 width being part of the total 16:9 aspect ratio.... mostly. I still don't fully understand why.

Some interesting reading here. (pdf)
The origins of the 4:2:2 DTV standard (https://tech.ebu.ch/docs/techreview/trev_304-rec601_wood.pdf)

[Logan9778]

Thanks Hello! Sorry to get back so late.

Well, I'm looking at Doctor Who stuff again, and it looks like most have the BBC blanking. About 12 to 13 pixel columns of black on each side. I still have a hard time finding a decent circle to look at though, and some just don't come out whether I use non-ITU, ITU, or BBC Blanking PARs. Finally found a good circle at the beginning of "The War Games" Disk 2, where they shoot some kind of force field out, and its a bunch of blurry circles. Looks like Ghitulescu may be right about the 788 stretch, as it only looks circular if you stretch it out to 788.

The whole line is, after proper 4:3/16:9 DAR, 788 pixels wide in 4:3 and 1050 in 16:9.

EDIT: In the end, after a lot of circle testing, I decided on going with 720x788 for Doctor Who. PAR for 720 x 788 is 197:180. I cropped it to 704 x 572 (cropping doesn't affect PAR), then went with a PAR of 197:180, to get 704 x 770, which seemed to give the best circles. Doctor Who NTSC videos, 720x480, seem to be ITU, with a 10:11 PAR, which produced the best circles.

There do seem to be parts of Doctor Who sometimes that appear to be 720 x 864, but examination of the rest of video shows circles at this DAR to be wildly off. You should stick with 720x788.

Thanks for your help guys!