|
Vintage Television and Video Vintage television and video equipment, programmes, VCRs etc. |
|
Thread Tools |
16th Jul 2024, 3:34 am | #1 |
Heptode
Join Date: Aug 2009
Location: Worthing, Sussex, UK.
Posts: 672
|
Teletext SAA50xx chipset
When Teletext uses double height characters these characters take up the next text line
below. In otherwords the top half of the double height characters are displayed on the current line, and the next line displays the bottom half. The BBC micro which uses the SAA5050 as a character generator in teletext (mode 7) requires that the double height characters are written into memory twice, once for the top half of the character and once for the bottom half of the character. However inspecting teletext data captured from video tape it appears that only the top half of the characters are encoded and the decoding circuit is expected to duplicate the text row for the bottom half of the double height characters. Does anyone know how a teletext receiver using the SAA50xx chipset performs this magic. I am asking this because I am looking at emulating a teletext card wt625 which is part of an 80 bus system. I don't have one of these cards to try it on. However it uses the timing chair part SAA5020. Is this the chip that does the magic for the double height characters? |
16th Jul 2024, 4:46 am | #2 |
Dekatron
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,361
|
Re: Teletext SAA50xx chipset
The data sheets are none-too-clear on this, but I wonder if it has to do with the TLC/ (Transmitted Large Character) output from the SAA5050 to the SAA5020 timing chain. Does this cause the latter chip to generate the address for the previous line when displaying the bottom half of the character?
|
16th Jul 2024, 12:49 pm | #3 |
Pentode
Join Date: Aug 2021
Location: Enfield, London, UK.
Posts: 139
|
Re: Teletext SAA50xx chipset
Yes, only the top row of characters that are intended to display as double height are transmitted, unlike the BBC Basic implementation. The relevant non-displaying control characters, Double Height / Normal Height, are then added before and after any double height sections during page composition. The decoder acts on those and "stretches" the same character glyphs (from the SAA5050) to also fill the line below. All new page lines are assumed to be normal height until a double height control character in encountered, so the normal height control code is only required if normal height text is to be returned to following a double height section in the same line.
Yes, I believe the SAA5020 is the prime mover in achieving this. Attached are some docs that may help. The relevant pages from the original 1976 Teletext spec and the same from a later ETS version. Also, the description of how Double Height is logically achieved from the book "Teletext & Viewdata" by Steve Money (recommended reading!). Hope it helps, and good luck with the project. Last edited by SunSPARC; 16th Jul 2024 at 1:09 pm. Reason: Fixed empty PDF |
16th Jul 2024, 7:47 pm | #4 |
Pentode
Join Date: Aug 2021
Location: Enfield, London, UK.
Posts: 139
|
Re: Teletext SAA50xx chipset
I knew I had this somewhere, but needed to search for Signetics and not Philips!
The datasheet (attached) makes it clear what role the SAA5020 has in Double Height duties. |
21st Jul 2024, 4:34 pm | #5 |
Heptode
Join Date: Aug 2009
Location: Worthing, Sussex, UK.
Posts: 672
|
Re: Teletext SAA50xx chipset
Thanks all - it looks like the SAA5020 handles the two rows (20 scan lines) of double
height text without duplication of data being required. The reason that other systems like the BBC micro/vt100 terminal etc require duplication is because they use crt controllers such as the 6845. However if anyone has experience of real hardware that can actually verify this. Ideally using a wt625 card or other teletext card using a SAA5020 that would be wonderful. Thanks all. |