AriX wrote:It would have been able to generate the text, but it would not have been able to encode it into the closed captioned format, which involves putting the captions into the VBI of the outputted video. Expensive hardware would be needed for that; normally it could just be encoded in the video delivered over satellite, but the Amiga's genlock renders that useless.
I'm not so sure it can't do it. The Amiga's video output was legendarily flexible, including being able to overscan all the way to the full signal resolution (rather than just the lines on screen) I'm doing some research on the matter, theres a few docs on the internet that seem to suggest it's possible but nothing concrete yet.
Also the genlock wouldn't necessarily render CC over satellite useless i thought? The code could just ensure that that bit of the video signal wasn't replaced surely? The more useless bit I would have thought would be that of only being able to send CC for one of the two video streams.
AriX wrote:It would have been able to generate the text, but it would not have been able to encode it into the closed captioned format, which involves putting the captions into the VBI of the outputted video. Expensive hardware would be needed for that; normally it could just be encoded in the video delivered over satellite, but the Amiga's genlock renders that useless.
I'm not so sure it can't do it. The Amiga's video output was legendarily flexible, including being able to overscan all the way to the full signal resolution (rather than just the lines on screen) I'm doing some research on the matter, theres a few docs on the internet that seem to suggest it's possible but nothing concrete yet.
Also the genlock wouldn't necessarily render CC over satellite useless i thought? The code could just ensure that that bit of the video signal wasn't replaced surely? The more useless bit I would have thought would be that of only being able to send CC for one of the two video streams.
I think at the very least they would have to redesign their genlock, and the new one would be pretty expensive. I really don't think it would work with the existing genlock - there's just no facility to encode VBI info into there, and when it combines the two NTSC feeds (Amiga and satellite/LaserDisc/local ad VHS), I'm pretty sure the captions would get destroyed.
Never tested it for actual captioning, but I'm pretty sure the Amiga itself can easily generate image data for line 21. The genlock hardware should have nothing to do with it - it would require no more than a 2-bitplane screen for that scanline, and one could easily switch between C-band captions or local captions by using color 0 on line 21 to feed C-band captions through, and using any two other colors on line 21, with no color 0, to cover the C-band captions and generate them using the Amiga. Line 21 caption signals aren't really anything special in terms of the actual signal itself, only in the way the data is encoded.
The hardest part of such type of captioning, from a Prevue standpoint, would be choosing between two different sets of captioning for both video feeds, left and right. Theoretically, the Amiga could be sent two sets of caption data and convert the desired one on-the-fly. However, as Prevue was coded in C, it's likely that performance wouldn't have been the best doing so (we already know how much of a pile of crap the code is anyway, given its fragility). Some tightly coded assembly routines would likely have no problem spitting out the proper line 21 data for each frame.
LocalH wrote:Never tested it for actual captioning, but I'm pretty sure the Amiga itself can easily generate image data for line 21. The genlock hardware should have nothing to do with it - it would require no more than a 2-bitplane screen for that scanline, and one could easily switch between C-band captions or local captions by using color 0 on line 21 to feed C-band captions through, and using any two other colors on line 21, with no color 0, to cover the C-band captions and generate them using the Amiga. Line 21 caption signals aren't really anything special in terms of the actual signal itself, only in the way the data is encoded.
The hardest part of such type of captioning, from a Prevue standpoint, would be choosing between two different sets of captioning for both video feeds, left and right. Theoretically, the Amiga could be sent two sets of caption data and convert the desired one on-the-fly. However, as Prevue was coded in C, it's likely that performance wouldn't have been the best doing so (we already know how much of a pile of crap the code is anyway, given its fragility). Some tightly coded assembly routines would likely have no problem spitting out the proper line 21 data for each frame.