Digital TV display scanning

[sophiecentaur]

Since the transition from CRT to digital displays, has there been any move to modify the displays to pseudo random scan for TV pictures, as opposed to raster scan? I have tried Google but the only references I can find refer to vector graphics, which is not what I wanted.
There must be someone around who just 'knows' the answer to this one.
I can't imagine that the original digital displays were not raster scan but memory is cheap now and an alternative to a conventional line-by-line raster could be much better for artifacts and resolution. Will this be the next selling feature, once everyone has 3D?

 

[Kholdstare]

This depends on video cable. How they encode the video data. Still this will have some serial manner and raster scan is only the way.

 

[sophiecentaur]

This depends on video cable. How they encode the video data. Still this will have some serial manner and raster scan is only the way.

It doesn't depend on the data link at all. Serial to parallel conversion is done all the time. All you need is a frame store (a very few £). It's done in the original digital coding of the signal and we could now afford it at consumer level. The MPEG decoder uses a pretty hefty amount of memory, in any case and the picture information isn't truly "serial", but in 16X16 pxl blocks.
I Am looking for information from someone who actually 'knows' about this.

 

[jim hardy]

I guess you've seen the A53 standards here? Probably no help.

My interest would be in building a device to interpose in the , say, S-video line and recognize the content identifiers. Content identified as commercials would have the volume lowered ~40db and the brightness turned down to near dark.
In US there's laws against blocking ads but so far as i know you can attenuate them.

Might it work, do you think ?

 

[sophiecentaur]

I guess you've seen the A53 standards here? Probably no help.

My interest would be in building a device to interpose in the , say, S-video line and recognize the content identifiers. Content identified as commercials would have the volume lowered ~40db and the brightness turned down to near dark.
In US there's laws against blocking ads but so far as i know you can attenuate them.

Might it work, do you think ?

Jim
I grew up on PAL, which is based on 625 line interlaced scanning. It just can't be optimal for a display which could do anything you wanted. I just wondered about how MPEG could benefit from an alternative way of displaying moving pictures on a pixcel (?) sequential basis.

Do any displays utilise this?

 

[jim hardy]

Jim
I grew up on PAL, which is based on 625 line interlaced scanning. It just can't be optimal for a display which could do anything you wanted. I just wondered about how MPEG could benefit from an alternative way of displaying moving pictures on a pixcel (?) sequential basis.

Do any displays utilise this?

Sophie I'm still NTSC 545 lines.
My brief excursion into video was in days of the TI-99. I fount its internal board drilled for an alternate RGB video processor IC and bought a couple . But the field was changing so fast i decided not to get into it.
So I'm sure not fluent anymore.

From what I've seen in TI DSP offerings i cannot imagine they aren't. Will see if i can find a tutorial suitable for us, er, engineers in transition...

old jim

 

[nsaspook]

I guess you've seen the A53 standards here? Probably no help.

My interest would be in building a device to interpose in the , say, S-video line and recognize the content identifiers. Content identified as commercials would have the volume lowered ~40db and the brightness turned down to near dark.
In US there's laws against blocking ads but so far as i know you can attenuate them.

Might it work, do you think ?

Take a look at the code in MYTHTV commercial skip functions.

http://www.mythtv.org/ http://www.mythtv.org/wiki/Commercial_Detection

 

[jim hardy]

Thanks nsa haven't found the code yet but will persevere.

Sophie - i forgot to post the link for those standards.

http://www.atsc.org/cms/index.php/standards/published-standards/50-atsc-a53-standard

more here
http://www.atsc.org/cms/index.php/standards/published-standards?layout=default [Broken]

 

[jim hardy]

Sophie i spoke too soon.

All the TI stuff i can find appears to output raster based analog, despite its considerable computing power.
If i understand correctly, TV is still 30 frames/sec but with the fast computer DSP chips they can generate intermediate frames to smooth motion and compensate for response time of whatever device makes individual pixels. Hence 120 and 240 hz TV's.

http://www.ti.com/product/ths8200-ep
http://www.ti.com/product/tms320dm6446


their OMAP line for mobile devices seems to be pushing the frontier.

Just out of my field. Sorry to waste your time. But I've been curious too.

 

[nsaspook]

Thanks nsa haven't found the code yet but will persevere.

https://github.com/MythTV/mythtv/tree/master/mythtv/programs/mythcommflag

It's not the easiest code to step into cold.