Double the frame rate

Forum for discussion of narrow-bandwidth mechanical television

Moderators: Dave Moll, Andrew Davie, Steve Anderson

Double the frame rate

Postby Andrew Davie » Sat Apr 28, 2007 1:10 am

Just something I've been thinking about --- we're playing a video signal from a CD, and the pulses embedded in the signal are used to synchronise the video such that the Nipkow disc is rotating 750rpm. We get a very flickery picture at 12.5Hz, which is roughly half what is needed to give a 'flicker free' image.

So I was thinking -- what would be the consequences of modifying the input waveform (ie: preprocessing with some software) to increase the frame rate. If the waveform were 'scrunched', for example, I would assume that up to a point the 'standard' club circuits would cope. They'd (at a guess) be happy working at 15Hz. Maybe 18.... what's the limit?

Of course, just 'scrunching' the waveform would also increase the video speed. One would need to write some software that would effectively transpose the 12.5 fps to 25fps (ie: keep the same speed, increase the frame rate).

I do suspect that one of the limitations will be the synchronisation with the timing holes on the Nipkow disc. I recall reading that some sort of timer was used, which kept getting reset *except* for the missing hole, and this was used to generate the frame pulse. Of course, if the next hole comes along too soon, we miss that pulse. But leaving that problem aside...

I am also aware that we would be significantly reducing our vertical resolution. But this is so 'good' on my monitor, I'm wondering how it would look at half resolution but double the frame rate. I would assume that brightness would not suffer, as although we're seeing each hole whizz by in half the time (so half the total light seen), we are also seeing each hole twice as often (so the overall effect is same brighness).

So my question is -- abandoning the ideal of the club standard 12.5Hz, just how much better would images look at higher frame rates and lower vertical resolution? Am I missing something in my assumptions?
User avatar
Andrew Davie
"Gomez!", "Oh Morticia."
 
Posts: 1590
Joined: Wed Jan 24, 2007 4:42 pm
Location: Queensland, Australia

Postby DrZarkov » Sat Apr 28, 2007 3:42 am

Well, maybe it looks better, that's why television improved to the standard we have now. But that has nothing to do with Baird's old TV standard and narrow bandwidth TV. If it's just picture quality and narrow bandwidth and not history, you could use digital compression to transport the signal, and I bet you could get with the same bandwidth 60 lines or more and 25 frames. For reproduction forget about Nipkow-disc, but take a computer, or if you want a stand-alone monitor take a TFT-display which holds the pixel until new pixel-information comes - no flicker at all, doesn't mater how many frames you transmit. The question is: Do we want this, or is that another kind of sport?
User avatar
DrZarkov
I think I've had a cranial implosion.
 
Posts: 1041
Joined: Mon Feb 19, 2007 11:28 pm
Location: Kamp-Lintfort, Germany

Re: Double the frame rate

Postby Roland » Sat Apr 28, 2007 10:00 am

Andrew Davie wrote:Just something I've been thinking about --- we're playing a video signal from a CD, and the pulses embedded in the signal are used to synchronise the video such that the Nipkow disc is rotating 750rpm. We get a very flickery picture at 12.5Hz, which is roughly half what is needed to give a 'flicker free' image.

I am also aware that we would be significantly reducing our vertical resolution. But this is so 'good' on my monitor, I'm wondering how it would look at half resolution but double the frame rate.


Curiously this is very similar to something I've been thinking about recently. I too noticed that the vertical resolution was more than adequate and wondered if I could double the lines up to 64 by halving the vertical resolution.

My proposal would be to keep the disk speed of 750RPM and the concept of 12.5 fps - however just pack the 2 halves of the image above each other in the single frame. The 2 sets of 32 holes would be interleaved on the disk as each of the original 32 lines would contain the nth and the nth+32 line of the 64 line display.

With a little bit of care on the hole spacing a 4:3 aspect ratio picture could be achieved which is much more useful for viewing old films/cartoons etc

I've not sure if I've explained it very well - but the important point is that the principle works equally well to doubling the fps as you propose. Instead of packing the 2 halves of the image above each other in the single frame - you would pack sucessive frames.

The Nipkow disk would then have 2 non interleaved spirals of 32 holes - each hole 5.625 degrees away from the next one.

All the electronics and synchronisation holes/sensors should be able to remain exactly the same as for the club standard.

Much as I'm in this for the mechanical apsects of old TV - I suggest that using a PC/Mac to pack the frames would be simpler way to generate a large variety of test data - rather than to have to build a special camera for each special monitor you might make.


Andrew Davie wrote:So my question is -- abandoning the ideal of the club standard 12.5Hz, just how much better would images look at higher frame rates and lower vertical resolution? Am I missing something in my assumptions?


I don't see it as abandoning the club standard - merely trying something different in addition to it. I'm currently looking at an 8 line sytem with 4 frames for each disk revolution (for non electric TV) . Its not Baird standard - but then from what I've read about him - I don't believe he saw the 30 line system as anything more than a stepping stone to better systems.

:-)

Roland.
Roland
Master Craftsman
 
Posts: 56
Joined: Wed Feb 21, 2007 11:56 am
Location: Hampshire, UK

Improving frame rate.

Postby Stephen » Sat Apr 28, 2007 11:22 am

John Logie Baird proposed a means for improving frame rate in the early 1930s that recognised that you could intermesh high definition low frame rate information with low definition high frame rate information. I wrote an article about this, the first portion of which is in the last newsletter.

Mr. Baird used a modified scanning disc to accomplish this. I came up with a scheme that uses standard scanning discs by "packing" three low definition frames and one high definition frame within the area of a standard 30 line frame. The low definition frames line up along the top of the high definition frame and are low definiton by virtue of being smaller.

On the face of it, this scheme would give you low definition frames at the rate of three times 12.5 fps, or 37.5 fps, to eliminate flicker whilst adding the detail at a rate of 12.5 fps, still sufficient to impart smooth motion. With a little trick, namely reverse imaging the low definition frames along the top of the high definition frame so that the low definition frames scan in reverse order, the low definition frame rate actually increases by four, to 50 fps!

I suggest one hypothetical way to optically project the images on or from a scanning disc, but I am sure that there are better ways. See the full article at http://www.taswegian.com/NBTV/images/SDS.pdf . Mr. Baird's patent for the scheme is in the "Patents and Articles" section of the forum. It is British Patent 391,924, filed 2 February 1932.
Stephen
User avatar
Stephen
Anyone have a spare straightjacket?
 
Posts: 427
Joined: Fri Jan 26, 2007 7:00 am

Frame rate conversion.

Postby Steve Anderson » Mon Apr 30, 2007 12:34 pm

Gents,

This very thing was done in Vol. 29 No. 3 of the newsletter by Graham lewis, 'NBTV Frame Rate Convertor'. The input waveform goes in at 12.5Hz and comes out at 50Hz.

This is still at 32 lines, it's not a conversion to 625 TV standard.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand


Return to Mechanical NBTV

Who is online

Users browsing this forum: No registered users and 37 guests

cron