Aurora CBS Sequential Output

[peter scott]

Thanks Terry, I'll buy into that explanation.

Peter

[tubesrule]

Hi Peter,
The CBS output is field sequential so it is actually running at a 144Hz rate. It takes 6 fields or 3 frames for a complete color picture so the color frame rate is 24Hz. You would need a 6 segment color disk spinning at 1440rpm. CBS used a RBG sequence rather than the more modern RGB.

The MultiStandard converter did not include CBS color although there was a special firmware build that was a CBS only converter. The MultiStandard and SCRF CBS converters can only accept an NTSC input. The World Converter supported sequential color on most standards for people who wanted to experiment.

Darryl

[peter scott]

Thanks Darryl,

I think you've given me the key to understanding it. The disc is rotating at 24 rev/sec but each rev has 6 colour changes. I was stupidly thinking that there was a difference between the luminance signal from a colour disc camera and that of the Aurora output. I had a picture in my mind of the complete camera luminance changing instantaneously to every change in the scene viewed which of course doesn't happen in a scanned image. Each field is just a snapshot only changing at field rate. The decaying phosphor image is not revisited faster than field rate.

Peter

[Jac Janssen]

The decaying phosphor image is not revisited faster than field rate.

Peter

Which nevertheless is 2.4 times faster than the normal 30 Hz B&W frame rate, so I'm not sure about the lagging phosphor effect muddling the colour. Darryl and the USA television collectors should be able to comment on that.
I regret that I have never seen a CBS color system!

This JVC monitor is interesting: https://www.youtube.com/watch?v=z-q8ehzHeQQ


Jac

[tubesrule]

Peter,
Yes, that is correct. Each field is a complete color, changing on the next field. Since there are an odd number of color segments (three) each color in the RBGRBG sequence exposes alternating fields to make a complete frame, but at a displaced time. (more on this below)

Jac,
The phosphor persistence is of concern as any lag from one field to the next pollutes the color. In operation with standard P4 phosphor this doesn't seem to be much of an issue to normal viewing. I don't believe anyone has done actual tests on one of the operational sets, but I'm sure there is some amount of measurable effect.

What is of concern and much more easily visible is color fringing on movement. While registration is inherently perfect with sequential color, since two fields of the same color are seperated by three fields in time, the image could have moved a considerable amount causing color fringing. This is noticeable on fast motion, but I would say not necessarily objectionable. Compare this to poor color registration on a standard tri-gun crt and you could argue about which one is really worse?

Darryl

[Mike Watterson]

Likely less nasty than 3:2 pull down comb effect on movement transferring film to NTSC. Which is why Progressive and Component was important on USA DVD players, but irrelevant on PAL, which usually simply played the 24 at 25.
I hate getting VHS or DVDs and discovering they are badly frame rate converted from NTSC with the artefacts instead of being a Film to PAL transfer.
Even without 3:2 pull down comb effect on movement, some NTSC to PAL conversions (say video camera originated) have disturbing shimmer on panning.

DLP does often use a colour wheel. Only expensive DLP uses three displays and combiner.

I'm tempted to try a colour wheel on a mono camera and the 2" or 6" TV (all three are mono 625). Somehow I ended up with a USA portable 6" mono TV. There is no video baseband in so it can only be fed with US RF modulator! I bodged a modulator and it does display 625 but with wrong height (naturally).

[ppppenguin]

Colour wheel cameras were used on the Apollo moon landings. A 3 tube camera would have been too big and heavy.

3:2 pulldown can be removed from a recording. Sometimes called reverse telecine. I think this was first developed in the late 1980s or early 1990s. It has to recognise the 3:2 cadence, then correct it. If the cadence jumps too often, which can happen if the video is edited after the orignal TK, the problem is a difficult one.

The modern version of colour fringing is failure of de-interlacing. A lot of older LCD TVs had poor quality de-interlacing giving truly horrible effects on moving objects.

[Mike Watterson]

Also single tube stripe filter vidicons were really poor quality and tended to go out of registration. Ghastly things. I'm still baffled that they even worked at all.

Yes, inverse Telecine can work well. Some TVs (even from VHS) and some bluRay players playing daft DVDs can do it now.
Indeed bad de-interlacing gives a similar appearance to 3:2 pull down to interlaced NTSC, except worse.
It boggles me the stuff on sale today originated in USA where I could do what they didn't do, 15 years ago on my PC.

[Geordie McBoyne]

From the ETF: http://earlytelevision.org/field_sequent...types.html

CBS model 12CC2:
http://earlytelevision.org/cbs_12cc2.html

Geordie McBoyne.

[peter scott]

I was very impressed by the Gray Research set and the time that Steve took to demonstrate it for me when I visited 5 years ago. My short video doesn't really do justice to it.

Peter