Philips PM5644 "PTV" Test Pattern generator. Anyone actually got one?

[inaxeon]

I've been poking around with one of these units. It is a PAL version however has a weird unexpected pattern with an outline of a native american in the circle.

The pattern is stored in a bank of 9 EPROMs on the main PCB. The EPROMs only store a single pattern so no chance of changing it, unless I can find another set of ROM images.

Does anyone have one with the EPROMs for the traditional pattern?

[inaxeon]

On further investigation. Looks like EPROMS 0-3 are for Luma and possibly 4-8 are Chroma.

Ooooo. This looks like an interesting project. See if I can come up with a tool to generate my own completely custom test card for this unit...

[inaxeon]

Already made some progress.

I've written a crude tool which is able to parse the chroma EPROMs although I'm not 100% sure how brightness is encoded yet.

The result is an image 2048x583 pixels in size (attached). It looks like perhaps even the blanking periods are encoded into this image.

Full sized image here:

https://xpander.mattmillman.com/files/PM5644.png

[Diabolical Artificer]

That's an interesting pattern, there must be a story behind it: why the 1st nation chappy?

Andy.

[dazzlevision]

The "Indian Head" test pattern was very popular in the USA, dating back to Monoscope camera days.

https://en.wikipedia.org/wiki/Indian-head_test_pattern

[Vintage Engr]

Agreed, it was also available on diascopes, both for 1 inch & 2/3 inch vidicon/saticon/ plumbicon tubes. Also many of our test tapes had the Indian Head test card on them.

You've now reminded me that I have a couple of diascopes stashed away somewhere, I will now have to try & find them.

David.

[inaxeon]

Originally Posted by dazzlevision

The "Indian Head" test pattern was very popular in the USA, dating back to Monoscope camera days.

https://en.wikipedia.org/wiki/Indian-head_test_pattern

So what's it doing programmed into a PAL PM5644? This is bizarre.

[inaxeon]

I added an improved rendition of it to Wikipedia. Looks like this is the first ever digital reproduction of a PTV pattern too:

https://en.wikipedia.org/wiki/Philip...cal_PM5644.png

(on the PM5544 page)

The Chroma ROMs appears to be empty so it's B&W. That is how it appears on-screen - but in 4:3. It's probably longer in the ROM to give the DACs a little more resolution when drawing the lines.

[Maarten]

Are there any labels on the ROM chips? The original ones probably would have had some kind of Philips code numbers on them.

Also, many facilities labeled their equipment with their own name somewhere on the outside. No such label?

[inaxeon]

These ROMs (with the "Indian head" pattern) are all labelled up with "4008" part numbers - the work of Philips Denmark I think you said in another thread?

https://xpander.mattmillman.com/file...M5644_Open.jpg

These are the luminance ROMs:
4008-102-58802
4008-102-58812
4008-102-58822
4008-102-58832

I assume the remaining are chrominance as there's no pattern data in them.

[inaxeon]

Originally Posted by Maarten

Also, many facilities labeled their equipment with their own name somewhere on the outside. No such label?

It has a Protelevision Technologies sticker on it - as I understand this is what PTV renamed to when Philips sold it? No other third party stickers.

[kan_turk]

Originally Posted by inaxeon

So what's it doing programmed into a PAL PM5644? This is bizarre.

Nothing bizarre - indian head pattern has been, at one time or another, used in many countries other than US and is, along with other test patterns, in no way irrevocably linked to any particular TV system - it’s a matter of choice by broadcasters

[inaxeon]

Originally Posted by kan_turk

Nothing bizarre - indian head pattern has been, at one time or another, used in many countries other than US and is, along with other test patterns, in no way irrevocably linked to any particular TV system - it’s a matter of choice by broadcasters

Yes having researched it I found there were a small number of PAL regions who used it. Hopefully someone who recognises it puts their hand up someday.

I bought it from a European scrap dealer who claims to have no record of where it came from.

[inaxeon]

Have learned a little bit more. The unit I have doesn't output a colourburst at all. I guess this is correct for a B&W generator.

I found that if I remove the jumper in the top centre of the main PCB the colourburst appears on my scope. The image remains in B&W (as expected) but now with a few fringes of colour scattered here and there. Random sampling errors on the LCD TV I'm using perhaps.

The EPROMs appear to be split into three banks. The first four are Y. The next three are R-Y and the last two are B-Y. I think. I saw some phase swapping on the DAC attached to the bank of three and I believe this is expected for R-Y.

Really interesting that B-Y has less resolution (two EPROMs) than R-Y (three EPROMs).

Next step is to do some experimentation to work out how the colour samples are actually laid out.

[inaxeon]

Originally Posted by inaxeon

Really interesting that B-Y has less resolution (two EPROMs) than R-Y (three EPROMs).

Maybe explained by the attached. B-Y axis seems to be smaller, so perhaps needs less samples to accurately represent.