PIC Replacement for the Philips OQ0500 used in the Philips PM56

[Chris56000]

Can anyone suggest whether it is possible to program a PIC or GAL  to be functionally equivalent to a Philips "OQ0500" Bit Reduction" IC?

Thus device is used in the Philips PM5654 Vertical Interval Test Generator and Inserter, and this device is IC12 of Unit 6 "Test Signal Generation", it reduces 8 bits to 7 bits for a T.R.W. TDC1016 Video DAC  – Page 114 of this PDF :–

https://github.com/inaxeon/PTV_Preservation/blob/main/PM5654/PM5654_Manual.pdf

(There is absolutely nothing whatsoever comes up on looking up this device, only Google's stupid nonsense results!)

Chris Williams

[macboy]

A PIC? No. Maybe a programmable logic device. A GAL might be too small/simple, so a CPLD will be needed, or a small FPGA.

I don't think that IC12 does what you think it does. The input is 8 bits and output is 8 bits as well, so it is not reducing the word size of the data stream from 8 bits to 7 bits.

My understanding of the signal generation is that the test signals are stored in a PROM, but they use what they call a "Bit reduction" scheme to reduce the amount of data stored in PROM. From the description in "17. Unit 6 - test signal generator", this seems to be a kind of RLE - run length encoding. As the unit steps through the addresses in the PROM, the 8-bit data values are passed through the bit reduction IC (IC12) and into the DAC, generating the output signal for luminance (and another for chrominance using a similar circuit). But when IC12 detects a special "hold" flag (a specific reserved 8-bit value) on its inputs, it does not pass this to the DAC, but retains the previous value instead. The next value from the PROM after this flag is the length (in clock pulses) that IC12 must hold this value. It has an output called ENLC ("ENable Luminance Counter" I assume) which it uses to inhibit the PROM Address counter ICs (IC7,8,9 : 74F161 counters) from counting for the duration of the hold. In this way, most test patterns can be dramatically reduced in size, just like simple bitmap graphics which use RLE compression.

Designing a replacement would not be trivial, as you would need a deeper understanding of the IC's function, especially the specific word used for the hold 'flag', and the relative timing and function of the inputs and outputs: ENLC, CP, CLOAD, LLOAD, LDACK, etc.
If this IC works as I think it does, it will need to contain 8-bit latches, a comparator (detect the 'flag' value), and a counter (8 bit or less... we don't know the maximum hold length they used) with preload capability or resettable counter and another comparator, plus some miscellaneous other logic.  You might be able to design a replacement with several discrete logic ICs, fast ones like 74F or 74ACT series because of the 20 MHz clock.

[Chris56000]

Hi!

. . .Member "Macboy", thank you for your excellent detailed explanation!

. . .Philips spares like these devices have long been unobtainable as spare parts for years and such Philips agents as did exist in the UK only dealt with Consumer spares – I believe it was only Pye Unicam that could easily get Philips T & M spares!

. . .Obviously it's not possible to duplicate a PM5654 without access to one itself to measure all the various timing waveforms and get the PROM/EPROM dumps, etc., and I've never seen one on eBay for years!

. . .I do have a much simpler fully transistorised instrument that generates a Vertical Interval Test Signal–like pattern but the last part of the active line is a simple unmodulated linear ramp rather than the  grey scale plus superimposed 4.43 MHz subcarrier generated by the '5654 and some of the Tektronix instruments!

. . .The V.I.T.S. pattern used to be transmitted for about fifteen minutes in full–screen on the former UK analogue BBC2 channel before morning startup , between 8 to 8.30 am – this pattern remains a very useful one for fault–finding video and oscilloscope amplifiers, as it covers almost the full limits of the 625–line analogue TV channel bandwidth in one composite signal, hence there have been many enquiries on the UK Vintage Radio Forum about circuits to generate it!

Chris Williams