1
\$\begingroup\$

In the effort to understand early computers (systems revolving around the 6502 CPU). I have started studying their makeup, while the operation of the CPU & memory makes sense. I'm left stuck and confused when it comes to how all of this was interfaced and displayed on a TV.

Analog video as I've found out is a complex problem all on its own. So I've decided to start simple.

I want to design a simple Monochrome Composite Video Driver - Its only function will be to produce a white screen - However, to really understand the functionality I want to use technology only available during that time period.

My theoretical understanding (PAL):

  • I need achieve composite sync. To do this I need to address:
    • Horizontal Sync
    • Vertical Sync
    • Color Burst --- Can possibly leave this one out?

General PAL Timings:

  • Line Period: 64 us (Micro Seconds)
  • Line blanking 12.05 +- 0.25 us
  • Line sync 4.7 +- 0.1 us
  • Front porch: 1.65 +- 0.1 us
  • Burst start 5.6 +- 0.1 us after sync start.
  • Burst 10 +- 1 cycles

General Specifications:

  • CCIR/PAL standard video signal has 625 lines/frame and it repeats @ 25 frames/sec.

  • Each frame is split into 2 fields; - each consisting of 312.5 lines, called odd and even fields. Thus field rate is 50. i.e. CCIR /PAL std has 50 fields/sec rate.

  • Interlacing: The lines of odd-even field lie alternately . This method of scanning is called interlacing. This interlaced scanning is used to reduce flicker while displaying the image on a monitor.

  • At the start of each Horizontal Line a sync pulse is fired.

  • There are 625 Hor Sync pulses per frame.

  • There are 50 vertical sync pulses per second.

To achieve a monochrome display:

  • I need to address intensity/luminance (Y Signal --- Don't have too much info on where this 'Y' signal comes from)

    • 0 Volts = Black Level

    • 0.7 Volts = Peak White Level

What I don't understand:

Lets assume I can put together the V/H sync pulses and the timings are perfect.

How do I assign a 'White' value to each and every pixel? Where does that 'information' come from?

Lets say the first H sync pulse is fired and the TV is now aligned and ready to draw the very first pixel on the first line (Hopefully I have the understanding right).

It is now time for the TV to see a voltage of +0.714 V to make that pixel turn white.

Does the electronic circuit now switch to a different sub-circuit to retrieve this picture information?

I have a digital mindset at the moment and see the operation in this fashion: The driver does the heavy lifting analog work to align the TV...and then the circuit looks elsewhere to see what color this pixel needs to be and thats my source of confusion where does that +0.714 V come from?

Disclaimer: Aside from what I've included here - Assume I know nothing more. I don't have any prior analog video experience.

\$\endgroup\$
7
  • \$\begingroup\$ Analog NTSC uses 1 volt PP voltage; the black level is 0.3 volts; the sync tips are 0.0 volts; white is 1.0 volts; the impedance is 75 ohms; cables are both source and load terminated; this requires 2.0 volts be generated. \$\endgroup\$ Mar 30, 2019 at 3:41
  • \$\begingroup\$ So you're saying if I can continuously inject 2V it will 'set' pixel values to 'High'= White? \$\endgroup\$
    – SheerKahn
    Mar 30, 2019 at 3:46
  • \$\begingroup\$ You cannot "continuously inject 2V", because the sync tips need to be at 0V. If PAL sync goes the same way as NTSC sync, you also need 0.3V for the horizontal sync pulses during the vertical blanking interval. \$\endgroup\$
    – TimWescott
    Mar 30, 2019 at 3:51
  • \$\begingroup\$ If PAL works the way NTSC does, you don't need the color subcarrier -- if you leave it out, you'll get black & white. But then, you don't say which "PAL" you want -- there's a whole bunch of them. I'm not sure what was the most commonly used for computers, because I'm from NTSC country. \$\endgroup\$
    – TimWescott
    Mar 30, 2019 at 3:52
  • \$\begingroup\$ @SheerKahn One place to start is Geoff's Monochrome Maximite. (Also see the Monochrome Maximite Schematic.) He references the book "Programming 32-bit Microcontrollers in C", where you should jump to page 333 and read forward from there. If you start by generating just black and white (no gray, no color), then the hardware is quite easy (three resistors, is all.) His Color Maximite might be your next step to examine. \$\endgroup\$
    – jonk
    Mar 30, 2019 at 4:23

2 Answers 2

1
\$\begingroup\$

There are hundreds of websites regarding analog video left over from the good old days, and it would be better for you to search for ones that explain things at your level and then come back with more focused questions.

To answer your two questions today:

  1. For a pure monochrome image (black, various shades of gray, white), color burst is not required. Depending on the country, it might be required in a broadcast signal if the monochrome segment is part of an overall color program, but that does not apply to a non-broadcast test pattern. Your receiver/monitor should not have a problem with missing burst.

  2. As you noted, a composit video signal is not continuous video information the same way an audio signal is continuous. Video is punctuated by the H and V sync pulses. The visual part of the signal is continuous only for the length of one line. During that time, a continuous analog voltage of 0.714 V (here in the US with NTSC rules) will be interpreted and displayed as pure white.

In its most basic (and slightly non-compliant) form, a white mono video signal needs three voltages:

0.0 V - blanking level (we'll skip the pedestal for now)

-0.286 V - sync tip level for H and V

+0.714 V - white level

http://www.broadcaststore.com/pdf/model/793698/TT148%20-%204053.pdf

https://www.maximintegrated.com/en/app-notes/index.mvp/id/734

\$\endgroup\$
1
  • \$\begingroup\$ I have edited the original question to clarify my confusion. \$\endgroup\$
    – SheerKahn
    Mar 30, 2019 at 4:06
1
\$\begingroup\$

The analogue TV signal (composite video) doesn't use the concept of pixels, if you are only envisaging a white display you don't need the concept of colour either. TV started a long time ago,but by the 1930's a fully electronic camera, transmission and display system emerged as the dominant system. Some earlier systems had demonstrated colour pictures but the new purely electronic system was just varying shades of grey (B/W, monochrome). The display system was a cathode ray tube (CRT) which had an inside coating of a phosphor compound at the screen end and an electron gun unit to provide a variable strength stream of electrons going towards the screen, between the electron gun and the screen there were arrangements to deflect the electron stream to hit any part of the display screen.

To make a working TV system the camera had to analyse the image it made and output a signal that could be reliably transmitted and received by 1930's equipment, and the receiver had to take the received signal and paint it onto the CRT display screen by moving the cathode ray from the gun over the screen and varying the strength of stream to vary the brightness of the spot of light produced where the stream hit the phosphor. For obvious business reasons the receiver had to be reasonably cheap whereas the camera and transmitter could be more expensive.

The technical solution chosen was a raster system for analysing and displaying the picture, which resulted in the camera scanning a fine line horizontally across the image and outputting a continuous signal representing the variation of image brightness along that line. The camera used an electron beam firing, focussing and deflection system similar to the cathode ray tube (because that was the only solution available at the time). At the end of the first camera line the camera brightness signal had to be blanked (forced to zero level = black), and the beam position yanked back rapidly to the beginning of the line, moved down slightly, the brightness signal un-clamped and the whole process repeated until the bottom of the screen was reached whereupon the camera brightness signal was clamped and the position of the beam yanked back to the start of first line in the picture.

The receiver did all this too but the camera brightness signal was used to vary the display beam strength and thus the brightness of the glowing phosphor dot as the beam was scanned.The display beam strength was also clamped to zero (black) during the yanking back time (flyback).

There were two neat tricks invented to enable reasonable cost receivers. The first was sawtooth scanning. If, instead of scanning horizontally you scan slightly downwards, the beam can be automatically in the correct position just under the previous line after flyback. The beam (camera and display)has to scan at a regular pace as it moves from left to right and then fly back as quickly as possible to start another scan, the control voltage for this movement is referred to as a sawtooth waveform. The vertical movement of the scanning beam need only be a regularly stepped signal to position each horizontal line just under its predecessor but if it too is a regular downward movement it will tilt the horizontal lines slightly downward and obviate the complexity of a stepped signal. The second trick was a way round the problem of painting the picture onto the screen, getting to the bottom of the screen and then starting all over again sufficiently quickly to avoid flickering. This was the 1930's, valves (tubes) were expensive and ordinary valves (ordinary cost valves) couldn't operate fast enough. The solution was to scan the whole image at a sufficiently fast rate (no visible flickering) but only scan half the number of lines (for example the odd lines 1,3,5 etc) and then repeat the whole process but scanning the other lines (2, 4, 6 etc). That way the eye sees a moving image flickering at twice the frequency of a fully scanned image but with all the detail of the complete scanned image. Two to one interlacing was enough to have a nearly invisible flickering and simple (cheap) electronics. Of course all these scanning mechanisms need to be kept running in sync so synchronising signals are also transmitted as well as the brightness signal. You can see that although the brightness signal is a pure variable analogue signal it is a bit chopped up in time to get everything to work.

The above description covers all the different country standards and preferences etc for black and white (monochrome) TV, only the signal timings and voltage levels vary between the systems. This system was developed, broadcast and viewed in the 1930's. Colour TV came later and the extra analogue information and timing signals had to be fitted into the old system without ruining the monochrome performance and without exceeding the allocated transmitting bandwidth. They did a great job. Then digital electronics came along and had to deal with a really ancient TV system - no wonder digital TV systems were invented.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.