I also try to answer the question "Why do this at all?" near the end.
This signal is included as part of the Luma (Luminence, or brightness) signal in S-Video signals. The diode here lets the signal do its thing, and the diode itself provides sufficient isolation, so we don't need a resistor on this line (also because the signal doesn not need attenuating.) The signal is run past the brightness control potentiometer since we don't want to have the signal be too weak if we have to turn this potentiometer up to a moderately high resistance level.
Each bit is either on or off. How much each bit's electrical signal affects the total luminance value is based on how much that signal is attenuated (decreased) by the resistor it goes through. The high-order bit, bit 3, gets attenuated half as much as the next most significan bit, bit 2. This means that bit 3 affects the brightness of the pixel twice as much as bit 2. And bit two is twice as important as bit 1, and bit one lords it over lowly bit 0, which will change the brightness by only a paltry amount.
You'll notice the resistor values are roughly doubled with each lower-order bit. You could get really persnickety about the values of the resistors, and get resistors that are precisely the right ratio of values, but there really won't be any noticable improvement in the image over just using whatever is pretty close value-wise and easily obtainable.
Once we've done the digital-to-analog conversion, there's a potentiometer to provide an overall brightness adjustment. I've tried building this circuit without a pot here, but the variability of the components used, and the display you'll be using, require some adjustment be available.
The next part of the circuit is the amplifier, a rather simple one with only a single transistor that is perfectly adequate for a wide range of tasks. If you are an amplifier fanatic you may want to insert your own favorite amplifier circuit here, but once you get the signal to your monitor you'll wonder why you bothered. This one is perfectly adequate for what we need.
The resistors R9 and R10 balance the amplifier to drive an impedance of about 75 ohms, which is about what video systems typically use. I could go into a whole bunch of technical detail here that would leave your head spinning unless you're the sort who deals in such things. In either case there's no real need to go into impedance matching and all here.
The 100uF cap gives us a little power storage and decoupling for the circuit. Its value is not critical, but I'll go into that in the construction section, where I discuss component substitutions.
This part of the circuit is pretty simple. Each signal is attenuated a bit, and in the case of the 2600 chroma, we are isolating the interference it produces when it's not being used from our glorious 7800 video signal. If you were to jumper out this resistor and start up a 7800 game, you would see some diagonal lines running up and down through your 7800 video image.
Since this is exactly the sort of thing we're working to get rid of here, I included the resistor in the circuit. I'd hate to have you spend a couple of your valuable evenings putting this together only to get a clearer version of a rotten video image.
In the 7800 the EAUDIO signal (or External Audio signal) allows a cartridge to generate its own sound and add this to the sound generated by the 7800 itself. For example, if a game manufacturer decided they wanted to include a Computalker chipset on their cartridge to enhance their software, they could do so and use the EAUDIO line on the cartridge port to bring the output out of your regular TV speaker along with the audio generated in the 7800 itself. The same would go for putting a POKEY or SID chip in the cartridge.
One approach to getting audio out of the 7800 would then be to simply pick up the EAUDIO signal and ignoring the AUDIO signal entirely. The thing is that the EAUDIO signal is seldom used by any cartridges (there are probably a couple out there, I don't know what they are, personally), and the EAUDIO signal would need another amplifier since it's not enough to drive the audio input of a normal TV or monitor by itself (unless you want to turn the volume way up and get blasted through the far wall when you switch to another input on the TV.)
So I take a compromise approach and mix the AUDIO signal, which is more than strong enough to drive an audio input on a typical device, with the EAUDIO signal. This is also the approach used by Jay Tilton earlier, and after going back and forth on whether to just pick up the EAUDIO and amplify it, I finally decided to just do it this way.
If you want to do the amplifier thing, refer to the 2600/7800 FAQ for a method which works fine.
The idea is that we could just find the right point to pick
these signals off inside the 7800 and put them out through a
connector, rather than building our own circuits for D-to-A
conversion, amplification, and mixing of signals.
The problem is that I've tried this, and here's why I ended up
building things as I have:
Why Do All This At All?
One could make a case that this circuitry already exists, pretty
much, inside the 7800 itself, so why go to the trouble of
duplicating it all?
The idea is that we could just find the right point to pick these signals off inside the 7800 and put them out through a connector, rather than building our own circuits for D-to-A conversion, amplification, and mixing of signals.
The problem is that I've tried this, and here's why I ended up building things as I have: