Apple IIe PAL Color Deep Dive and Repair 

You're welcome! I'm glad you found it useful!

Wow, thanks! That means a lot!

Thank you Chema! :-)

Really happy to hear that! 😃👍

Glad you enjoyed it!

Thanks so much! I'm really glad you like it. I also love computers made mostly out of off-the-shelf parts. You know you can always fix them, not matter what happened. It's almost the same thing with early ZX Spectrums and Amstrads (except for the pesky ULA). But once you get to ASICs and things like that... things get more complicated.

Quite ironic that the modern Apple now deliberately use custom chips to perform functionality easily available from off-the-shelf parts, in order to discourage or prevent repair!

Thanks! I'm glad you enjoyed it even if it was unexpected :-)

Glad it helped!

Thanks! It's very satisfying when I manage to fix what I set out to do, which is not always the case :-)

Thanks! Much appreciated.

Hi Roland! Good to see you here. Yeah, I've been doing this as a hobby on top of my normal gamedev. Really fun!

Thanks. Glad you liked it. It took me a while to dig all the way there, although it's one of those things that once you figure out what it is, it seems like it couldn't have been anything else :-)

@@NoelsRetroLab Aaand... I spent most of my time finding out most of the info about PAL composite video on my Apple //e myself... the problem I have is similar, mine shows yellow only, and seems to be in color mode even when it shouldn't (e.g., in text mode). I just started studying how the 6600 works, but looks like you have quite a good description here (and way better schematics than the one I have!) I'll rewatch your video soon, with the board in one hand and the oscilloscope in the other :)

@@marco_foco Make sure you read the PAL section in Understanding the Apple IIe (linked in the description). There are two crucial pages about it which helped me immensely following the logic and tracking this down.

@@marco_foco What you're describing is really weird though. Yellow in text mode? I doubt it's generating the color burst in that mode (easy to check), and if it isn't... that sounds more like a TV issue. Drop by the Discord if you want and we can chat there. Easier to post pictures and stuff like that.

@@NoelsRetroLab That's the problem, or at least one of them: it is generating the color bust in that mode. The color is greenish/yellow in text mode, in graphic mode you can see random pink/green patterns sometimes, and the black isn't really black but often very dark red. As you did in your video I tested with several devices and adapters (a TV, a portable projector, and a screen with a composite to HDMI adapter, exactly like yours, but white :)), they do have some differences, but the problem is always there. I even tried to swap one of the transistors (Q9) that seemed bad in circuit, but revealed good once removed, and the variable capacitor (which broke in two when I tried to move it). Absolutely no change. Unfortunately I didn't have time recently to check it again, I hope I'll have some time on Sunday.

Thank you! Yeah, it's kind of crazy they managed to build a circuit to emulate the results on NTSC. I was blown away when I learned that.

@@NoelsRetroLab Before the integration on the PAL IIe mainboard they essentially made a very similar expansion PAL card for this. The older European Apple II+ was also called "Europlus" back then.

Thanks! I'm glad you enjoyed it :-)

Thanks! Glad you liked it!

Thank you! I'm glad you enjoyed it. I'm really bummed that I haven't been able to get more Apple IIe content, but I've been swamped with other things. I have a few things planned though, so it'll reappear for sure 👍

Thanks! Yes, it's often that way: Lots of searching and then the actual fix is really simple :-)

Thanks!

Thank you. Yes, I agree that in retrospect it was useless. I just just grasping straws in case there was something "slightly" incorrect about the composite signal and somehow the RF modulator would be less sensitive to it than the TV.

Thanks! Glad you found it :-)

Thanks! The plan is to still be around and release videos, just not as many of them.

Thanks! And yeah, good point. I've been meaning to get one of those for a while. I'll bump it up in my priority list.

Thanks! I don't have enough of an instinct developed for the Apple IIe to be able to tell where faults are located, so I just had to take the brute force way and follow the whole color-generating circuit. Which in the end was really interesting and learned a lot, so I'm glad I did it. Good luck with your Apple IIe! It's a fun system!

Merci! That means a lot from an Apple II expert like you! :-)

Haha thanks! I'm glad you enjoyed it in "spite" of being about an Apple computer. Here's the list of some of the games I've created since I went independent: www.snappytouch.com/ The new one isn't announced yet though. I'll probably mention it in the channel whenever we announce it though.

@@NoelsRetroLab Thanks for sharing.

I don't have any experience with Apple IIc, but from what I can read about it, it seems very similar to the IIe. I did also read problems with Apple PAL monitors showing as black and white only, so it seems it's not an unusual problem. If I get my hands on a IIc I'll definitely check it out and make a video from it. These old Apple computers are really neat from an architecture point of view.

Yeah the IIc is nearly identical to a IIe, just has 128k of ram and a built in 80 column card and disk drive and more ports, that’s about it. I could get full clear colour on an AppleColour monitor just using a single composite cable. So the information must be in that signal for colour. Interestingly I read that the AppleColour monitors automatically switched to black and white for 80 column text mode. So perhaps that’s why they only showed monochrome for other video signals?

Haha, thanks! No, I'm a game developer, not a teacher :-)

Really good insight on that. I hadn't thought about it that way. Thanks!

Yes, I'm a game developer. You can read some stuff I wrote about gamedev on my web site gamesfromwithin.com/. I don't know if i'll go as far as making an autobiographic video though. It might be a bit too much navel gazing :-) As for the electronics, I did study Computer Systems Engineering, but I really learned all practical electronics by myself when I started repairing my own computers a few years ago. So it's never too late to get started with that if it's something you're interested in.

Thanks! As you could tell, I wasn't sure if there was a specific ideal setting.

Yes, you're right. I'm in the habit of taking them out without cutting them because many times I'm either not 100% sure the IC is faulty, or it's an interesting chip that I want to keep for future tests. But in this case it's just a dumb 74xx, so yeah, that would have been much better.

Yeah, kind of amazing they managed to pull that off with such a simple circuit actually.

By an RGB card do you mean a generic card that takes composite and generates RGB, or is there an Apple II-specific expansion that generates an RGB signal. I'd love to check that one out.

@@NoelsRetroLab There are RGB cards for the IIe that fit in the aux/80col slot, including one sold by apple itself. They were available with digital or analog RGB signals.

Oh really? That's really intriguing after seeing how convoluted the IIe PAL color was. Now I'm really curious. I'll have to see if I can find one 👍

Interesting! I don't know much about Apple IIe history and details like that. I would imagine the European models could have been different. Also, maybe the board says 82, but the model itself was a later one? Apparently this one was rescued from a big bonfire (!!!!) pile many years ago, so we have no idea of its history.

CGA has actual color information built into it (analog RGB) which can be displayed directly on a CGA monitor or encoded in some other form (composite). The Apple II didn't really have any color information in the signal. It was a monochrome signal with high frequencies and timings in such a way to trigger color artifacts. The Apple II didn't control the palette of possible colors, it's just whatever NTSC produced, whereas CGA did at least make some (bizarre) color choices.

@@NoelsRetroLab That's a little hand-wavy. Also, what mode are you referring to with that explanation? Mode 4 on composite? I was actually referring to the not-super-officially documented but common 160x200 composite mode, which needs bits 4 and 1 set in the 03D8h register. And in that commonly known and not excessively tweaky composite colour mode, the CGA does fundamentally the same thing as the Apple II: It uses 640x200 "monochrome" data ("b/w" by default) to generate a composite colour image via NTSC artifacting. Yet if you look at the 16 colours which composite, 160x200-mode CGA produces by default (w/o further tweaks and 1337 hacks), you'll see that the CGA's palette is very different from the Apple II's default composite colour palette. (No hardware? Just look at the corresponding Wikipedia articles; they both have palette pics.) So that unexplained difference between both systems' respective composite palettes is what I asked about, and if anyone can give a non-hand-wavy answer, I remain very interested in hearing it. It is a given that I do of course know that both 160x200 mode CGA and Apple II use composite artifacting to create their chrominance info out of little more than "thin error" (if you'll forgive the pun).

Everything about monitor television signals is about precise timing. Chrominance information is effectively sent on a the phase of a subcarrier mixed onto the luminescence signal. The angle represents the vector between the Y-R and Y-B for a particular color. Color artifacting works by sending digital signals on the chrominance subcarrier during the duration of a particular pixel.The monitor/television's color decoder interprets these signals as having a particular phase based on some PWM/PDM filtered result of those bits and specific timings. The computer doesn't generate the colors, it just generates bits at randomish times and the monitor interprets those as colors based on it's color gamut. As long as the signals are always generated with the same phase referenced to the color burst signal, a series of bits will always be interpreted as the same color -- but the computer has no idea what color that will be. The Apple ][ and the IBM CGA graphics cards both use a similar method to generate color; but, their timings are completely different. While Apple displayed 140x192 pixels do to other timing constraints on the Apple ][ board, CGA displayed 160x200. The vertical and horizontal intervals were therefore different and different in reference to the colorburst. Thus the timings of when phase the digital signals were sent over the chrominance signal were completely different. Therefore, the palettes of the machines are also completely different.

@@timharig Thanks for the detailed explanation. Your last paragraph resolved one of (retrocomputing) life's little mysteries for me. :)

Thanks! Actually, if you start with the premise that you can't change the video circuitry from NTSC to keep everything compatible, this was remarkably clever. I'm still kind of amazed that it works the way it does! :-)

Yes, that's a good idea. I usually don't do it because I'm never 100% sure that the chip itself is faulty (sometimes it can be shorted from somewhere else) and there's something satisfying about testing them after you desolder them to confirm they're indeed faulty. But yes, that's a much more practical approach, especially with cheap ones like this.

Definitely! That was the first thing I did with this one when I got it a couple of years ago. And it was totally cracked, ready to burst too. The wDrive is awesome. That's what this video was originally going to be about until I ran into the no-color issue. So I'll have a video about the wDrive coming up soon :-)

@@NoelsRetroLab you in the apple 2 group on FB ? few euros are in it.. not loads though haha

I believe so (unless there are many different ones). I posted this video on a couple of them.

That's very true. I'm in the habit of desoldering them without cutting them because half the time I'm desoldering potentially damaged chips or valuable ones that I want to keep even faulty (to run experiments with them). But in this case you're right, I would have been better off snipping it off.

The scope I usually run it at 10x precisely to avoid that and being able to see the signal coming from crystal oscillators. The metal screwdriver... there you got me. I still haven't gotten a plastic one yet! 🤷‍♂️😃

@@NoelsRetroLab Even at 10x the probe still introduces some capacitance which de-tunes the crystal. Normally you want to find a buffered output when tuning while measuring. It's all good I'm just being nit picky. You do a nice job finding the faults. You really need to get a bit nicer scope IMO however. Something with a gradient display and faster update.

@@jeffm2787 Agreed...a crystal should be measured after a buffer. The PAL crystal freq is 4.43361875MHz. The colour burst is ten cycles of subcarrier which changes phase by 180 degrees every line. PAL is a lot more involved than one might expect but Noel did a good job explaining the problem and fixing it.

True! I'm based in Spain at the moment and there don't seem to be any around here. I'm planning to work on some Dragons, which are pretty similar though.

Haha, you had me go back and look at the video to see if it really showed I'm wearing shorts and flip flops (which I probably was) and then I realized what you meant! Well played 👏 👏 👏 🤣

Actually Choplifter was developed for the Apple II, then ported to the other systems.

True. I guess they were one of the first (if not the first) to do it? I'm not sure what the historical sequence was there.

@@NoelsRetroLab Thank you for all the repair videos, always excellent. A nice touch having a diagram inset where you are setting the probes for testing. Thank you again.

Especially for 74 series chips which are easily sourced.

Agreed. I usually don't do it because the desoldering gun does such a great job. I think I stuck with it because I wanted to have the option to go back and test the faulty chip afterwards.

:-b

Right. That whole circuit is only active when the switch is on the color setting.

Seriously! 😕