World's First Single Chip Apple II Boots! 

Huh, I just got the strangest craving for a peperony and chease pizza...

@@Otakunopodcast pepperoni* cheese* ; sorry my OCD kicked in haha

Brings back memories of using the HD64180 in a dual processor console for stage lighting control as my final year project at Uni in 1988. Wire wrapped a 6502 VDU board inspired by the bbc micro's design and wrote firmware for character based overlapping / movable windows, connected that to one of the 64180 serial ports, and wrote a multitasking OS so I could have different independent processes to manage the various parts of the console, which had a mouse, sliders, keys, LED display etc. , similar to some of the consoles that Strand electric made at the time. The main board was also wire wrapped with a 64 pin adapter that adapted the small pin pitch to standard 0.1". Really nice CPU and fun project to design.

Thanks!

will the Tandy be the "Extended Basic Version", with all the in/out port of the original machine, is it based around the Motorola CPU Based design ( which was basic why the TRS-80 and the Dragon, where nearly 100% compatible with between each model, with no connection of the companies making them, is there a weblink this new CoCo 2 Project:

@@dh2032 I was inspired to do this when I saw a video on the Just CPC 4 ATX which is a modern Amstrad CPC clone with expansion slots and original chips. I thought what computer could use some mondernization that ran off of mostly off the shelf parts? The Tandy Color Computer came to mind. The CoCo 3 was my first computer, but I decided to do a CoCo 2 instead, because the advancements of the 3 was out of my range. The 63C09 is 100% backwards compatible with the 6809 that the CoCo used. It just has some extra features that can be used by programmers, which some have already programmed for, because there are people who have upgraded their CoCo to use a 63C09 instead. The 63C09 can run at a faster speed too. I have it set up right now to run at 3 different speeds. Original .89 mhz, The CoCo 3 fast mode at 1.79 mhz, and the new 3.58mhz mode. It has expansion slots, which will mean if someone wants to make it a CoCo 3, they'd be able to, not to mention add more modern features to it if someone wishes. It will be capable of 256 banks of 64K, which means it'd be capable of 16MB of RAM. Right now I'm testing with both ECB and CB (The two color BASICS) to get it up and running. Once all the bugs are figured out and fixed, I plan on taking that out and have the OS load from an SDCard straight into RAM. There will be a ROM BIOS so when you turn it on it will start up pretty much like a PC and load the OS from the SDCard. This will also mean that people can create different OS' for it if they wish. I'm omitting the cassette ,bit banger (Serial), and joystick ports because of modern tech replacements. I would like to try to use Wii Nunchuck remotes instead as joysticks, and use a more modern serial interface. It will also have a RTC and use PS/2 Keyboard and mouse. Ultimately would like have a GUI OS made for it as well. Sorry, there is no video or really any links yet, as the project is still in its early stages and it keeps changing. I have a few pictures I have taken, but even since then the design has changed a lot. But I can at least share those if you'd like to see it.: drive.google.com/drive/u/0/folders/1etaFqm7GWb3E364tAR15PYyjaJGTs7lf

TL;DR: I think your project sounds _quite_ achievable. I can see that given my limited electronics experience and lack of background with the CoCo 2. The schematic is NOT complicated. The CoCo 1 and 2 and the Dragon machines are literally implementations of Motorola's reference design for the 6809. When people ask why the Dragon wasn't an exact clone, that's why. It doesn't look like it should be hard to clone the system, honestly, aside from sourcing the video chip. If you have a 6845 and a 6309 … everything else to clone the CoCo should be pretty easy. A programmable logic device can replace _all_ of the glue logic, and you can program one originally made by Atmel in a clone minipro. Doesn't get much cheaper than that for dev gear. an SRAM and ROM later, wire up the analog video stuff, and you should have enough to boot. Bank-switched RAM adds a little bit of complexity … but your glue logic chip should handle that for you. You could build this thing on breadboards if you've got good quality ones handy. From there it just becomes what's still missing (ports?) and where can you reduce your part count? The video chip is going to be the sticking point unless you've got a CoCoVGAs laying around or can borrow a 6845.

So just make the Gerber file and send it in for fab

@@knghtbrd My experience with PLA's is with GALs and using a ROM as a decoder. Nothing complicated. Would love to go that route if I had more knowledge. Right now I'm using the 6847 VDG, found in the CoCo 1 and 2, for the display chip. I have several working ones here and you can still find them out in the wild. The system will be upgradable, as the video portion is on its own card in one of the expansion ports on the mainboard. If someone decided they wanted to make a CoCo 3 compatible video/memory unit for this system, they'd be able to. Someone could even develop something that is completely CoCo 3 compatible and offer more advanced graphics and memory features too, which is part of making this project exapandable and upgradable via expansion "slots". I do plan to have at least 4 of these on the board. I'm also designing it so the cards can bypass my bankswitching method, using their own method and DMA. Would love to see what people could make for it like an advanced sound card, Wifi Card, and perhaps even cards that will allow you install the original bit banger, cassette port, joystick ports, etc, if you wish.

Yup. I originally considered building a circuit to do that, but I didn't want to end up spending time debugging it since I knew I'd be getting rid of composite video anyway. That's why I just used the IIGS VGC circuit. The latest version of the project uses RGBx to produce video (via RP2040). RGBx is a de-serialized copy of SERVID.

MEGA-II also provides the /WINDOW signal which the VGC doesn't use. Turns out, it's a great way to determine when vertical blanking is happening. /SYNC from the Mega does the usual inverse polarity during blanking, but /WINDOW is way easier to detect.

Reverse engineering is way easier than designing from scratch. Everything is there for you, all you need is time. When you design from scratch you have to evaluate and test every detail on paper then hope it all works when you physically make the chip. Then you have to test all the inputs and outputs on the chip.

@@common_c3nts It depends what you’re reverse-engineering. For example, reverse-engineering a PS5 such that it can play original PS5 games is essentially impossible, due to DRM. It would be easier to design a new console from scratch, although that’s not exactly “easy”.

Don't forget the Woz worked at HP so he had example computers to work from. And he did own hobby projects (including a video terminal for TV) before the Apple I. And then he was asked to work on the Breakout arcade game and that wasn't only usefull for Atari... because the he got a good inside look on what Atari was building already before. Wozniak is an example of an extra ordinary gifted talent being lucky enough to be at right time (time of rapid development of ic's and thus be able to get the obsolete/subpar ones relatively cheap to play around with) and place (HP, ATARI and not to forget the Homebrew Computer Club). So you can't say he did it 'from scratch'.

I don't think a cap would quite cut it. There's a circuit involving a couple of passives and transistors that makes a standard PC pot work on the Apple II … but if you're assuming it doesn't have to be fully Apple II peripheral compatible, you might recalculate the cap values and just use a standard modern thumbstick for the joystick. Makes sourcing paddle pots easier too, actually-though I seriously kind of wonder about using weighted spinner connected to a controller and a digital pot for paddles. It's too bad IBM and Apple went with cap recharge timing instead of Tandy's direct A/D solution. Tandy 10k, 100k, 150k, 1M? Tandy doesn't give half a crap.

@@knghtbrd it was a resistor/capacitor A/D charge circuit using the time to charge the circuit based on 150K variable resistor as the paddles/joystics. Putting a ceramic capacitor between the V+ and ground on the bottom of a 100K made it work, the wrong values either less than a full swipe for 0 to 255 or less than a full 255 ( ocasionally only 192 with the wrong capacitor ). I helped him wire up 6 pairs of paddles, he did the soldering.

They didn’t follow up with this and the channel has remained dead. Disappointing.

You're telling me. I've been trying to divide my personal projects into "interesting for me" and "useful things for others" categories, but nearly all of the "useful for others" projects are completely stalled because nearly all the FPGAs have vanished off the face of the Earth for the time being unless you're a multi-billion dollar corp. Another thing that bit me was that nearly all of the galvanic isolation op-amps, current sensors, mosfet drivers, etc. are all gone as well.

and price.. an FPGA on its own is useless without the dev kit surrounding it. The popular DE-10 Nano kit is $200 when it's available. The regular DE-10 is over $450.

Yes, the IIGS's chipset and ROM provide the backward compatibility. Arguably, video modes are a big part of what makes the Apple II the Apple II. But very few of the Soft Switches in the Mega-II are used because they are implemented in the IIGS's chipset.

@@bald_engineer - Do you know which IIGS chipsets specifically? That's the million dollar question. :) As for firmware (ROM) that is something necessary even with the Mega II to function, so nothing unusual there. Interestingly, the 12 ASCII unique characters for various international languages, plus regular ASCII and Mousetext character sets are not in the GS's 128/256K ROM firmware, but the Mega II itself and used by the GS, along with Apple II graphic video modes generated. I'm also curious what functions of the Mega II the GS does use. Classic 8-bit Apple II graphic modes, check. All ASCII character sets and international languages symbols, check. What about 1-bit speaker toggling audio? Does the GS generate that, or the Mega II? I'm really curious, because for decades I just assumed the Mega II handled everything with regards to emulating an Apple IIe. It's a part of the technical inner workings of the GS I glossed over.

@@Apple2gs For the most part the CYA/FPI. Regarding firmware, the big clue is that that Mega-II's ROMENable signal isn't tied to anything. The only way it knows what is going on is by the CYA/FPI shadowing stuff into its memory.

tell me you've never used a IIgs before

I've used a IIgs. It's amazing that a machine released in 1986 can be that bad at running Lemmings.

That's mostly stuff we learned in E.E. school back in the 80's.

I emailed "Team Woz." I got two responses back. I think one was him. But the Mega-II was not "his" design, so even he may not have answers to my questions.

@@bald_engineer Dan Hillman is still around but above the working level I believe (big shot) but I'm not sure if Jay Rickard is around. Can't wait until the next part.

@Lassi Kinnunen 81 It's not dubbed the "Vintage Commodore Federation" for nothing.

I wish we could do that!

I emailed "Team Woz" with a detailed question. I got one response that was thoughtful but not helpful. About 3 hours after that I got one that just said "see if you can contact ." I strongly suspect the 2nd one was Woz.

Gotta be upfront. Your comment had the fastest "troll to laughter" ratio! :) If you're amazed, imagine how I feel! j/k

I assume he's talking about the Mega II - it's basically all the logic from an Apple II integrated into a single chip.

Yup. It's a play on how many sources claimed the Mega-II was a "single chip Apple IIe." (Also the reason for the question mark in the thumb...)

@@bald_engineer A single chip "set" perhaps? It's no different then early PC chip sets that consolidated all the IBM AT's 74xx logic, PIC, timer, and DMA controllers into one big ASIC. Those machines still needed RAM and ROM chips to operate.

I need tungsten to live. TUNGSTEN.

@@bald_engineer They were mostly confusing the Mega-II with the SWM (Super Woz Machine) chip used in Macs, which *was* a full II/e on a chip.

@@Stoney3K Well, no, that isn't the case at all. The SWIM, not SWM, was an update to the IWM which included MFM support. But both were limited to the logic of the original Disk ][ controller (plus extensions.) The Mega-II and Gemini are the only (known) chips that implement Apple IIe logic.

Yup.

It's part of his style. Never change, James! Love the enthusiasm.

R/woosh