Hi

I just started building an Apple 1. The boards available on ebay looked really nice, so I just went for it..

Obtaining components is the real pain (obviously) and I want to keep them as close to the original as possible. That is even with old date codes. Obviously ebay is a nice source, but still it looks like its going to take years.

I was thinking of building it with "newer" components (still TTL/NMOS) and upgrade as I see things working. So its mostly going to be plugin sockets.

Anyone else here done a similar build and have some good advice?

There's this:

https://www.unicornelectronics.com/Apple1/Apple1kit.html

No idea if this is still available; I found the page through Google, after remembering that Unicorn Electronics offered a kit of parts for an Apple I build some years back.

Ok, thanks for the link. I already have the memory so I am covered there.

I have looked through their inventory and they do have a lot of 7400 TTL in stock. No date codes though.. but its a good place to start.

The Signetics prom's seems to be the hardest to find.

Finally managed to get hold of a 6502 from 1977. Not too expensive, but its in a black plastic packaging. Those white 6502 were too expensive. I hope it still works..

When was the ROR bug fixed?

Supposedly in mid-'76. And supposedly there was also a "die shrink" at the same time, which might reasonably have heralded a change from CDIP to PDIP packaging.

Software written in '77 - notably M$ BASIC - was sometimes provided in two distinct versions, to accommodate people who had buggy 6502s. The version for the KIM-1 was built with such workarounds. Versions for later micros were not.

(Do you have a reference for that die shrink? I'd love to know more.)

As for ROR workarounds, this little story is interesting:
https://www.pagetable.com/?p=45

It was just a random fact I spotted in passing while researching methods to distinguish different 6502s.

** EDIT **

Sorry, the following was for an Apple II rev board. Not Apple I.

Someone feel free to delete.

But here it was originally:


FYI, I'm not sure if this is OT, but it looks like some of those replica boards can be bought here too:

https://www.reactivemicro.com/2019/05/2 ... therboard/

In fact, it seems to have an interesting story behind it.

_________________
Cat; the other white meat.

LOL!

My first Apple II was a clone ... back in 1980. I built the thing without sockets and it worked 1st time!!! One of the only times that has ever happened.

It's interesting to see they can sell a clone PCB board for $40 now, but it cost $120 back then.

It's also interesting to see that this board, more than twice the size of an Apple I clone bare PCB, sells for 1/10 of the price!

Market dynamics I guess.

I'm going to have to buy one of these boards and build it ... without sockets ... just to see if I can repeat my previous triumph.

_________________
Bill

Just to follow up:

I have managed to get sockets and most of the ICs. A few still remains, and passives as well.

Anyway, I recently found out that the MK4096 DRAM is a troublesome component with up to 50% fail rate. For that reason I had to build a MK4096 memory tester (as I couldn’t find any). The 4096 was the first address multiplexed DRAM with CAS and RAS pins. It might be the reason for its strickt timing spec which meant that I had to get a really fast MPU to test it. My choice was the STM32 with the Nucleo board (F401 version) that had an arduino-like shield port and ran at 84MHz. I was able to use the arduino IDE with STM32duino library and access the IO ports directly using assembly code.

A couple of weeks work, and I had all pins connected and a 12ns IO control accuracy that was enough to get the CAS low to RAS low strobe within the needed +/- 25ns. All-in-all, 10 million random write/read cycles take around 10 seconds, and about 35 seconds for nearby flip-bit checking.

The 18 memory chips I bought for 175USD all passed the tests, which was a reward in itself.


The proto MK4096 tester with 12ns IO resolution. Probably useable to test alot of other RAM configurations.

Rather impressive side project!

I hope to draw together a schematic with an STM32 F410RB (cheaper than the F401RE I used).. when I get the time. Maybe add a second ZIF so I can also test the 2114 SRAM I often use.

A while back I tried to use an ATMega2560 to test 6502 and 6522, but I never got it up to the required speed.. (NMOS).