jbailey wrote: ↑
03 Jun 2019, 22:34
500 kits shipped is a big accomplishment itself. However I suspect the bigger one is the ratio of kits assembled and working vs abandoned attempts.
I always hear how a lot of electronic kit projects have a 25-40% failure rate. I suspect yours is exceedingly low with the excellent manual, step by step YouTube videos, and further guidance you both provide to fellow builders on this forum.
Since the Gigatron I have done a few Z80 kits, an Apple-1 kit, Relay computer, and PiPDP-11.... none of their build documentation even comes close! I doubt I would have been able to complete a couple of them if I hadn't started with this project.
Thank you for a fine kit.
PS: I can’t wait for the SPI / SD Card reader accessory to come out!
Thank you for the nice words! It was great fun to develop it all, learn new things, and make something that so many people can enjoy in their own way.
P.S.: About the expander, the devil is in the software. As you know, we've gotten the SPI circuit hammered out
back in March, and we got 128K of banked memory as a bonus. I sent out a few prototype boards to those interested. The hardware was relatively
easy (it wasn't easy at all actually). The real mountain ahead is the programming, and there was no good plan for that other than doing it all in GCL again. However, that felt completely wrong and more like a bad day job instead of an interesting and fun hobby project.
Then suddenly the C compiler arrived on the stage
. To me it was immediately clear that LCC provides the best way forward: we have to mature this C compiler so that memory card and FAT handling can be done with it, or even ported directly from existing code. And we get a proper programming language for those who don't want to bother programming everything in assembly or in esoteric GCL.
While an incredibly impressive achievement, the C compiler isn't fully mature yet. It can do a lot of party tricks, and we have some of the C library implemented. But there are a few bugs (not many) that prevent it from handling quite a bit of existing code. At this point we're a bit stuck at a nasty register allocation issue
. I feel we're close to solving it, but reality is, you need the accompanying book to really understand what LCC is doing and why. I'm still waiting for my book to arrive.
After C, we can go back to do memory card init and block reading/writing over SPI, perhaps simply by using Joerg's sdcard.c code. Then add FAT by using FatFs
. That's my personal ambition for this year. By then we can see if there's interest in an upgrade (or if Gigatron interest has waned in general).
If the above all works out, enthusiasts might want to see if they can port MINIX 2.0 or FUZIX
(Alan Cox's Unix-like OS for 8-bit processors). We have banked memory, so something like this might fit and run...