The display specs are very advanced for an 8-bit machine, even an 8MHz machine.
Using an opaque black chip that says “FPGA/w Video Stuff loaded on to it” vs a black chip with some random numbers on it is pretty moot from a system design perspective.
The primary motivations for these projects is not simply to have “better graphics”, but because the 80’s chips and hardware are more and more difficult to source.
Add in the connectivity to modern displays, where the '80s stuff falls flat, means that if you want any “cheap” adoption, “something must be done”. Even if someone built an FPGA to mimic an 80’s chip that Just So Happens to have a VGA interface violates the “made from 1980’s desgins” tenent.
So, may as well toss the baby out with the pond water and do a modern design for modern equipment that at least provides an 80’s style development aspect (memory mapped, VBLANK and HBLANK), and some blocky sprite capability.
Mind, my preference is for an updated IIGS, a bigger chip, more speed, more RAM, rather than hamstring a slow processor to small memory spaces. Not even for just games, but just for everything else that sucks computing dry today.
I’ve to admit, I am a bit sceptic about the project in general for its emphasis on the C64 heritage. Don’t get me wrong, I had (and still have) a C64 of my own and learned much from this system. However, the C64 was an influential users’ machine for its huge discrepancy in system software and hardware capabilities, which forced users to explore the machine. Making the C64, despite its ROMs, the last of the bare metal machines, or, rather, rendering the ROM routines extensions to the bare metal. This just worked out at a rather specific state of specs and technology. I’m not so sure, if this may still work as well, if you step up the specs or the software. It may just become yet another boring system. (E.g., compare the rather moderate success of the C128 in C128 mode, which was by any means a much better system.) As you begin losing simplicity, there are much better options as well. E.g., the graphics aren’t that great or user friendly, fandom besides, there are systems with better color palettes, the Moog synthesizer like SID is nice, but there’s FM synthesis as well, personally, I liked the PET font, but never liked the fat C64 font that much, which was an inevitable consequence of adjustments for RF video output. All these features are engaging, while there’s still simplicity and while strict constraints on the hardware communicate to the user. If you start adding complexity and abstraction, none of this does make sense anymore. – I think, the scepticism towards using any FPGA emulation may be related to this, because any constraints automatically become somewhat artificial, while on “real hardware” interfaces are more like fate, it is what is.
There’s a tension, I think, between appealing to nostalgia and making an excellent machine. And there’s a tension between the different levels of purist: whether a machine should use only mid-80s parts, whether it’s OK to use microcontrollers or CPLDs or FPGAs. There can’t be any reconciliation between purists, because that’s what it is to be a purist. And since you can’t please everyone, a project has to be able to proceed amid a crowd of displeased would-be customers.
There are a few measures, after the fact: did the people in the project have fun? Did they feel it was worth it? Did anyone lose money? Did any friendships or marriages break up? Did the project bring more joy to the world?
I am still on the fence… but my take is that I would like to build one of the systems… I am not a prurist, but I would like a retro system built on the ethos of retro. By that I mean ‘Built with todays components that fulfill the same design goals as the original’ The system should be able to teach how computers and interfaces work and layed out in the classic BUS type topology.
The Harlequin ZX 128 to me is a good design as its a complete replacement - with compatibility and no special components. We all have to remember that back in the day we bought computers based on how much software was available for it, and I think this is true today. I remember the day I got my ZX Spectrum with only the Horizons cassette… I was very disappointed until I started getting magasines and trading with friends.
That’s my ideal, but we are all different… that is what makes this world wonderful.
Thanks @RichardP - I think saying what we like and why we like it is a good way to keep an interesting conversation going.
I too like to see visibility of components and busses and mechanisms. As it happens, I like to see computers computing, so text-only output is fine for many of my purposes. The RC2014 project and its variants appeals to me for these reasons.
As it happens, I can solder, in a technical sense, but it’s best if I don’t, so that also influences what kind of project would appeal to me as a buyer.
And as it happens, I like the process of designing CPUs, and I like to see results quickly, so there are a number of FPGA and emulation projects which appeal to me to, for different reasons.
After some playing around with the inner interpreter possibilities for a direct threaded 65C02 Forth, and after they did a BIG clean up of the zero page leaving basically 128 bytes of zero page space available (unless you want to use the GEOS Kernel or the graphics keywords that make use of the GEOS Kernel), I’m tilting more toward Go than toward No. “INX: INX: JMP ($0000,X)” as a self-modifying NEXT residing in the zero page is pretty appealing.
But I have no particular interest in getting the kit-buildable “Stage 1” … it’ll be the priciest of the three, and for my interests having the smaller cost reduced surface mount parts, and possibly a CPLD or two replacing a lot of the glue logic, is no loss for a substantial cost saving.
Other than the C64, though, my other 8bit system that I remember the most fondly is the Geneva PX8 that I took with me to the Eastern Caribbean in the mid-1980’s as a Peace Corps volunteer teacher. I was excited to get a C128 when I got back to the States … but very shortly after getting my C128D, I fried its processor by plugging the Datasette power tap for my User Port printer interface upside down, and I don’t know what happened to the Geneva.
What would be really cool would be to adapt one of the modern CP/M boards, either the RC2014 or the ez80 based MakerLisp, into an expansion board card for the CX16 … especially the MakerLisp, as it would finally, and only 40-50 years too late, correct the main flaw in the CP/M side of the C128D design, that it was running at 2MHz when even with wait states it would have been a much more capable Wordstar/Multiplan system at 4MHz or 8MHz.
I’d want that expressly to see what someone could do within the deisgnspace allowed by a version of the arcitecture unfettered by ‘lets hamstring it so it will die on the alter of mac.’
Perhaps someone who knows and loves the IIGS could post a thread explaining the virtues of the machine? (Perhaps, even better, they could also comment on the Apple III, which might be viewed as the other approach to making a bigger better Apple II.)
(Edit: @whartung took up the challenge - I’ve split his post into a new thread, as it’s no longer about the original subject.)