Epic! 8-bitguy uses 1 weird trick to detroy rare prototypes!

  • 🐕 I am attempting to get the site runnning as fast as possible. If you are experiencing slow page load times, please report it.
Same, in my opinion, with all these solutions that give you modern outputs (HDMI etc.) for old systems. I don't know what's the newest there but I can promise you they all probably work by just reproducing what the original graphics chip did and then piping it to a digital output.
I can speak to the RGBtoHDMI when it comes to compact Macs (a hat for a Pi that’s been brought up ITT) it steals the video signal coming off the logic board and converts it to a digital signal. It’s hardly a mod in the same line as what you described, because the rest of the computer is not vestigial at all. On top of that, the video signal will still make it to the CRT unmolested and you can use the machine as if it were unmodified.

Sure, that means I may have TWO Pis in my Mac now (it also has a pico-driven BlueSCSI) but nothing about any of the mods “obsoletes” it, if that makes any sense.

As for processor upgrades, I know people are recreating accelerator cards of the time like the Carrera040 and such. You could definitely do it all with an FPGA. So what if you did?

I guess what I’m saying is that modding to expand the capability of these old machines makes sense, in the context of helping them run stuff they could back then.
 
Now there's a can of worms. How far do you go before your vintage machine is no longer a vintage machine? I know some enthusiasts are all about "muh purity", but at the other end of the spectrum (no pun intended) there are dudes who throw everything they can at making their 30+ year old machine behave vaguely like a modern PC.

imo there are some computers where modern upgrades are very helpful e.g. a BlueSCSI on a Mac where the original spinning rust no longer spins, and some are even within the spirit of the original machine. The PiTubeDirect for the BBC Micro is a classic example, as the ARM chip is a direct descendant of the original Beeb (there's even a version of Elite specifically written for PiTubeDirect equipped Beebs). I don't find this anywhere nearly as difficult to reconcile as a PiStorm on an Amiga.

Then there are some computers where even the most inoffensive of modern hardware just seems a bit wrong. Why anyone would want to load ZX Spectrum games from an SD card is beyond me; the tape loading noises are part of what makes the Speccy so iconic.

Eh. Using modern things like SD card readers and USB-to-network adapters and stuff is one thing that doesn't bother me. Disk rot is a thing, it's getting hard to find older disks and good quality tapes, tape players are breaking down, etc.

Basically, tape drives and floppy drives are some of the weakest components of old computers. Plus they're often in formats that are really tricky to make readable on modern hardware, so getting stuff on or off them from the internet can be a pain. An SD card just makes it so much less of a pain in the ass.
 
Basically, tape drives and floppy drives are some of the weakest components of old computers. Plus they're often in formats that are really tricky to make readable on modern hardware
Yup. Have fun trying to write to an 800k floppy. Also, CD drives seem to be failing en mass with me, though the CDs themselves still work.
 
  • Agree
Reactions: Never Scored
Gotta say right now, the classic compact mac scene is low energy compared to even the Apple II. For example, a guy reproduced the entire Mac Classic logic board with full files to be used to manufacture new boards on your own. The Classic II, which is superior in every way and basically a SE/30 at 70% power? Fucking nothing, if the clock battery did jihad to the board you get nothing. No one's even done any kind of accelerator for the II or found ways to do cool shit like 640x480 video with either model or get past the 4/10MB memory limit either, best you get are FPU cards that all have varying degrees of tindie jank. Meanwhile in the past five years the AII has gotten VGA video, HDMI video, SD card storage, processor boards using modern RPi type solutions, universal floppy controllers and more I can't even remember right now. Are all the old macs doomed to the LGR "durrdurr windex and Dark Castle" treatment or something, where the fuck is the actual support.
 
Old compact Macs were already a bit expensive even before YouTubers drove up the prices of retro tech, so i'm not surprised about them having very little new hardware getting made compared to other old platforms.
 
Gotta say right now, the classic compact mac scene is low energy compared to even the Apple II. For example, a guy reproduced the entire Mac Classic logic board with full files to be used to manufacture new boards on your own. The Classic II, which is superior in every way and basically a SE/30 at 70% power? Fucking nothing, if the clock battery did jihad to the board you get nothing. No one's even done any kind of accelerator for the II or found ways to do cool shit like 640x480 video with either model or get past the 4/10MB memory limit either, best you get are FPU cards that all have varying degrees of tindie jank. Meanwhile in the past five years the AII has gotten VGA video, HDMI video, SD card storage, processor boards using modern RPi type solutions, universal floppy controllers and more I can't even remember right now. Are all the old macs doomed to the LGR "durrdurr windex and Dark Castle" treatment or something, where the fuck is the actual support.
It’s a scene that’s actively emerging. The Classic II gets the cold shoulder *because* the SE/30 exists. People are actively designing and prototyping replacement boards for other battery-bombed Macs such as the IIci, as well as various other interesting projects.

The main reason why the Apple II scene has more stuff developed for it is due to its inherently easier homebrewing, massive expansion capability, and the fact that the homebrew scene was encouraged from day 1 of its existence. Compact macs, on the other hand, have little to no native expansion capability (another reason the SE/30 is popular, it actually has an expansion slot.)

Desktop 68k machines have nubus and such, and true, I don’t see much happening with those yet. But I think a more fair comparison to the compact macs would be the Apple IIc, which has missed out on 95% of the fun that Apple II modders can have, because of lack of native expansion capabilities.
 
I'd write more about how you can easily build 8 and 16 bit systems time appropiate with cheap off the shelf parts, but this post is long enough.
I'd love to build a Harlequin 128k one day. It's a ZX Spectrum clone board designed to use discrete logic chips in place of the original ULA. Most (if not all) ICs are still in production, and those that aren't are very common. You can even buy a brand new reproduction ZX Spectrum case and keyboard.
Basically, tape drives and floppy drives are some of the weakest components of old computers. Plus they're often in formats that are really tricky to make readable on modern hardware, so getting stuff on or off them from the internet can be a pain. An SD card just makes it so much less of a pain in the ass.
I have no objection to modern storage systems either, and many of my machines either have a modern storage device in them (such as the aforementioned BlueSCSI in my Macs) or I use a tape emulator for machines where tape was the dominant storage media. It's way easier to store thousands of ZX Spectrum games on a TZXDuino than it is in cassette cases (not to mention way cheaper and more reliable).

Tape emulators are also handy when you have a bunch of old machines but can't justify the cost of buying storage devices for each one.
 
  • Like
Reactions: Kalsarikännit
I can speak to the RGBtoHDMI when it comes to compact Macs (a hat for a Pi that’s been brought up ITT) it steals the video signal coming off the logic board and converts it to a digital signal.
yes, there were quite a few systems that produced digital video, it's not really a modern invention. (e.g. CGA/EGA, the C128, etc.) It was always more optimal, analog video was more of a compromise because of technical limitations. The Amiga also has digital RGB output, but only at 4 bit (for CGA Monitor compatibility, for the most basic of technical reasons). It was always trivial to convert such a signal to VGA (as long as you had hardware that could deal with a 15 kHz refresh) with a few passive parts but nowadays actually interpreting them with something like a Pi and some programmable logic is probably even easier on account of how slow they are. I mean, what still has VGA nowadays and in a way that it's compatible to these frequencies? That said, the older amiga graphics chips produce a similar "digital output signal" internally (4 bits per color channel for a range of 12 bit color) that goes to the DAC and you could probably apply the same solution to it too even though it'd need a little more circuity than a 4 bit signal. With AGA/PC VGA replacing the DAC like this would get a lot more complicated. (Side note: The OSSC was mostly advertised for consoles but also works really well for really old VGA graphics cards (ISA vintage) if you have a tolerant monitor, you get much higher picture quality with it than with whatever legacy VGA connector you might find on an LCD) Of course replacing the DAC like this and getting the signal "from the source" will always lead to a better quality signal than the pipeline digital -> analog -> digital.

If I ever set up my ancient early 286 system with age appropriate EGA graphics card, I'd probably go that route too.

I'd love to build a Harlequin 128k one day. It's a ZX Spectrum clone board designed to use discrete logic chips in place of the original ULA. Most (if not all) ICs are still in production, and those that aren't are very common. You can even buy a brand new reproduction ZX Spectrum case and keyboard.
PLAs like used in the Spectrum are basically just "truth tables". In a pinch you can replace them with an EPROM even, although that is not ideal for electrical reasons and most likely will lead to problems because eproms glitch when used like the ULA is used in the Spectrum. (You could work around that but why would you when you're already designing these systems and the parts cost cents now, contrary to smaller EPROMs which are probably not that easy to find anymore?) The speccy is interesting because it was so easy to build, even back then. There were a lot of clones all over the world, especially in the soviet union. I use that audio chip from the parts list as tone generator in my own design.

I designed and built a Z180 system two three (jesus time flies) years ago which was supposed to be a forth machine where every part of the system could be modified on the fly (similar to a lisp machine) then I got sidetracked, first by designing an (period appropriate) graphics output, then by playing with AI all of last year. I'll eventually get back to it (I hope), but I have some other projects in the pipeline first.
 
Last edited:
  • Like
Reactions: Pee Cola
truly a man of culture
Do I see another connoisseur? My very first experience with Forth was on a C64 and I basically knew nothing about programming then. I remember distinctively that I was fascinated by the ability to redefine numbers to the point that I could write a program that would tell me that 3 + 3 = 10. (Which then later made me grasp I could actually redefine math words themselves for any math operation to be sent off to external hardware) Also reverse polish notation which frankly, just makes sense. What's not to like?

Later then, JForth for the Amiga. (site doesn't look like it but is still lovingly maintained, last new entry is from 2023) Forth always vibed with me. It's incredibly simple and efficient, usually it's interactive and you get a very cool REPL (which I think is what people truly love about lisp) but contrary to lisp it's very low level, can run in a handful of bytes and on the right system, gives you very close access to the hardware. I can recommend starting FORTH. It's kind of a dead language for good reason on modern systems but for the retro aficionado who wants to do a bit more with his computers I can only recommend playing around with it. It's usually always easy to find a decent forth package for them.
 
Last edited:
  • Feels
Reactions: Pee Cola
Do I see another connoisseur? My very first experience with Forth was on a C64 and I basically knew nothing about programming then. I remember distinctively that I was fascinated by the ability to redefine numbers to the point that I could write a program that would tell me that 3 + 3 = 10. (Which then later made me grasp I could actually redefine math words themselves for any math operation to be sent off to external hardware) Also reverse polish notation which frankly, just makes sense. What's not to like?

Later then, JForth for the Amiga. (site doesn't look like it but is still lovingly maintained, last new entry is from 2023) Forth always vibed with me. It's incredibly simple and efficient, usually it's interactive and you get a very cool REPL (which I think is what people truly love about lisp) but contrary to lisp it's very low level, can run in a handful of bytes and on the right system, gives you very close access to the hardware. I can recommend starting FORTH. It's kind of a dead language for good reason on modern systems but for the retro aficionado who wants to do a bit more with his computers I can only recommend playing around with it. It's usually always easy to find a decent forth package for them.
My first experience with Forth was actually just me figuring out a final project for my Computer Architecture and Assembly class in college. That class used a fake architecture and assembly language made up by one of the professors (luckily we did have a simulator to actually test programs on) and I wanted to see if I could write a BASIC interpreter in that assembly language. I quickly figured out BASIC would be a bit much for me to do in the couple of weeks I had to work on the project but then I discovered some guys on retrocomputing forums talking about Forth.

Anyway, I threw together an interpreter and implemented a bunch of the common words and was able to hack together a good enough demo to impress the class. It wasn't a big thing but it definitely left a profound impression on me.
 
Relevant to @AmpleApricots and @Fcret's interests:

Jupiter_ACE_(restored).JPG

Think ZX81 but with Forth. It was one of the many early-mid 80s 8-bit machines that sank without trace after about a year or so. Only a few thousand units were made and now they're worth a fortune if you can find one.
 
Old compact Macs were already a bit expensive even before YouTubers drove up the prices of retro tech, so i'm not surprised about them having very little new hardware getting made compared to other old platforms.
A little trick I learned is to not buy computers from thrift stores, ebay, or fb marketplace but rather directly from e-waste centers. They'll typically just charge you by weight.
 
ZX81 but with Forth.
Well Forth is easy on the old memory but 1 kb? Yeesh. That spec sheet isn't impressive, even for the 80s. To be fair though, Forth is kind of an aquired taste, BASIC is a lot quicker to get into so I can imagine that machine was kinda doomed from the start, even if it had better specs. If you look at any "serious" BASIC program from the time though (and yes, they existed. People pretend nowadays everyone who wrote anything of value then was an assembler-whiz, but that just wasn't the case, there are good reasons BASIC was so popular) I'm not sure I'd call BASIC easier to maintain or more legible, especially on those systems. To be even more fair, developing on these systems was painful.

My Z180 has an luxurious 512 kb of RAM, while it of course can only address 64kb at once it has an MMU with which you can do banking. (Max. address space is 1 MB) The Z180 was attractive because it has all the useful/necessary components (serial etc.) already integrated on-die and is also reasonably fast for an 8-bitter (mine is clocked to ~18 Mhz with ways to double that and some instructions use less cycles than on a true Z80) while still being easy to use.

I also googled the Speccy reproduction cases. A transparent one with the Harlequin would look nice!
 
  • Like
Reactions: Pee Cola
Why am I telling you all this? We're now in the world of tomorrow and both the 68060 and the amiga chipsets are trivial, downright primitive and easy to implement in an FPGA. And that's what the FPGA accelerator people did, they just implemented the entire computer in the FPGA and avoided all these complicated speed bumps and timing problems.
Yeah. First, I only now realized I made a slgihtly evolved dupe of Pee Cola's post somehow. Wacky.

But this is what I'm talking about. In some sense this is time-honored: (some) Amiga and PS/2 accelerators have always to some extent been completely separate computers using the host as more or less a passive backplane. Our good (100% alive) friend byuu gave one example of cycle-accurate emulation in software... What do you have when your C64 is an ARM based emulation of a SID talking to an RPi pretending to be the PLA talking to one of those new FPGAish VIC II replacements? Something that hipster collectors will still say is totes more real actual hardware than an Ultimate 64, for example.

Well Forth is easy on the old memory but 1 kb? Yeesh. That spec sheet isn't impressive, even for the 80s. To be fair though, Forth is kind of an aquired taste, BASIC is a lot quicker to get into so I can imagine that machine was kinda doomed from the start, even if it had better specs. If you look at any "serious" BASIC program from the time though (and yes, they existed. People pretend nowadays everyone who wrote anything of value then was an assembler-whiz, but that just wasn't the case, there are good reasons BASIC was so popular) I'm not sure I'd call BASIC easier to maintain or more legible, especially on those systems. To be even more fair, developing on these systems was painful.

My Z180 has an luxurious 512 kb of RAM, while it of course can only address 64kb at once it has an MMU with which you can do banking. (Max. address space is 1 MB) The Z180 was attractive because it has all the useful/necessary components (serial etc.) already integrated on-die and is also reasonably fast for an 8-bitter (mine is clocked to ~18 Mhz with ways to double that and some instructions use less cycles than on a true Z80) while still being easy to use.

I also googled the Speccy reproduction cases. A transparent one with the Harlequin would look nice!
When I looked over that system a while ago my whole thought was 'why did they think home users were going to pick up Forth?'
 
  • Like
Reactions: Pee Cola
What do you have when your C64 is an ARM based emulation of a SID talking to an RPi pretending to be the PLA talking to one of those new FPGAish VIC II replacements
"You will never be a real C64"

Amiga and PS/2 accelerators have always to some extent been completely separate computers using the host as more or less a passive backplane.
The Amiga actually had even IBM compatibles you could stick into it, an 8088 one I have that went into the Trapdoor on the A500 even used the 68k and Amiga chipset in various ways. If you put a faster CPU into these, you also have to put some fast RAM with it, otherwise the fast CPU is pointless because it'd just finish instructions quicker and then wait on slow memory accesses. (These things had no internal cache worth mentioning) Then eventually you also throw a SCSI chip in because hey fast DMA and it just kinda goes from there. You generally still need the chipset, although you wait often for it.

It's still a far cry from just reimplementing the entire computer in an FPGA and just reading electrical signals from the inputs. It's just overkill IMO. I do understand how you can fall down that hole from an engineering standpoint, especially with today's tech, but still. There's no software truly taking advantage of all this anyways. Nothing that's worthwhile to run on the Amiga really needs this. It'd at least be cool if this was a roundabout way of a community to diverge from the "original" path it all took and build a truly open platform with new software but even that isn't really happening, all these FPGA accelerators are proprietary last time I checked, as is AmigaOS, as is a lot of the "new" software. Do we really need another one of those?

A worthwhile and time appropiate expansion for an old chipset Amiga in my opinion is an 68000-68020 at 14 Mhz (so you're still easily synchronous to the chipset) with 1-4 MB FastRAM. That's how that computer should've been originally built. 14 Mhz will be quite an noticable speed boost for old programs too, without being so fast that it's pointless. The original 68k in these was a bit anemic, even in the 80s. The later AGA A1200 had such an 14 Mhz 68020 but they didn't give the processor it's own RAM here either, again because of the cost. 2 MB for the entire system was just entirely too little. I brought my A1200 back to the store and got a 486.

I've off and on worked on an ITX compatible design of an ECS chipset Amiga, supplied via 12V barrel, 14 Mhz CPU with 4 MB FastRAM, proper serial UART chips and an i2c bus, IDE. That's a modest machine with which you can do a lot with the classic software that actually exists for it. One day ima finish that one too. I always got stuck around the decision to either cut out analog video completely or not. Even if you integrate a flicker fixer, nothing really takes analog video anymore, this just gets truer each year. There's a bunch of cheap HDMI chips you could use Denise's output with but it's tricky and these chips are huge and do a ton of things.
 
  • Like
Reactions: Pee Cola
To get the thread about David Murray back on track talking about David Murray:


Featuring an appearance by """"June"""" from Nybbles and Bytes after David realizes he doesn't know what the fuck he's doing (again) in trying to fix an European C128. He also doesn't even try fixing the monitor and tells the guy to send it to Adrian Black (who actually does know what he's doing, sometimes).
 
Featuring an appearance by """"June"""" from Nybbles and Bytes
I went to """her""" channel and was like, "Wait, is this an ugly woman or a tranny. Something isn't right here" .. so of course I come straight to the farms and so far this is the only post mentioning that channel. It does seem like it's a man, right? I mean with a name like "June," seems likely.

One thing that I was surprised about was that there was no disclaimer in this video at all about working on power supplies. Both GamersNexus and Adrian's Digital Basement will always have a standard disclaimer whenever they have a video taking apart a AC->DC power supply (danger high voltage; capacitors can contain a lethal charge even if unplugged, etc.) David has included warnings when working on CRTs and discharging them in the past, so did he just forget? or does he not realize how dangerous it is to work on these even unplugged?
 
I went to """her""" channel and was like, "Wait, is this an ugly woman or a tranny. Something isn't right here" .. so of course I come straight to the farms and so far this is the only post mentioning that channel. It does seem like it's a man, right? I mean with a name like "June," seems likely.

One thing that I was surprised about was that there was no disclaimer in this video at all about working on power supplies. Both GamersNexus and Adrian's Digital Basement will always have a standard disclaimer whenever they have a video taking apart a AC->DC power supply (danger high voltage; capacitors can contain a lethal charge even if unplugged, etc.) David has included warnings when working on CRTs and discharging them in the past, so did he just forget? or does he not realize how dangerous it is to work on these even unplugged?
Half seriously: On the other hand I'm sick as hell of those constant disclaimers and if you're too stupid to realize that working on a CRT or line voltage PSU is dangerous I want you to find out the way we did in the old days.
 
Back