The C256 Phoenix Project aka What Commodore 256 could have been like | Hacker News Reader

The C256 Phoenix Project aka What Commodore 256 could have been like

c256foenix.com

108 pointsJerry266 comments

Its been really interesting for me to observe this re-focus on older technology, having been involved in the retro computing scene for the last few decades .. it has grown and grown, to the point where there are now titles being released for machines in the 80's that are nowhere like what our expectations were, back in the day. We have titles on some systems that are simply mind-blowing these days, in comparison.

My favourite 80's system, the Oric-1/Atmos range of computers, has even gotten a few re-designs as well:

The Oric-1/Atmos machines: https://en.wikipedia.org/wiki/Oric

The SuperOric: http://andre.cheramy.net/super/hardsuper.htm

The AtmoStrat: http://forum.defence-force.org/viewtopic.php?t=1692 -and- http://oric.club/atmostrat

The Twilighte expansion: http://orix.oric.org (^^ named in honour of one of the best programmers for the Oric systems that ever lived...)

One wonders whether or not we won't see, in a few short years time, some sort of return to the retro-computing ethos of simpler computers, more easily maintained by their interested users, for more advanced tasks.

I could imagine, for example, building an Internet capable suite of software for Atmos machines equipped with the Twilighte card, and do away with (for example) using my MacBook Pro for email/casual news reading. The idea of moving to a less powerful architecture for the most common daily tasks, as perverse as it sounds, really appeals to me. There's no reason I couldn't use the same machine as a Twitter client, also...

I wonder if there are two reasons:

1. there is a generation coming to age now that never got to know these computers first hand.

2. More and more "computers" are getting locked down, supposedly for our own good. Even FOSS is bitten by this bug.

80’s computers are also much simpler both in hardware and in software. You can actually fix a computer beyond say redoing caps on a motherboard and more importantly you can fix it in a lot of creative ways for example if you have a buggy IO chip you can bypass it, and while you lose some functionality it will work.

The software is also much simpler both in concepts/design and implementation due to the limits of the hardware.

I think this make it much more approachable to many people as a hobby.

I’m old enough to have used everything from a ZX81 onwards. The appeal used to be the sheer excirement of having a relatively simple machine do clever thinga in a way that felt open and which promoted imagination - a bit like Lego, but in code.

They encouraged you to get involved, learn, and have a go.

Modern systems can do a lot more, but the cost of entry in money and especially time is so much higher. They’re more like closed appliances optimised for consuming experiences created by others than playpens for creating your own experiences.

The play and imagination element has largely disappeared from modern computing. Most experiences are trying to sell you something, and if you make something new you’re supposed to sell it - and sell it hard - too. Even FOSS projects are measured mostly by stars and forks and not by fun and enjoyment.

IMO the nostalgia for old hardware can’t recreate the culture that surrounded it. It might be more interesting to design new software that has more of that just-for-fun feel to it - although it’s not obvious what that would look like today.

>It might be more interesting to design new software that has more of that just-for-fun feel to it - although it’s not obvious what that would look like today.

PICO-8: https://www.lexaloffle.com/pico-8.php

Worse, my read is that most people in computing these days fetishize complexity. Why do solve problems the simple way when we can keep adding layers of abstraction?

Your read is correct; they want the world to know how smart they are, but are going about it in the wrong way. As it turns out, those who “fetishize” complexity know deep down that they aren’t smart enough to make those things simple, because that is too hard for them. Hence the computer industry and IT and particular have gone to shit.

Could be both, I think.

> We have titles on some systems that are simply mind-blowing these days, in comparison.

Sounds interesting, could you name any of these?

I sure can! Check out this great collection of games for the Oric-1/Atmos, released in the 21st Century, which nevertheless would have blown several thousand users' minds if they had've been released in the 80's. Titles such as this were never even dreamed of back in the day - yet here we are, in the 21st Century, and the amazing titles just keep coming:

http://www.defence-force.org/index.php?page=games

For a bonus, check out anything published by Twilighte in the Oric archives: https://www.oric.org/index.php?page=software&fille=softwares...

.. he was (RIP, fuck cancer) one of the greatest Oric developers of all times, and the inspiration he set off in the community over 15 years ago has shown no signs of dying out .. in fact, we keep getting stuff on this platform 'as a tribute to Twilighte', 5 years after his passing ..

EDIT: a recent example I thought worth mentioning - Twi's unfinished O-Type project, I link here to a mid-thread demo video:

http://forum.defence-force.org/viewtopic.php?f=20&t=345&hili...

Anyone who was ever an Oric user is bound to watch that video and go "WTF, I had no idea the Oric could do that..." Graphics and Music far beyond what we considered a norm back in the day.

Simply mind-blowing.

Never heard of the Oric before. Very interesting, thanks.

It’s a 1980’s British computer manufactured into the early 1990’s, was fairly popular in France.

Speaking of people making new games for old platforms I just came across an Atari 2600 demake of Unreal Tournament on Reddit https://www.reddit.com/r/gamedev/comments/8q238x/about_to_re...

Unlike the ones you mentioned it’s not mind blowing but it’s still neat and I figured you might be interested in knowing about it since you are interested in these things.

Definitely - thanks for the kind thoughts. This is precisely the sort of thing that motivates me to write assembly code again. ;)

In fact, a couple of the more elite Oric hackers are working on a Wolfenstein-like release for the Oric .. that is going to be amazing when it happens.

I’ve been thinking about this topic a lot, myself. Growing up, I never had access to a c64—we had a pcjr at home and apple 2e at elementary school—but I am, now as an adult, really interested in these types of projects and could easily see myself picking one up.

What intrigued me though is two fold: one, I find myself also remembering my father when I was growing up, he was not interested in computers, they were more or less just a tool to get a job done, but he loved cars. He had a VW beetle, late 50s early 60s model (though they didn’t change much) and he would tear it down, rebuild the engine, do whatever. But I mean honestly, I can’t see much difference between what he was doing then, and what I am doing now (building a home-brew Micro8088). In fact, probably for a lot of the same reasons...

There is something about that era of computing, I can hold a lot what is going on in my head and there are very few abstractions between myself and the machine. The limitations of the architecture inspire creativity and doing things “the wrong way”, but in a way you can gain a more significant understanding of the underlying system. But, part of that is just nostaligia, and I imagine if i could time travel back and have a beer with my dad over his car, he’d probably say something very similar about the lack of fuel injection computers and the simiplicity of a two stroke engine...

And secondly, and this isn’t intended to be a scoff at “millennials”, quite the opppsite actually, I wonder what the current generation will look back on as this level of “things were simpler then”-retrospection. Because I’m sure that my dad never thought that I would look back on the 8088s and z80s of my youth as a manifestation of “simplicity”. So, what am I not seeing about my present?

An actual Commodore 256 was designed, but never built, thank goodness: https://dfarq.homeip.net/commodore-256-the-8-bit-that-never-...

But they did release the CBM 256, with an optional 8088 or Z80 daughterboard: http://www.zimmers.net/cbmpics/cb3s.html

Oh, to be a fly on the wall at all of those meetings of computer manufacturers where they asked the engineers "why can't we run PC software too?"

The C128 came standard with a Z80 as a secondary processor. (Would have been more useful for me if they'd also had an 8" floppy drive so I could have easily gotten stuff to my parents' CP/M machine at work.)

The C128 is a really bizarre beast - it actually "boots" with the Z-80 in control initially, and uses that to check for a CP/M disk in the drive, a C-64 cartridge in the cartridge port, or if you hold down the Commodore key, and uses that to determine whether to go into CP/M mode, C128 mode or C64 mode on startup...

During CP/M operation, some BIOS functionality actually causes a context switch from the Z-80 to the 8502 and back...

The C128 is a really bizarre beast

It really is. It's a fantastic looking machine but the internals are just bonkers.

The reason for the internal Z80 was that Commodore had sold all these CP/M plugin cartridges that actually had a Z80 in them, but they drew too much power and the C128 engineers pretty much unilaterally decided it was cheaper to just build in the second chip than sort out the power issue [0].

The story of how this machine was designed and built is quite hair-raising and the likes of which I doubt we'll ever see again!

[0] https://www.reddit.com/r/c128/comments/682da4/c128_ama_from_...

> An actual Commodore 256 was designed, but never built, thank goodness

I have to disagree with "thank goodness". David's source is undoubtedly Brian Bagnall's "Amiga Years" follow-up to his Commodore book. He leaves out many tantalizing details. The C256, codenamed BMW, was being designed by Dave Haynie with Dave DiOrio on the team. These are really brilliant guys and they were putting together a beautiful machine that was never intended to replace the C64, mind you. It was another product family altogether.

I appreciate the 80's hardware ethic, as well as the 'art by constraint' philosophy. But I do see a place for devices where those constraints are just programmed into into a modernish FPGA - no silly scouring old warehouses for obsolete parts - way less routing, fraction of the size pcb etc, all the 80's angst. The words of a Philistine perhaps, but I do think we can get the constraint and creativity benefits without the hardware headache, and I'm saying this a EE.

I am glad you appreciate the discipline I am trying to put on the design. However, I do have to take into account these facts. This is 2018 and there just so much I can do to emulate this philosophy. So, I will do as much as I can without destroying the practical aspect of things. Also, the idea is to create it in a way, someone could actually build it (solder it) by hand... So I need to take that into account as well...

This is a dupe (sort of) of https://news.ycombinator.com/item?id=17273954

Page is unreadable on a tablet.

Getting quite tired of comments like these. The web was made for computers, not for tablets. If tablets are dumbed down and handicapped to the point that they can no longer partake in the web your beef is with the tablet manufacturer or the browser vendor. Either one or both has all the tools at their disposal to fix these issues. All the webserver does is serve up a bunch of data, the interpretation of that data is up to your client.

Page was unreadable on a desktop computer. To read the actual content, I had to disable Javascript, and disable CSS rendering (Firefox: View->Page Style->No Style). Otherwise, any interaction with that page resulted in the Beach Ball Of Death, as my computer struggled to keep up.

Perhaps it's my own fault though, for not using an 8-way 64-bit core machine with 64G RAM to view a site for a hypothetical machine from the late 80s.

EDIT: It also looks like the main home page is attempting to load every font from webfonts.fonts.com. At the very least, it references 271 external fonts. It also takes 1.8M of HTML to render some 4K of text.

“Unreadable on tablet/mobile” is almost always a misguided attempt to accommodate those platforms, though. Sites made without even thinking about them tend to work just fine.

Yes. I think this is what people are actually aggravated by. Somebody did extra work, however well-intentioned, that resulted in a worse experience than had they done nothing. On Chrome Android, you can "Request desktop site" which changes the User-Agent to look like a desktop.

Sorry (and apologies for a tangent completely unrelated to the OP), but this is a terribly reactionary way to view web technology. Moreover, it ignores what was an endemic problem in web design up until the proliferation of responsive layout.

If CSS had never become a layout language, sure. We could all be browsing on opinionated clients that render everything arbitrarily. But that's not what happened, and front-end people for much of the past 20 years instead wrote opinionated layout. These layouts generally assumed quite a few things, such as a landscape-oriented display and a certain resolution or point density.

So yes, these sites were, as you say, made for computers, not for tablets – quite in opposition to the original, client-agnostic vision of the web.

Modern, responsive web design attempts to bring the layout-oriented web closer to the client-agnostic place it started by assuming fewer things about the client. You're advocating for more assumptions.

Them not implementing scaling correctly means one should have a beef with the hardware vendor? really?

It also scales poorly on my 4K display in landscape mode and it’s a unreadable as it is on my phone on my portrait 1440p computer displays.

Who should I blame?

The web was made on a NeXT workstation for any device, on any operating system. Sir Tim Berners-Lee would likely be extremely grumpy if he read your comment.

Phones are computers these days, so are tablets. I'm not aware of any mechanical interpreters for web pages but I'm open to being surprised.

To be fair, it's just not a very good Web site. There's some 3D stuff in the background making it very slow, too.

I usually don't comment like this, because it's noise. I just downvote and move on. But if no-one tells the author, they'll never know.

A very similar project is nearing completion for the arch rival (in the UK anyway) ZX Spectrum.

https://www.specnext.com/

Uses FPGA to ensure backwards hardware compatibility, while providing extra clock speed and new video modes.

I grew up with the ZX Spectrum, initially starting out with the ZX80 and 81 which I soldered together myself with the help of my equally clueless father.

One of the most enduring and engaging aspects of the old micros is the inherent limitation of the hardware. You're constantly battling with extreme optimisation and utilising often quite hacky hardware tricks to do things that would otherwise be impossible to achieve while mastering the arcane arts of assembler. I think above all that's the beauty of them and why people are still supporting these projects.

I'm very sad about missing out on the Spectrum Next which is nearly due to ship their completed cased model after a successful release of only the board. Complete with keyboard and design from the recently late original designer of those same machines I had in the 80s, Rick Dickinson.

is this guy aware of the Mega65 Project ( http://mega65.org/ ) which actually already has working machines, a case and is on the brink to begin mass production?

Yes, I am actually a girl! ;o)

Yes, I am aware of the Mega65.

Like I wrote before. 2 different computer with 2 different approaches...

I don't know why people come back with this comment... Like one 8bit music creator will stop creating music because someone else is doing it too? What's up with that?

Great to see a follow-up to our Gigatron computer!

I wonder what you will use for video generation in the C256, because the greatness in 8-bit gaming came from these chips. So you'ld have to differentiate there or run the danger of becoming yet another retro knock-off (well, the merits of compatibility should not be underestimated. Most of the effort will be in software otherwise). For the Gigatron it was ok to have limited video effects, because its insanely low chip count was its primary objective: so we have no tiling, no hardware sprites and no color indirection. But we do have indirection per scanline and you can do neat things with that, such as the bending road in the Racer game.

One thought is to design your own video circuit and put it in an FPGA. ASIC is also an option if you go with a 0.18um/0.25um foundry (X-FAB). You can also do it with discrete logic, but it will become big. There is the Vulcan-74 guy doing TTL (well, CMOS) video circuits: http://forum.6502.org/viewtopic.php?f=4&t=3329 and http://sleepingelephant.com/ipw-web/bulletin/bb/viewtopic.ph... (video: https://youtu.be/EIghk1BUG98)

Another idea is to do a hardware version of the PICO-8 platform: https://www.lexaloffle.com/pico-8.php

For sound you should really consider the awesome SAA1099 chip.

>For sound you should really consider the awesome SAA1099 chip.

Yea !! As far as I know only the SAM Coupé used it, and nobody really pushed it on that platform.

> I don't know why people come back with this comment... Like one 8bit music creator will stop creating music because someone else is doing it too? What's up with that?

Yea, I have a similar project, I've heard this so many times...

Girl.

Also, two other Commodore remakes/redos, C64DTV[0] and C-One[1] were done by non-guy Jeri Ellsworth[2]

[0] https://en.wikipedia.org/wiki/C64_Direct-to-TV

[1] https://en.wikipedia.org/wiki/C-One

[2] https://en.wikipedia.org/wiki/Jeri_Ellsworth

> this guy

girl, actually.

Always bothers me how common it is for people to gender everything male.

English doesn't have a third-person gender-neutral pronoun and traditionally uses "he" when the gender is unknown, so it isn't really surprising that people, at least English speaking people, assume "male" until they know otherwise.

"They" and "Their" are perfectly acceptable as gender neutral pronouns. A stickler might object, but also bear in mind that in English there's no "proper" second person plural ...

You’re describing a historical norm, but it’s one I feel is rooted in a historical bias. Men ran the world and defined the word “man” to mean a person of any gender, but that was always an act by men. I’d like to see this kind of bias removed from our language, and it takes individual choice to make that so.

It's an easy trap to fall into. I'm looking around my office here and I can count about 20 men and 5 women. When I was in university the ratio far more male dominated. It's unfortunate that such an imbalance can make things uncomfortable for women (e.g. oafishly gendering things as male) in the industry sometimes.

Maybe because they don't pay attention! ;)

The site is kinda sparse on details so far - Wondering if the 256 is going to be along the lines of the initial C-One concept where it would have something like a Monster SID (multiples of 3 voices) and a Super VIC (massive resolutions, copper list, etc.)

These were the Specs Jeri Ellsworth had hoped for initially: http://c64upgra.de/c-one/s_specs.htm

Unfortunately costs/manufacturing decisions curtailed much of that.

I don't believe the 256 will be like a C-one. Graphic Controller wise, it ought to be closer to an A500 than anything else. For the sound, I too would like to come up with something that would be more than just 3 voices. A 6 Voices SID Per Channel, like a 6586 or 8586 would be great!

As far as know, the C-one was abandoned or never completely finished... Was it?

Otherwise, I am sorry if the information is sparse right now, the whole SnapEDA interview caught me off guard. I wasn't quite ready to deal with that kind of attention. Not now, anyway! Cheers, Stefany

These are very exiting times! Old machines and old custom chips can be re-imagined on FPGA (and then custom ASIC given enough demand) if NOS isn't available. Seeing artists work within the constraints of these 8-bit games to make new games that push the limits of the technology beyond what anyone thought possible (like moving video).

How about building a machine to run TempleOS natively? The crazy guy who wrote it says explicitly that the OS was meant to implement the commodore 64 ideal of one big flat address space and no memory protection, everything is rung 0, no protected io, etc.

I'm glad to know I'm not the only one with this engineer's nostalgia for the kind of systems I first learned everything on. When Terry compares TempleOS's design inspirations to the C64, I know exactly where he's coming from. I've thought once or twice about writing a 64-bit DOS in Rust but just couldn't justify the time and trouble.

I thought about doing the same thing with AmigaDOS

Buy yourself a Raspberry Pi and install RISC OS (Full or Pico version, depending on whether you want a GUI or not). It runs BBC BASIC, so it can run old school type ins and has an inline assembler and modern hires graphics.

Doesn't TempleOS already run natively on a rather bare x86-64 PC?

Is making a good sid chip clone something within reach of hobbyists today?

There are issues with the analogue aspects of the SID.

No problem with the digital, but the oscillators, filters etc rely on imperfections of the silicon manufacturing process back in the day. I'm given to understanding that they weren't so much designed as were just "there" and exploited to make sounds.

This is evident in how different generations of SID sound slightly different.

With a FPGA: http://www.fpgasid.de/

How about with the actual analog circuits :)

Probably not I guess, if it even depended on the manufactoring process

The Super Nintendo was probably the closest thing to a C256.

No head, no tail... what are the specifications?

https://www.c256foenix.com/forum/the-specifications/early-sp...

You will find them there... Cheers Stef

I was really tempted to buy that C64 (C64 mini?) remake the other day, eventually decided not to buy it but this is very intriguing.

After reading http://blog.snapeda.com/2018/06/06/building-the-commodore-co... I doubt pretty much about this. Besides, we had the C-one which IS a successor and more...

I love your enthusiasm! C'mon! let's stay positive! The worst that can happen is that I will fail miserably and I will be forgotten immediately and then people will rejoice around their Mega65! cheers! Stefany

Reply on news.ycombinator.com