04/11/07: Thoughts on a CoCo 4 by boisy, | Category: VR CoCo 4 Project | 34 comments - (Comments are closed)

Thoughts on a CoCo 4

I jotted down this mini-essay just after this year’s fest. Hopefully others can comment and add to it.

The idea of a CoCo 4 has been around for over almost two decades. Attempts have been made to create systems and dub them the worthy successor of Tandy’s line of Color Computers.

Fast forward to today: emulation and the virtual creation of older computers has reached a point in modern computing that the design of such virtual vintage systems raises confusion and deep questions. At what point does a CoCo (or any other vintage computer) become just a PC with an emulator? What truly makes a computer unique and special with a sense of identity?

[b:4a28514c0b]What Makes A CoCo?[/b:4a28514c0b]
Ask any Color Computer user this question, and a variety of responses can be expected, but all would agree that what makes a CoCo a CoCo is a combination of several factors:
the exterior: the case, keyboard, ports and general layout
the interior: the motherboard and all of the chips/parts that make it up
The unique combination of common integrated circuits, along with special chips like the GIME and the stylized case make a Color Computer what it is. Years of using this oddly shaped computer known as the CoCo has created a bond that endears us to this hunk of plastic and metal.

So if there is a soul to a CoCo, it is embodied by its design, its look and its feel.

[b:4a28514c0b]Can There Be A CoCo 4?[/b:4a28514c0b]
In name, there will probably never be a CoCo 4. The Color Computer was a creation of Tandy and sold through Radio Shack Corporation. The relationship between the CoCo and its progenitor is a connection that is embedded in the brains of all Color Computer owners. It is historically significant and a part of the CoCo’s legacy that cannot be changed.

If Tandy were to somehow miraculously decide to create a new CoCo, then it would be fair to consider that the CoCo 4 could exist. However, the likelihood of that happening is very close to nil.

In spirit, there can be something akin to a CoCo 4. The idea of a community coming together to design what they wish they had in a successive CoCo, but never got the chance to make, could be the foundation for a new and unique computing system. This new ideal could be a computer which would retain compatibility with the CoCo 3, yet gently but firmly push the envelope in ways that would enhance the user’s experience and still keep it a CoCo experience.

[b:4a28514c0b]Does a PC emulating a CoCo make the PC a CoCo?[/b:4a28514c0b]
No. A PC emulating a CoCo does not make that PC a CoCo, no more than a pauper pretending to be a rich man makes him rich. The CoCo exists in reality as a product born solely of Tandy. Emulating something that already exists is simply that: emulating it.

[b:4a28514c0b]What About Emulating a CoCo successor?[/b:4a28514c0b]
The answer is not so cut and dried here. Much like the GIME made the CoCo 3 what it was, could an x86 based chip running a low level operating system and the MESS emulator become the heart and soul of a so-called CoCo 4?
Emulation is emulation only if it emulates something that exists. The CoCo 3 exists, so hence it can be emulated. The CoCo 4 does not exist; hence, it is reasonable to conclude that a PC configured to be a CoCo 4 could indeed be thought of as a real CoCo 4.

It takes some time to get used to this way of thinking, but if you abstract out the fact that an Intel (or AMD, or whatever) processor is under the hood, and think of that as the fabric of the whole computer, just like the GIME or the 6809 is the fabric of the CoCo 3, then it is easier to picture. The computer is the sum of is parts, not just the individual parts themselves.

An x86 emulating a Hitachi 6309 is, from a purely physical point of view, no different than a real 6309. Both are made of silicon, both have conductive paths and transistors. The difference is the make up and organization of each chip. Programming an x86 to act like a Hitachi 6309 could be akin to a PAL being programmed with specific equations for a specific application. Even the Hitachi 6309 is a microcoded microprocessor, meaning that it has some internal programming to maintain flexibility.

We still think of it as a 6309.

It is easy to become hung-up on identities of specific chips, and no doubt, the lack of love for the Intel architecture may keep many CoCo users from seeing the abstract beauty of this chip as a basis for a CoCo 4. However, there is no denying that the scaling up of power in modern computer processors has provided ample room for them to morph into whatever creative minds can comprehend.

This type of power and flexibility wasn’t at everyone’s fingertips 20 years ago when the CoCo 3 was introduced. Today, this power is everywhere, and it opens the doors for all sorts of creations.

[b:4a28514c0b]Is Emulation An Artificial Experience?[/b:4a28514c0b]
As pointed out earlier, emulation can be found at the hardware level and has been around in some form for a long time. How close one is to the levels of abstraction that make up a product will change the view that that person has to what is being considered. For the engineers of the CoCo 3, they probably did not see a lovable, easy to use computer that would win the hearts of thousands, nearly as much as they saw a collection of ICs, capacitors, resistors, and other parts.

In much the same way, the builders of a CoCo 4 will see the creation of such a computer as a collection of hardware and software which will come together to create a unique system. Whether or not this new machine will have the same appeal as the Color Computer is to be seen, but it is not likely. That’s because the CoCo 4, as an emulated system, could conceivably be run on any PC with minimum hardware requirements. Gone is the small hardshell plastic case and unique keyboard. There is no unique external identity which marks a computer as a CoCo 4.

And that is probably the sense of loss that any CoCo 4 would experience at the hands of its user. The idea of a CoCo in the sense of what Tandy gave us is probably gone forever. That idea is replaced by a newer, more modern version of what a CoCo successor can be: a mostly software invention that pulls together the collaboration of many people to make it what it is. That is truly the empowerment of technology that was unattainable in the days of the original Color Computer.[b:4a28514c0b][/b:4a28514c0b][b:4a28514c0b][/b:4a28514c0b][b:4a28514c0b][/b:4a28514c0b]

34 comments to Thoughts on a CoCo 4

  • RobertGault

    I’d like to add to Boisy’s thoughts a few things that I think were missed.

    From my point of view, a critical distinction between the Coco models and most of their contemporaries was the relationship with the user. Most other computers whether home built or not were/are used the way a calculator is used. You get what you pay for and use the device as it was designed. That is not how the Coco is used.

    The Coco makes a perfect platform for user hardware hacking, system code modification, and original code generation at a level that can’t be matched any other way unless you have the equivalent of a degree in electrical engineering. We know exactly what each chip in the machine does, even for the most part for the mysterious custom GIME. We know exactly how each byte of system code works with the hardware. And we have a machine that, with the exception of the GIME and perhaps the 6309, still has all parts available from electronic parts suppliers.
    We have had dedicated magazines publishing hardware and software projects. And we still have many many web sites dedicated to the same subject.

    So in this context, it is not too hard to imagine what a Coco4 would be. There are even some hints from ex-Tandy staff on a proposed unit.
    I think it is a good bet to say, the Coco4 would look much like a Coco3, would used the 6809 but perhaps with a faster CPU clock, would include a 256 color mode, and probably would have better I/O in terms of sound, RS-232, etc.
    This would have been the only commercially practical design Tandy would have sold that could still be called a Coco4. Anything else that actually was real such as the MM1 (non-Tandy) might have had the flavor of the Coco but clearly was not.

    Can the Coco community create a Coco4? I don’t think it is likely. Whatever might be sold would need to be adopted by all remaining Cocoists and supported with the same enthusiasm for publishing programs as happened with the originals.
    There is no evidence of this happening with any of the hardware not sold by Tandy even though Cloud-9 and several other third party sources have made an impact. I’d say the only item that came close in this regard is NitrOS-9 which I’d bet has been adopted by all serious users of OS-9. However, OS-9 never made the impact on the Coco community that Disk Basic or straight ml programs had. If not for several blockbuster games in OS-9, I doubt that OS-9 would have had any significant Coco penetration.
    Could any Coco4 be designed which could compete with the current crop of computers and be a Coco in anything other than name? I don’t see how? Would anyone buy a computer that could not compete with a current PC for any reason other than nostalgia? I think not, and how far can nostalgia take you?

    The Coco community would be much better served by users contributing ideas to Mary’s magazine at the level The Rainbow engendered. There needs to be interaction between all in our group even if remotely or we won’t be any different than a large “swap and shop” HAM event. Even the current NitrOS-9 project (which I contribute to) will not be more than a personal project of Boisy’s unless users start demonstrating what they are doing with the software.

  • Randy

    I think a coco 4 would have used at least a 6309, very possibly a 68xxx, better graphics, ega/vga display, a ‘real’ serial port, 1 Meg standard, at least some backwards compatiblity with the coco 1-2-3. It would have been roughly comparable to other computers of the early ’90s. I have had a coco 2 and 3 for a long time. While I have never really been in touch with other coco users, I think os-9 is just as much a part of a ‘coco’ as Basic and Coco-Dos.

  • RiJoRi

    I have a PIM, RedBox Organizer. Its tag line is “Other PIMs are used. RedBox is loved.” I think this is the CoCo’s charm: “Other computers are used. The CoCo is loved.”

    Part of the love is because it is user-modifiable, from the first memory expansion to plug-in cards. Another part is that the average user can understand the hardware — no need to worry about cache memory, etc. Finally, there is a cameraderie between the users, which is rare these days.

    As to a CoCo4, we should ask, “What do I use my CoCo for? Games? Letter writing? Record-keeping? Nostalgia?” and “Could an improved CoCo make my life better?”

    The things I’d like to see in an improved CoCo are: (1) a plug for a PC-style keypad, (2) a plug for a PC-style mouse, (3) a built-in real time clock, (4) VGA output, (5) a real serial port (or two!), (6) a parallel I/O port (or three!), (7) a USB port. Of course, firmware for all these features would need to be supplied with the CoCo. And as to firmware, I’d like to have an AUTOEXEC.BAT type of file — one that is automatically run at startup. Hmmm– How about auto-running a file named “DOS”, to autostart OS-9???

    (1), (2) and (4) are there simply to take advantage of the cheap PC hardware. Using interrupts for more things would be nice, because the “interruption” should speed things up somewhat. Speaking of speed, how about using DMA to write to / read from the drives? The less the 6809 has to do, the faster it will appear.

    Enough for now!!!
    –Rich

  • RobertGault

    RiJoRi wrote:
    “And as to firmware, I’d like to have an AUTOEXEC.BAT type of file — one that is automatically run at startup. Hmmm– How about auto-running a file named “DOS”, to autostart OS-9???”

    This already possible if you have RGB-DOS or HDB-DOS. With either, if a program named AUTOEXEC.BAS exists on drive0 of the Disk Basic portion of the hard drive, then it automatically runs when the system is turned on. You can use that program to boot into OS-9 or NitrOS-9 from the hard drive or present a menu.

  • MSBB

    I’d say th63xx and the 68xx might under go one or two more logical upgrades,(keeping the following:

    to make a 6809 compatible chip. When they finished the design,
    they found there was a lot of unused space in the chip. With this in mind
    they added extra registers and expanded on the instruction set, but due to
    the licensing agreement with Motorola, they were unable to release the
    information about the extra features.

    Not only does the chip have an expanded instruction set, but it also
    has a native mode that will run many of the instructions in fewer clock
    cycles and a mode select for the FIRQ (Fast Interrupt ReQuest) that will
    enable it to opperate the same as the IRQ.

    In fact, all new instructions will execute in emulation mode, which
    was originally seen when ‘illegal’ 6809 instructions produced odd effects
    when run on a computer with a 6309 installed

    ) but trying to keep the essence of the CoCo still in thare, aslo enhance the ECB, as well as a biultin edtasm. Oh almost forgot the sound could use improvements.

  • RetroRick

    Sound – both music and sound effects can play in the “background” rather than stopping all other functions dead in their tracks. This feature can be turned on and off by writing a 2-bit value to a certain registry.

    Graphics – I’d like us to have four screens of hi-res graphics available rather than just one. It swayed my interest in CoCo 3 graphics programming when I found out the HSCREEN modes wouldn’t let me screen-swap.

  • courtesi

    “back then”…

    I would have suggested the CoCo 4 as:

    - dual 6309 (be the first mass-produced low cost computer with dual cpu’s). this would at minimum given a total of 8mhz of processing power to take on the amiga / atari st. also this computer could be sold to schools to learn multi-processor programming in an affordable package.

    - a custom ensoniq synthesizer processor derived from the apple gs ensoniq chip. “custom” meaning its configured to be in the same “spot” as the 6 bit dac, but also allowed to breathe with enhanced abilities. another possibility would be to license the sound chip from the nec turbografx / pc-engine. less sonic quality, but mass produced and makes stuff cheaper. also with the proper programming can sound fantastic.

    - graphics, obviously compatible with the 1, 2, 3, but not that extensible. In other words, a slightly larger resolution, with more colors (32 on screen at one time out of a palette of 255.) The goal is to keep the memory bandwidth down. For advancement however, go forward with plenty of hardware sprites that will take all the big IO off the 6309′s.

    - disk drive, high density 5 1/4′s. compatible with the older stuff, plus plenty of room for the future.

    - switch to “atari” / “commodore” controller pin outs to expand joystick/mouse possibilities.

    - operating system: RSDOS in ROM, OS9 optional. (keeps price down)

    - rgb, composite, stereo output

  • MSBB

    Looks like Motorla ether splite or sold thier chips :? thier new name is freescale, and it looks like thier were several upgrades the 68XXX family compared to the 68XX so far no 09′s, so far I’ve not found any 63XXX yet :) , some of the chips are pretty impressive, I keep looking through thier vast inventory.

  • YukonKid

    Well for me it was the unravelling of the Motorola machine language, as well as Intel, as I had an intel powered machine as well.. not PC..
    I would only think that a Coco 4 could evolve with the 68xxx chips.
    I do have a KIX30 which I thought was a step in the right direction. The only thing missing was a way to play the RS-DOS programs. OS9-68K was a nice step up from level II. But alas KIX30 did not catch on.
    Coco sold me on the nice things about Motorola programing and I turned into an Apple user because of it.

    So for me a Coco 4 would have to be backward compatible for software I have in the Coco 1, 2 and 3 as well as use the new hardware in the marketplace.

    Could this be done in software, sure it could because the emulation of the older coco’s would be easier than in hardware.
    Booting up to a Cursor in living colour waiting for an input command would be a blast. lifting some of the limitations such as mem and sound and cpu speed, while giving access to new hardware would be welcome from where I sit.
    poking a screen to get upto a 1280 x 1024 would be easy on the eyes.
    but nothing is as fun as watching the dancing Devil.

    Coco 4 would have to bring the same feel as the previous Coco’s no matter what hardware it used.
    Coco 1 and Coco 3 used the same family of chips but coco 3 did more things than the coco 1, this is how I would see Coco 4, enhancing the Coco 3.

    Just some thoughts.

  • RetroRick

    Here’s my belated response on one of the first questions in this thread:

    Would anyone buy a computer that could not compete with a current PC for any reason other than nostalgia?

    If there’s the software for it, and the machine’s affordable, that’d be a definite “yes”, actually. Good software and economical price helped the PC become more popular than the CoCo, and I’ve always thought (and still think) the PC isn’t anywhere near as good a computer as any of the CoCos were.

  • MSBB

    [url]http://www.freescale.com/files/shared/doc/package_info/98ASS23461W.pdf[/url]

    8) this is the chip MC68EC060RC50 OS-09 level II or Nitr09 theoretically those should work, low cost as well. By my calculations the proto type should cost about hmm.. $600.00 +-, don’t wary about coufing the money I’m doing it my self :D , been wanting to do it any ways for a very long time.

    Well since I’ve ran across this [url]http://www.microware.com/products/Microware%20OS-9.cfm?CFID=1024000&CFTOKEN=59073941[/url] :? I guess I’ll bag my project.Nha, I still do the project :o .

  • rescot

    To be a “coco” you have to have a heart, a 6809 compatible heart. IniCore sells a 6809 compatible inicpu that runs at 40Mhz…. check it out at
    http://inicore.com/core_lib/inicpu.htm

    Michel Julien

  • jdiffend

    I’ve let my feelings on this be known elsewhere but I have a couple comments I obviously need to repeat.

    Emulation *isn’t* a coco and if you are going to base the technology on an Intel CPU it’s a lot cheaper to buy a PC motherboard than to produce any custom board. You might as well just use MESS. Isn’t that what a lot of us are already doing? (BTW, I’m building a PC with a 3400+ that will cost under $200 complete with X1300, 512MB RAM and HD. It could have been $65 for motherboard+CPU+Graphics if I used a board based on the new ATi integrated chip)

    If you want a single chip system with an upgraded CPU other than a variant of the 6809/6309… then the only real alternative is ARM. You can buy FPGA’s with ARM CPU’s built in. There are even ARM CPUs the size of a fingernail that run at 60MHz or more.

    I don’t see the 68000 as a good option. The only models with any real speed are Coldfire chips. They are *not* 680×0 chips… the instruction set is not 100% compatible. One of the biggest problems with these chips is there is no CPU based on them, only micro-controllers. This may or may not be a problem but it definitely impacts the design. I’d rather have just a CPU and put hardware I want in an FPGA. Speed is competitive to ARM but you end up with a more expensive design do to a larger board, additional chips, more complex logic, more power, etc… so that is definitely a drawback. I love the 68000 and have even had to deal with the coldfire a little on my current job… but the only way I see anyone ever selling a hardware Coco cheap enough is use an FPGA/ASIC + RAM + FLASH.

    Licensing a 6809 CORE will cost too much to be practical… so using the free core as a starter may be the only option there. If you think I’m wrong you are welcome to check pricing.

  • MSBB

    @ rescot looks fairly interesting, but probably a cracked product, but as to licencing the chip, currently no I’m building a prototype, only in mass production would some licence it, on a diffent note I’m making sinificant progress, unfortunatly the MC68060 is no more nor any replacements, but the MC68000/MC68000/MC68008/MC68010/MC68HC000/MC68HC001/MC68EC000 are slowly become no, but thare are replacements the MC9S08GB60 which carrys on the instuction set that the MC68XX family had though there are a number of improvements, soon I’ll be adding memory [url]http://www.freescale.com/files/microcontrollers/doc/train_ref_material/MRAMWC.pdf[/url] 8) , but the parts are cheaper the I thought.

  • YukonKid

    jdiffend

    I would tend to agree with your comments except.
    As I said in my above post it would be easier to emulate the coco 1 thru 3. To have a “GIME” chip or compatible chip does not make sense. To be able to use new hardware this would be the area that would be important. A Coco 4 would have to use a good graphic card. I was not really saying that the cpu should be any old thing and your points about the 68K family are well taken.
    Just think about the new things that the coco 3 gave us. Better Graphics, some speed and a new keyboard.
    A 40mhz inicore system would be great. it would really let things fly.
    Just think that your Coco 4 needs to be able to connect to the net and render videos and web pages so you could navigate easier.
    I have travelled the net in text only mode once upon a time and I do not want to do that again just to look around.
    But to play whatever games I want from any model Coco would be a dream. It would one up the coco 3 as it didn’t play everything made for the coco 1 and 2. Graphic Spectrum Analyzer comes to mind, one cart coco 3 wouldn’t play.
    This brings up the input area, USB is where it is at. from keyboard to joystick to mem and other devices.

    I could ramble on and on but I think you get my point.
    I have a coco 1, 2, and 3 and I would love to have an updated version to play on.

  • tpreitzel

    I’ve always thought the CoCo represented a sleeper system for its day. In other words, I viewed the CoCo as a plain Jane sedan with a hemi under the hood. I’d like to see a repeat of this concept today. The sheer brillance of programming the CoCo was everywhere, and I’m sure this talent remains to this day with many former CoCo programmers nearing “retirement”. What would separate the modern CoCo from the pack? One person suggested multiple CPUs and I think that person is right. Even with the relative novelty of multiple core CPUs beginning to erode, the design of a multiple CPU based computer with the associated operating system extended for modern hardware (USB, etc.) while maintaining compatibility with the past would be a massive undertaking for even the most brilliant and dedicated. 8)

  • 6809er

    Before everyone gets off track here, let’s define what we are talking about for this project.

    First of all, the name of the project is V.R. CoCo 4. The V.R. stands for Virtual Reality, not reality. (Not a real CoCo 4.) This is a SIMULATION of a CoCo 4 could be.

    Why a simulation?

    Costs. It would cost too much money ($1,000′s) to make real CoCo 4 for the small quantities that the market could support.

    Why do this project in the first place?

    First, our race of CoCos are facing extinction. Without new CoCos being manufactured to replace the ones that fail, they soon will be gone. A simulation running on another (modern) computer will keep the CoCo environment alive.

    The other point, why limit the simulation to a CoCo 3?

    We can fix the simi-graphics 24 bug in the CoCo 3 so it all coco 1&2 games work on the V.R. CoCo 4. What about adding the two missing bits from both the Memory Management Unit and the Color Palette Registers? This will give us for 2 Megabytes of Memory and 256 colors in place of 64 color palette. There are just a few of the quick fixes that could be done.

    Why not expand the simulation to include the capabilities of modern computer?

    Let’s take advantage of the modern computer’s hardware like hard drives, flash memory, Networks, Mouse, Keyboard and much more. There will be backwards compatibility so the new hardware devices can be used by old software. But let’s not forget about new software taking full advantage of the hardware. Let’s create a Basic Input/Output System (BIOS) for a standardize software interface.

    What about those that like to do a little hardware hacking?

    There are many USB to I/O ports interfaces on the market and it would be easy (via the BIOS) to use them under DECB or OS-9. Yes, you heard right, you can make the V.R CoCo 4 talk to the outside world in a way a real CoCo never could (without extra hardware.)

    While most will run the V.R. CoCo 4 on their PC (Windows, Linux) or Mac. You could put hardware like a PC-104 or Micro form-factor PC motherboard inside an old CoCo case. These board are so small that there is also room for other hardware like a hard drive, power supply and small AT Keyboard too. For all those people that want a easy to carry “all-in-one” computer with a build in monitor, just use a laptop computer.

    Let’s get back working on what the V.R. CoCo could be and stop talking about what it is not. We all know that it won’t be the end all “CoCo 4″ but than again, it’s a V.R. CoCo 4.

    Steve “6809er” Bjork

  • pastageek

    A PC can never amount to a coco4 since the memory and IC relations are so different as I have just been learning, but some features of some PCs that I have come across would be nice or maybe a nice adaption for a true type of coco4. I like being able to connect multiple cards into the PCI and ISA. I think a multipack that has both the traditional cartridge and the PCI 32/64 and ISA 8/16 (bus risers perhaps?) should be altogether and still in a very traditional case sense that beats opening up the PC. An old keyboard from the 8″ dsk drive dummy terminal makes a beautiful keyboard since some of its hot keys could be used for the cassette drive. And of course the coco4 should keep the same types of colors. Nothing is a coco without that and the unique chips. I like how when I open the color computer there is no tangled junk like I have in my pc. I think all the extra hard ware should be outside as always. Otherwise it can’t be a coco. Even though TRS has betrayed us (lol just exaggerating) and will probably never make one again, I still think that a group project to make the coco4 is still a coco since they have spent their lives with this machine.
    I personally was born too late to enjoy the homebrew days but I can still appreciate the majestic wonders of this machine. Sorry, I know that was corny. Right now I am still a newb at the coco but I think we should work towards the coco4 even if it never sells in mass quantities. We can still say it exists. I think that at least will be enough reward. To say the coco4 runs at 16GHz and that it has 1 Tb of SCSI HD and almost that in RAM and have OS-9 on the ROM.

  • RetroRick

    I think the PC would never amount to a CoCo for very different reasons.

    1 – The PC’s keyboard scanning is very different, and – IMO – is completely inferior. You need a separate keyboard handler just to read more than one key at a time. Windows interfaces do this automatically now, but that’s only because it has its own keyboard handler that someone else did, and has bugs the manufacturer will never fix.
    2 – The built-in sound capabilities of the PC just simply don’t add up to the CoCo’s in any way. The only thing the PC does reliably from OS to OS and machine to machine is one-voice beeps with no volume control whatsoever.
    3 – Both the OSs and the tools associated with PCs are frustrating to work with, and takes all the fun out of designing software for the computer. Plus, you have OS incompatibilities no matter what OS you program for. It’s a no-win situation.

  • pastageek

    No arguments there. On my PC now my sound is tied directly to the fate of YIM. The coco would never stand for this. I think that the bus architecture of the Altair should be adapted to the coco in that it wold be nice to have 16 cartridge ports and maybe 16 of every other standard. The coco is by far my favorite computer. It would be nice to see what a 64 bit coco is like.

  • PunkMaister

    It is sad that we will never really know how a color computer would have looked like if Tandy did not cancel it’s production and had kept up production to this day with version after version, considerinvg what the CoCo was it would be not unlike a Mac but with tons and ton of much more software, games and what have you… Perhaps it does exist in an alternate reality who knows!

  • sbatson

    I’m amazed at all the philosophical discussion on this. I think Steve Bjork hit the nail on the head with response to keep things on track.

    Other than reminiscing about days gone by and how much fun “The Good Ole Days” were, it seems more appropriate to get something in place that keeps the CoCo community alive before all the hardware dies. Yes the little machine had/has charm, but I’d rather be able to preserve my software and ability to use it than worry about what it’s running on.

    My 2 Cents:

    1) Doing it VR/Emulation on today’s machines makes total sense to me. Hardware is plenty powerful enough.
    2) Ability to run on Windows/Mac OS/Linux are a must in my opinion.
    3) Open Source would probably help it stay alive and grow
    4) 100% Compatibility to Coco 1/2/3 seems an absolute must to me. There is absolutely NO reason today’s hardware can’t achieve that. Previous emulators have failed at that task, it’s something that should be addressed
    5) New features of course limited only to the imagination of programmer, but it should be able to do most of what people do today (Internet, large storage, USB devices, photo editing, etc.)
    6) Extensible – The system should be able to be enhanced through plug-in’s and maybe some type of hybrid mode where the programmer can pull from the features and APIs of the native OS if desired.

    In any case, it seems that for this project to really succeed, it needs to attract new users and programmers, not just support the existing but ever shrinking base of CoCo users.

    On a side note, I personally like the idea of the project being VR since I could use it on my existing PC or Mac and not need another place for a system, wires and other devices.

    Steve

  • PunkMaister

    ^With all due respect if all you are really worried about is running 80′s CoCo software current emulators like VCC and MESS more than suffice for this task, it would be nor represent any progression toward the future. Yes a future CoCio would have to be virtually developed to keep costs down but if the end game is to truly make a new actual phisical 21st century CoCo the emulator would have to run on really, really high end systems so a truly marketable machine could be created otherwise is a total waste of time!

  • sbatson

    ^With all due respect if all you are really worried about is running 80′s CoCo software current emulators like VCC and MESS more than suffice for this task, it would be nor represent any progression toward the future. Yes a future CoCio would have to be virtually developed to keep costs down but if the end game is to truly make a new actual phisical 21st century CoCo the emulator would have to run on really, really high end systems so a truly marketable machine could be created otherwise is a total waste of time!

    I’m not sure if this message was directed at me, someone else, or the entire thread…Howerver, I don’t think anyone is saying that they wouldn’t want more. I certainly didn’t indicate that. My primary concern is that it should run all the old stuff, and properly. I’ve tried about every emulator out there. There are always coco games that the things barf on or not run well. After that is addressed, I’d love to see a more powerful coco with new features and software.

  • PunkMaister

    I’m not sure if this message was directed at me, someone else, or the entire thread…Howerver, I don’t think anyone is saying that they wouldn’t want more. I certainly didn’t indicate that. My primary concern is that it should run all the old stuff, and properly. I’ve tried about every emulator out there. There are always coco games that the things barf on or not run well. After that is addressed, I’d love to see a more powerful coco with new features and software.

    A bit of a tall order considering that some CoCo 2 aofware never ran properly on the CoCo 3 in the first place, not without some modifications of some sort. But yes current emulators need to be vastly improved but that is in many ways a separate issue from a next generation CoCo project. It would be better if the same people that worked on the currently avaiable emulators worked on this project and their emulators simultaneously, having said that I have yet to see an emulator of any kind of computer or xonsole that works with all of that particular machine’s software be it a CoCo, Commodore, Nintendo, Collecovission you name them there is not a single emulator in existence right now that works with the target machine’s software in a 100% capacity so if it is achieved it would be a world first!

  • sixxie

    … I have yet to see an emulator of any kind of computer or xonsole that works with all of that particular machine’s software be it a CoCo, Commodore, Nintendo, Collecovission you name them there is not a single emulator in existence right now that works with the target machine’s software in a 100% capacity so if it is achieved it would be a world first!

    I’m sure we’re always interested in information about the edge cases though!

  • PunkMaister
    … I have yet to see an emulator of any kind of computer or xonsole that works with all of that particular machine’s software be it a CoCo, Commodore, Nintendo, Collecovission you name them there is not a single emulator in existence right now that works with the target machine’s software in a 100% capacity so if it is achieved it would be a world first!

    I’m sure we’re always interested in information about the edge cases though!

    Sure! But that is not what Mr/Ms sbatson is talking about, he/she clearly implies that he/she wants the world’s most absolutely perfect emulator ever written. One that runs all previous CoCo software regardless of what processor 6809 or 6309 it was written for. Now think about this not even console emulators which have a very simplified/standarized architecture based on their hardware counterparts and most certainly do not use 2 different processors have failed at this! The task to build such an emulator is nothing less than monumental and in a scale never done before. Again it would require that somehow the emulator can distinguish between 2 different processor codes and switch automatically and invisibly to the user to the right one so a program that has been written for either processor runs smoothly.

  • sbatson
    … I have yet to see an emulator of any kind of computer or xonsole that works with all of that particular machine’s software be it a CoCo, Commodore, Nintendo, Collecovission you name them there is not a single emulator in existence right now that works with the target machine’s software in a 100% capacity so if it is achieved it would be a world first!

    I’m sure we’re always interested in information about the edge cases though!

    Sure! But that is not what Mr/Ms sbatson is talking about, he/she clearly implies that he/she wants the world’s most absolutely perfect emulator ever written. One that runs all previous CoCo software regardless of what processor 6809 or 6309 it was written for. Now think about this not even console emulators which have a very simplified/standarized architecture based on their hardware counterparts and most certainly do not use 2 different processors have failed at this! The task to build such an emulator is nothing less than monumental and in a scale never done before. Again it would require that somehow the emulator can distinguish between 2 different processor codes and switch automatically and invisibly to the user to the right one so a program that has been written for either processor runs smoothly.

    It’s Mr. Batson if you wish to be formal ;) …the name is Steve. Actually with all of this, I’m just adding to the wish list…what happens, happens based on the time and talent put into the project. As to your statements about nothing perfect has been written, well I think that pretty much goes without saying since Man isn’t perfect. But it’s no excuse to leave big holes in it as has been done with previous emulators regardless what the reasons are.

    If one wants to look at an awesome emulator, just take a look at “Rosetta” built into the Mac OS from Tiger on. It does things you seem to indicate haven’tt been done. Rosetta invisibly detects the processor the software was designed to run on for the Mac OS. If it was a PowerPC based app, it just runs the program with no messages or dialogs to ask you what to do or tell you what it is doing, the program just runs as if it were designed to run on the Intel Processor based Macs.

    Steve

  • PunkMaister

    Rosseta should not even be called an emulator because you never run the power PC Mac programs on an emulator window of any kind, they simply run withing MacOs itself. In fact they do not bill themselves as an emulator but rather as a translator. The closest thing I can think off is a program that converts X-box games to run under windows and not even that aproaches the functionality of Rosetta but the point is it is far, far more than an emulator in this case.

  • dentman42

    Definitely an emulator. Programs are using instructions not present on the Intel chips, therefore it’s emulating the PowerPC environment.

    I can name another perfect emulator – the old “Williams Arcade Classics” for the PC that emulated Defender, Defender 2 (Stargate), Joust, Sinistar, Robotron, and Bubbles. It was done in Assembly, ran ok on a 486-33, and even the original Williams programmers said it was a perfect emulation. (I’d love to see the guys that wrote that become Mamedevs).

    Actually, they might be good for a CoCo emu, considering that those games were based on 6809 and 6821 (custom sound a vid though).

  • sbatson

    Definitely an emulator. Programs are using instructions not present on the Intel chips, therefore it’s emulating the PowerPC environment.

    I can name another perfect emulator – the old “Williams Arcade Classics” for the PC that emulated Defender, Defender 2 (Stargate), Joust, Sinistar, Robotron, and Bubbles. It was done in Assembly, ran ok on a 486-33, and even the original Williams programmers said it was a perfect emulation. (I’d love to see the guys that wrote that become Mamedevs).

    Actually, they might be good for a CoCo emu, considering that those games were based on 6809 and 6821 (custom sound a vid though).

    Agreed Dentman42.

    Williams Arcade Classics is Awesome.

    Punkmaster, in this case, emulator, translator…I don’t think anyone really cares what you call it or how you classify it. What people will care about is the end result. The point is, it has been done with more than one well known product and more than likely in many cases none of us even know about. And regardless of where the emulator/translator resides (embedded in the OS, running in a program full screen, windowed, from the command line or whatever has absolutely NOTHING to do with if it is an emulator or not….I’m not sure where that came from. No offense, but it seems you are more concerned with being argumentative in this thread then really talking about what the possibilities could be (which is what I thought this section was really about).

    Just my 2 cents.

  • PunkMaister

    Agreed Dentman42.

    Williams Arcade Classics is Awesome.

    Punkmaster, in this case, emulator, translator…I don’t think anyone really cares what you call it or how you classify it. What people will care about is the end result. The point is, it has been done with more than one well known product and more than likely in many cases none of us even know about. And regardless of where the emulator/translator resides (embedded in the OS, running in a program full screen, windowed, from the command line or whatever has absolutely NOTHING to do with if it is an emulator or not….I’m not sure where that came from. No offense, but it seems you are more concerned with being argumentative in this thread then really talking about what the possibilities could be (which is what I thought this section was really about).

    Just my 2 cents.

    It was not my intention to sound confrontational it just seems like a daunting task to make an emulator that runs all previous CoCo software in a 100% capacity that’s all, but I hope they can indeed make such a thing.

  • sbatson

    It was not my intention to sound confrontational it just seems like a daunting task to make an emulator that runs all previous CoCo software in a 100% capacity that’s all, but I hope they can indeed make such a thing.

    No problem :) I hear you on the amount of work it will take to achieve 100%. It is possible with the power of today’s hardware. I too hope it can be achieved. The greatest asset to the coco is all it’s support and software to that point. If that was abandoned in favor of new features, then a lot is lost. The main reason the PC and DOS/Windows became such a hit was not only the open architecture that encouraged developers to take advantage of, but also the strong commitment to backwards compatibility. The PC and Windows unfortunately have grown so out of control with so many products they have conflicts and compatibility issues where there should be none…anyway….

    If the Coco 4 or other future coco products protect what has come before and add new stuff it seems it would stir up excitement. If the future coco’s hardware or emulators leave stuff in the past, there doesn’t seem to be much of a point. All new software would be needed. The market/audience would be very small. But having a new system that runs a ton of stuff, runs it properly, and can do new stuff..that would be cool! I really would like to see something good happen with this. I think we all do.

  • Emtwo

    I’m coming into this late but my thoughts on a viable Coco 4 emulator are the following:

    1) it should still run the Coco3, Coco2 and Coco1 programs. I think thats one of the places the MM/1 and other 68000 machines failed. They left behind their base. And the base was left choosing between a new machine that didn’t run any of their old software and didn’t have any new software to speak of and an Intel machine that didn’t run any of their old stuff but had thousands of programs available. Some choice.

    2) since its in emulation, it should be able to access the PC hardware. ie, printers. Now with the the 1001 different printers available, I’d suggest it be a PDF printer and that the resulting PDFs be saved “outside” the emulation, where the PC can get it and print it.

    3) The Coco4 should be an upgrade to the Coco3. Being an upgrade, we don’t get to goto a 32bit processor (virtual). Instead we should look at all the nice things the 6309 did for the coco3 and try do the same thing again virtually. ie, a “6409″ with the same 8/16 bit registers and command set as the 6309 PLUS some more. Like 4 extra 16-bit registers that could be ganged as two 32-bit registers, a tru math-coprocessor and a video co-processor that would add-on to the GIME or optionally replace the GIME at BOOT time.

    4) The emulation needs some way to talk to the rest of the universe. I just bought the bluetooth pack available on this website. Now that I’ve paid for the hardware, I’d like to see some way for the virtual machine to get files to and from the PC or the rest of the universe with out jumping thru hoops like we do right now for the Coco3 under emulation.

    Those are my thoughts for right now.

    Boisy, this is a great idea. I hope you are still working on it.