Dreaming of a Next Gen Atari - one can Dream!

[AtariSociety]

Anyone else dream that we had for purchase a more powerful (in today's standards) Next Gen Atari TOS based system?

Not a emulator or environment that lives on a Windows or Mac box, but like in Amiga world, X5000 that was built to run their Amiga OS.

Having been raised Atari and jumped ship mid 90s for professional purposes to Mac, I do dream of such things.

I have a nice collection of older Atari I enjoy but do wish we had a more powerful Next Gen Atari. Nothing against Firebee, nothing against the Hatari and like, all really cool stuff, but lets face it, sub 300MHz for Firebee will not get far in this multi-media world we live in. Or Atari evironments like Hatari on Windows or Atari just does not have the same feel for me.

I now still remain Atarian at heart but also am an Amigan, since 2009/2010 when I heard about the X1000 Next Gen Amiga. Owned one of them and now own the newest X5000. I see some fun in the Amiga community how the Amiga OS is still professionally worked on and A-Eon have thrown some good money at creating newer PPC systems. It is all rather cool. Even these Next Gen Amigan's are underpowered compared to Windows or Mac boxes but still, close enough to do some real work stuff, fast enough to get things done.

Just dreaming it would have been cool if somehow TOS as we know it could be advance in a way to be more modern and that Atari specific or multi-platform systems that are native to Atari could come about.

I wonder if all the current crop of Atari guru's that make upgrades and goodies for Atari could get together and create something like this. Would be a dream come true.

I know I know.... takes lots of money and this is all a hobby. But one can dream.

I mean, look at Vampire.... they are building these things and selling as fast as they make them. From what I hear almost 1600 people in line.... Thats 1/2 a million $$ and its happening. Tells me if the cards were right, something like that could happen in Atariland too.

One can dream.

TJ

Edited February 19, 2017 by AtariSociety

[leech]

I think for this to happen we would need the OS to catch up more. But on the upside we do have EmuTOS and FreeMiNT which is still being worked on. The thing is though, even in mew Amiga land, it is the software support that counts, and with few exceptions, the only software being made for OS4 is the same software I can install in a Linux machine on a ridiculously fast and cheap x86 system.

 

I personally would love it if someone could fins and release the source code for TOS 4.92 and get TOS 5.0 out the door.

 

There is a thread over on atari-forum.com about building expansion ports/graphics adapters to get all of the STs out there to be at the same level of features, but if we could push out a Vampire based stand alone with upgraded TOS and SuperVidel included, we could have a Modern ST clone.

 

We are already partially there wint MiNT running on WinUAE and a fVDI driver to support it.

 

When you have such small gropus of people though, it takes years to get things out there.

[AtariSociety]

Is there some rights to TOS 4.92 that prevents this from ever happening?

 

The land of Amiga is a mess with who owns what but stable enough where it seems Hyperion have a lock on Amiga OS 4.1 at least and they have developers that continue to move it forward, even if slowly.

 

Is anyone working on newer TOS or is it a dead subject?

 

I am not up to speed on the whole FPGA thing but does it tap out at a certain speed meaning does FPGA lock in at lower MHz speeds and not GHz like modern day machines do?

 

Just curious.

 

TJ

 

 

I personally would love it if someone could fins and release the source code for TOS 4.92 and get TOS 5.0 out the door.

There is a thread over on atari-forum.com about building expansion ports/graphics adapters to get all of the STs out there to be at the same level of features, but if we could push out a Vampire based stand alone with upgraded TOS and SuperVidel included, we could have a Modern ST clone.

[exxosuk]

The best base for this IMHO is the Milan. The guy was working on a 060 version, but never finished it due to lack of demand. Though if that design was updated to even take PCI cards instead of ISA, then that would be pretty much a complete system. I did ask about doing a run of the original milan, but never got a reply about it

 

Annoying thing is, speaking from a hardware point of view myself, Everything has to be re-invented all the time. Also while I do a lot of stuff as people know, I have to do all this work and research pretty much all by myself. There is Rodolphe who helps me get up to speed with how the 68000 bus operations work and helps with with some coding, though at the end of the day, I am having to learn everything from the ground up. This results in endless years ( yes YEARS) of trial and error to figure it all out. I'm not one for just ripping off designs which are already produced, I do research all designs and then do my own design and have stuff working how I want it to. So this in itself means even more work for myself.

 

As already mentioned, there is a thread on atari-forum about it. Though there is no where enough active hardware development people to do much past what is already been developed. I've talked to a lot of people about problems I have with various stuff, nobody has a clue what I am talking about, so nevermind anyone being able to actually help with development of stuff.

 

It's not just that though, again its software, the more hardware changes, the more old software breaks. Its why I am trying to max out the 68000 series of CPU speed wise. The 030 adds a nice boost due to caches, but even so eventually you hit a wall of raw MHz power. Even with fast-ram, it makes a fast system, but I don't know what the solution is after that.

 

I myself while I do a lot of work, I won't be doing stuff indefinitely either. Unless more people start learning how to solder and learn how the 68000 systems work, then I fear in the near future there will simply be nobody left producing new hardware at all.

[AtariSociety]

Which I may also add, seems the Firebee may be the fastest from the ground up designed Atari compatible then. 260MHz or whatever it is. Has FPGA and DSP if memory serves. Although not a modern machine in terms of real CPU power, could be quite usable if FPGA and especially DSP were working. Seems so much potential but not enough of those that know how to work on this stuff are willing to help.

 

Wonder if those making the Firebee should of, if they have not already, offered big discount to those that could help, get a Firebee as long as they promised to work on these core items.

 

Would be cool that Firebee sold into the 1000s and not the low 100s like it has.

 

Just some thoughts.

 

But seems Amiga platform made that next step. Amiga specific machines in the 2GHz range.

 

Guess they had more passions that did whatever it took to keep the platform alive and growing over those in Atari, qty wise.

 

Must admit, being an Atarian now looking at what Amiga was back then, their OS did seem more capable. I still like the Atari more for the classics but for modern, Amiga has leaped ahead.

 

Still love though the few left in Atari-land making neato stuff. I applaud there work. Atari 8 Bit seems VERY alive!

 

TJ

[+Gemintronic]

I think reviving a platform while dealing with non-commodity hardware is an excuse to fail. I wish people would cobble together a distro based on EMUTOS and FreeMINT and get it to work on cheap, generic ARM boxes. Heck, maybe even a cut down version for Raspberri Pis.

 

Being able to run RISCOS on Raspberry Pis is a good example of what I'm talking about.

[exxosuk]

The FPGA stuff could be the only way forward. With real hardware failing like mad these days it is at least a alternative. Though it again gets back to another thread about the future of the machines, in terms of new people using Atari ST's etc. Likely they will buy a "cheap" ST and see if they like the games, then they may invest in a hard drive and maybe other things to make it faster such as the boosters. Though I am not really sure if new people are going to spend big bucks on a firebee right off.

 

Its also one thing to have games run a bit faster and smoother, though after that, then what ? Not exactly any games being developed which would take advantage of the speed of a faster system. Of course everyone uses their machine for various things. No way to really generalise anything. Though my hope one day is to produce a booster which will be cheap enough and give enough power that people may want to start coding again for the Atari's. Though that would need mass production and people willing to buy the stuff, though I don't think the Atari market will ever be large enough for that to actually happen.

[AtariSociety]

I would go for a very hopped up FPGA based Atari too.

 

I am one of the crazy ones I guess. I have spent thousands on Next Gen Amiga and I was not even an Amigan. But the excitement in that group is addictive and my pocket-book has suffered from it. LOL.

 

Just paid $2K for an X5000. Would probably do the same for a blown up native Atari too, although I don't see something like that ever coming.

 

But there are some of us out there. We do spend.

 

An interesting example of something more affordable is the new AmigaOne A1222, or Tabor. They will sell for less than $500. 1.2GHz, high speed video card slot, etc.... So there are some folks making specialized boards in certain markets.

 

I am not sure how fast an FPGA system can get, but if a MonSTer model came out with Atari in mind, and could one day feasbly view videos, say DVD quality minimum but better 720P and use Netsurf for somewhat real web work, it could allow some to use Atari daily.

 

That is what the X5000 is doing.... although I use Macs and Loonix boxes for my day to day work, I use my X5000 all day too as I work from home. I watch videos, play music, web browse, etc.... things I expect computers to do for me.

 

TJ

Edited February 19, 2017 by AtariSociety

[pixelmischief]

I think the right first step for this endeavor is to spend some time thinking about what it is you want to achieve. I other words, what is it you are yearning for and what will a new thing have to be to satisfy that yearning. I'll start with some questions:

 

1. Does it have to bear the name Atari?

2. Does it have to be unique hardware?

3. Does it have to have a unique OS?

4. Does it have to run existing Atari software?

5. If so, to what extent does that support have to be seamless?

6. Will it have to support MIDI in hardware with the precision that 68K Atari hardware systems do?

7. Will it have to support cartridges?

8. Will it have to have ACSI support?

 

I will answer these questions for myself.

 

1. It must bear the name Atari, but not officially. Not from a corporate ownership standpoint.

2. It does not need to BE unique hardware, but I think it would need to include unique hardware.

3. It doesn't need to have a unique kernel, but everything after that would probably need to be unique. The primary focus would need to be highest levels of TOS compatibility and unobtrusive expansion of it's architecture. I envision TOS "versions" available via a boot loader.

4. It must run as much existing Atari software as possible. This must be it's primary focus. Games, demos, and desktop software must all run on the system. This, even to the extent that it can reproduce the experience of booting from a floppy as closely as possible.

5. Software running in the extended feature space of the solution must run side-by-side with legacy software and applications without any visual or functional boundary.

6. It would have to support MIDI in hardware with the same precision as legacy Atari systems. I envision this being accomplished by a combination of hardware and kernel support.

7. It would have to support cartridges to the extent that existing cartridges had an entry point into the system; even if this meant some kind of USB-style device that was read and inserted into the memory space at boot.

8. It would have to have ACSI support and support for Atari file systems.

 

There are an enormous number and variety of other questions. But for me, this is a start. I think the right solution starts with PC hardware. It would need a flashable BIOS. It would probably start with a Linux kernel or FreeDOS. From here, a great deal of the required software is already developed and source-available.

Edited February 20, 2017 by pixelmischief

[leech]

I think reviving a platform while dealing with non-commodity hardware is an excuse to fail. I wish people would cobble together a distro based on EMUTOS and FreeMINT and get it to work on cheap, generic ARM boxes. Heck, maybe even a cut down version for Raspberri Pis.

 

Being able to run RISCOS on Raspberry Pis is a good example of what I'm talking about.

There is this;https://sites.google.com/site/emaappsarch/news/beekeyandbeepibeta9areavailable

[leech]

I would think an upgraded MiST capable of handling a Falcon core with 060@100mhz and SuperVIDEL would potentially fit this bill the best. That would give us a high end, reconfigurable system. Does TOS/multiTOS/MiNT, etc need multi ghz to do things? Granted, I haven't tried browsimg the internet on my mostly stock Falcon yet, but most other things are fairly fast because MiNT+Teradesk is so light weight.

[joska]

The best base for this IMHO is the Milan. The guy was working on a 060 version, but never finished it due to lack of demand. Though if that design was updated to even take PCI cards instead of ISA, then that would be pretty much a complete system.

 

The Milan has four PCI-slots. It also (optionally) has a 060. The main difference between the Milan and MilanII is that the MilanII was an ATX design, not AT like the Milan.

 

The MilanII was finished and prototypes made, but it never made it to the market.

 

Anyway, I've had my Milan060 for more than 15 years now, and while I really love that machine I have to say that going in the same direction with a new piece of "Atari" would be a big mistake. First of all, PCI has been obsolete for years. Secondly, going that route means having a GEM-only machine. Forget about running any sort of legacy software that's not pure and clean GEM. So no cool pixelpainters, no music applications, no games, no demos. Just "business"

 

The Milan is the best "GEM-machine" out there though. Rock solid and very, very stable. I've spent thousands of hours with PureC on that machine the last 15 years

 

I also have a Firebee and it's pretty much "just" a GEM-machine too. A lot faster than the Milan, but also a lot more fragile. As a new "Atari" I think it has failed. I don't think people realize how backwards compatible new hardware must be to use legacy software in a stable and usable way. Even when using a TT or Falcon you'll quickly get in trouble.

[phoenixdownita]

I fail to understand the interest in GEM/TOS in modern day and age. Linux works and it's rock solid, there's BSD, AROS, Plan9, Inferno and other free OS flavors and most of them run on x86/x64, ARM or whathaveyou modern CPU.

Whats' the point to have an obsolete OS modernized? What need would it satisfy?

 

If instead the dream is to have a backcompat 1+Ghz superST with 16M color 3D accelerated gfx support .... ain't gonna happen, buy modern hardware and use modern software.

I am pretty sure all needs are satisfied one way or another ... am I wrong?

[AtariSociety]

As stated..... "one can dream!".

 

No sense needed. Dreams are just that.

 

There is a reason Amiga has survived all these years and went onto a more modern gear PPC gear. Expensive mind you but still happened.

 

Someone had a dream.

 

And for some of us, $ is spent on this dream for sure.

 

TJ

I fail to understand the interest in GEM/TOS in modern day and age. Linux works and it's rock solid, there's BSD, AROS, Plan9, Inferno and other free OS flavors and most of them run on x86/x64, ARM or whathaveyou modern CPU.

Whats' the point to have an obsolete OS modernized? What need would it satisfy?

 

If instead the dream is to have a backcompat 1+Ghz superST with 16M color 3D accelerated gfx support .... ain't gonna happen, buy modern hardware and use modern software.

I am pretty sure all needs are satisfied one way or another ... am I wrong?

[leech]

Sad thing is, the ST line more or less died out before it could make any jump away from the 68k. The Amiga only jumped to PPC, because of OS3.9 support, which then turned into pure PPC for OS4. But even 3.9 wasn't made by Commodore. Would Motorola have continued the 68k line of processors if Atari and Commodore hadn't died out? Who knows, seems to me they had already decided to go with PPC with IBM at that stage, so the next ST potentially would have ended up in the same semi-PPC/68k setup that the last Accelerated Amigas did.

 

With that in mind, I do have an old iMac PPC G5 that is pretty much worthless to me unless I put an alt OS on there (I hate Mac OSX, and it's stuck at Snow Leopard or whatever that one is, and unfortunately has an nVidia, so MorphOS isn't supported).

 

Speaking of Mac, they went 68k -> PPC -> x86. So even they finally gave up having a unique architecture. Granted, FreeMiNT ported to a Power8 would be fantastic! Not that us normal folk could ever afford such a machine.

[joska]

 

There is a reason Amiga has survived all these years and went onto a more modern gear PPC gear. Expensive mind you but still happened.

 

Whether the Amiga has survived or not is debatable If it did survive, it's despite of the PPC, not because of it. Amiga today is still mostly about 68k.

 

The move to PPC was a failure. The Amiga community focused only on the hardware, but failed to realize that 99% of the effort would have to be put into software. It's the software that brings the system forwards, not the hardware. Why do you think the Vampire is such a big success in the Amiga world? Because it combines two things - raw speed AND backwards compatibility.

 

If we ported TOS+MiNT to ARM or Intel today, there would be two major differences:

 

1 - Much faster hardware.

2 - A lot less software.

 

Who will write the new software that exploits the new hardware?

[joska]

Sad thing is, the ST line more or less died out before it could make any jump away from the 68k.

 

It could never had made that jump. Atari never had the resources or visions to create and develop a rich OS, neither did they provide developers with usable tools. The result was that a major part of ST software was developed like it was an 8 bit. Developers invented their own solutions to stuff the OS and dev-tools should take care about, but didn't. Writing applications for GEM is hard, much more work than writing the same stuff for Windows or MacOS. And the developer tools - like TOS itself - never progressed. The result is software that's so tied to the hardware that jumping away from the basic ST cause big problems for the users.

 

Remember that Apple - who were big and filthy rich - barely made it. A small company like Atari with the philosophy of never spending a dime never had a chance.

[AtariSociety]

I consider the AmigaOne made by A-Eon to be an Amiga, as much as say Milan is an Atari.

 

Both machines designed to run their respective OS of choice. So I consider them the real deal.

 

But indeed, both parent companies are long gone, but I am OK with someone else picking up the ball and playing with it

 

Commodore/Amiga/many new Owners moved on but in my opinion and we have seen many Next Gen Amiga's, X1000 and X5000 to name 2. These were designed to be next gen Amiga. Also now a 3rd from A-Eon being beta tested A1222. All PPC, all much more modern and capable.

 

I don't consider that failure. I consider that persistence and just getting it done no matter the odds. A-Eon been around before 2010 so they seem to be the real deal. They have some $ and goals and just plain get them done. I would welcome that in Atari-land and think many others would too.

 

Software has also moved forward. There are apps on Amiga NG that move forward, play much of today's media, and even web browser, that although not on par with those in Mac and Windows land, are not really that far off. Odyssey is pretty good web browser and I can do most my daily web work from it, at decent speeds. Far from what Atari can do in that arena.

 

As much as I love classic Atari, I am very happy to see a company like A-Eon put their money where their mouth is and move forward.

 

I would not turn down someone doing the same in Atari-land either.

 

We can enjoy our classics and NeXT Gens all the same.

 

TJ

 

 

Whether the Amiga has survived or not is debatable If it did survive, it's despite of the PPC, not because of it. Amiga today is still mostly about 68k.

Edited February 21, 2017 by AtariSociety

[AtariSociety]

What I would like to see is a dual system.... one that can as close as possible be the most compatible classic box able to run older Atari apps as they were designed but yet on the other hand, an eye to the future, where maybe new developers with a retro flair can build upon and add to Atari, one that is more modern with Atari look and feel.

 

One can dream...

 

TJ

[zzip]

 

Anyone else dream that we had for purchase a more powerful (in today's standards) Next Gen Atari TOS based system?

 

Not really, the Atari eco-system was already getting terrible support in the 90s while Atari was still in the business. By then, TOS was late to the party with features other OSes had adopted. Same with software. I liked my ST, but the grounds for moving to another platform were just too compelling.

[AtariSociety]

Yupper, I moved to Mac mid 90s.... but looking back in 2017, I can dream of fun today, as side hobby machine.... one maybe more powerful that I can do some more modern things in yet still enjoy the field of old Atari.

 

TJ

 

 

Not really, the Atari eco-system was already getting terrible support in the 90s while Atari was still in the business. By then, TOS was late to the party with features other OSes had adopted. Same with software. I liked my ST, but the grounds for moving to another platform were just too compelling.

[zzip]

I fail to understand the interest in GEM/TOS in modern day and age. Linux works and it's rock solid, there's BSD, AROS, Plan9, Inferno and other free OS flavors and most of them run on x86/x64, ARM or whathaveyou modern CPU.

Whats' the point to have an obsolete OS modernized? What need would it satisfy?

 

If instead the dream is to have a backcompat 1+Ghz superST with 16M color 3D accelerated gfx support .... ain't gonna happen, buy modern hardware and use modern software.

I am pretty sure all needs are satisfied one way or another ... am I wrong?

 

Same here. Sure I get nostalgic sometimes, I've booted up aranym with emutos, which would simulate a newer more modern Atari system that never existed, but then I get I bored with the whole thing with after 10 minutes, because:

 

1) The OS feels archaic

​2) There aren't a whole lot of great applications to run on it that don't have better alternatives on the host OS.

 

Why run mint when you can have Linux? why stick to GEM in 2017 when you could have MacOS, Windows or any of the UI's that Linux provides with an ST emulator to run your old software?

[zzip]

What I would like to see is a dual system.... one that can as close as possible be the most compatible classic box able to run older Atari apps as they were designed but yet on the other hand, an eye to the future, where maybe new developers with a retro flair can build upon and add to Atari, one that is more modern with Atari look and feel.

 

One can dream...

 

TJ

 

Sounds like you want a Raspberry Pi with a custom Atari-like case running the hatari and aranym emulators then.

Also it wouldn't be hard to take some Linux desktop and reskin then to have a GEM look.

Edited February 22, 2017 by zzip

[leech]

 

Same here. Sure I get nostalgic sometimes, I've booted up aranym with emutos, which would simulate a newer more modern Atari system that never existed, but then I get I bored with the whole thing with after 10 minutes, because:

 

1) The OS feels archaic

​2) There aren't a whole lot of great applications to run on it that don't have better alternatives on the host OS.

 

Why run mint when you can have Linux? why stick to GEM in 2017 when you could have MacOS, Windows or any of the UI's that Linux provides with an ST emulator to run your old software?

You know what I find funny is that TOS, being a single tasking operating system, would probably be a LOT better for people with ADHD. Like me, I can't concentrate on doing one single thing on my computer at a time, hell I usually browse the Internet while I'm watching something on Netflix. But if I were to sit down and write out short stories on Atari Works on a ST/TT/Falcon, I'd be able to concentrate on it a lot more.

 

Not to mention the UI simplicity of GEM programs being desirable in this day and age of tabs and no 'file/edit' menu unless you hit 'Alt'.

 

While yes, it'd be nice if some of more modern ideas of TOS 4.92 were implemented and we got a maintained TOS 5.x with some multitasking built in, you'd still have a more focused environment to work in because you wouldn't have any of that worthless ads that are pushed onto us through flash or javascript.

 

While most of us stick around with the Atari computers out of nostalgia, there are some applications that simply aren't as good out there (like some of the extremely tight timing with MIDI). While they may have no features, that integrated hardware still is hard to beat from everything I've read about it.

[AtariSociety]

I have seen this now a couple times about TOS 4.92.

 

Give me the cliff-note version about this. I had never heard of it.

 

Guessing the last Atari TOS before they went belly-up that was never installed on any systems?

 

What benefits over the last TOS would it have brought?

 

TJ

 

 

While yes, it'd be nice if some of more modern ideas of TOS 4.92 were implemented and we got a maintained TOS 5.x with some multitasking built in, you'd still have a more focused environment to work in because you wouldn't have any of that worthless ads that are pushed onto us through flash or javascript.