Atari Portable

[SainT]

I've been pondering for ages building a 68000 based computer, for no real reason other than why wouldn't you want to? But building a completely custom thing which only I would really use is a bit frivolous, so the next thing that came to mind was an ST compatible machine. Well, there's a few of those already... so finally I've been pondering an ST compatible portable. This idea I quite like.

You can get small 7" (800x480) LCD's cheaply like you'd get in satnav systems etc... which have resistive touch sensors as well. That could form the basis of quite a neat little STPAD style machine (with keyboard / mouse ports). Interfacing SDRAM, LCD and the ST chipset could be done in a CPLD / FPGA with an 68sec000 as the actual CPU (which can be overclocked quite substantially).

Apart from novelty factor, I'm wondering if anyone would actually use a device like this. Are ST's actually still used musically these days? It might make an interesting portable Cubase machine for those that want microsecond accuracy.

[Atarieterno]

You said the magic word: Cubase.
I still use Atari for music, so your portable project does have an interest for me. The issues are (among others):
- Would he have ROM port for dongles and accessories?
- Would the price be affordable?

Regards.

[tonma]

For me, it would be mainly for games and to be able to program games on it everywhere.

[SainT]

You said the magic word: Cubase.
I still use Atari for music, so your portable project does have an interest for me. The issues are (among others):
- Would he have ROM port for dongles and accessories?
- Would the price be affordable?

Regards.

Indeed a cartridge port was what I was thinking too for dongles etc...

I'm not sure on price as yet, in thoery the parts aren't too bad, but minimum run for getting boards built at the fab house I use is 100 units. I'm a bit doubtful I could shift 100 of these... lol. But I would think perhaps around the 250 ukp mark would be doable.

This is a future project for me as I'm just working on a Jaguar SD cartridge at the moment, but something like this has been at the back of my mind for ages and I'm building up experince in all the tech required through other projects.

It could be something I could take to Kickstarter I guess once I have a working prototype.

I keep thinking of neat things like mechanical keyboards to go with it too. I quite like the idea of building a keyboard too... But just a main unit you could plug a ps2 (or USB) keyboard and mouse into makes more sense.

[penguin]

Wolfgang Foerster's (Suska) implementation of the ST/E chipset should be pretty much complete. For MIDI apps, you'd absolutely need the cartridge slot. Maybe call it the STE Pad, to give it a nice marketing spin (the ST tablet is here!) although I doubt that the special TOS version of the original ST Pad would run on it.

Switchable TOS version would also be necessary. Your tablet would obviously be released with EmuTOS which is nice but may not be compatible with all music software.

A special keyboard would be a good second Kickstarter. I could imagine that FireBee users who don't own an Atari keyboard might be interested in one.

[SainT]

The separate keyboard project idea is interesting. I'd always quite fancied an Atari keyboard for general use...

TOS switcher is interesting, that does make sense. EMUTOS is naturally far more legal, but its up to the end user if they wish to use a TOS from their old machine. I could certainly make the touch interface look like a mouse drag pad for old TOS versions.

Out of interest for the cart port for cubase use -- is this mostly for the dongle, or also for additional MIDI ports etc?

[penguin]

Out of interest for the cart port for cubase use -- is this mostly for the dongle, or also for additional MIDI ports etc?

Both actually - many music programs came with a ROM port dongle, but it was used for all kind of extensions as well, e.g. the Midex: http://stcarchiv.de/tos1990/11/midex

Other programs used the parallel or joystick (!) port for dongles, the first version of Live had a joystick dongle, though they got rid of that later. The ROM port was by far the most popular port for dongles, especially for music software.

[SainT]

Holy crap that's huge!

[Mindthreat]

I would definitely be interested in a board or two for a mini ste of sorts =)

[Frank B]

A mini ste would be great. A mini falcon would probably make a fair bit of money though Hint hint

[Atarieterno]

- I remind the respectable colleagues of this forum that there is already a project of "Super STfm" that is underway.

- There are also USB ports for TT that need the help of a programmer to complete the possibilities.

- All this does not mean that the idea of a portable ST (beyond STacy or STbook) is not equally attractive.

But I wonder why there is no collaboration between all Atari gurus, it would be easier, faster (and cheaper) to reach the goal.

Regards.

[SainT]

- I remind the respectable colleagues of this forum that there is already a project of "Super STfm" that is underway.

- There are also USB ports for TT that need the help of a programmer to complete the possibilities.

- All this does not mean that the idea of a portable ST (beyond STacy or STbook) is not equally attractive.

But I wonder why there is no collaboration between all Atari gurus, it would be easier, faster (and cheaper) to reach the goal.

Regards.

You tend to find people do this sort of thing because they enjoy it, as soon as collaboration comes into play (especially when money is potentially at stake) things become far more complex and ultimately not so fun.

I've started doing retro hardware actually as a way of earning a living. It's hardly mega-bucks, but it affords me a far more flexible way of life (which at the moment is what I require) and is enough to pay the bills. I've found having the time to concentrate on these types of projects, and enjoy what I'm doing, is exactly what is required to get them done. With the added incentive that if I want to earn any money, things have to get done... I've so far produced the Lynx SD cart and the NeoGeo Pocket SD cart with a Jaguar SD cart currently in development.

At the moment I'm just weighing up other projects to tackle. There are a few other consoles which require the SD treatment, but I'm up for anything retro, and Atari is my favorite kind of retro. Especially the ST. The potential market for something like this is pretty darn small though I would think. However I can't help but love the idea...

A portable Falcon type machine would likely be out of the spec for this -- I'm trying to keep costs as low as possible. As you can still get a 68k compatible CPU for reasonable money, an ST portable makes sense. Getting a higher spec 68k starts getting pricey and more complex.

[Atarieterno]

It is very sad the situation that you comment.
I think there are people very capable and with the union of knowledge it would be possible to continue with a new compatible Atari machine, but finally the money sends in all the humanity and the Atarians are good people ... but they are still human.
Greetings.

P.S. If any programmer wants to collaborate on the USB for TT, I will put it in contact with the hardware developer.

[leech]

Holy crap that's huge!

That's what she said...

Sorry, I couldn't resist..

This is a brilliant idea. What I'd love to see is a Falcon Portable... one could dream. It's be far easier to do an STe though. I always wondered why someone didn't make a portable MiST. With modern batteries, that thing would probably last forever.

[leech]

- I remind the respectable colleagues of this forum that there is already a project of "Super STfm" that is underway.

- There are also USB ports for TT that need the help of a programmer to complete the possibilities.

- All this does not mean that the idea of a portable ST (beyond STacy or STbook) is not equally attractive.

But I wonder why there is no collaboration between all Atari gurus, it would be easier, faster (and cheaper) to reach the goal.

Regards.

You tend to find people do this sort of thing because they enjoy it, as soon as collaboration comes into play (especially when money is potentially at stake) things become far more complex and ultimately not so fun.

I've started doing retro hardware actually as a way of earning a living. It's hardly mega-bucks, but it affords me a far more flexible way of life (which at the moment is what I require) and is enough to pay the bills. I've found having the time to concentrate on these types of projects, and enjoy what I'm doing, is exactly what is required to get them done. With the added incentive that if I want to earn any money, things have to get done... I've so far produced the Lynx SD cart and the NeoGeo Pocket SD cart with a Jaguar SD cart currently in development.

At the moment I'm just weighing up other projects to tackle. There are a few other consoles which require the SD treatment, but I'm up for anything retro, and Atari is my favorite kind of retro. Especially the ST. The potential market for something like this is pretty darn small though I would think. However I can't help but love the idea...

A portable Falcon type machine would likely be out of the spec for this -- I'm trying to keep costs as low as possible. As you can still get a 68k compatible CPU for reasonable money, an ST portable makes sense. Getting a higher spec 68k starts getting pricey and more complex.

I would think the 030 would be fairly easy to source, and maybe even the 56k DSP, but I haven't really looked. The hard ones to come by seem to be the 060s, since they were only available in accelerators and some high end Macs, if I recall. But then again, I don't know if you'd want to put an 060 in a laptop sized system, since they are pretty huge. One of the low power variant 68000s would work great though, I bet.

I'd love a portable STe. I know exxos had decided he wanted to do the STFM because that's the most compatible system out there. But with all of the patched games out there, I'm not sure anyone would be missing much if we just went STe with a switchable 2.06/1.62 TOS.

[Ragstaff]

The 56001 isn't easy to source, but it has present-day descendants which, according to freescale's website, are "100% object code compatible" with the 56001. These are dual core DSP's that run at speeds around 300mhz. And they only cost about $10.
But making a Falcon compatible would be difficult and expensive.

@Atarieterno, I agree something feels hollow here when as a scene we have so few resources and yet many people are working on similar projects that cancel each other out. They all do it for fun, they are under no obligation for anyone, I understand.

[leech]

I think the argument can really go either way. If they banded together, they could work on a fantastic design and quite possibly get it done quicker. Or, it could turn into a battle between various egos, and then the project gets halted and split up, to possibly return years later after temperatures have cooled. (I don't know for sure what happened with Natami, but from reading the forums it kind of sounded like that's what happened, but now we have the Vampire.. sort of).

Edit: I do have high hopes for this getting done eventually though. I have two of the SD cards for the Lynx, and am definitely going to order the Jaguar one when it's done. SainT makes fine hardware! Even considered picking up a Neo Geo Pocket, but then saw it had less great games than the Jaguar (though I only looked briefly).

If you build it, I'll buy one

[SainT]

I think the argument can really go either way. If they banded together, they could work on a fantastic design and quite possibly get it done quicker. Or, it could turn into a battle between various egos, and then the project gets halted and split up, to possibly return years later after temperatures have cooled. (I don't know for sure what happened with Natami, but from reading the forums it kind of sounded like that's what happened, but now we have the Vampire.. sort of).

Yep, this would be my concern. The simple fact for me is that if I'm doing this as a source of income, collaborating with others makes things far more complex. If I set a price point, this is enough to allow me to live. If 3x the number of people work on a project and income is split, you're talking 3x the price. I can generally keep things reasonably priced for the small number of units I sell because of the simple fact its just me!

For a portable, the 68sec000 makes perfect sense -- It's designed for low power embedded systems and is incredibly efficient, easily sourced and works nicely with modern IC's being 3.3v. The 68030+ are not efficient, not easily sourced and are 5v. The best approach would be an entirely FPGA based system for something like this with a softcore like the one used in the vampire (which is very impressive!). But then this becomes expensive and less efficient again.

So for a realistic portable ST, the 68sec000 paired with a modest CPLD and small LCD makes perfect sense. You can run the new 68k's up to ~50Mhz without issues, so there's plenty of scope for making it run reasonably quickly, especially when you consider you'd be working from SD cards for storage.

I tend to work on the most realistic design possible rather than a dream-machine. It means it's likely to happen.

One question of the cart port..... it may make more sense to include additional MIDI ports on the device if this is a concern and also the main use for the cart port. It may end up that the cart port determines how thick the device is, making it much more of a brick than is necessary.

[dbsys]

One question of the cart port..... it may make more sense to include additional MIDI ports on the device if this is a concern and also the main use for the cart port. It may end up that the cart port determines how thick the device is, making it much more of a brick than is necessary.

The ROM port is essential for most MIDI based applications such as Steinberg Cubase or C-Lab Creator/Notator or Emagic Logic, because they all need a dongle to run.

On a "regular" ST, MIDI I/O is included anyway on board.

The Steinberg MIDEX device, someone else pointed you to, expands the number of available ROM ports from 1 to 4, and adds additional MIDI I/O. The MIDEX Plus device adds SMPTE I/O for synchronizing a tape recorder etc.

Here is an example, showing a C-Lab Creator dongle attached to a 1040ST:

atari_1040_stfm_midi_ports_big.jpg

[leech]

What are your thoughts on adding a higher resolution mode for this (or even better, higher color / higher resolution) for GEM stuff? Getting a 50Mhz system with a high resolution desktop would be awesome for things like Pagestream/Calamus.

[nagual]

Somehow, I'm not a guy for gaming, I always had more interests about apps, system tools, utils, programming etc...

So, IMHO, will be really nice to have something like this with all beautiful Atari software for every day work like editors, text processors, databases etc.. - no PC, no Windows, no Linux - pure GEM in all glory:-)

Let's be honest, I really don't need any kind of network/Internet, it's always complicated main idea, and MiNT is definitelly not my cup of tea, no matter what, N.AES, Geneva, even MagiC:-)

According to leech, high resolution is really must, but, I'll be really satisfied with something like 1280x1024 or 1280x720 in 16 or 256 colors, a lot of older software don't working good with more than 16 (256) colors, like Phoenix:-)))

[SainT]

What are your thoughts on adding a higher resolution mode for this (or even better, higher color / higher resolution) for GEM stuff? Getting a 50Mhz system with a high resolution desktop would be awesome for things like Pagestream/Calamus.

Yep, I already have a native resolution in mind. So 800x480 -- could be in 16 colours or possibly 256. There's no reason not to have upto 12mb of RAM as well if memory serves me right on the memory map for the ST.

Lots of possibilities.

[leech]

Hmm, out of curiosity. If there was a Tri-TOS, 1.62, 2.06 (maybe even 3.06 if we could get the TT resolutions, though I don't know if that requires an 030), and EmuTOS.

This way you could have Compatible mode, Desktop Mode, Mega Memory/Desktop/Productivity mode. I'm not sure if EmuTOS allows more than 10mb of ram (pretty sure it's 10mb, due to the address space?)

Oh, now that I think about it, I think the 68000 can only address 16mb of RAM.

[SainT]

Actually looking again, you should be able to support RAM upto $E00000 (which is where TOS resides), which is 14mb. Some lower addresses are used for the VME bus, but I'm not sure which TOS versions touch that.

TT resolutions are not possible. Support would strictly be for ST and STE chipset with extended video and potentially audio (audio in would be an easy option).

[Mindthreat]

This topic really wets the appetite... =D