Balance of Values - A Plea for the Next TI-99 Supercomputer

[Switch1995]

Long time lurker, first time poster.

 

I just signed up for an account yesterday so I could join the feeding frenzy for the Extended Basic v2.7 Cartridge (unsuccessfully). The demand for that product surprised a lot of people and it got me thinking about the values we place on our hobby. Namely, "Nostalgic Value"; "Functional Value"; "Collector Value"; and "Actual Value/Cost".

 

I feel this product was so successful because it had a very good balance of these values. It was housed in an actual cartridge case (Nostalgic); it made programming easier (Functional); it didn't exist before now (Collector); and was only $40 (Cost)

 

You could dip a Parsec cartridge in gold and "maybe" a die-hard collector would want one, but that is where the demand would stop because the other values would be out of balance. We've all seen the TI-99/4a on Ebay for $800+ graded inside an acrylic case

 

Now, to the point....

 

Please check out this link:

 

http://www.cnet.com/how-to/25-fun-things-to-do-with-a-raspberry-pi/

 

The Raspberry Pi's dimensions are 3.35" x 2.2" and cost around $40. In comparison, a TI-99 Command Module dimensions are 4.15" x 2.75".

 

If you search "Raspberry Pi" on Ebay, you can see the enormous amounts of ready made peripherals that are already on the market.

 

I'm excited for the Geneve 2 to come to market....I will probably try to purchase one if I am able to. However, I ask the community, for the next "supercomputer" could the values be more balanced? When you work with a Geneve, you gain tremendous Functional value, but you lose a great deal of Nostalgic Value...The very symbol of why so many of us are in the hobby....and why so many will take up the hobby in the future. Hint: Texas Instruments sold 2.8 million of them.

 

Theoretically, if you removed the TI-99/4a motherboard and replaced it with a Raspberry Pi running an emulator mounted on a simple daughter board could we have the best of both worlds? Wireless USB keyboard/mouse, internet browser, massive storage, etc. but still maintain the ability to use the console keyboard, joystick port, and Cartridge port with physical cartridges?

 

As with most vaporware, I don't have the skills to know where to start, but if you looked at the link above, these guys made a keyboard out of beer cans. I would like to think our simple keyboard would be a step up!

 

Thanks for reading.

 

Bill

[Gary from OPA]

There was a thread already here a while ago, about a video that was released that shown a Raspberry PI inside a TI99/4a case using the actual ti99 keyboard for it to operate.

 

Sadly with that project, the current systems that Raspberry PI can emulate, the TI99 is not on list, and the project only emulated the actual ti99 keyboard, it didn't do the same with cartridge port or joystick port.

 

I would like to see a similar project that would allow a way to replace the original ti99 motherboard and use existing real cartridges, and even add-ons like the nano peb' or actual ti99 peb flex card interface.

 

A Geneve 2 has been so long coming it is almost like it will never happen, and again you need a PEB and those are hard to find compared to the ti99 consoles, which even limited the original Geneve 9640 market.

 

Personally I would like to see a new hardware to use an actual 9900-based chip like 68689 or a licensed FPGA core of the 9995, then using a emulator design running on arm microprocessor.

[Switch1995]

Thanks, Gary. The expansion peripherals for the Pi include a breadboard. That might be a cost effective way to add a vintage processor?

[Ed in SoDak]

Welcome to the forum! Forward thinkers are continuing to drive development for the TI and other great vintage computers. I'm no way near what others are doing, but I have my hands in the TI99, the Apple II and Timex/Sinclair in a round-robin fashion. I'm using both hardware and emulation, porting programs back and forth just using the minimal hardware and mostly what I already have laying around.

 

Breadboards are handy for prototyping small circuits or building a limited one-off add-on, but they aren't really well-suited to anyone hoping for a more plug'n'play approach to hardware upgrades. Perhaps someone proficient in the architecture of the Pi and advanced TI chipsets could potentially design a daughtercard that would insert into the breadboard, but the complexity of what would be required to "emulate" a TI in hardware on a Pi would likely tax the limits for power as well as sheer size and scope. 40-pin chips plus support circuitry do not translate well to a single plane and tiny breadboard. TI did not make it easy for outsiders to make major changes to any of its facets, this fact has hindered the TI since birth.

 

Maintaining backwards compatibility to retain the nostalia also crippled the Apple II. Even the last and best IIgs continued to require that old 140k disk-based programs be used via a 5.25 floppy drive, even though the IIgs might have both 3.5 drives and a hard drive attached. All was done to maintain "backwards compatibility."

 

I considered buying a Geneve when it came out and was finally gaining some support that hinted to me it might stick around awhile. I was very serious about my TI in those days and did all my computing on it, which was basically everthing I still do today. But when I took serious look, the $500 investment for the card alone was money I felt would be better spent going to a more capable platform. In my own case, I changed over to Mac, but I never forgot my roots and the TI has come along for the journey as much as I could muster via emulation on the Mac and connecting the two physical machines.

 

It's a great germ of an idea that could take root. Maybe not quite as you envision, but these little Raspberry machines and their ilk are bringing these old dreams of a "Super TI" closer to perhaps being more than a pipe-dream. There's a lot of us out there still exploring the possibilities of this vintage hardware, no matter which brand(s) you happen to favor.

 

Just keep making history. It gives historians something to do!

 

-Ed

[+mizapf]

When you work with a Geneve, you gain tremendous Functional value, but you lose a great deal of Nostalgic Value...

 

 

When you ask Geneve users, they would disagree. At least I do; I never felt like losing anything when I switched to the Geneve back in 1990. At that time I had my TI console for almost 8 years.

 

In contrast, because of the good compatibility, everything I had before was still working; no more cartridge port issues; much faster; better graphics; mouse support ... Not a single minute of regret. From then on, the Geneve was my platform, and there is lot of nostalgia towards that one.

[TheMole]

 

When you ask Geneve users, they would disagree. At least I do; I never felt like losing anything when I switched to the Geneve back in 1990. At that time I had my TI console for almost 8 years.

 

In contrast, because of the good compatibility, everything I had before was still working; no more cartridge port issues; much faster; better graphics; mouse support ... Not a single minute of regret. From then on, the Geneve was my platform, and there is lot of nostalgia towards that one.

 

I still regret not having been able to get a Geneve back in the day. I actively used the TI as my main computing platform until sometime around 1996 (although we did get a 486dx around 1992, but that was the family computer, the TI was mine). It would have been a very natural feeling upgrade for the TI back then, so I'm sure I would have felt just as nostalgic about it as I do about the TI now.

 

Hell, I'd pay a pretty penny to get my hands on one today...

[+mizapf]

Theoretically, if you removed the TI-99/4a motherboard and replaced it with a Raspberry Pi running an emulator mounted on a simple daughter board could we have the best of both worlds? Wireless USB keyboard/mouse, internet browser, massive storage, etc. but still maintain the ability to use the console keyboard, joystick port, and Cartridge port with physical cartridges?

 

<heretic>Theoretically, why should you want to do that?</heretic>

 

Of course, I do understand the motivation why to put a Raspi inside a console shell, but thinking a bit further, what is the ultimate goal? There are many emulators out there for all kinds of configurations; as one of the MESS maintainers I won't even try to look neutral in that sense, and I also got used to work on a virtual TI. Maybe this is also a result of using the Geneve; I got weaned (had to look up that word ) from my TI console 24 years ago.

 

If you envisage to loosen the constraints of the TI-99/4A, you eventually get another computer. Not a good thing in nostalgic terms. It will make all shortcomings of the good old TI all too plain to see ("this would have never been possible with it" etc.).

 

There is nothing to say against having e.g. a USB interface for storage (because it still looks like a device), but when you change things like speed, video, audio, what we could call inherent properties, you get a new platform. The Geneve is a platform of its own, with compatibility to the TI console. I would also call an F18A-enhanced TI console an own platform once its specific features are used (beyond the simple video output).

Edited September 29, 2014 by mizapf

[Switch1995]

Not sure who maintains the Cafepress Store...maybe this could be a product with some Hi-Res TI-99 graphics on it?

 

https://www.modmypi.com/raspberry-pi-cases/raspberry-pi-case-branding/raspberry-pi-case-branding-and-customisation

[Omega-TI]

 

If you envisage to loosen the constraints of the TI-99/4A, you eventually get another computer. Not a good thing in nostalgic terms. It will make all shortcomings of the good old TI all too plain to see ("this would have never been possible with it" etc.).

 

There is nothing to say against having e.g. a USB interface for storage (because it still looks like a device), but when you change things like speed, video, audio, what we could call inherent properties, you get a new platform. The Geneve is a platform of its own, with compatibility to the TI console. I would also call an F18A-enhanced TI console an own platform once its specific features are used (beyond the simple video output).

 

I agree with you about eventually, "you get another computer".

My personal philosophy is this, "If it's hanging off the ORIGINAL console, you're good to go!" (Other's may think differently).

 

For an explanation; Many of us expanded our TI's in the old day's, and we still considered them TI's. Did a PC become something else when you replaced the CGA video with an EGA or VGA? Did adding a hard drive somehow turn your XT into something else? I don't really see the F18A changing the TI into a new platform anymore than the 9938 or 9958 based cards that people installed.

 

I'm sure that 'back in the day' if someone could have figured out how to get 64K of RAM working on the TI, many of us would have jumped on that MOD without a second thought and would still consider it a TI afterwards.

 

To get serious here for a minute, I really believe that there is a closing window of opportunity for new hardware on the TI. The generation that grew up and loved this machine are getting older, many have already passed away. When those with the technical skills and knowledge of this machine are gone, who will care about keeping them running or expanding it's capabilities? I don't know if it's 5 years, 10 years or as many as 15, but our time is limited. I guess that is why I sometimes seem a little 'overly motivated'. I see the people dying off and the writing on the wall.

 

Nothing makes me happier than to see guys like Matthew Hagerty or Greg Allen Warner step up and add a new piece of hardware to the lasting legacy of the TI. As our machines get older, some of the old hardware is going to die too, these new items may be our last hope to keep things running. I've already read a couple of threads where older TI's had video that went TU and the F18A saved the day.

[jstimson]

What I would love to see?

 

A production of a TI 99/8 as it was meant to be. That would give use a machine with 99/4a software and hardware compatability as well as an evolution in the machine (at least as TI saw it at the time).

 

I understand that emulation can manage this, but I love working on the real thing. I much prefer using my real TI with floppies as opposed to using a TI emulator on my PC.

 

Heck, even if somebody gave us something that behaved like a 99/8 on the outside (cartridge port, expansion port, etc) but was implemented with modern technology in the inside I would be happy.

[+hloberg]

Take one TI-99/4a. add a f18a card, a NanoPEB + HDX or PEB with HDX. throw in a XB 2.7 suite cartridge, RXB (when comes out?) or a XB + E/A + (pick your disk manager) on a Navarone cartridge switcher. Over clock with a new crystal and bingo! you now have as much if not more power than the 99/8.

And if you want to get really wild mount the whole thing in a PC case with a wired keyboard.

[jstimson]

Take one TI-99/4a. add a f18a card, a NanoPEB + HDX or PEB with HDX. throw in a XB 2.7 suite cartridge, RXB (when comes out?) or a XB + E/A + (pick your disk manager) on a Navarone cartridge switcher. Over clock with a new crystal and bingo! you now have as much if not more power than the 99/8.

And if you want to get really wild mount the whole thing in a PC case with a wired keyboard.

 

Except the 99/8 ran at 10Mhz as opposed to 3Mhz in the TI, and could address 15 Megs of memory with XB II being able to handle it. I appreciate the add-ons and upgrades the old 4a has received over the years, but they are still coping with existing limitations in the machine hardware and architecture. Throw me a machine with complete backwards compatibility and give me the upgrade that we all wanted in 1983 and I'd be more than happy.

[+Ksarul]

Well, as Rich noted, RXB does handle larger memory spaces, and a HyperAMS card is expandable to 15MB, so there is a way to access huge memory spaces with your 99/4A. Don't get me wrong--I love having my pair of 99/8s, but it is possible to get a lot of their capability into an expanded /4A.

[+mizapf]

Wait ... this fictitious machine reminds me of something I saw recently ... where was it? Ah yes, there it is, on my desk. The name's Geneve, Geneve 9640.

 

12 MHz TMS9995, v9938 video, 640 KiB banked RAM, PC keyboard, mouse, ABASIC, DOS-like command shell, cartridges on hard disk, 99/4A compatibility mode.

 

If we talk about a "supercomputer" I would say we should not start from the 99/8 that we never had (except for some prototypes) but from the Geneve that we (or some of us) actually have. Unfortunately, by now most of us will be able to get in touch with both 99/8 and Geneve by emulation only.

 

---

 

Sorry, could not resist, don't want to bore you with my praising, but this is what I just feel when I hear that we'd like to have such a kind of "Super-TI". We had it, and we were just too few to exploit its potential. What would be the fate of a new computer of that kind?

 

---

 

Nevertheless, it could be interesting to build a Super-TI based on emulation. I'm still dreaming of a TMS99105-based system with a v9938 or v9958. But before I would have to add the 99000 family to MESS.

Edited September 30, 2014 by mizapf

[+Ksarul]

That's not a bad idea either, Michael--although you might want to base it on the 99110 to get the extended math functions in the macrostore of that chip. . .

[Omega-TI]

 

We had it, and we were just too few to exploit its potential. What would be the fate of a new computer of that kind?

 

 

Exactly, spot on. I believe the huge expense at the time for one single item is part of what moderated their numbers. These days I believe the most viable expansion method is a series of individual and affordable cards or modules. After all this is a hobby, it's not a primary computer or a necessity. From a pure marketing and human nature aspect, more people are willing to take a chance, part with discretionary income or do an impulse purchase under a set psychological limit. In this way they can 'personalize' their systems to suit their tastes.

 

Many of the projects and upgrades people have purchased prove this out, F18A's, HDX's, Nano-PEB's, Cartridges, and other items. Odds are, if it were not for those reasonably-priced items, many would never have upgraded at all, and their interest may have waned. Look at the excitement that was generated over that one XB v2.7s cartridge, it even brought out the lurkers. Now once you start buying things, you always want to go a little further, get one more item and make your system just a little better. You know the old saying, "In for a penny, in for a pound". After a while, a person looks back and say's, "How the hell did I get so much stuff!" One piece at a time!

 

I'm looking forward to gregallenwarner's new speech in the box card, as are others. I'm also sure that there are other's out there lurking, waiting to pounce on the next NEW item that they can afford to plug into their TI. What will it be?

 

 

As a community, how are we going to keep people once they come in to take a peek?

[+hloberg]

well said

[+Gemintronic]

As a community, how are we going to keep people once they come in to take a peek?

 

Port ZX Basic and have at least one resource for producing carts? I'm thinking the symbiotic relationship the Atari 2600 has with AtariAge and batariBasic.

[+mizapf]

One piece at a time!

 

Right. This is certainly feasible and nothing to say against. I'm just cautious about dreams of new "disruptive" evolutions. They sound too good to be true at first, but then either do not get anywhere, or if there is something in one's hands at the end, don't find a community for it.

[Omega-TI]

 

Right. This is certainly feasible and nothing to say against. I'm just cautious about dreams of new "disruptive" evolutions. They sound too good to be true at first, but then either do not get anywhere, or if there is something in one's hands at the end, don't find a community for it.

 

I agree, vaporware is always a possibility, as is a failed product. Hopefully the builder would ask if anyone is interested or take a simple poll to judge the market and get some input before proceeding . On the whole, I think the majority of the items seem to have worked out fairly well.

 

Now days some of the standards are not as important as they once were. One example is the storage device, since the major vector for dissemination of software is no longer diskette, inter-compatibility is not as important, rather that the device works seamlessly with the computer is. So now we have a myriad of different storage methods that suit the individuals needs, but all can still enjoy the TI. Personally, I'd love to see a disruptive evolution in an affordable SSHD myself, but until then, thankfully, there are other alternatives that can fill the gap.

[matthew180]

What I would love to see?

 

A production of a TI 99/8 as it was meant to be.

The problem is, TI never finished the 99/8 and you will never get a consensus as to what the final hardware configuration would have been or what the firmware would have done.

 

Heck, even if somebody gave us something that behaved like a 99/8 on the outside (cartridge port, expansion port, etc) but was implemented with modern technology in the inside I would be happy.

Doable, but requires detailed technical specs, schematics, and firmware for the original system.

[matthew180]

I agree, vaporware is always a possibility, as is a failed product. Hopefully the builder would ask if anyone is interested or take a simple poll to judge the market and get some input before proceeding . On the whole, I think the majority of the items seem to have worked out fairly well.

This has been tried, and as it turns out everyone wants something different. Being somewhat pessimistic tonight, just because you build something new does not mean anyone will use it. More RAM seems to be one of the biggest commonalities though, i.e. some insane RAM card (64MB or something) using the standard AMS (or SAMS?) bank-switch scheme, or whatever. However, even if someone built that, I would be willing to bet, doing-the-hardware, that no one would write new software to take serious advantage of it.

[Omega-TI]

This has been tried, and as it turns out everyone wants something different. Being somewhat pessimistic tonight, just because you build something new does not mean anyone will use it. More RAM seems to be one of the biggest commonalities though, i.e. some insane RAM card (64MB or something) using the standard AMS (or SAMS?) bank-switch scheme, or whatever. However, even if someone built that, I would be willing to bet, doing-the-hardware, that no one would write new software to take serious advantage of it.

 

You're right of course, everyone seems to want something different or has a different vision of the way "it should be", which is another reason I believe smaller, less expensive, 'function specific' items sell better. You are right again, just because someone makes something does not mean anyone will write something for it. One example of this, the F18A, it's been available for some time now, and if I remember right more than half of the people who hang out here use one, it's become the new standard, but I've yet to see a single F18A specific program or game that exercises that devices potential.

 

We all have visions, dreams, ideas and even desires for our little hobby box, the reality is not everyone shares the same vision or enthusiasm. All we can do is hope that someone out there in our small pool of users does shares our vision. Sometime the wait to see the fruits of someone else's vision is agonizingly long, if at all, and then sometimes we get lucky, like with Greg and his speech-in-box card.

 

In the end, sharing our ideas with each other builds excitement, it opens others up to the possibilities and might just infect that one person capable of making it a reality. The journey we take along the way is a large part of the fun, who wants to squander that?

[gregallenwarner]

Necessity is the mother of invention, so as I see it, a better question than, "What features do you think should be in the next TI?" would be, "What is it you wish you could do on the current TI but can't?"

 

Then we can hone in on the most critically needed upgrades.

[Bill Loguidice]

What I find fascinating is that a similar discussion has been raging for YEARS in the CoCo community about a possible "CoCo 4" to the point where not much has gotten done. As was stated here, it's almost impossible to get everyone on the same page of what the features should and should not be. Most efforts that actually amount to something are around FPGA recreations of the original specs, but even that lacks sufficient enthusiasm for relatively small communities like the CoCo's or TI's to properly rally around. I think in the end, what most people really want are just ways to make the original hardware more useful, since it's the original hardware where the main point of interest resides, i.e., once you start to get too far away from the original capabilities you have what amounts to a new platform without the built-in nostalgia value. To put it another way, once you start making significant changes, you start to splinter an already small community.

 

One thing that I always point to that I find fascinating and might be a starting point (though is admittedly yet another idea in the mix, further diluting things), is doing something like the Chameleon: http://wiki.icomp.de/wiki/Chameleon . It not only plugs into a real C-64, extending its capabilities, but also works by itself as a stand alone C-64 computer. It's those kinds of projects that the larger communities have in relative abundance. To further tie back to the C-64 community, efforts are underway to recreate the C-65 in an FPGA, which could also be done for the TI-99/8. Both are the true next gen successors (ignoring things like Geneve) to the main hardware of interest that few people will ever get access to.