Hardware Limits for Tms based Computers and/or the 99/4A mainboard

[kl99]

I am curious to discuss possibilities, theoretical and practical, about what could be changed/improved while staying with the 99/4A mainboard.

And second what are the possibilities of a new Tms based Computers.

 

We have seen the efforts that went into 99/8 and 99/2, the 99/4B, the 99/5, the 990 mini computers, the geneve, the 9995 breadboard by Stuart, the SGCPU by Snug, the TMS99105 fpga based computer by Erik.

I am wondering on what could be achieved, if we are free to redesign everything around the CPU family.

 

As far as I read it, TI did decide for many compromises when designing the 99/4A. We only have an 8 bit bus to the PEB. Most memory and devices are connected via 8 bit.

Many Cpu signals are not passed onto the I/O Bus, or the signal is inverted.

 

Afaik all the Cpus have an address range limitation of 64KByte. There are memory mapping possibilities that could bring us 16 (like the 99/8) or 64 MByte (like planned on the geneve 2).

 

What would the best TI computer have as specs and can we create it?

 

This also goes a bit into the direction when does it stop feeling special to all of us, when does it stop feeling like a TI.

 

So some questions I can think of to start:

 

Which Cpu should we pick? Will it be Fpga based to be able to overcome limitations that even the best Cpus have?

Can we do a full 16 bit data bus and design a PEB using that or could we even use the existing PEB for a 16 bit data bus by switching to another interface card?

How do we deal with the Grom approach used by most 99/4 software? We could only reuse the Grom0-2 chips from the 99/4A in higher numbers, but that would introduce dependencies to the grom system software part of the OS.

Would the F18A be the video chip we want? Afaik the only limitation in comparison to the 9938 and 9958 is the video ram size.

What would be the advantages/disadvantages or the VDP to be using Cpu Ram (like the 99/2).

 

How would the memory map look like? I always feel a bit disappointed seeing on the 99/4A so much memory space is permanently assigned to Cartridge space, Console Rom, DSR Rom and the user (or an user program) can not use all 64k memory for a certain program.

Can we copy the system rom from an Eprom chip into the RAM when the computer is turned on? Or is there a nicer approach?

Are CRU based DSR the thing to go for? Or is there something smarter already?

 

To also discuss the 99/4A topic:

I am thinking on things like replacing the System ROM by fixed versions. Providing some Grom 0-2 replacement pluggable circuit. I was amazed by Matthew on his F18A. He stayed compatible to the mainboard while suddenly providing the 16K of Video Ram from his FPGA board, providing additional 2K of ram and providing direct access to the registers and video ram for the virtual Co Processor, a Tms9900. Or think of the solution that Erik Piehls showed where we attached his device to the I/O port and simulating a 640KByte cartridge (i think it was the XB2.7 suite) with having zero at the cartridge port and with having zero mods done to the mainboard.

 

This intro might sound a bit fuzzy but so seems to by my knowledge on this hardware-dominated topic.

[+9640News]

If you want to maintain the interest of this community, any real hardware would need to be compatible with with minimally existing TI-99/4A software. Compatibility with the Geneve would be secondary. The concern I have would be are there enough programmers in this community to have a desire to write anything to use the extra features for a new computer. Maybe some would modify some existing programs, but how much new stuff from scratch? I get the perception people are pulled pretty thin right now with the current projects that are pushing through.

 

From where I sit, this community has seen some recent projects be successful. Not sure if any have exceeded the 100 or 200 unit marks, but still successful and very well received. So, this begs me to question if the hard $$ cost for any design would allow someone to break even on something that would likely be much more expensive. It is one thing if you are doing this as a personal hobby/interest, but if your intent is to make it available to others and if your satisfaction is going to be gauged by how many users commit to getting a unit and using it, be careful As an example, there is someone on here, do not recall who, that has a redesigned motherboard I believe would go into a PC case with loads of features that was TI-99/4A compatible.

 

It is something I would purchase but not sure where it stands as of late as I have not heard any reports in the last couple of months. What that guy did, and I do not recall his name, was develop what looks by pictures to be an amazing piece of hardware he should personally be satisfied in what he accomplished. Now, regarding any commercial introduction, if that was his intent to cover his time and efforts, he will likely be hard pressed to cover all his costs. I gather though this was a project for himself with the primary goal, and offerings to others were secondary.

 

Something I have thought about, and I do not know what the specific limits of MESS or MAME's use of the source code may be, but if you are thinking about a greater computer, design it in emulation mode first. Cut out all of the MESS or MAME source fluff for hardware you would not use. Then, add or expand video processor capability with more memory, more graphic features, etc you want to explore in your hardware design. I'm guessing there are plenty of pieces of code one could use out of another piece of hardware source code in MAME/MESS to make that all work. Once you have a working emulation, and you know how things are supposed to work, I would think a hardware design then would be appropriate. You could also gauge people's interest if you had some neat demo software for it without having invested in the hardware and prototype costs which is real $$$ out of the pocket.

 

For example on mods, expand the 9938 to have 9958 capability where more displayable colors are an option. Also, you could tie in with the 9958 the ability to "bring in" an external video source. This would then give the ability to have an emulated TI-99/4A with 9958 capability with more video capability perhaps even streaming video. You could add some newer hardware for mass device storage whether it be USB or some other option.

 

Something with more speed, more CPU memory, and more Video memory may lay the groundwork for a web browser.

 

Anyways, just throwing out some thoughts. Minimally, I think such an emulated system would need to be Raspberry PI and Windows compatible in the software. Probably lesser so with a Mac version.

 

I'll be following the thread. Just my penny's worth.

 

Beery

[kl99]

It's Fabrice Montupet and the TI(ny) 99/4A project you refer to. He did it solely for himself. People asked if they can have one of these once he did go public about it, but that was not his intention.

He achieved to analyze the 99/4A down to the last resistor and recreated the schematics based on his findings which revealed a few errors in the schematics we know.

 

To be honest I am not interested in money. Nor do I feel capable of realizing such a project on my own.

Even if the result of this thread is we are having some consens on some specfications and people can learn more about the inner workings of TMS computers, it is totally worth it.

And with FPGA as hardware option there are no big expenses I see coming.

 

First step would be having some specs, that can be picked up by anyone interested.

 

I am not so keen about extra features but rather getting a clean design where we no longer have waitstates or unnecessary compromises.

A computer where one can switch between original TI gaming speed and maximum speed like intended with the 99/8 and 99/5.

A computer where one can load the system like a 99/4A, or use the whole 64K RAM for his own program.

We have two successors that got produced and some which were prototyped. They all have one thing in common. They are no longer produced/available.

Basically if there is one to be produced, which features should it have.

 

Instead of big new features on hardware side I rather feel like on the software side we will have new ways of doing things.

Using the internet to search and load software/data into the memory and stuff like this.

 

If we would have a 16 bit PEB connection it would be a solid basis for every new hardware project.

[Omega-TI]

 

This also goes a bit into the direction when does it stop feeling special to all of us, when does it stop feeling like a TI.

 

 

To me (other opinions will differ), if it plugs into and supports the TI-99/4A it's all good, but I personally have ZERO interest in supporting a new system built around an old CPU that carries no nostalgic value for me.

 

Even IF a system like that interested me, the amount of money I would have to invest for a system with arguably little support, software variations and a lack of computing power (by today's standards) would be a deal breaker. I'm just not a hardware or software guy.

 

I also don't have the time to split my attention down yet another path, so I'll stay focused on my cherished 99/4A and continue to support it with as much of a laser like focus I can still muster,.

[+adamantyr]

The only other project I'd like to see is maybe a "TI-99/8 reborn" kind of thing, where we put an F18A in the design, use the TMS9995 processor, and maybe add a more advanced sound chip into the framework. Essentially start with the prototype and finish the job.

[Keatah]

Anything much beyond the original is not interesting to me.

[Casey]

I agree with the post above - if there was a machine that gave me the chance to use some of the prototype designs, I'd be interested. (Perhaps a machine that has all of the 99/x series ROMs available and the user can select what they want the computer to be) - but a brand new design based on the 9900 wouldn't really spark my interest.

[+mizapf]

I'd be really interested in a design with a 99105 or 99110, and then write a multitasking OS for it. Ah, wait, I forgot that this takes a lot of time. Anyway.

[fabrice montupet]

Beery Miller, thank you for your compliments :-)

 

 

It's Fabrice Montupet and the TI(ny) 99/4A project you refer to. He did it solely for himself. People asked if they can have one of these once he did go public about it, but that was not his intention.

 

It is not exactly what I said.
The TIny-99/4A is a project that I had in mind since many years but the project was constantly delayed due to lack of time (I was working on other projects that I had to finish). When some people asked me if it was possible to buy the computer once finalized, I said that this project was a personal project made just for fun, I love the 99/4A and I wanted to create the computer version that I dreamed to have in 1983. However, I have also said that If someone also wanted to play with and wanted to get it, I would be happy to propose them. I am not interested in money too. However, it must considered that this computer has anyway a price: Even I will give the hundred hours I spent debugging TI schematics and to design the computer (and I assure you that it took a lot of my time), the parts have a cost: The TIny99/4A only uses 1983 components and new ones (NOS). Due to the numerous components on the board, the soldering time is consequent.
The TIny-99/4A v1 is finalized since several months, it works fine and I use two of them very regulary. This computer was a first step, in the topic of this 99/4A forum dedicated to the project ( http://atariage.com/forums/topic/266585-tiny-994a-computer/) , I announced that I was working on the TIny-99/4A V2 which brings many other (interesting) features. The V2 is also finalized since some weeks. I have ordered 5 first boards to my PCB manufacturer. I am waiting their arrival.
You can see a 3D view of the Tiny-99/4A v2 here:

 

EDIT: A picture added:

Edited April 25, 2018 by fabrice montupet

[iKarith]

I can't say I'm interested in a new machine with none of the limitations of the existing TMS machines, mainly just because as Omega said, a new system to cater to this old processor just doesn't have the draw of a nostalgic machine like the TI-99/4A. Where you could go with it that would interest me is a board that that maintained compatibility with the TI-99/4A but made it more portable particularly in the case that you wanted a pretty heavily expanded system you could just pick up—and possibly a way for new people to have a lower cost of entry into a more tricked-out system. Enable a FinalGROM with a switch if there's no other cart plugged in. No need to plug in 32k, and it's fast RAM unless you disable that or turn it off entirely. Want to speed up the console? Easily done. And both card edge for PEB and pin connector for TiPi. Larger RAM upgrade possible too! F18A features standard would also make sense. Probably would be a couple hundred dollars, but that's still less than all of those upgrades would cost, especially the ones that'd require a PEB.

How much of this to do in FPGA vs. real hardware … I dunno, I like real hardware but if an actual TMS chip becomes a difficulty in making it affordable, I don't care if the whole thing is FPGA-based honestly. For me the whole point of this would be to make a better all-in-one system, or all-in-one-if-you-wanted-it-to-be. It could run directly of a modern power supply without needing to modify anything, and I could see some of those 60% keyboard designs being popular to complete the system, albeit at a high cost if we're using one of the ones I was just looking at.

[kl99]

Our local TI-99 Usergroup in Austria started talking about making a TI-99 Laptop.

While one guy prefers to cut an original 99/4A Mainboard using the F18A and a nanoPEB, I would prefer to design a dedicated PCB. Doing that would simply allow learning a lot more from such a project.

From this (still not working) TI DS990-1 Commercial Computer I am learning so much about Hardware, Diagnosis and Electronics.

And I am keen to do more in this direction.

 

From any design it would be key to maintain compatibility with existing operating systems (Geneve, 99/4, 99/4A, 99/8, 99/2, TI-990).

I am really suprised there was so few interest shown in this thread.

[Nick99]

I'm more into a complete Geneve-system that don't need a PEB, with some modern features like USB-support for keyboard, mouse and storage and HDMI video output...

I guess that the Geneve could have had more success if it didn't needed the PEB from the beginning, but I also understand that it was considered an upgrade of the TI-99/4a.

 

And I also wish someone could make a Pro Tracker music making program, so I can make some serious music making on my favourite computer!

 

Well, that´s what I would like to see...

[+mizapf]

The Geneve was not an upgrage of the TI console, since it replaced the console completely, but from the view of the TI user at that time, it absolutely made sense to design it as a PEB card, since all the peripheral system that you had was the PEB with its cards. Imagine we would not have been able to re-use the PEB cards...