Ksarul

You might want to look in the TI subforum, as I believe there was a thread in there discussing the TI Diagnostic program relatively recently. The memory tests can be loaded and run using the Minimemory, IIRC. One other thing that sometimes happens to cause problems like this: the card slot in the PEB goes bad. Try the memory card in a different open slot to eliminate that possibility (assuming you always put them into the same slot). Other than that, you've done pretty much everything to isolate the problem. . .other than testing yet another 32K card.

Nice, thanks for putting this up!

Many thanks! I'll have to set up my burner and do some tests later this weekend. . .or next, if the family doesn't leave me time this weekend. I just got a Powertran Cortex and a Marinchip S9900 in the mail. This is one really happy retrocomputing weekend for me, as I've been trying to find a Marinchip since 1987 or so, when I first found out there was actually an S-100 computer using the TMS9900. . .

I can also do that set of changes to the layout if it isn't there yet, Matthew, this is a very good idea for folks wanting to develop stuff. . .

That would come in useful, Rasmus--and for much more than testing. It would make builds for both Inverted and Non-inverted cartridges a snap, as you'd just have to run something built one way through the Java app to turn it into the other. . .

I've only seen a few copies of the Hagera books in the last 25 years, so they might be hard to find. There are about 250 copies of the Koppelmann books (I talked to him about this when he was distributing the second edition (60 copies, IIRC), which I have). The Minimem book shows up on eBay Germany once or twice a year, as do Spezial 2 and the TI Revue Assembler Spezial.

I can do that once some of them have been thoroughly tested, Acadiel. I haven't got anything larger than an 8K ROM image to work with right now. . .I'm good at designing things, programming, not so much. . .

The biggest problem with this board was that it just didn't sell well enough to keep Foundation alive, Their memory cards were pretty common, as they were the first TI card that had more than 32K, and they had a degree of third-party software support as well. When later, battery-backed cards came out, sales plummeted, and with the low sales of the Z-80 and 80-column cards, Foundation folded their tent and disappeared. . .unfortunate, as it was a great idea, for all of the reasons Gary mentioned and more.

Now that IS a sweet idea, Matthew! You could put a bunch of these on an HRD set up for 32Kc8 chips to make it nonvolatile without having to worry about rechargeable batteries too. . .you'd have to put a switch or a jumper on the write enable for the cart board to use it this way though. That's an easy change. . .and one I could implement this weekend.

Note this is for the standard bank switch mode, Acadiel--the updates to Classic99 and MESS recognize the Inverted ones (in addition to the standard, non-inverted types). I'm just not sure what the selection is to set the build to a standard type in either emulator. Tursi or Michael can give a better answer here. . .

I believe that all of them will, Rasmus, as that was the TI-standard switching method prior to the development of the 16K/64K/128K bank switched boards. They would need to support it just to support al of the pre-existing carts that switched between up to 4 banks (the only TI-manufactured cart that used four banks was TI-Calc, which is why most GRAM devices won't load it).

It is--it expects its own disk drives and it expects a terminal for the video. It will also use the Foundation 80-column card as a direct connect using the 25-pin serial connectors on the back of each card. Like the MorningStar board, it is basically a complete single-board computer inserted in the PEB that pretty much ignores all of the TI peripheral cards. Note: the 80-column card is monochrome. . .great for text applications but not so great if you were trying to do graphics.

Acadiel's already answered for the V3 boards, and I put the V4 Gerbers up here on AtariAge, so those are available for folks to make as well.

It appears that several of us on the board have one of them too, but unless OPA has the RP/M disk for it, none of us have a fully functional setup for it. . .

I have one of these cards as well. I also have the 80-column card from Foundation that works well with it. What I DON'T have is the manual or a functional CP/M disk for it. . .one note on this card--it DIDN'T come with CP/M, it came with RP/M, a CP/M workalike. I've seen a MorningStar card, but I do not have one.

The biggest problem with doing that is space, Kevan. If you're running a board with the Atmel, you have to deal with a 40-pin chip, and that eats up almost half of the board. Add another DIP chip for the memory and you have pretty much no real estate left to run the necessary traces to connect everything together. It would be extremely difficult in a TI cartridge case. It might be doable on a ROMOX ECPC footprint though--but cases for those are a bit more difficult than TI cases. I've been toying with the idea of doing a new version of a ROMOX-like modified case though, as it will facilitate some other ideas I have.

That's a very good idea Acadiel--and I can always use the leftover GAL-capable boards for GROM-only multicarts. . .waste not, want not.

I don't use cell phones, but this is indeed a bit of good news!

There are several. There's an Assembler Special from TI Revue, another from 99-Journal, Assembler Kurs I, II, and III from Hagera (Rausch & Haub), the Editor Assembler Handbuch from Wickanderware, TMS9900 Assembler auf den 99/4A by Simon Koppelmann, and TMS9900 Assembler Handbuch für das MiniMemory by Rainer Bernart. There is also some useful stuff in the 99 Special II volume from Texas Instruments. There are a couple more in French, if you can read that one. . . I have most of these (I'm missing Assembler Kurs I).

On the V7 and V8 boards, both of them were GAL variants, with the V7 using Tursi's original spec for the Atmel pin-out, and the V8 using an updated pin-out. A small number of an earlier iteration of the V6 boards using the original Atmel pin out were also produced (20, all used as test articles). When the GAL on the V8 boards (no V7's were ever produced, but the layout is complete) exhibited issues, I went back to the V6 board and updated it to the new Atmel pinout and made a few other board optimizations we'd incorporated into the V8 boards. This generated the V6H boards (I made 50 of those, and I've used about half of them as test articles). Of the V4 boards, Acadiel made 4 of them and I made another 23, of the V9 boards, I made 52. There are several hundred of the V3 boards in circulation. I also incorporated a few lessons learned into the V9 boards (and in my current iteration of the V6 board), as I've shaved a very small amount off of the cartridge connector to make it a little easier to insert and added a bit of an angle to the connector corners to make it seat more easily.

Here's something that will clarify a lot of things with the Über-GROM, the WIP manual. The GAL is primarily to eliminate the issue with start-up banks exhibited by the 378 or 379 chips, as you surmised, Gazoo. Your solution would probably work quite nicely as well, it just hasn't been tried. 512K ROM-GROM Cartridge Board Manual V0.7.doc

Correct on the buses--but they also used some nonstandard memory chips in a lot of their cartridges (GROMs). The regular ROMs were also in a lot of them, so you had boards with the appropriate set of pads set up to install a ROM. You also have the option of getting one of the after-market cartridge boards of the 64K or 512K bank-switched variety and using that as your test board instead. I should probably build one with a ZIF socket on it for my own cartridge experiments. . .

Great job, Stuart!

It's either that, or try to find one of the beige Speech Synthesizer cases, Tempest. Then the two will match (although they also have a different shape than the regular synthesizer). I may have to check a Hex-Bus case to see if a regular synthesizer fits in it and still lets you connect the rest of the peripherals. . .as it is the same size as the beige synth case.

I haven't found a good switch with the right spacing, Acadiel--I just found some nice small switches that I could attach to the case at the end of some wires, which also allows for all kinds of creative switch placement. On the black ones, I just did a really small batch of them as a test run to see how this vendor would work out, but even there the price per board was really good--about $7.50 each, shipped, for a group of 20. I made 50 of the red boards, to see how much of a difference a medium-sized run would make of the cost--and it cut it to under $4 per board with the enhancements I ordered. (Red solder mask and gold contacts). The red boards are a cheaper build than the black ones too--they have the same component count as the 64K boards. These make an excellent evolution for the 64K board, once the ROMs are reversed--and it gives a LOT more space. It also lets folks build and test the ROM side for one of the Über-GROMs if they want to build a massive multi-cart with an integral XB variant and a handful of other GROM carts in the other GROM slots. I'm hoping to finish my tests on the Über-GROM boards this weekend. . .so Atrax can get his test board back. And you've been part of these board projects since the beginning, along with Tursi--those were sent as gifts to recognize your contributions to the project! I would have never begun doing cart board layouts if it hadn't been for you letting me look at the original 16K boards to start playing with--and I wanted the 379 to fit in a cartridge case even when it was socketed (the engineering issue I was trying (unsuccessfully) to explain in Chicago last year).