ClausB

There was Castle Communications on E. Michigan in Lansing in '82 or '83. I knew Rob there - he was at New Dimensions in Computing (on Grand River in East Lansing) before that when he got me into Ataris.

It looks like one of these: http://www.myatari.com/ebay/48k400.pdf Thanks for posting the photos!

I designed the original upgrade (see: http://www.atariage.com/forums/index.php?s...t&p=1481893 ) and put it into the public domain. ICD commercialized it but I had no relationship with them. I was pleased to see it on the market but I hated the name (thought the Rambo movies were stupid).

Actually, I covered the inside of the 400's shielding with electrical tape to prevent shorts. It was pretty tight. Axlon was kind enough to provide a spare IC position on the board so I could add the extra selection gates.

No, $E000 thru $E7FF is free space on the 5200 and the required signals to decode that space are brought out to the expansion connector. See: http://www.atariage.com/forums/index.php?s...t&p=1580659 and: http://www.atariage.com/forums/index.php?s...t&p=1578682 So I guess that implies that Atari made a Votrax expansion module for the 5200, at least as a prototype. Anyone know more about this?

No offense taken. Just correcting your misinformation.

That's just not true. The cartridge address space is completely defined as $8000 thru $BFFF. Plus there is another space at $D500 thru $D5FF (mentioned above) where you would map control registers for bank switching. There is no "colliding". Also not true. I designed the "Rambo" on an 800XL and had no intention of making it work on the 800. The 800XL expansion port could have hosted the upgrade but it was easier to put it inside the computer, in place of the original RAM, than to have circuit boards made for external connection. This is not another one of those mine vs. yours threads!

Warerat, is it too late to suggest a solderless version of your upgrade? I think the only thing you would give up is the separate ANTIC access to extended RAM, a rarely used feature also missing from many popular XL RAM upgrades. Many potential users would prefer not to solder connections to their 800's motherboard and CPU board. Among the factory connections on the Personality Board Slot (PBS) and on RAM Slot 1 (RS1), you have the entire address bus. You have A12 thru A0 on both slots, A15 on the PBS, A13 on RS1, and you can regenerate A14 using the select lines on both slots, -S0, -S1, -S6, and -S7. You also have -EXSEL on the PBS, so you can selectively disable ROMs, even the cart ROMs, as long as your SRAM connects to RS1's data bus. You have full I/O decoding on the PB, so you can capture accesses to PORTB without wires soldered to the PIA. So the upgrade could be a pair of boards that plug into the PBS and RS1 with a cable between them, and an optional connection to the CPU board for separate ANTIC access. What do you think?

Anyone know what VOTRAX is? Is it this speech synthesizer? http://en.wikipedia.org/wiki/Votrax Was it used in any 5200 games?

How can we get Joe to come over to the A8 forum and give some Colleen development history? BTW, Bryan, you nailed the 2:16 timer answer five years before Joe confirmed it. Nice work!

At Program Design, Inc., in 1984, I wrote educational software for the Atari mostly, but also some for the Apple II and C64. I enjoyed working on the Apple but not so much on the C64. I loved the Atari and respected the Apple but, to me, the C64 was the young upstart already threatening Atari's dominance. And it saddened me that Commodore was winning via marketing instead of engineering. (Don't flame me - this is what I thought back then). I had to do one project for all 3 platforms. It was a multiple-choice, SAT-prep, quiz program, IIRC. I decided to develop it on the Atari and port it to the others. But they all had different disk formats and I didn't want to retype all the text data. So I made a little transistor circuit and cable that connected the Atari SIO port to the Apple's or C64's serial input. I then transmitted the text to the other computers. Only the quiz engine differed across the platforms. One thing I disliked about displaying text in the Apple's graphics mode was the color artifacting. So I transferred the Atari character set over to the Apple and used that. The characters were a little too close, because the Apple could only show 7 bits per byte in hires mode, but I though it looked nice and white. I heard later from Sales that several Apple customers were angry about the character set switch! I guess they saw the Atari as the upstart.

My memory (not my RAM) failed me when I posted this. Later on I opened my 400 and found the RAM Sandwich intact. Now I re-remember that I did the 48K 400 mod on a friend's machine in exchange for another RAM-CRAM which I later modded into a RAMdisk.

I'm just glad to help us all keep the Atari spirit alive!

OK, thanks. Too much going on though. No progress on the video thing, sad to say. How are things in the sunny West? Oops. Only up to 7 apps.

My original 256K XL upgrade used 8 32K banks but I quickly changed it to 16 16K banks after the 130XE came out. Now, in retrospect, I wonder if 32K might have been better. Each 32K bank had its own page 0 and stack page, so task switching would have been easier. The upper, non-banked, 16K could have held a task switching kernal and DOS and maybe common screen RAM. You could have had up to 8 applications running in their own 32K spaces taking time slices of the CPU. Something like that for the Atari in 1985 could have been a big deal.

Absolutely! I don't know if the Newell upgrade works that way. I was describing the Buchholz/RAMBO/Wizztronics 256K upgrade.

It's really the RAM test programs themselves that are at fault. Those upgrades existed in 1985 and the test programs came later. So, the test programs failed to work properly with existing upgrades. My original 256K upgrade instructions explain clearly that there are only 12 16K banks of extended RAM. Just because there are 4 bank-select bits does not mean there have to be 16 valid values. The Atari chipset itself has some hardware registers where not every bit combination is a valid one.

It's really the RAM test programs themselves that "cause" the behavior. Those upgrades existed in 1985 and the test programs came later. So, the test programs failed to work properly with existing upgrades. My original 256K upgrade instructions explain clearly that there are only 12 16K banks of extended RAM. Just because there are 4 bank-select bits does not mean there have to be 16 valid values. The Atari chipset itself has some hardware registers where not every bit combination is a valid one. Also, the 4 "invalid" banks are really just copies of the base RAM. That in itself can be useful, especially the copy of the RAM under the OS. So, upgrades that add RAM are not superior. One could even argue that they are inferior because they have more RAM chips and circuitry that could go bad.

Or maybe stronger (smaller) pull-up resistors on the GTIA luma pins? Something for the video mod designers among us to ponder.

Thanks! I'm glad to contribute a little bit to the fine job you're doing. Back in the day I modified Letter Perfect to work with ACE-80 but it was not optimal. What you're doing should be the final word (pun intended).

Back in the day, we were not spoiled by crisp VGA displays, so it was a small minus next to a huge list of Atari plusses.

I've seen it on my Ataris. I think it's normal.

That's what ACE-80 is.