warerat

Any word on this? I think it would be a nice mod to do especially since I already have the switch free. Tempest Pretty easy, but it is important to make sure the ribbon cable is disconnected when you switch back to stock mode (you'll have two signals driving the same line if you don't-- bad). If you have a switch with three lugs (one lug is common to both positions): First, disconnect the wire from the ribbon cable that goes to U14 pin 16. O ---- from U11 pin 1 O ---- to U14 pin 16 (this is the common) O ---- from ribbon cable that used to go to U14 pin 16 that you desoldered.

Somewhere in the world, a 130XE motherboard has breathed a sigh of relief. Having some of the tools and knowing how to use them are two separate concepts. I'm almost volunteering at this point to try to fix the damage for free to your two boards if you promise never touch an iron to an Atari again.

Maybe your time and money is better spent having someone more experienced with better tools install this in the machine rather than destroying motherboard #3.

Nice work! I take it adding Bit 3 emulation wouldn't be too difficult?

VBXE has the advantage of being a new device, versus having to source these old CRT controllers for AF80/Bit-3 but even then I'd bet they'd still cost less. But make no mistake, the font looks great and is crisp. This is output from RGBI interface on TTL monochrome.

Demos for Austin Franklin. Run with BASIC enabled. AustinFranklinDemo.atr.zip

This is the RGBI daughter card that plugs in. The monitor connector is a DB-9.

Not following you here? This cartridge is useless without the AF board plugged in slot 3 in an 800. It contains the firmware for the hardware that didn't fit on the board. Could it be converted to a left cartridge? Sure. Just a few pin swaps on the cartridge connector. But then an 800 owner has no way to plug in BASIC, Assembler Editor, etc. to use the board in the first place which makes it crippled. Could it be converted so you could plug it into an XL/XE? Sure. You'd have to recode the ROM to not jump to hardcoded OS-B locations and then interface an Atari 800 memory card slot to the PBI bus to make it work. Not to mention the fact that very few people even own the board kind of puts selling anything for it out the window, unless you plan on sourcing the long obsolete ICs and recreating the board.

Sorry for the awkward format. Unzip both files then un-RAR to get the scanned PDF for the manual. The board is accessed via $D5F6. Bit 0-3: Video bank select ($0-$F) Bit 4: 0 = Enable 2K character RAM Bit 5: 0 = Enable 2K attribute RAM Bit 6: 0 = Enable CRTC registers (Bit 0-3 must be 0) Bit 7: 0 = 40 column output, 1 = 80 column output Only one of the bits 4-6 can be 0 at a time, the other two must be 1. Video memory/CRTC registers are seen at $D600-$D67F (with foldback). Austin Frankin 80-Manual.part1.rar.zip Austin Frankin 80-Manual.part2.rar.zip

It seems this board wasn't very common and there isn't much information out there about it short of some magazine reviews. Were the cartridge dumping project was still alive, this is one of the ROM dumps that was missing. So for the sake of historical reasons I scanned both manuals (operations and RGB daughter card install) and am including ROM dumps of the cartridge, the character ROM on the video board itself, and the custom PROM binary. I reverse engineered the cartridge board as there is an IC that has its part number ground off. It wasn't hard to surmise what it was given it ran really hot and the lack of any discrete logic for the decoder, so my assumption was that it was a bipolar PROM. Turns out I was right and it is a 84S129/74S287, 256x4 PROM (CCTL- is used as an enable and A0-A7 are the eight inputs). Only two of the outputs are used, one to clock data when writing to $D5F7 and the other to enable the ROM. On power-up, the cart ROM is briefly mapped to both $8000-$9FFF and $D500-$D5FF. This allows the 800 OS to jump to the right-cart initialization vector at $9FFE which points to the same bank in $D5XX which then enables the RAM back the way it was. As a consequence, you can't use any 16K carts with this. Here's the technical stuff on the cart bank switching that isn't discussed in the manual: ROM size is 4K, is visible at $D500-$D5EF & $D5F8-$D5FF ROM disabled at $D5F0-$D5F7, single write-only register at $D5F7. Cartridge ROM Bank register at $D5F7: Bit 0-3: ROM Bank select ($0-$F, 16 banks) Bit 4: ROM output enable (0 = Enable ROM, 1 = Disable ROM) Bit 5: Right-cartridge RD4 control (0 = $8000-$9FFF cart ROM, 1 = $8000-$9FFF system RAM) Bit 54 -- 00 = ROM enabled both at $D500-$D5FF and $8000-$9FFF (foldback occurs every 256 bytes) 01 = ROM disabled, system RAM disabled at $8000-$9FFF 10 = ROM enabled at $D500-$D5FF, system RAM enabled at $8000-$9FFF 11 = ROM disabled, system RAM enabled at $8000-$9FFF Going to have to lower the resolution on the manual as it is ~87MB in PDF format. Austin Franklin 80-ROMs.zip Austin Franklin 80-RGB Guide.pdf

Ironically I just came across an incredible find-- an Austin-Franklin board, with RGB connector, cart, manuals, and box-- all in working condition... I am in the process of scanning everything I have for this as I've noticed documentation for this is pretty much non-existent except for a few magazine reviews. I've already dumped the character ROMs and the program ROM on the cart. I also got a Bit 3 and a Axlon Rampower 128K in the same lot. The way the cart and board work is unique, but somewhat similar to the Bit 3 in the sense that they take up all of $D5XX and $D6XX (which are mapped to Slot 3 on the 800). On the Bit 3, the firmware for the display drivers is on the card itself. On the Austin-Franklin, you get twice the display memory (4K vs 2K for Bit 3), but there's no on-board ROM for the firmware-- hence the cartridge. The cartridge goes into the right slot of the 800 maps its 4K ROM into the $D5XX address space, 248 bytes at a time. There's a fixed eight-byte window in the page where the cart registers are accessed, but essentially the last 16-bytes of the page aren't usable. Also looks like they did direct OS calls for the screen editor in the ROM. The Bit 3 uses a 6545 CRTC, the Austin-Franklin uses a SMC9007. The font on the Austin-Franklin is larger than the Bit 3 and is pretty nice. I just hooked it up to a TTL monochrome monitor and the quality is excellent.

Final release firmware v1.4.0 with original ICD menu. Next revision needs a hardware mod to allow access to unused ROM space (the MIO normally comes equipped with a 16K 27128, but only the top 8K is used) because this ROM is FULL. Minor changes from 1.4b3: - Arithmetic routines fixed to allow displaying >65535KB (assumes we can create partitions that big one day) - Holding down HELP + RESET does coldboot, just like Black Box - SIO "beep" can be turned on/off for HD access Hope it's useful... mio140.rom.zip

Excellent work. I wish I knew Altium Designer like that-- it's a complex piece of software. Want to route a card for my 800 project for me?

55104 shows 16. Text color did change to green. When I run CMAP.COM I get the same banding you did on yours.

Yes my crystal is still there. But pulling up and connecting directly to pin 2 on FREDDIE does the same. Just a different way of thinking..

Looking good. This is the output from a Viewsonic P815 21" CRT.

nS4, S4n, S4_n, S4-, S4*, and S4(with over-bar) all mean the same thing- active low signal S4. Just different notation.

unfortunatly it is already zero'd i tried to read it last night after i took the pic... i havnt opened the thing in 20 years so i didnt know the eprom had lost its sunscreen... sloopy. Here you go. v3.02 of the ROM. ATR8000_.ROM.zip

First off, I'd like to thank Larry White in the community for trusting a complete stranger off the 'net and letting me borrow his Bit 3 FullView-80 to test in this modified 800 rig. So I've validated that in 800 mode you can still run your slot 3 80-column boards (and keep 48K/52K/512K Axlon modes). This has interesting implications in 130XE mode as it would not normally be possible on a real XL or XE computer. The 6545 CRTC hardware registers are still visible at $D5XX so you could make a driver in 130XE mode to control the 80-column display. The downside is there are lots of devices using that same address space. The Bit 3 maps its 4K ROM a page at a time at $D6XX but from what I disassembled of it, there are many references to OS-B ROM locations. What makes the board unique to the 800 is a signal on the bus called EXSEL- (not to be confused with XL/XE EXTSEL-) that disables the databus buffers that connect the personality slot module to the CPU. I'm still learning how this board interacts with the memory map but under real conditions you have to ignore EXSEL- in XE mode as it's technically not valid for that architecture. If I do that I don't see the ROM get mapped in for XE mode. As an experiment, allowing EXSEL- to work in XE mode makes the Bit 3 ROM completely visible. Except for all those OS-B references that hang the machine... Good stuff. Thanks, Larry!

R/W- is on pin 1 of Z402. Hint: You can't have pins 18, 20, and 21 connected to the socket. I pulled out pin 18 of SRAM and connected to pin 20 and left pins 20,21 in the FP ROM socket, but I couldn't get Atari to boot up. I thought it would read the data at least and I would modify pin 21 (W*) for write later. What connections do you suggest for pins 18,20,21 of SRAM (for read/write)? It's not going to boot up because the pin-out obviously isn't the same and the SRAM is getting enabled when it shouldn't be. You need additional logic to make it work: WE- = not (PHI2 and not RW) CS- = not (A11 and A12 and not Y6) A11 = pin 18 on socket Y6- = pin 20 on socket A12 = pin 21 on socket CS- = pin 18 on SRAM OE- = pin 20 on SRAM WE- = pin 21 on SRAM CS- and OE- are tied together. Simplest is to use a single 'LS139. 1 = GND 2 = PHI2 3 = RW- 5 = WE- (to SRAM) 8 = GND 9 = CS- (to SRAM CS- and OE-) 13 = A11 14 = A12 15 = Y6- 16 = +5V

R/W- is on pin 1 of Z402. Hint: You can't have pins 18, 20, and 21 connected to the socket.

Yes if it has older 64Kx1 DRAM and no if it is expanded. I had a 600xl once, took its Antic and placed it into an 800xl. IIRC this indeed caused refresh problems. The 600xl of course has only 16kB, so 7bit Row/Column is enough. ...just to make clear that the refresh of the Antic chip in the 600xl is (If I'm correct) also 7bit. Early 4416 allowed 7-bit refresh. Newer 16Kx4 (like TI TMS4416) devices and others (64Kx4, 256Kx4) need the 8-bit 4ms refresh. But yet I have 1200XLs with 64K and the old ANTIC.

The older 16Kx1 and *some* 64Kx1 DRAMS needed to refresh 128 rows in 2ms. The newer ANTIC does work on the 400/800 because while its address counter during the refresh cycle updates A0-A7, only A0-A6 exist on the 16Kx1 DRAM. Yes if it has older 64Kx1 DRAM and no if it is expanded. This is why there are some XLs with 64K and ANTIC "D" with no problems depending on the type of DRAM installed. NEC/Fujitsu 64Kx1 DRAMs needed only 128 rows. Consequently some of the earlier memory upgrades for the XL had option to add the extra bit of the address with an external counter and register. The row count are refresh requirement are proportional. ANTIC "D" (CO12296/CO14887) were used for 400/800/early XLs. ANTIC "E" (CO21697/CO21698) was the same for XE's and newer XL's. XE's usually had the newer Micron MT4264 which needed 256 rows refreshed in 4ms (like the newer 256Kx1). So you can't plug in an ANTIC "D" for an XE with 4ms refresh DRAM.

This is the email quote I got yesterday: MOQ is 10 pieces 3D7323Z-70 $8.68 each 1 week to ship MDU3C-70 $11.55 each 4-6 weeks This is the part: http://www.datadelay.com/datasheets/3d7323.pdf The delay tolerance is 2%. The ideal delays are 69.8 ns for NTSC and 70.5 ns for PAL. They differ by less than the tolerance. Would something like a IDT ICS570 work to generate the clocks?

Well, the schematics for this are surprisingly simple (four IC's-- two SRAMs, one ROM, one 100-pin CPLD, handful of pullups, a 3.3V regulator, and a Maxim reset IC). But completely useless without the CPLD core logic which controls the whole thing and is where all my time, sweat, and tears have gone. So yes, this has consumed 100s of hours of my time to perfect it to where it is today while still having a life that doesn't involve Atari or computers. As a result of changing my design philosophy on this several times since I first started I decided to make some hard decisions about how this gets installed. As a result, this is an obtrusive (but not destructive) install and once it's in there, you can't go back to bone-stock (original OS and memory cards will not work) unless you remove the entire upgrade. You have to pull two IC's from the motherboard (and have jumpers in place on a few pins) as I re-purpose some of the pins on the OS slot to reduce the external wiring necessary. As an 800 owner I know it's impractical to have to pull out cards if you want to go back to stock so I've successfully implemented BOTH architectures at the hardware level and they are jumper selectable-- I think you'll agree that makes up for it. So if you want to go back to 4-port 800 mode you just move the jumpers. There are a few things I need to verify as a feel-good measure to validate specific portions of the hardware implementation but for all intents and purposes this is an All-In-One Personality module that pretty much can make the 800 act like any 800/XE model. Remember ports 3/4 don't work for joysticks in XE mode, but you can still use 8 paddles and no trigger on port 4.. So what it looks like now: Stock 48K 800 OS-B (verified OK) Stock 52K 800 OS-B (verified OK, RAM at $C000-$CFFF) Axlon 512K 800 OS-B (verified OK, with Claus Buchholz Tron animation, Axlon diags) Stock 128K 130XE ANTIC-enhanced (verified OK) Peterson 320K 130XE 8ACE (verified OK) Peterson 576K 130XE 8ACE (verified OK) CompyShop 576K 130XE 26AE ANTIC-enhanced (looks OK, but has trouble with some demos-- I'd have to validate with someone who has this). But then again I could just add another SRAM for full 1088K, but not much software to use it. XE mode is 100% PORTB compatible, including built-in BASIC C. And yes, it should be VBXE compatible but I'll know for sure when I get my VBXE board here this month. There would be major logistical problems squeezing it on the CPU board and getting it back in the metal housing. It could be plugged into a custom board that connects to one of the free memory slots with handful of connections to ANTIC as I provide all the XE ECI signals and hardware I/O from the OS board. PBI is implemented, but again, logistically difficult getting that ribbon cable out of the metal cage, but not impossible. Thanks for being patient.