AtariGeezer Posted October 30, 2016 Share Posted October 30, 2016 I hope the rev. B board was stored in darkness. The eprom window is not covered by a sticker, so if not, its contents may already have become corrupted from UV light present in daylight. If you have the means to do so, dump the eprom with a eprom programmer ASAP. Looks like a 2764 to me. Pin 26 is NC on a 2764, but here it's used as a route to get +5V to pin 1, 27 and 28. BTW: exciting wiring with the piggybacked IC on top. I'll try drawing a diagram of the wiring going to the eprom. To me, it looks like a 24-pin PAL. If its copy prohibit bit is set, no real chance of dumping the contents. IIRC, the mask rom was ordered to be manufactured with bank switching logic included on-chip, so instead of dumping it with a eprom programmer, Avery's program will probably be the only way to get a valid rom dump. re-atari The 2764 looks to be wired as a 2732 and I have been able to pull a dump of it on the willem which appears to be ok. The Masked ROM is a 2732, but the willem can only read the lower 4k cleanly. This ROM is a Version 7 as it's written on a label on the bottom of the ROM. I'll use Avery's program to complete the dumps later on today as long as one of the 810's works okay... Quote Link to comment Share on other sites More sharing options...
AtariGeezer Posted October 30, 2016 Share Posted October 30, 2016 Interesting notice in the v7 dump: 2 Quote Link to comment Share on other sites More sharing options...
ijor Posted October 30, 2016 Share Posted October 30, 2016 The 2764 looks to be wired as a 2732 and I have been able to pull a dump of it on the willem which appears to be ok. The Masked ROM is a 2732, but the willem can only read the lower 4k cleanly. This ROM is a Version 7 as it's written on a label on the bottom of the ROM. I'll use Avery's program to complete the dumps later on today as long as one of the 810's works okay... The 2764 is not Ver 7? I thought the PAL/GAL was there for implementing the banking. If so, it should be rev 7. Not that it matters too much, but not sure the other is actually masked ROM. I might be OTP Rom. One time programmable, not erasable. If it were actually masked I would expect a Happy copyright label. OTOH, it is labeled "HART". Hart was the nickname for the custom chip on the Discovery Cartridge. Just by chance? Quote Link to comment Share on other sites More sharing options...
AtariGeezer Posted October 30, 2016 Share Posted October 30, 2016 The 2764 doesn't have any text in the rom dump (so far) and it's on the Rev B board. The HART chip did have the Happy Copyright label on it, but I removed it to see what kind of chip it was. It does have another Happy Copyright label on the bottom side of the rom with a "7" written on it. We'll see what's what after I run Avery's program on these... Quote Link to comment Share on other sites More sharing options...
ijor Posted October 30, 2016 Share Posted October 30, 2016 The HART chip did have the Happy Copyright label on it, but I removed it to see what kind of chip it was. Yes. But I meant that a masked ROM usually (not always though), being a custom chip after all, comes with the firmware copyright MARKED on the chip encapsulation, not just with a label attached. Quote Link to comment Share on other sites More sharing options...
AtariGeezer Posted October 31, 2016 Share Posted October 31, 2016 Yes. But I meant that a masked ROM usually (not always though), being a custom chip after all, comes with the firmware copyright MARKED on the chip encapsulation, not just with a label attached. Sometimes it does. I was looking at the original Happy 1050 ROM and it's marked as HART ECMT-R2 with a part number of 94P64B-2583 where the Happy 810 v7 has a part number of 94P64B-0585 and marked as HART 810-R1. So it looks like the Happy 810 v7 is also a 8k ROM? Anywho, I was able to run HappyRead on the Rev B board with the modified 2764 and came up with this: HAPPY810-B.EPR.ROM The Happy 810 v7 94P64B ROM isn't working on the Rev D board or Rev B board, even with a different 810 Side Board. I'll have to check the other 810 with a Happy Rev D board to see if that one works... 2 Quote Link to comment Share on other sites More sharing options...
ijor Posted October 31, 2016 Share Posted October 31, 2016 (edited) So it looks like the Happy 810 v7 is also a 8k ROM? Yes, of course. That's why it is banked. Anywho, I was able to run HappyRead Dump seems good. Many thanks! The Happy 810 v7 94P64B ROM isn't working on the Rev D board or Rev B board, even with a different 810 Side Board. Seems that they are not compatible with older boards. See some earlier post on this thread mentioning B&C selling two different rev 7 upgrade roms, depending on the serial number. The Rom content is likely the same, but the pinout is probably different. Can you post the two 4K dumps you made with the Eprom burner? Even when the 6K dump is probably all we need for technical operations; for historical reasons it might be interesting to get the remaining 2K that are not accessible by the CPU. Surely 1K of this inaccessible area you already dumped, it's the one with the copyright notice. And possibly the other 1K is in the other 4K dump taken from the 2764. Edited October 31, 2016 by ijor Quote Link to comment Share on other sites More sharing options...
AtariGeezer Posted October 31, 2016 Share Posted October 31, 2016 Using the Willem, this is from the 2764: Happy810-B(0).ROM This is the partial from the v7: Happy810-D.ROM And if you compare the first 1000 bytes of each, they are the same. This is a guess of combining the two: Happy810-84.rom Quote Link to comment Share on other sites More sharing options...
ijor Posted November 1, 2016 Share Posted November 1, 2016 Thanks once again, AtariGeezer ... I don't know exactly what to make out of this. Since you said the 2764 dump has no text, I was hoping you got a different bank than the rev D ROM. But that doesn't seem to be the case. As you say, some of the dump is identical and is from the same bank in both dumps. Cleary, the copyright text is the 1K missing at one of the banks. But we can't be sure about the other 1k missing at the other bank. For actually functionality, it doesn't matter. Those 2k, 1K on each bank, are not visible by the CPU. Quote Link to comment Share on other sites More sharing options...
phaeron Posted November 1, 2016 Share Posted November 1, 2016 The 6K dump we got with happyread works in emulation but is failing the v7.10 ROM test. However, I'm suspicious of the test since it only checks for one set of checksum values. AtariGeezer, does the drive that you got the full dump from pass the ROM check diagnostic with any version of the Happy software? And thanks for the work you've put into this! Quote Link to comment Share on other sites More sharing options...
ijor Posted November 1, 2016 Share Posted November 1, 2016 The 6K dump we got with happyread works in emulation but is failing the v7.10 ROM test. However, I'm suspicious of the test since it only checks for one set of checksum values. Why do you think there are multiple versions? I'm not claiming there are not, I just assumed there is only one precisely for that reason. Do you have any reason to believe otherwise? Quote Link to comment Share on other sites More sharing options...
tep392 Posted November 1, 2016 Share Posted November 1, 2016 I'm going to dump my V7 rom but want to confirm the pinout. Is it the same as a 2732? Quote Link to comment Share on other sites More sharing options...
ijor Posted November 1, 2016 Share Posted November 1, 2016 I'm going to dump my V7 rom but want to confirm the pinout. Is it the same as a 2732? I'm not sure we really know. So far we haven't seen a Happy Board rev D with a standard eprom. Somebody would need to trace the board to check the pinout. But note that rev 7 ROM is banked, you normally can't dump it with an eprom burner anyway. You need some software (or logic) to select the banks correctly. Or, connect the Happy drive and use Phaeron's program. AtariGeezer managed to make a partial dump with his eprom burner. But it is not even a full bank for some reason. Quote Link to comment Share on other sites More sharing options...
AtariGeezer Posted November 1, 2016 Share Posted November 1, 2016 (edited) Good News, my 2nd v7 ROM dumped okay using Avery's program: This also has some text that was missing from the homemade/beta v7 ROM. happy810-v7.epr.rom And with the dump from the eprom reader catching the 1k copyright notice, I pasted it to the beginning of the rom dump. Then added an empty 1k block at the beginning of the 2nd half and now we have a 8k v7 Happy 810 ROM happy810-v7.rom Edited November 1, 2016 by AtariGeezer 3 Quote Link to comment Share on other sites More sharing options...
AtariGeezer Posted November 1, 2016 Share Posted November 1, 2016 (edited) The 6K dump we got with happyread works in emulation but is failing the v7.10 ROM test. However, I'm suspicious of the test since it only checks for one set of checksum values. AtariGeezer, does the drive that you got the full dump from pass the ROM check diagnostic with any version of the Happy software? And thanks for the work you've put into this! The drive with the 2764/GAL does not pass the ROM Test, it does pass the Initial and Exhaustive test though. Both original v7 ROMs do pass all Diagnostic tests though using a Happy 7.10 disk... Glad to help Edited November 1, 2016 by AtariGeezer Quote Link to comment Share on other sites More sharing options...
AtariGeezer Posted November 1, 2016 Share Posted November 1, 2016 I'm going to dump my V7 rom but want to confirm the pinout. Is it the same as a 2732? Very similar pinout, but I'm guessing (haven't traced it out yet!) that the GVpp (program) line of the 2732 is actually an A12 line instead on the v7 ROM. It would be nice to know who dumped the Happy 1050 v7 ROM and what they did to get the full 8k... Quote Link to comment Share on other sites More sharing options...
tep392 Posted November 1, 2016 Share Posted November 1, 2016 Very similar pinout, but I'm guessing (haven't traced it out yet!) that the GVpp (program) line of the 2732 is actually an A12 line instead on the v7 ROM. It would be nice to know who dumped the Happy 1050 v7 ROM and what they did to get the full 8k... Looking at the back of the board, It looks all pins except 18,20,21 go to the same corresponding pin onthe 6116 ram chips, so I would deduce that gnd, 5v, d0-7, a0-10 all match the 6116 pinout which is also the same for 2732. I can't tell where pins 18,20,21 go because the traces are blocked by the CPU pass thru connector. Quote Link to comment Share on other sites More sharing options...
ijor Posted November 1, 2016 Share Posted November 1, 2016 Good News, my 2nd v7 ROM dumped okay using Avery's program ... Both original v7 ROMs do pass all Diagnostic tests though using a Happy 7.10 disk... Fabulous. Great work! So now we know why the GAL version doesn't pass the ROM test. It's a self made rom on an original happy board. And they erased all the copyright text. LOL Then added an empty 1k block at the beginning of the 2nd half and now we have a 8k v7 Happy 810 ROM We still don't know if that 1K block is empty or not. My guess it isn't. But we won't know until somebody, somehow, will dump the whole 8K ROM. It would be nice to know who dumped the Happy 1050 v7 ROM and what they did to get the full 8k... The memory map is different. The 6507 is replaced with a 6502, so the CPU has access to the whole 4K per bank, not just 3K. Then it is possible to dump the whole ROM with the CPU. The Happy VAPI imaging tool (there is no version for the Happy 810) can dump the whole ROM. Quote Link to comment Share on other sites More sharing options...
phaeron Posted November 2, 2016 Share Posted November 2, 2016 It's possible that whoever burned that EPROM got the firmware the same way we did, and might not have even been aware of the hidden message. Took a look at the 2764 dump, and it does indeed look like the programmer tripped a bank switch. It read almost all of the first bank, except for the last six bytes which came from the second bank. That's right after the banking registers. If the programmer has the ability to read a user-specified range, it should be possible to trip the bank switch at $1FF9 and then read out the hidden $0000-03FF range on the second bank. It's not necessary for functionality since it's inaccessible to the drive, but I'm curious as to the message that may be in it, since the messages seemed to get more interesting with successive firmwares despite few people ever seeing them.... Quote Link to comment Share on other sites More sharing options...
sup8pdct Posted November 2, 2016 Share Posted November 2, 2016 I plugged the rom into a cart board, put it in to my XL and used a monitor to copy selected parts then did a load from the switching address and read selected bits again. Easy James Very similar pinout, but I'm guessing (haven't traced it out yet!) that the GVpp (program) line of the 2732 is actually an A12 line instead on the v7 ROM. It would be nice to know who dumped the Happy 1050 v7 ROM and what they did to get the full 8k... 2 Quote Link to comment Share on other sites More sharing options...
AtariGeezer Posted November 2, 2016 Share Posted November 2, 2016 I plugged the rom into a cart board, put it in to my XL and used a monitor to copy selected parts then did a load from the switching address and read selected bits again. Easy James Ah, okay, the old cartridge board trick that's how I copied the 810 Archiver Chip before I had a eprom burner... Quote Link to comment Share on other sites More sharing options...
re-atari Posted November 3, 2016 Share Posted November 3, 2016 The 2764 looks to be wired as a 2732 and I have been able to pull a dump of it on the willem which appears to be ok. The Masked ROM is a 2732, but the willem can only read the lower 4k cleanly. This ROM is a Version 7 as it's written on a label on the bottom of the ROM. I'll use Avery's program to complete the dumps later on today as long as one of the 810's works okay... I traced how the PAL/GAL is wired to the eprom. It goes like this: Eprom signal Gal --------------------------------- p1 Vpp p24 p2 A12 p21 p3 A7 p3 p4 A6 p4 p5 A5 p5 p6 A4 p6 p7 A3 p7 p8 A2 p8 p9 A1 p9 p10 A0 p10 p14 GND p12, p13 p20 /CE p11 p21 A10 p14 p22 /OE p1, p22 (1K resistor between eprom and GAL) p23 A11 p15 p24 A9 p2 p25 A8 p23 p26 NC p24 p27 /PGM p24 p28 VCC p24 I think it takes removing the PAL/GAL's VCC line to be able to read out the entire eprom's 8K adress space (if it's a 2764). BTW: sorry about the column layout, it looks like using tabs between them does not format the text properly. re-atari 1 Quote Link to comment Share on other sites More sharing options...
re-atari Posted November 3, 2016 Share Posted November 3, 2016 How far did you get with this? If you send the results to me then I will go ahead. I finally finished tracing the wiring and drawing a schematic of the mod in post 27. The wiring is as follows: Note that I drew the wiring as the PCB is seen on the photo's: from the solder side. Because of that, the IC's pin numbering is mirrored in my drawings. I'm fairly certain the wiring is drawn correctly (checked 2x). The edges of the PCB are drawn in a thick line, so that it stands out from the wiring lines. From some photo's posted by Jon (FlashJazzCat) several years ago, I learned the 2 unknown IC's are a 74LS00 and a 74LS32. The resulting schematic looks like this: Note the remarks about 2 cut traces and a wire routing on the 810 Side Board. Sadly, I haven't had time yet to dump the 810 rev. B rom. Hopefully I'll get a chance for it during the coming weekend! My TL866 programmer can't program the 2716's and 2732's I have in my sparebox. They are all types that require 25V programming voltage, where the TL866 won't go beyond 21V VPP. So, I'm thinking of replacing the rom with a 28c16 eeprom. re-atari Quote Link to comment Share on other sites More sharing options...
+Stephen Posted November 3, 2016 Share Posted November 3, 2016 I traced how the PAL/GAL is wired to the eprom. It goes like this: Eprom signal Gal --------------------------------- p1 Vpp p24 p2 A12 p21 p3 A7 p3 p4 A6 p4 p5 A5 p5 p6 A4 p6 p7 A3 p7 p8 A2 p8 p9 A1 p9 p10 A0 p10 p14 GND p12, p13 p20 /CE p11 p21 A10 p14 p22 /OE p1, p22 (1K resistor between eprom and GAL) p23 A11 p15 p24 A9 p2 p25 A8 p23 p26 NC p24 p27 /PGM p24 p28 VCC p24 I think it takes removing the PAL/GAL's VCC line to be able to read out the entire eprom's 8K adress space (if it's a 2764). BTW: sorry about the column layout, it looks like using tabs between them does not format the text properly. re-atari There you go. To preserve spaces, surround stuff in the code tag. Quote Link to comment Share on other sites More sharing options...
re-atari Posted November 3, 2016 Share Posted November 3, 2016 There you go. To preserve spaces, surround stuff in the code tag. Thanx, this looks precisely how I wanted it to be! re-atari Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.