cetics Posted February 2, 2020 Share Posted February 2, 2020 Four 8k games on a 16k eprom how is that possible? I have this board that has the games (Time Pilot - HERO - Moon Patrol - Phoenix) So I researched the atari document SIZES.TXT all these games are 8k. But when reading the NMC27CP128Q eprom, with the TOP3000 reader it resulted in a 16K file. Is there some kind of compression in the file to fit? Or is it a modified version of the original games? Pins 26 and 27 of the eprom are used by the keys as game selectors. However, in the datasheet, pin 27 is marked "AR" I did not understand correctly what it is for, because the datasheet is old and in image format. This makes it difficult to translate into my language. Does anyone have any ideas on how this works? Could it be a remarked eprom? All games went perfectly into the console. * In the photo the eprom pins are raised but they were used correctly when reading the TOP300. Attached image obtained from eprom and Datasheet. Thanks. NMC27CP128-NationalSemiconductor.pdf tpilot-mpatrol-phoenix-hero.bin 1 Quote Link to comment Share on other sites More sharing options...
Thomas Jentzsch Posted February 2, 2020 Share Posted February 2, 2020 (edited) The 16K dump contains H.E.R.O and Phoenix, both using a variation of UA bankswitching. This will be supported in the coming Stella 6.1 release. The ROMs are very much identical to the known versions (99%), just with the UA bankswitching, swapped banks and "RENTACOM" text. Phoenix (tpilot-mpatrol-phoenix-hero.001).bin H.E.R.O (tpilot-mpatrol-phoenix-hero.002).bin Edited February 2, 2020 by Thomas Jentzsch 1 Quote Link to comment Share on other sites More sharing options...
alex_79 Posted February 2, 2020 Share Posted February 2, 2020 (edited) 5 hours ago, cetics said: Four 8k games on a 16k eprom how is that possible? I have this board that has the games (Time Pilot - HERO - Moon Patrol - Phoenix) So I researched the atari document SIZES.TXT all these games are 8k. But when reading the NMC27CP128Q eprom, with the TOP3000 reader it resulted in a 16K file. Is there some kind of compression in the file to fit? Or is it a modified version of the original games? Pins 26 and 27 of the eprom are used by the keys as game selectors. However, in the datasheet, pin 27 is marked "AR" I did not understand correctly what it is for, because the datasheet is old and in image format. This makes it difficult to translate into my language. Does anyone have any ideas on how this works? Could it be a remarked eprom? All games went perfectly into the console. * In the photo the eprom pins are raised but they were used correctly when reading the TOP300. Attached image obtained from eprom and Datasheet. Thanks. According to the datasheet, the NMC27CP128 is indeed a 16k x 8 eprom. On page 7 it states that it's "half" a NMC27C256, only containing the upper 16K: Quote [...] Since the NMC27CP128 employs the last 131,072 bits of a 262,144 bit memory array, program- ming must be started at address 16,384 to provide cor- rect data read. [...] The NMC27CP128 is a partial NMC27C256 and therefore is not program compatible with most 128k EPROMs. [...] Pin 27, which in a 27c256 is the highest address line (A14), is labeled "AR - Block Select" and it's supposed to be held HIGH while accessing the 27CP128 eprom. Anyway, as you say, if there are 4 8k games in this cartridge, the eprom must be 32K... I would set the eprom programmer software to a 27C256 and try reading the chip again. It should be safe to do so, and, if it's a real 27CP128, the lower 16k of the resulting binary should only contain garbage data. Anyway, if there are four 8k games in this cartridge, the eprom must be 32k, so maybe it's indeed a mislabeled 27C256, and you'll find the remaining two games in the lower 16k of the dump. Edited February 2, 2020 by alex_79 1 Quote Link to comment Share on other sites More sharing options...
cetics Posted February 2, 2020 Author Share Posted February 2, 2020 I had done a reading as if it were a 256, and it resulted in a 32k file, but looking superficially it seemed to me that it was different in hexadecimal, I thought it was all garbage .. I will try to read it again and post the result here. Thank you all. Quote Link to comment Share on other sites More sharing options...
cetics Posted February 3, 2020 Author Share Posted February 3, 2020 I made a new dump now using one based on a 256 eprom. Test if you can play the games (here only one worked partially) I tried to separate the games the folder follows with the full version and the separate games. NEW RENTACOM AFS.zip 1 Quote Link to comment Share on other sites More sharing options...
alex_79 Posted February 3, 2020 Share Posted February 3, 2020 You need to install the new Release candidate of Stella to run those roms, as previous versions don't support their bankswitching scheme: https://atariage.com/forums/topic/301342-stella-61-rc1-released/?tab=comments#comment-4448110 The dumps are good: game 1 is "Time Pilot" and the rom is identical to the one you dumped from the other multicart some time ago. game 2 is "Moon Patrol". You must set the starting bank to "1" in Stella "game properties" for this rom, else it crashes. That's because whoever back in the day converted the rom from the original "F8" bankswitching method overlooked to change the hotspot address that is triggered when the rom starts in bank 0 (which is Stella default starting bank). game 3 and 4 are identical to the "Phoenix" and "H.E.R.O." roms posted earlier in this thread So it was indeed a mislabeled eprom. Weird! Thanks for sharing the dumps! 2 1 Quote Link to comment Share on other sites More sharing options...
Thomas Jentzsch Posted February 3, 2020 Share Posted February 3, 2020 How do you select a game from the cart? I can't see any switches. Quote Link to comment Share on other sites More sharing options...
alex_79 Posted February 3, 2020 Share Posted February 3, 2020 The switches are on the two sides of the cart. They're partially visible in the picture of the PCB. They connect to pin 26 and 27 of the eprom (yellow wires), which are address A13 and A14 2 Quote Link to comment Share on other sites More sharing options...
cetics Posted February 5, 2020 Author Share Posted February 5, 2020 Thank you all very much for the explanations. I will try the new version of stella. I was thinking about reusing these cards to put other F8 games, but as they mentioned that they use a modified UA bank, I believe that recording a normal F8 rom from the internet will not work directly. I believe that I will need a converter (if any). What do you think? If this is not possible or simple to do. you have some F8 cartridge scheme with common logic circuits (without using PLD). Thanks. Quote Link to comment Share on other sites More sharing options...
alex_79 Posted February 5, 2020 Share Posted February 5, 2020 (edited) 7 hours ago, cetics said: I was thinking about reusing these cards to put other F8 games, but as they mentioned that they use a modified UA bank, I believe that recording a normal F8 rom from the internet will not work directly. I believe that I will need a converter (if any). What do you think? F8 games won't work on these boards. You need to alter the roms to use the different scheme, which requires some skills in 2600 programming. There's no an automated way to do it. Quote If this is not possible or simple to do. you have some F8 cartridge scheme with common logic circuits (without using PLD). Thanks. This one uses three logic chips: http://www94.pair.com/jsoper/bankswitch_f8.html Edited February 5, 2020 by alex_79 1 Quote Link to comment Share on other sites More sharing options...
cetics Posted February 5, 2020 Author Share Posted February 5, 2020 Thank you very much for the tips. Quote Link to comment Share on other sites More sharing options...
R.Cade Posted February 5, 2020 Share Posted February 5, 2020 The 27CP128 is an odd beast. It has the A14 address line redefined as a "block select" switching between the two banks. Why they sold it as a "dual block" 128 is beyond me. There must be some reason in the specification somewhere that makes it different. 1 Quote Link to comment Share on other sites More sharing options...
+batari Posted February 6, 2020 Share Posted February 6, 2020 Very interesting. I wonder how many so-called 27128 EPROMs are really a 27256 in disguise? 1 Quote Link to comment Share on other sites More sharing options...
R.Cade Posted February 6, 2020 Share Posted February 6, 2020 3 minutes ago, batari said: Very interesting. I wonder how many so-called 27128 EPROMs are really a 27256 in disguise? Just the "CP" variants, I would assume. 1 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.