Jump to content
Have You Played Atari Today?
2600|5200|7800|Lynx|Jaguar|Forums|Store  

acadiel

+AtariAge Subscriber
 View Profile  See their activity

  • Content Count

    2,108

  • Joined

  • Last visited

  • Days Won

    3

Posts posted by acadiel

  2. Posted

    I was wondering if some of you expert assembly programmers could redirect the DSK1 or DSK.TELCO (whatever it looks for) to a small subroutine in ROM that instead would bank to something, and copy the bytes to where Telco wants them, and then pass control back to Telco. :-)

  4. Posted

    I had to post this for atrax on FB earlier, but here you go, for anyone wanting the source code Stuart, Tursi and I made to bank switch a 128K ROM (i.e. Fred's Megamod cart with all three of his utilities). It should read well if you know assembly, if not, let Tursi or I know and we'll explain.

     

    Source code: http://pastebin.com/LL1Lqm3t

     

    Copy/pastable object code: http://pastebin.com/METtGvwy

     

    All the routine does is take code in certain locations of the EPROM of a certain size, copy it to a certain address in 32K, switch to the next bank when it need to continue and has run out of data (obviously the data starts after the header), and continues to the next bank as needed, then branches to >A000 to start the program.

     

    Enjoy. :)

     

    • Like 2

  10. Posted

    I usually get my refills from Meritline. I have a Canon MX850 (probably about 6 years old). I usually order a bucket of refills at once from Meritline:

     

    http://www.meritline.com/showproduct.aspx?ProductID=29252&SEName=canon-cli-8-ink-compatible-with-new-chip-combo-5

     

    Four sets sets me back $19.95 plus shipping (sometimes they have free shipping codes), and I've really only had maybe one in probably 50 go bad. For that price, I just tossed it and put another one in.

    • Like 1

  11. Posted

    The problem has been identified. It seems that the Winfred Winkler GROM0 for the HSGPL "diddles the bits" in VRAM and that seems to be the problem. When I switched back to TONY Knerr's TKG0, it runs fine. However, that causes a problem running TurboForth from the HSGPL. Tony has promised a fix from ROM-only programs and I hope to have everything back to a semblence of normalcy shortly.

     

    I don't run anything other than TKGROM0 in my SNUG system (when I boot it up, which is rarely). Tony did a great job making everything work OK.

     

    Is there any detriment to running TKGROM0 vs the stock GROM0 that comes with the HSGPL? I've never seen a list of pros and cons for the stock one vs. the Winkler one vs. TKGROM0. All I know is that Tony's ran what I wanted it without fuss.

  12. Posted

    I can't wait until we let this thing loose in the wild. Between the design that Ksarul did on the PCB, Tursi did in the Atmega GROM simulation, the 512K ROM part, and people putting stuff together like this... it's gonna be awesome.

     

    +1, everyone. Great job!

    • Like 2

  14. Posted

     

    Here's a Youtube video:

     

     

    Gives me more hope....I have a feeling if you want a professionally looking built one, we could take one of the Amiga ones and put a different end on it:

     

    http://amigakit.leamancomputing.com/catalog/product_info.php?cPath=21_72&products_id=226

  15. Posted

    Classic99 will only support inverted ROMs today. I'll just comment I argued against inverted ROMs all those years ago for precisely the reasons we are fixing today. ;) I was basically waiting till I got around to a revamp of the cartridge system, because with all the variants it is starting to get rather hacky. The XML format MESS uses helps solve a lot of that. I suppose one more hack won't hurt though. ;)

     

    Yep, and you can blame me for following the Databiotics inverted example for that. :-) Knowing them, they did it probably because it was a pain in you know what to deal with.

    • Like 1

  17. Posted

    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.

    Can we put up the V9 Gerbers since they handle the whole 512K without the two switches? Folks will probably want to make those over the V4's.... Thoughts?

  21. Posted

    Here's what I got back from CaDD:

     

    Thank you for discovering this error. The original Tigercub disks were copied by Charles Good and those are the ones CaDD supplies. I've gone back to the earliest known copy and it is the same. So it was either a bad transfer, or a bad original, most likely the latter since there is still catalog information on the copy. I just checked and do not have a 5.25" version of that disk. I will have to try other sources. If a copy is found, CaDD will email you the .dsk file.

    And Bill Gaskill:

     

    No luck in getting 972-2 to work, and I looked at 3 locations, from separate sources. They all show the same, which is to provide a disk catalog, but not much more.

     

    I never bought any of the public domain disks from Jim Peterson, so I never had any, other than what was given to me. I think Mike Wright still has the originals, but we didn't discuss that topic when I visited him in Aug 2010.

     

    • Like 2

  22. Posted

    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.

     

     

     

    I remember having a headache the night after I contributed my section to that document... LOL

     

    We probably want to deprecate the GAL information since we aren't using it.

  23. Posted

    Correct me if I'm misunderstanding this.

     

    There seems to be an issue with the 378 & 379 chips not always selecting the first bank upon powerup, and that's why a Gal is being tried in their place?

     

    If so, it should not be an issue as long as the Grom part of the board is in place. just use the Grom powerup to write to c>6000.

    Boom, no more problem. Is there a reason why this couldn't be done? Seems a real simple solution.

     

    If I were to setup one of these boards, the Rom section would only contain the two XB banks and E/A 5 program images. All the Menuing and moving of the program data in the Roms to its correct position for execution would be done in GPL as it's much simpler to do there. After using 40k of Grom for XB and 40k more Grom for the combined TI Writer/Editor assembler cart, that would leave 48k for the menus and code moving/execution from Rom - plenty of space.

     

    Last question (for now), the first bank of Rom selected by writing to c>6000 - is that at the beginning (offset >0000) or the end (offset >7E000) of the eprom?

     

    Gazoo

     

     

    Great idea if the Atmel AVR is going to be used... but if someone uses it as a ROM only cart (which it can be), then the 74LS378's I have tried are all powering up in the lowest 8K bank. This would need to be tested with new batches to make sure, but it's been pretty predictable. The same as the 74LS379's almost all powered up in the highest bank (being that we were using the inverted outputs on powerup.)

     

    For ROM only, the user would just need to make sure they verify their 378 is powering up in the lowest bank. We have a test program and EPROM image that writes to each bank to verify the correct bank contents, and we can likely also add a small section that tells you what bank it powered up in to verify operation.

  24. Posted

    You might want to keep up with this topic:

     

    http://atariage.com/forums/topic/193163-512k-cartridge-status/

     

    Bottom line summary of where we are at with cartridges:

     

    2007 or so, Carlos Randolph laid out an 8K ROM Board (v1) from a ROMOX board in ExpressPCB.

    2009 or so, I took that layout and made a 16K ROM board (v2) with a 74LS379 (inverted) to do the bank switching. I designed.

    2010 or so, I took that same layout and made a 64K ROM Board (v3) (you could also use 8K, 16K, and 32K ROMs with it.) I designed. The Gerber file was released for this.

     

    The 64K Boards are the last ones we've done any batches of (we being me, Matthew, and other people like Arcadeshopper). I have no concerns with others duplicating the work as long as my name or website stays on it (i.e. I get credit).

     

    Others that have been done but not mass released:

    • V4B - 128K x 4 board - a prototype 128K ROM board that used all the inputs/outputs on the 74LS379 (27C010 EPROM). Ksarul helped design. Had space for two switches (i.e. four ROM banks of 128K).
    • V5 never left the ground. It was going to be a GAL bank switched 512K cart with a 49F040 512K EEPROM. One rough draft exists in ExpressPCB.
    • V6H - Uses a 49F040 512K ROM, a 74LS378 to bank switch, and an Atmega 1284P for GROM simulation. I believe this is close to the current iteration of the one that we are going to produce.
    • V7D - Did get designed and printed; this used a GAL to replace the 74LS378/9, the 49F040 512K EEPROM, and the Atmega 1284 GROM simulator. We had issues with the GAL and decided to go with a 74LS378. Ksarul can elaborate why we moved from the V7 to the V8; I believe it was due to a different pinout on the ATMega 1284P.
    • V8J - A 512K ROM board (49F040 - PLCC EEPROM) with an optional Atmega 1284 controller that can simulate the GROM in conjunction with the ROM with the GAL 74LS378/8 type bank switching. Shelved due to issues with the GAL.
    • V9a - A 512K ROM board that uses a 74LS378 to bank the whole 512K (2764/128/256/512, 27C010/020/040) in 8K segments. An evolution of the V4B board with the 74LS378. Ksarul helped design.
    Ksarul can help correct anything that I missed. :)
AtariAge
Forums
Store
General
Follow AtariAge

Copyright ©1998-2021 AtariAge

×
×
  • Create New...