DiagROM V0.8 Released

Yet another time it is time to call it “Stable”

 

What have happend since last time?

Well. Now I have added exceptioncode handling, so if there are a crash or something like that, the machine will try to handle it, also telling the user what happend.

Chipmem detection in the beginning is now totally rewritten.
Now while detecting (actually done by writing a series of data to one longword every 32K block and reading it back) it will show on the serialport what address it is working on.  AND if there are a failure it will print what it wrote to the memory and what it read so you can see biterrors.  to make detection somewhat fast only combinations of 00 and $ff is used.

If there are errors in the beginning it will check until 2MB is checked or memory is detected, if errors comes after memory is detected it will take that as there is no more memory avaible.
DiagROM will use the end of detected memory as workspace.

There is a CIA Check aswell. this have space for improvements and will HOPEFULLY give correct result, I have atleast sucessfully tested it with removing a CIA and got fail.  also by removing the TICK signal on the A3000/2000 will also fail the test so. guess it works. 🙂
(OF some reason, if you remove the ODD CIA, the machine will crash in the start, will try to figure out how to handle this)

At last stable version I told that I will open the source. and YES I will someday. I just need to figure out how to handle it.  YES,  GIT etc exists. I know.  it is not really about that point. the point is to handle how to implement changes etc.  so I might instead just do so that I will include the source, no verisionhandling etc. and if someone does a change, this person must hand me the original source AND changed source. so I can diff it out.  not sure yet.

I have got question about supporting my project with donations.

Well sure. if you want to. just use: http://paypal.me/JohnHertell

I also got questions about sending preprogrammed ROMs. Sure also, no problem BUT think if that this is a program in constant change. so unless you have a eraser and epromprogrammer you will have a very outdated version real fast. but if you still want this.  contact me.

It will now be some time until next stable version as I need to finish more tests first.

Anyway, Enjoy and please comment.

 

And if you cannot find it: Downloadpage is located at: http://www.diagrom.com/index.php/download/

14 thoughts on “DiagROM V0.8 Released”

  1. I’ve just programmed an EPROM with DiagROM 0.8 and tested it in a Amiga 500 Revision 5 – works great! Thank you very much! I’m looking forward to get more functions. 🙂

  2. Hi John.. thanks for starting this nice work!! i tried out your “daily”-version on my Amiga 1000 (using a Kickstart-Disk, which seams to be the easiest way to use your ROM).
    i tested with various variants (rev. a, rev. PAL, with and without FastRam) and i encountered a lot of problems 🙁
    1.even though i have fastram, and it is detected and shown in the bottom-line of your screen, the FastMem-Test quits with “no memory found”.
    2. neither CIA nor IRQ-Tests work .. the tests start but immediately crash saying “oOoooops Something went borked” (whatever that means).
    3. your new “GFX-Test” gives only bizarre patters on the screen and in 1/3 cases it crashes the whole maschine.
    4. sometimes keyboard-inputs are not displayed although the key is supposed to be working fine (the keyboard was tested before).
    i can be a beta-tester for Amiga 1000, if you like.. so please feel free to contact me.
    regards, Rene (aka fook42)

    1. I know the fastmem check needs some work as it actually says that no memory is found on all machines except A3000/4000.. will soon redo that handling.

      that errormessage us a processor exception. and this is on a working machine?

      yeah gfxtest sure needs rework. it works in uae but on real machines it just goes bananas..

      for betatesting. well.. that is the daily versions.. 🙂

      1. the chipmem-testing works fine on these machines, so maybe its not a big bug with the fastmem-tests… 🙂
        yes, the machines (two A1000) are working fine – both equipped with the 68000 at 7MHz. and i never got any Problems with this.. so maybe your routines are accessing some memory-areas where the A1000 has some specialities?
        i really appreciate your work! keep it up! 🙂
        whenever a new version drops out, i’ll test it and let you know the results or is there anything i can share to you in order to find the bugs?

  3. Hi, could you recommend some online documentation for EPROM programming (for dummies)?

    Also, what’s wrong with github? Just make sure to choose the right OSS license that suits your requirements first. Handling change would be the same as for most projects on github: forking + PRs.

    Cheers

    1. Nothing wrong with github.. it is more how I should handle stuff.. 🙂

      Anyway. guide how to program? well.. load file. put in prom in programmer. “program” 🙂

  4. Thanks a lot for this piece of software!
    I have a few questions:

    1. What hardware do you recommend for burning the EPROM?
    2. Is it possible to reflash an existing ROM?
    3. Do you think a ROM checksum error (red screen at boot) can be caused by a leaked capacitor?

    1. 1. I use a Tl866a with a 27C4096->27C400 adapter. but the genius G840 also works fine.
      2. yes. by erasing it in UV Light.
      3. weell eveerything is possible but nothing I hacve noticed.. but on the 2000. VARTA can do that.

      1. Thanks for your reply but when I try to go to the url chrome says the webpage can’t be reached (I’ve tried several browsers)

  5. I just managed to flash my original two EPROMS with the help of a collegue. I have backed up the kickstart beforehand and it seems just one hex value was corrupted (!). As you know the code is copied twice on the chip, and one copy was good. Too bad there isn’t a fallback on the copy when checksum is bad. Anyway, diagrom works great thanks again!

Leave a Reply

Your email address will not be published. Required fields are marked *