Buddha Flash (gold edition 1998) not working after flashing with latest V52.101

Caution: Non registered users only see threads and messages in the currently selected language, which is determined by their browser. Please create an account and log in to see all content by default. This is a limitation of the forum software.


Also users that are not logged in can not create new threads. This is a, unfortunately needed, counter measure against spam. Please create an account and log in to start new threads.

Don't Panic. Please wash hands.
  • Hi,


    This Buddha Flash (gold edition 1998) with 28F512 Flash chip works before flashing perfect. (original old EB device)


    After flashing "flashBUDDHA Buddha_ROM_52-101.bin" (flashing without any problem, Buddha found, ID 003, 64kB ROM detected too OK) and reset is all ok in early startup (vendor, working status) BUT in Workbench already no any visible buddhascsi.device (HDD connected)


    After check flash chip 28F512 direct in programmer found probably problem - flash chip is OK BUT data is "corrupted" - after compare with original V52.101.bin is flash chip without any buddhascsi.device and too rest content is curious.


    Flash chip is deleted direct in programmer and new loaded with V52.101.bin file (32kB)


    After return to Buddha Flash but still no any buddhascsi.device in Workbench. (HDD connected)


    Next test with others flashing in Amiga - same problem with "curious content" and after compare with first flashing in Amiga have identical "chaos" files/ROMs


    For test with old ROM (original) need original old device for my flash chip 28F512 - no any backup here. (EB 103.8 from Bflash archive?)


    Please any help about flash and idea with "curious content" in flash chip after flashing in Amiga?


    Why not working/help too direct programming in programmer with latest V52.101? (here content with buddhascsi.device after programming OK)


    Thanks Pavel

  • For test with old ROM (original) need original old device for my flash chip 28F512 - no any backup here. (EB 103.8 from Bflash archive?)

    the ebfinal.lha archive contains all the Elaborate Bytes ROMs. For the 64k chip, just duplicate the 32k content (load to $0000 and to $8000 of the buffer, then program).


    Please any help about flash and idea with "curious content" in flash chip after flashing in Amiga?

    Only very few Buddha controllers with 64k flash were delivered - probably less than 100. The tool does exactly what I described above: It duplicates the 32k contents, so the upper address line does not play a role. However, after 23 years, these early flash chips may have trouble with the direct 12V connection that I gave them on these controllers: Today's power supply madness on the Amiga market drives people to using unsuitable power supplies, and a digital chip requiring 12V that is directly driven from a noisy switchmode PSU may result in funny behaviour of the chip.


    The circuit itself and the flash tool are known-good, as the exact same thing was done on the Catweasel Z2 MK2 (the black "S-Class" model). This has survived lots of flash updates in the field, and it was a very popular controller at the time.

    Why not working/help too direct programming in programmer with latest V52.101? (here content with buddhascsi.device after programming OK)

    There are other customers who were successful flashing the new device to the old controllers, so it's not a software issue. Please re-try by duplicating the 32k contents - this may already do the trick.

  • Thanks for info about archive. I know 64k chip need duplicate the 32k content. I have test with programmer and duplicated V52.101 to flash 28F512


    I have reccaped original A2000 PSU and test too with others original PSU.


    I know and have too first Buddha 1997 IDE (eprom 27C256) in my collection - here works V52.101 great - informed too here in forum.


    Maybe is problem at my duplicate the 32k content - please is possible post here or via email V52.101 as finished 64k bin file direct for programmer? (compare with my 64k bin file V52.101 and for new programming test in programmer)


    Thanks for support.

  • Now have test with 103.17 from ebfinal.lha archive direct in programmer and probably right duplicating 32k to 64k but too not working.


    AMD flash chip is in programmer ok then what is wrong on card when before first flash in Amiga Buddha works perfect. MACH or GAL chip?


    Buddha is still recognized in Amiga as working and all is ok "only" without any device.


    Thanks.

  • Could it be that this controller was inserted the wrong way round at some point? This will blow up one of the LS245 drivers. Note that I only started adding the "resistors of shame" way later.

  • 100% NO


    Problem come really first after flashing im Amiga. Never inserted wrong. No problem public my 64k files (v103.17 + v52.101) for check but here probably all ok. (standart duplicating 32k to 64k ($0000 + $8000))


    Possible problem/contact in PLCC socket for flash chip? when card is still recognized in Amiga as working, vendor ok and flashBUDDHA tool too all ok:


    found BUDDHA controller number 0 at address $00E9...

    serial ID $00000003

    ROM size is 64kB.

    flashing now...


    or bad flash AMD chip after flashing in Amiga? (but in programmer all ok)


    On this Buddha works only AMD flash? I have here new 28F512 but from ST (probably ST vendor not working at direct flashing in Amiga but with programmer and after programming back to Buddha?)


    Thanks.

  • Possible problem/contact in PLCC socket for flash chip?

    If it's that, you can always load the Elaborate Bytes drivers from shell.


    On this Buddha works only AMD flash? I have here new 28F512 but from ST (probably ST vendor not working at direct flashing in Amiga but with programmer and after programming back to Buddha?)

    I have only designed it for AMD - you'd have to verify with the data sheets if the pinout and voltages are the same. Please understand that I won't spend any extra time on this after the sale of this product is more than 20 years ago.

  • Looks like there's more broken than just the flash chip. Once again, try loading the Elby drivers from shell and see if drives are identified on scsi.device (or 2ndscsi.device, depending on what computer you have).

  • Hi,


    Sorry for delay. Now ready for test but please possible detailed written what execute about "loading Elby drivers from shell"


    Started manually Elby Buddha rom file?


    Thanks!

  • OK. Thanks.


    After start from shell "BuddhaIDE" is here scsi.device 103.16 and buddha_atapi.device 120.5 (8.12.00) BUT still no any unit (HDD/CD)


    Test with or too without (empty flash plcc socket) flash chip.


    Please any idea where is possible hw problem? (GAL/MACH?)


    Many thanks for support old hardware.

  • This controller is about 23 years old. Has it always been in your possession, so you can really say that it was never inserted the wrong way round?


    LS245 drivers are not really expensive, and they are fairly easy to exchange. I'd try that first.

  • OK. First then preventive exchange LS245 IC in SOP-20 package - right?


    Buddha Flash has 4x this LS245 then all exchange or you have please recommendation on direct 100% defective (only one?) LS245 (when e.g. Friend at test really inserted wrong)


    What is interesting - Buddha works ok and problem come really first after flashing with latest device in my A2000D.


    Many thanks and after exchange write here news. First must find/order right package LS245 ICs.

  • The narrow package is hard to find on the market - you can take a normal SO20 and bend the pins inwards. The chip will stand up a little from the board, but it'll work.


    I don't have the chip designators in my head, and I used an entirely different CAD system 24 years ago, so "quickly looking it up" is not an option. The chip most prone to failure is the one that handles the odd byte of the 16-bit bus. Since I must have placed the chips "suitable for routing", it should be the chip that's "bottom left on the board" if you have the Zorro connector facing you, the IDE ports away from you and component side up.