Buddha Flash phoenix edition - problem after flash with latest 52.101

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.
  • Hi,


    I have here TWO Buddha Flash phoenix edition - original both with factory 2nd.scsi.device 103.16 and buddha_atapi.device 120.4 (3.6.2000)


    Here works ALL OK.


    After flash with new 52.101 works one Buddha (flash IC CSI) OK (buddhascsi.device 52.101 in system) with IDE HDD BUT second Buddha (flash IC Samsung) NO - no visible any buddhascsi.device 52.101 in system. (HDD is of course 100% connected on IDE)


    I know buddhascsi.device 52.101 has "sleep" (not visible in system) when no any device on IDE port BUT have really problem with connected HDD on IDE port then here no problem or exists "bad detection" in latest 52.101 ?


    I have test with programmer (check Samsung eprom) and after flash (flashing in Amiga) is eprom OK - verification flash 52.101 file is in eprom ok.


    After re-flash with programmer back on old original 103.16 works too this second Buddha again OK.


    Test with same IDE cable and too HDD, same system and same A4000D


    Where is problem please at second Buddha Flash together with latest flash 52.101? (buddhascsi.device 52.101 not visible in system, Buddha as hw in system - ShowConfig or Buddha flashing tool/sw OK)


    Thanks. :)


    Regards, Sveta

  • This may sound weird, but please try to flash both phoenix edition controllers to the old version and then operaten them under Kick1.3 - is there any difference between the two?

  • Hi,


    Thanks. Both Phoenix edition flashed (external in programmer) with old (original previous version buddha_atapi.device 120.4) and works (device visible on system) both fine too under Kick 1.3 (A2000) then Zorro II


    One question - ShowConfig - Buddha Flash Xicor but on second controller (where is problem with latest 52.101) is Samsung Flash IC


    Btw. What MACH IC - identical content at all Buddha Phoenix or what fix with Phoenix X-Surf?


    Regards, Sveta

  • Both Phoenix edition flashed (external in programmer) with old (original previous version buddha_atapi.device 120.4) and works (device visible on system) both fine too under Kick 1.3 (A2000) then Zorro II

    This confirms that you do not have an old MACH version that has a false IRQ-readout bit. The error only shows on Kick1.3 and was quickly fixed after the controllers were released into the field. With the old version working under Kick1.3, you have confirmed that it's the new/fixed MACH revision.


    One question - ShowConfig - Buddha Flash Xicor but on second controller (where is problem with latest 52.101) is Samsung Flash IC

    Back when the flash tool was written, we used "Xicor" as a synonym for 32k flash space. The Samsung chips are software-compatible with the Xicor ones, so that's just a display flaw, but nothing we need to follow up on.


    Btw. What MACH IC - identical content at all Buddha Phoenix or what fix with Phoenix X-Surf?

    Only the MACH type is identical, but programming is of course different from the X-Surf für Phoenix. If you want to get to the bottom of the difference between the two controllers, you could try to exchange the MACH chips and see if the behaviour moves with the MACH chip - this may give hope for a change if you get a new MACH (programmed for Buddha-Phoenix of course).

  • Hi,


    Thanks for fast reply. OK test with exchange the MACH chips no problem for me and test with latest 52.101 ? When is here any positive change (device on second Buddha visible in system) is possible buy from icomp new latest MACH IC?


    Thanks Sveta

  • is possible buy from icomp new latest MACH IC?

    You have already confirmed that both are the "latest" with the Kick1.3 test. I suspect that there's a defect in one of the chips.


    Yes, I can (still!) program MACH chips here, so I can make new ones for old products.

  • Hi,


    Thanks for support! All my Amiga computers (ca. 150 pcs.) :-) has hardware from icomp.


    I must too detailed tested under 1.3 (have only "visible device test" and no e.g. booting or test with IDE unit on Buddha)


    Regards, Sveta

  • Hi,


    After swap MACH ICs second Buddha Phoenix works OK! and 52.101 is here in system. :-)


    First Buddha Phoenix with MACH from second - not visible 52.101 then confirm MACH IC from second Buddha is problem BUT works fine with factory buddha_atapi.device 120.4


    Why not working together with latest 52.101 - probably not damaged MACH IC


    Under 1.3 is not possible booting from partition - fix problem?


    No problem for me send MACH IC for reprogramming.


    BTW. "Not working with 52.101" MACH is 215 (Buddha from AmigaKit) and working 210 (Buddha with sticker icomp)


    Thanks Sveta

  • BTW. "Not working with 52.101" MACH is 215 (Buddha from AmigaKit) and working 210 (Buddha with sticker icomp)

    The 215 just has a few additional (async-)features that my design never used. The code in there is just a re-compiled version of the original 210 design, just with a different target chip.


    If you really see a difference in booting under Kick1.3, then the MACH215 obviously has the older version programmed. You could double-check that by reading offset $f80: I suspect that the value has the MSB set (use Megamon, CMON or any other debugger/monitor to read that address). The other MACH chip should have the MSB of that byte cleared.

  • Honestly? You've been ordering from iComp for years, and the address has not changed in the mean time. See any of the invoices you have, sender addresses on parcels, imprint of all web sites. Payment: I'll send an invoice after the service is done. Price: Make that a pay-what-you-want article, plus shipment, which is 11.31 EUR for international registered mail.