IO Stability Issues with Buffered IDE2CF Adapter, IDE Speeder Enabled, and Amiga 1200 Rev 1.B

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.


Please understand that you need to create an account to be able to post, guest posting was disabled as an anti spam measure.

    Hello,


    I’m encountering IO stability issues when using the iComp buffered IDE2CF adapter with IDE speedups enabled on Amiga 1200 motherboards with revision 1.B. Specifically, in this configuration, the system often fails to boot or experiences unstable IO behavior.

    * Amiga 1200 motherboard, revision 1.B (tested on two separate boards)

    * Buffered IDE2CF adapter (iComp A1200/A600)

    * IDE speeder enabled on the following accelerators: ACA1240/1260; ACA1234; ACA1221lc


    Working Configurations:

    * Unbuffered IDE2CF adapter with IDE speeder enabled on A1200 rev 1.B

    * Buffered IDE2CF with IDE speeder disabled on A1200 rev 1.B

    * Buffered IDE2CF with IDE speeder enabled on A1200 rev 1D1 or 1D4


    It appears that the board revision (1.B vs 1D1/1D4) is the key differentiator affecting IO stability.


    Are there any hardware mods, tricks, or configuration tweaks that could improve compatibility and stability in this setup? Any insights or suggestions are greatly appreciated!


    Best regards,

    Jaroslav P.

    First thing I'd look at is other components on the clock port, as that loads the same signals as IDE does.


    ACA1221lc has pretty relaxed timing - it should be compatible with just about anything. ACA1234 is a bit more aggressive, and ACA1240/1260 can be set to values close to the hardware limits, but also close to original speed. So in order to check if it's speed-related or just related to "speeder enabled at all", the ACA1240/1260 is the best test bed.


    Further, please write down the exact chip designations on all Gayle chips. Your 1.B boards may have an older version that's acting up during accelerated IDE access.


    Jens

    Boards of my and theirs GAYLE chips (are the same version):

    1B : 391424-02 / AA-GAYLE-R5

    1B : 391424-02 / AA-GAYLE-R5

    1D4: 391424-02 / AA-GAYLE-R5

    1D1: 391424-02 / AA-GAYLE-R5

    > First thing I'd look at is other components on the clock port, as that loads the same signals as IDE does.


    no CLOCKPORT devices are in use nor any PCMCIA


    > ACA1221lc has pretty relaxed timing - it should be compatible with just about anything. ACA1234 is a bit more aggressive, and ACA1240/1260 can be set to values close to the hardware limits, but also close to original speed. So in order to check if it's speed-related or just related to "speeder enabled at all", the ACA1240/1260 is the best test bed


    ACA1221lc:

    - just unstable, sometimes the "IO operation" fail if "speeder on" and I need to powercycle the Amiga

    ACA1234:

    - do not even boot from internal IDE + buffered ide2cf if speeder on OR

    - boot and works from ACA1234 CF slot like a charm till touching internal IDE slot

    ACA1240/1260:

    - do not even boot internal IDE + buffered ide2cf if speeder on

    note: there is no issue if I swap to non-buffered ide2cf


    > Further, please write down the exact chip designations on all Gayle chips. Your 1.B boards may have an older version that's acting up during accelerated IDE access.


    1B : 391424-02 / AA-GAYLE-R5

    1B : 391424-02 / AA-GAYLE-R5

    1D4: 391424-02 / AA-GAYLE-R5

    1D1: 391424-02 / AA-GAYLE-R5

    What's the warranty ID of the buffered CF interface? What speed are you selecting with the ACA1240/1260? What brand of CF card is this?


    Jens

    Quote from jaroslav.pulchart

    I try piomode 4 (no override)

    Please do try slower modes as well. The main question to answer is: Is this speed-related, or is it related to the underlying technology that we're using to speed up the IDE port? The observation that this is happening on 1B boards only indicates that it must be something special about that board. If the Gayle version is identical, I take it that none of them has the additional GAL chip installed that's meant to iron out the kinks of the older Gayle version regarding IDE? Please take pictures of the Gayle area and precisely describe if IDE speeder+buffered interface work in it or not (ideally, write that into the picture, so there's no confusion possible).


    Jens

    > Just to be complete: please also tell what CPU Speed Profile you are using


    ACA1234@50

    ACA1240@33

    Switching between: ACA1260@50, ACA1260@60, ACA1260@95, ACA1260@100


    > Please do try slower modes as well


    piomode 3 with the same issue


    > Please take pictures of the Gayle area


    Attaching pictures of the boards:

    * the problematic 1B board are with RED mark @ PCMI slot

    * the OK 1D board are with BLUE mark @ PCMI slot

    Quote from jaroslav.pulchart

    piomode 3 with the same issue

    that's still a very much accelerated mode. Please try even slower modes 2 and 1 and see if it changes anything. I still want to know if it's speed-related, or if it's the underlying technology that the 1B boards don't like.


    Jens

    We have another case where PCMCIA and IDE are not 100% stable, and it's getting more stable if the customer removes the RTC module. This case, and the other one have in common that they are 68040@33MHz.


    Do you experience the same problems when running at 25MHz?


    Jens