ACA1233n Hangs at boot with IDE speeder technology enabled

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.
Don't Panic. Please wash hands.
  • When switching on the IDE speeder modes The Amiga will hang on a cold boot. However The system will boot fine from a a soft boot and sysinfo shows the increase in transfer speed. If I continue to use the ide boots modes the file system becomes corrupts after sort amount of time, normally after a cold boot.


    The card is installed in a recapped 1d.1 revision a1200. I removed the E121C and E122C and I'm using the original power supply. The IDE to CF adapter is on a short IDE cable and the bottom RF shed is installed. Top shield not in use due to an indivision. The Cf card is isolated from the indivision to prevent shorting.


    any advise on that what steps I can undertake to get this stable? thanks

  • Hi Jens,

    apologies for the late reply. I've tried just IDEMODE 1 and two different CF cards and while the system seems to work fine after a soft reboot. Turning the Amiga off for about 30 - 60 seconds then back on results in the system posting, it then seeming to load ACATOOL, only for the system to hang with the activity light on the CF adapter permanently on. The only way to then boot is to hold down the left mouse button to bypass ACATool. Please see example video working link - https://photos.app.goo.gl/qeaXqRp6Xc7yJRKV9

  • Can't view the video. Please describe the config/Setup first - no need for a video. Include the type of adapter, type of power supply and other hardware in your system.


    We have ACATool V2.1 in the works, not sure yet if Timm has already uploaded it (I'm late trying it out). It does at least one fix for MapROM if not loaded from file, and it has multiple fixes for commandline options, which are always used if the tool is executed in startup sequence.

  • I fixed the link so it should work now... The power supply is the original Amiga 25w, its a 240v with a UK plug that I'm using with an adapter for use in Australia. The CF card adapter is one of the cf-ide 44/2.0mm vh2. The only other hardware other then the ACA1233n is a MK2cr and a RTC on the Amiga's clock port. The system is running Amiga os 3.9 with Classic Workbench.


    Yeah I saw 2.1 was out and did all my test with it last night before my last post. Please let me know if you need any other details I may have missed.

  • ACAtool V2.1 is online now. Please see if that improves your situation.

    didn't seem to make a differences. I'm wondering if it has to do with setpatch? It seems on a cold boot the Amiga locks up after the first reboot from setparch.

  • Setpatch never caused any trouble with accelerators "below" 68040. Maybe it's worth taking a closer look as to why one of the wire links was ripped off: Is there maybe a trace damaged, too?

  • Just the basic settings when turning it on in ACAtool, for example: acatool >NIL: skiponlmb force noerrors idemode 1


    I've moved it to just after "End OS3.9 Checks" and while it no longer crashes sysinfo shows the the transfer speeds for DH0 DH1 as same as stock.

    The only other hardware change from when I created this post is I swapped the kick start ROM out with a '"lobennto ROM 3.X. kick start" to see if that would make a difference, but it did not.


    Here is the startup sequence, minus the acatool code, its mostly stock classic wb a part from code for my pcmcia CF card.


  • Why didn't you attempt to launch ACAtool from the launched WB and see if the IDE speeder is on or off, like I asked? Is the tool crashing, or did this just slip your attention?


    That startup-sequence is full of patches that are of questionable use - FBlit should be removed for sure, and why do you use Cyberpatcher if you don't have an 060?


    What's the "maxmem" thing? Why is XSurfIDE even looked for?

  • Quote

    Why didn't you attempt to launch ACAtool from the launched WB and see if the IDE speeder is on or off, like I asked? Is the tool crashing, or did this just slip your attention?

    My apologies, when you said latter, I assumed you where talking about placing the acatool latter in the startup-sequence. Thus to confirm, if I run the tool after workbench has started, and the tool's removed from the startup-sequence, I can apply PIO mode 4- and activate with no problem or crashes. I do however have to perform a reboot via Ctrl-Amiga-Amiga before sysinfo reports faster read speeds.


    Quote

    That startup-sequence is full of patches that are of questionable use - FBlit should be removed for sure, and why do you use Cyberpatcher if you don't have an 060?


    What's the "maxmem" thing? Why is XSurfIDE even looked for?


    Like I said Its a ClassicWB startup sequence (apologies maybe I should have been upfront about the use of the "community pack"). I'm very new to Amiga ownership . I'm learning as I go, I guess you could say I know enough to be dangerous lol. Bloodwych's ClassicWB community pack seemed like the bang for ones buck after much trail and error.


    Quote

    That startup-sequence is full of patches that are of questionable use


    I also questioned if some of these patches may have been messing things up (do keep in mind XSurfIDE and Cyberpatcher are disabled by default, that's why they are raped in if statements, they can be turned off and on by the user and the startup sequence checks to see if they active).


    So I did a clean, stock install of workbench 3.9. I used ACAtool to add its self to the startup-sequence with the IDE boost turned on but the system still lock up on a cold boot. Given that this was tested back when I was using the 3.1 kick start, I decided to test again, before posting, with my newly installed 3.x roms.... and would you believe I can boot from a cold boot!


    While testing under the clean install I also noticed that I don't need to soft reboot ( Ctrl-Amiga-Amiga) to see the speed boots take affect. I also don't think I ever seen seen the tool remember the IDE Accel. setting.


    After all this time it looks like I have an answer. One of the patches in ClassicWB is possibly the root of the problem, or maybe the version of setpatch it uses... also switching to a different kickstart or some combo of all 3? I'll need to do more investigating. Once again Forgive me for not informing you about using ClassicWB at the start. These packs seemed so popular I just assumed most where using them *face palm*. I appreciate you sticking with me and providing support.

  • if I run the tool after workbench has started, and the tool's removed from the startup-sequence, I can apply PIO mode 4- and activate with no problem or crashes. I do however have to perform a reboot via Ctrl-Amiga-Amiga before sysinfo reports faster read speeds.

    This confirms that the hardware and our tool are working fine. The delayed activation is a miracle though, as the tool writes to the registers immediately and does not remain in memory. The reboot must trigger something else out of our control.

    Bloodwych's ClassicWB community pack seemed like the bang for ones buck

    Did you pay for that?


    While testing under the clean install I also noticed that I don't need to soft reboot ( Ctrl-Amiga-Amiga) to see the speed boots take affect. I also don't think I ever seen seen the tool remember the IDE Accel. setting.

    The ACA1233n does not have an EEPROM or Flash to store settings in. The ACAtool therefore writes all settings to the startup-sequence and also attempts to read it from there (parses it's own command line). So the operation should just be transparent to the user - as long as there is no other software that torpedos ours, which appears to be the case here.

  • Quote

    This confirms that the hardware and our tool are working fine. The delayed activation is a miracle though, as the tool writes to the registers immediately and does not remain in memory. The reboot must trigger something else out of our control.

    So rebooting with ctrl-Amiga-Amiga is no different (other then convenience) to turning the Amiga of at the power brick, and the registers should be getting whipped?

    Quote

    Did you pay for that?

    lol hell no, figure of speech, just implying it provided most of the pre-installed app / tools I was looking for and thus saved a lot of time as I experimented and learnt the OS and subsequently broke stuff ;)

  • So rebooting with ctrl-Amiga-Amiga is no different (other then convenience) to turning the Amiga of at the power brick, and the registers should be getting whipped?

    A keyboard reset will not change the register settings, but a power cycle will (hence the mechanism that stores the settings in the startup-sequence command line).

  • Hi.
    Exactly same situation.
    New ACA1233n-40 with IDE accelerator. It hangs up with ClassicWB 3.9 installation when PIO mode is active.
    I am also searching which patch or useful add of ClassicWB is the reason.

    regards Peter

  • As promise I did some more investigating... and well I'm not better off then when I started.


    I removed all the references to the extra patches from the ClassicWB version of os3.9 but with no luck. To make matter worse I was finding I could no longer get the ide boost to work at all. Plus now the clean os3.9 install was now locking up at boot as well.


    So I did a fresh install of both version of the OS on two different CF cards. This time I was able to get the IDE boost working even with out the need to Ctrl-Amiga-Amiga reboot in both installs. plus I hadn't removed any of the extra patches from the startup-sequence.


    The next day I was finding both installs where having intermediate locks ups on boot. First I suspected maybe it was the gotek drive that was contented to the rear floppy port. All most every time I powered on with it plugged in the system would lock up. removing it and the amgia would boot fine.


    Then I remembered I was still using a 3rd party PSU so I tested the original commodore PSU with the gotek drive and it would boot fine. back to the 3rd party would result in a lock up. ok cool it looked like I had a repeatable test that had the same out come. until I went back to the commodore PSU... and it locked up. not every time but some of the times, with and with out the gotek drive. same would happen with other PSU.

    I looked back at my past video and the gotek drive was not connected.


    In conclusion the Amiga boots more times then not with a clean install of both OS 3.9 and the classic WB version with the IDE boost set the startup-sequence. The usage of 3rd party PSU and a gotek drive on the rear floppy port result in more constant lock ups at power on. However switching to an commodore PSU improve the odds of booting but would still lock up with or with out the gotek drive, but more offtern with it not plugged in.


    The only other point of notice is I did see two yellow screens during boot up. The first time I had to reboot to pass it. The other time the system still booted fine.