Welcome, Guest. Please login or register.

Author Topic: A4000 problem: IDE = good, ATAPI = bad?  (Read 2765 times)

Description:

0 Members and 1 Guest are viewing this topic.

Offline spiranthoTopic starter

A4000 problem: IDE = good, ATAPI = bad?
« on: April 24, 2013, 09:19:10 PM »
Hi everybody,

I'm trying to repair an A4000. It's working more or less nicely now, except....

The IDE bus is acting very strange.
With an HDD (120MB) on master, it's fine.
With a CF (256MB) on master, it's fine.
With HDD (Master) and CF (Slave) it's fine.

However... put any ATAPI device on the bus....
Put a CD-ROM (Slave) on with the CF and the CF gives lots of read errors. The CD-ROM is not detected.
Put a ZIP250 (Slave) and the A4000 gives a red screen instead of booting. I had it booting once or twice - it never appeared on HDToolBox though.
I have one CD-ROM which is detected by the IDEFIX97 program (the other three I tried don't appear). The name string is garbage. The device always gives no disk present. Eject CD command does work though.

The A4000 is running OS3.9 at the moment. I've also tried with IDEFIX97. FindDevice program gives a recoverable alert of 0BADC0DE.

Does anyone know why this might be happening? Why my CF Card and HDD work, but anything at all ATAPI just blows everything up.

Thanks for any help!
--
Ian Gledhill
ian.gledhill@btinternit.com (except it should be internEt of course...!)
Check out my shop! http://www.mutant-caterpillar.co.uk/shop/ - for 8-bit (and soon 16-bit) goodness!
 

Offline Dr.Bongo

  • Sr. Member
  • ****
  • Join Date: Jun 2003
  • Posts: 342
    • Show only replies by Dr.Bongo
    • http://www.c64radio.com
Re: A4000 problem: IDE = good, ATAPI = bad?
« Reply #1 on: April 24, 2013, 09:23:08 PM »
Have you tried a different IDE cable?
38911 BASIC BYTES FREE, less when I`ve had a drink!

 *** http://c64radio.com *** Commodore 64 Radio 24/7 for free!
 

Offline spiranthoTopic starter

Re: A4000 problem: IDE = good, ATAPI = bad?
« Reply #2 on: April 24, 2013, 09:43:57 PM »
Yes, and the HDD wouldn't work if that was the problem...
It's a real puzzler!
--
Ian Gledhill
ian.gledhill@btinternit.com (except it should be internEt of course...!)
Check out my shop! http://www.mutant-caterpillar.co.uk/shop/ - for 8-bit (and soon 16-bit) goodness!
 

Offline Ami_GFX

  • Sr. Member
  • ****
  • Join Date: Apr 2007
  • Posts: 379
    • Show only replies by Ami_GFX
Re: A4000 problem: IDE = good, ATAPI = bad?
« Reply #3 on: April 24, 2013, 11:13:21 PM »
Try swapping the positions of the 2 devices on the cable. I had a similar problem with a CF and Atapi CD/DVD. It worked fine with a hard disk but it didn't with a CF card adapter. I switched the order of the devices on the cable and it worked fine.
A2500 owned since 1993 with A2630/DKB 2632, DKB Megachip, GVP EGS Spectrum, A2320 and GVP HC+8 on the inside and a DCTV on the outside. A4000D with CSPPC, Cybervision 64 and a Flicker Magic flicker fixer. A4000T Toaster Flyer & CSMKII. All systems completly retro and classic and mostly used to do geometic art as in my avatar.
 

Offline spiranthoTopic starter

Re: A4000 problem: IDE = good, ATAPI = bad?
« Reply #4 on: April 25, 2013, 10:41:42 AM »
It's not the cable, unfortunately. Tried three cables. The same happens.
Doesn't matter where on the cable it is, master or slave - no ATAPI device works - Zip, CDRW, CDROM.. they just don't appear. Also, the Amiga takes 30 seconds or whatever to boot, as if nothing is on the bus.  If there's a CF Card, IDE or nothing on the bus... same thing, ATAPI devices just don't get seen. Even a Zip disk with a valid RDB doesn't get seen.

Really puzzled...
--
Ian Gledhill
ian.gledhill@btinternit.com (except it should be internEt of course...!)
Check out my shop! http://www.mutant-caterpillar.co.uk/shop/ - for 8-bit (and soon 16-bit) goodness!
 

Offline nicholas

Re: A4000 problem: IDE = good, ATAPI = bad?
« Reply #5 on: April 27, 2013, 11:31:38 PM »
I had very similar problems on an A1200 a few years ago.  I can't recall how I solved it at the moment but once the subject matter has filtered through the old grey matter for a day or two I'll probably remember. :)
“Een rezhim-i eshghalgar-i Quds bayad az sahneh-i ruzgar mahv shaved.” - Imam Ayatollah Sayyed  Ruhollah Khomeini
 

Offline Rob

Re: A4000 problem: IDE = good, ATAPI = bad?
« Reply #6 on: April 28, 2013, 03:28:29 AM »
Stupid question but do have a mount list for e.g. CD0: in devs/dosdrivers.
 

Offline spiranthoTopic starter

Re: A4000 problem: IDE = good, ATAPI = bad?
« Reply #7 on: May 11, 2013, 08:58:09 AM »
Yes, it has a DOSdriver. :) It's just not detecting it at all, though - it's not finding the ATAPI devices on a IDE device scan.

After looking at it some more I've decided to leave it until I can get another A4000 in (no way I want to risk my own precious A4000!). I reckon it must be either Fat Gary or one of the support logic chips, maybe the PLA thingies too.

Very weird how it works ok with an IDE hard disk though. In the meantime the customer whose A4000 it is is borrowing my Oktagon 2008, thereby bypassing the problem.

If/When I finally find out what's been causing this I'll post again in case it helps other people.

Thanks for the help though!
--
Ian Gledhill
ian.gledhill@btinternit.com (except it should be internEt of course...!)
Check out my shop! http://www.mutant-caterpillar.co.uk/shop/ - for 8-bit (and soon 16-bit) goodness!
 

Offline danbeaver

Re: A4000 problem: IDE = good, ATAPI = bad?
« Reply #8 on: May 12, 2013, 01:27:22 AM »
The A4000's IDE is just a simple one:  it simply expects a HDD to take a read or write command, nothing with a tape drive, CDROM, ZIP drive, etc.  the extra instructions needed for these devices is in the software layer added by the ATAPI programs. They convert the control commands for non-HDD devices to work on the bus. More sophisticated IDE controllers have this in their firmware.

The fact that the HDD part of the controller works means that it is working at its basic level. The buffering may be affected, but I'm not able to tell you how to test it much less fix it. You could add a buffered device connector like they have for the A1200 and check it that way. I'm guessing that it is either something like that or you may need another ATAPI software program such as AsimWare's AmiCDFS Program (3.9 is the latest I've seen) to locate and configure the devices. There is also a difference in the IDEFix on Aminet and the full one on the Amiga Forever DVD that has the registration code -- I use the latter.
 

Offline Astral

  • Sr. Member
  • ****
  • Join Date: Nov 2005
  • Posts: 491
    • Show only replies by Astral
Re: A4000 problem: IDE = good, ATAPI = bad?
« Reply #9 on: May 12, 2013, 03:33:52 AM »
I've stuffed around with similar things before and had varying success - on A4000, 600 and 1200 - with a lot of hair-pulling frustration to go along with it! And if you google search and so on there's all kinds of non-matching opinions on what can and can't be done, and how to do or not do it!!! AAAAAAAAAAAAaaarghhh!

Maybe ATAPI is simply not compatible? Maybe you need a buffered interface?

Perhaps try with and without IDEFix? (My experience is that I've not needed it from memory).

With the "no disk present" scenario did you try "Diskchange CD0:" to see what happened?

What happens if you only have the CD drive connected and you boot from floppy to mount the drive? Does it work fine?

I have recently had a HD and CD on the IDE header on my 4000 while tinkering with 4.1 Classic and it worked fine....but...a few months before I had trouble getting a CF card and CD (not sure if it was the same drive as when the HD was connected) on the IDE header to play fair! I'm not sure on the spec of the CD drive that worked though (IDE/ATAPI ???).

You say there is a boot delay of 30 secs as "if nothing is on the bus". From memory if there is nothing on the bus you get the opposite effect - no (or very little) delay. Maybe the delay is because it is trying to "detect" it but can't?

One approach, given you have had SOME success, is to connect the only CD drive you say worked fine and tinker using that one in varying combos of software/configs and so on and see what happens?

If it helps I have a fair few hardware bit that I can try for you and report the results.

Good luck! :D
 

Offline danbeaver

Re: A4000 problem: IDE = good, ATAPI = bad?
« Reply #10 on: May 12, 2013, 05:13:20 AM »
The ATAPI program that has to run on top of the firmware controlling the IDE interface has the proper commands to rewind a tape, open/close a CDROM tray, read from tape & CD, etc. These are built in to the IDEFix program, the CDFileSystem's and such. They are built into the CS MKIII and CSPPC; when the latter in combined with OS 4.1, it can boot from a CD drive. Don't expect the A4000 to do anything more than boot from a HDD with 3.1 ROMs.
 

Offline spiranthoTopic starter

Re: A4000 problem: IDE = good, ATAPI = bad?
« Reply #11 on: May 12, 2013, 11:20:20 AM »
I tried with OS3.9 and IDEFix 97...

The problem is that the drivers can't even detect the drive. The IDEFix program FindDevice finds only one CD-ROM drive, and on that one the ID string is all corrupt. That's the one that mounts ok and can eject, but that's all. FindDevice in fact usually gives a recoverable error with the code 0x0BADCODE (which means an invalid code has been passed to the controller, I believe)/

So it's definitely not a driver issue, because the same set-up on my A4000 detects the CD just fine.

It's just really confusing that it boots ok off the old 120MB HDD and a 256MB CF Card, but no ATAPI devices at all. If I connect a Zip drive (ATAPI) I just get a red screen. If I boot OS 3.9 with just the CF card, it'll hang when it reboots from Setpatch - with a non-working CD-ROM drive as slave, it boots fine (but can give errors). It's just really odd.

As soon as I get another 4000 to play with I'm determined to track this one down, though!
--
Ian Gledhill
ian.gledhill@btinternit.com (except it should be internEt of course...!)
Check out my shop! http://www.mutant-caterpillar.co.uk/shop/ - for 8-bit (and soon 16-bit) goodness!
 

Offline Thomas

Re: A4000 problem: IDE = good, ATAPI = bad?
« Reply #12 on: May 12, 2013, 12:00:13 PM »
Might be a power issue. Perhaps the power supply is dying and does not provide enough power to run two drives.

Offline spiranthoTopic starter

Re: A4000 problem: IDE = good, ATAPI = bad?
« Reply #13 on: May 12, 2013, 12:51:42 PM »
Not the PSU unfortunately. I checked voltages which are ok, and that would be unpredicable, where as the symptoms I can see are consistent. Also I checked with just the A4000 and ATAPI drive and there was no change.
Thanks for the idea though!
--
Ian Gledhill
ian.gledhill@btinternit.com (except it should be internEt of course...!)
Check out my shop! http://www.mutant-caterpillar.co.uk/shop/ - for 8-bit (and soon 16-bit) goodness!