CDRom on Buddha 20th edition - Different behavior on each port

angelo.carzis · Sep 26th 2020

Jens,

I have this issue which I am curious as to why it's occurring ...

My system: WB 3.1.4, A2000 system, Buddha IDE 20th edition, latest driver

When my CDrom drive is attached to the southern most IDE port (south being the front of the A2000), the Amiga doesn't always detect it on startup.

When the drive is attached to the northern most IDE port, there are no issues.

In the devs mountlist entry, the unit number is obviously being changed accordingly.

Note, I also had this intermittent issue with WB 3.1 (prior to upgrading to 3.1.4) with the CDRom in the southernmost port.

It didn't occur to me to swap IDE ports at that time.

I can live with the current arrangement. All is fine.

I'm just curious as to why this is occurring.

Thank-you.

Jens · Sep 26th 2020

Quote from angelo.carzis

My system: WB 3.1.4, A2000 system, Buddha IDE 20th edition, latest driver

Does that mean there is no accelerator in the system? I'm asking because the accelerator has great influence on the Zorro timing.

Quote from angelo.carzis

When my CDrom drive is attached to the southern most IDE port (south being the front of the A2000), the Amiga doesn't always detect it on startup.
[...]
I'm just curious as to why this is occurring.

The ports are electrically identical, except of course for the select-signals, which are generated by the same TTL chip and the IRQ signals. Since it works "sometimes", I'd start by investigating if this is a cold-start or a warm-start issue, as it may well be the Amiga booting very fast from your boot medium with both ports occupied (=Buddha startup is not running into any timeout because it finds devices on both ports), but the CD drive didn't start yet.

If that's the case, you can just insert a "wait 3" command (for 3 seconds) in the startup-sequence before the extra devices are mounted. Or you can just leave it like it is, as you probably have better things to do than to fix a working system

angelo.carzis · Sep 26th 2020

My 68030 card is still on loan to a friend.

Right you are. Don't touch if it ain't broken.

I'm going to investigate anyways as I'm curious.

Thank-you Jens.

angelo.carzis · Sep 27th 2020

I experimented a bit yesterday and found some interesting results. Sharing this information.in case anyways observes the same issues that I faced ..

I decided to change the IDE channel on the CDRom from master to slave and place the CDRom back on the IDE port which was originally intermittent. The result now is, there are no boot problems.

What I observed is that if the CDRom is unit/device 0 in the mountlist, issues will arise.

If it is anything other than 0, there are no issues. This may be an OS issue.

Jens · Sep 28th 2020

Quote from angelo.carzis

What I observed is that if the CDRom is unit/device 0 in the mountlist, issues will arise.

If it is anything other than 0, there are no issues. This may be an OS issue.

I don't think so - when Atapi CD-Rom drives were "recent", they were almost always developed for the case "it's a slave to the master/boot HD". This means that these devices were almost never tested as "Master", so ever since I've made the Buddha controller (first prototype in November 1996), the general rule of thumb was "jumper CD-drives to slave, then they behave much more like IDE/Atapi devices".