Subject: Re: My problems with merged memory
To: None <>
From: Niklas Hallqvist <>
List: amiga-dev
Date: 03/07/1994 07:55:11
>>>>> "Niklas" == Niklas Hallqvist <> writes:

Michael> What was the address in a1?  I'm wondering if the long word
Michael> access is being made at an odd-word alignment.  Also, is the
Michael> access to 32 bit or 16 bit memory?

Niklas> The a1 address varied from run to run, probably due to such
Niklas> misbehaviour earlier in the process.  From what I recall (I
Niklas> have now unplugged the 16bit-mem, trying to build a sellable
Niklas> machine out of my 2nd A2000!) the address were indeed odd-word
Niklas> aligned.

I put the card back in and found that the address sometimes were long-
word aligned as well.

Niklas> I have tested the board both with and without the DTACK jumper
Niklas> to no avail.  Where did you get the tip about this jumper
Niklas> from?  And what does it mean?  I don't have but a german m68k
Niklas> reference, and I don't understand tech-german very well :-)
Niklas> Hell, I don't even understand common german :-)

I tried to decode it last night, and have now an idea of what it is
used for.  DaTa ACKnowledge is driven low as long as the requested data
is invalid, or if it were the opposite, in order to make slow memory
(or slow memory-mapped registers) safely deliver values to the CPU.
Is this correct?  I don't think the 16bit mem is to blame in this case
though, but rather the combo's bus driving logic.  Have there been
complaints of GVP accellerators not listening to the DTACK signal of
the bus?  That could explain why my bridgeboard doesn't work too good
in my '030 machine.

Still asking, anyone know an electronic address of GVP tech support?