Date   : Tue, 05 May 2009 12:41:09 +0100
From   : Charlie.Robson@... (Charlie Robson)
Subject: Strange Atom behaviour

I have been experimenting with my MMC code in order to see if I can increase 
the compatibility. I've had some success. Of my many different MMC cards
I now have a compatibility rate of 90%, up from 20% originally. Small changes
in timing and protocol have helped.

I developed the new driver code as an atom program loaded from card. Flushed 
with success I replaced the appropriate sections in my existing code and
burned it. Epic fail. No card is recognised at all, yet identical code can
successfully boot the cards when run from RAM.

Are there wrinkles to be born in mind when accessing ports from the utility 
ROM address range? Speed issues caused by contention? Will driving the 6522
from #Axxx cause unfavourable sunspot activities and hence break itself?
Is accessing the ports slower from code running in graphics RAM at #9000??

:( <- me, unhappy.

Thanks





Sophos Plc, The Pentagon, Abingdon Science Park, Abingdon, OX14 3YP, United
Kingdom.
Company Reg No 2096520. VAT Reg No GB 348 3873 20.