Re: 1551 Repair - 28 pin 6523 Wanted

From: Gerrit Heitsch <gerrit_at_laosinh.s.bawue.de>
Date: Sun, 09 Mar 2014 19:40:04 +0100
Message-ID: <531CB584.5020008@laosinh.s.bawue.de>
On 03/09/2014 07:18 PM, smf wrote:
>> Yes, unfortunatly. Would have been nicer to keep the static design
>> when it comes to address lines and run everything through the
>> multiplexers as well (as was done with TED later). But I think the
>> 6567 doesn't have enough Pins to do that since you'd still need _RAS
>> and _CAS.
>
> They wouldn't have needed RAS_ or CAS if they used the same multiplexer
> as the CPU.

Well, someone has to generate the address strobes for the DRAMs. You can 
do that in TTL as well, but it takes a number of gates to do it 
properly. The way it was done in the C64 is a bit of a hack, but it 
saved on expensive logic and pins on VIC at the same time.

The multiplexer for the CPU are controlled by _CAS from VIC by the way. 
VIC does the DRAM timing for itself AND the CPU, generating 2 complete 
RAS/CAS cycles for every single PHI0 cycle.


> It's possible it wouldn't have met some dram timing doing it that way.

Well, TED does it and it works.

  Gerrit



       Message was sent through the cbm-hackers mailing list
Received on 2014-03-09 19:01:07

Archive generated by hypermail 2.2.0.