Re: ROMs replacement

From: Jim Brain <brain_at_jbrain.com>
Date: Tue, 18 Sep 2012 13:04:48 -0500
Message-ID: <5058B7C0.9060604@jbrain.com>
On 9/18/2012 10:53 AM, Gerrit Heitsch wrote:
> On 09/18/2012 06:53 AM, Jim Brain wrote:
>> On 9/17/2012 2:25 PM, Gerrit Heitsch wrote:
>>
>> Well, I sent BASIC, CHARROM into the 0,1 lines of the '148.  !GS drives
>> !CE and !OE on the EPROM, while the A and B lines go to the A13 and A14
>> lines of the EPROM.  A12 is handled wither with the NAND gates or the
>> '257, nothing new to report there.
>>
>> Setting up 2 KERNALS means bringing '148 lines 2 and 3 to the ends of a
>> 3 pin header.  Middle pin goes to !KERNAL  shorting one half will shunt
>> KERNAL to one of the two address lines.  It does require 1 more resistor
>> than the '11 solution, as it requires both the 2 and 3 lines of the '148
>> to be pulled weakly high, but the wiring is simpler.
>
> There is another problem with the solution with the '148 that is not 
> present when using the '11. In order to switch KERNALs, you have to 
> route _KERNAL to one or the other input. _KERNAL is the signal used to 
> select the ROM, meaning it changes state twice per every cycle the CPU 
> access the KERNAL ROM. That makes it high frequency signal, not 
> something you want to route through a mechanical switch in a 
> convinient position at the end of a long cable.
As it was a late addition, I chose a quick and dirty solution.  I defend 
it as:

  * I optimized for soldering time, not purity of circuit.
  * If this was the basis for a production-quality design, wire routing
    would not be an issue, so another option would be better.
  * Routing 1MHz signals through a switch is not the end of the world
    (people route IO1/IO2 through switches all the time :-) Not saying
    it won't cause issues, but if someone is doing this as a one-off,
    there are worse things.  I was, though, assuming a short cable, etc.

>
I think a solution with all 2 input NAND gates is available as well, but 
that is getting overly academic.  It basically boils down to how much 
the OP wants to solder.

Jim

>
>  Gerrit
>
>
>
>       Message was sent through the cbm-hackers mailing list


-- 
Jim Brain
brain@jbrain.com
www.jbrain.com



       Message was sent through the cbm-hackers mailing list
Received on 2012-09-18 19:00:52

Archive generated by hypermail 2.2.0.