Difference between model boards for the 1200?

[Vars191]

What is the difference between the different mother boards for the 1200?

[Charlie]

Ooo, that's one of those simple questions with a long answer.
(I like long answers)

Here's a start:

All A1200's are basically the same design but there are detail changes, why?

-If you're C= (or AT) & you produce a cheap, mass produced computer the profit margin is very low.
-The price of the components will vary over time, also one manufacture vs another. So you buy from whomever can do you the best deal @ the time.
-Trouble is, for example, supplier A's DRAMs may have a slightly different spec from supplier B. If you don't watch out a change in supplier can bugger-up the timing of your mobo & it no-longer works.

-So C= designed their mobo's with some 'built-in flexibility' to allow for such detail changes. Sometimes by allowing a wide range of potential timings, sometimes with the addition/removal of 'tweaking' components - resistors / capacitors...
-Also with time any manufacturer will learn from their mistakes & hopefully correct them. While they are at it they'll usually find ways of doing the same thing cheaper.
Fixing stuff can sometimes be a bad thing for compatibility reasons. Part of the reason why the C128's FDD interface is as 'broken' as the C64.
-Occasionally parts dry up & wholesale changes to the design are needed to compensate.

...The result of all of this will be a number of motherboard revisions, that non-the-less are functionally identical. (or close enough)
Trouble is the testing to make sure the various revisions of a design are not only functional but behave in a similar fashion under all conditions is very expensive & extremely time-consuming...
...two things C= possibly skimped-on from time to time?
In C='s defence we do tend to push our Amigas way beyond their intended design.

The result is that not all A1200 revisions are considered equal.
-Unexpanded they are practicably all the same(ish), but as you load up your miggy with more stuff you start to 'expose any timing differences' resulting in 'unexpected behavior'.
-Some revisions are known to be 'troublesome' when pushed, though this is very variable. A mobo that's supposed to be bad can be rock-solid under all circumstances & vice-versa.
-I have two 1d4 revision A1200s; reputed to be a bad revision. One has been thoroughly modified with all the known 'fixes' prior to my ownership. I assume because it misbehaved - it's completely stable for me. The other, though the same revision has had no fixes & is just as reliable...

My own feeling is there is no such thing as a 'bad version'. (AT's bodge of the floppy drive aside)
I suspect it's more a case of the upgrades most likely to expose revision differences (fast accelerators for the most part) were designed with a particular version in mind & it's not surprising troubles pop-up when using said upgrades on other revisions.

Which revisions have what reputation, how to fix them, and the history of the A1200 design would be the subject of a much longer post...

Regards.

[gklinger]

You might find this thread of interest. It's filled with lots of great info.

[DDNI]

Nice post Charlie! I enjoyed reading it many thanks.