A1200 Motherboard Choice

[Peter]

I have got hold of a couple of A1200's this week - I will eventually only keep one... thing is, I would like to know what people think as to which one I should keep - either the 1D1 or the 1D4

Comments very welcome.

[fitzsteve]

I thought the 1D4 was the board of choice

I'm sure I read that somewhere anyway...

[prowler]

1-D-1 is the latest, most compatible variation of the A1200 motherboard for use with '040 accelerators and up, requiring no timing fixes. This is not true of the later revisions.

[Peter]

That's interesting, so was 1D1 the latest MB that Commodore made?
I may be getting an Apollo 040 so the 1D1 sounds like a good option for that...

Does the 1D1 have any known problems?

1D4 and 2B have the timing problems...right...?

[prowler]

Originally Posted by Peter

That's interesting, so was 1D1 the latest MB that Commodore made?
I may be getting an Apollo 040 so the 1D1 sounds like a good option for that...

Does the 1D1 have any known problems?

1D4 and 2B have the timing problems...right...?

AFAIK Commodore made ALL the A1200 motherboards. Amiga Technologies (Escom) just assembled machines using the parts that they had acquired from Commodore.

1-D-1 revision motherboards have no documented major problems.

Yes, the 1-D-4 and 2B revision motherboards require the timing fixes for full compatibility with '040 and faster accelerator cards.

[Peter]

Thanks for the info Prowler....

Thinking back I had a 1D1 in my Tower a while ago - it was a bit of an odd board as it had the clock port pins on the wrong side or had a full set, I cant remember which... the 1D1 board I have at present doesn't have this problem.

Steve - you cant have much more room for another Amiga

PS - just out of interest (probably only to me) does anyone know the order in which the A1200 motherboards were released?

[Bamiga2002]

Originally Posted by prowler

1-D-1 is the latest, most compatible variation of the A1200 motherboard for use with '040 accelerators and up, requiring no timing fixes. This is not true of the later revisions.

It's true 1D1 is the best choice, but it's not the latest. 1D4 was made after it, needing the timing fix with some turbo cards.

[prowler]

Originally Posted by Bamiga2002

It's true 1D1 is the best choice, but it's not the latest. 1D4 was made after it, needing the timing fix with some turbo cards.

Hi Bamiga,

You are right, of course. What I meant was that 1-D-1 is the latest revision of the A1200 motherboard to be fully compatible with '040 and faster accelerators, later revisions requiring the timing fixes.

I'm sorry that was not clear. I was preoccupied, bidding on several eBay auctions when I wrote that, and was not giving it my full attention.

[Bamiga2002]

No probs there prowler, just wanted to clarify it. I had one 1D1 and one 2B mobo. The other ones have been 1D4 revisions, two needing a timing fix and the one i have now seemed to have them factory-done.

Btw. what's the last (newest) A1200 mobo revision made?

[prowler]

Originally Posted by Bamiga2002

Btw. what's the last (newest) A1200 mobo revision made?

Revision 2B seems to have been the last revision made, immediately preceded by Revision 1-D-4.

The issue is generally clouded by the fact that there was always some degree of overlap in their availability, i.e., Revisions 1-D-4 and 2B were (and still are in NOS machines) available concurrently.

[DDNI]

I have used 1D1, 1D4 and 2B with Blizzard 1230 MkIV and MkII, Apollo 1240 and Apollo 1260, Blizzard PPC 060 @ 60 and 603e @ 330mhz

None of the motherboards had any timing fixes done.
I am sure that there may be some configurations out there that have issues.
However it is wrong to say that ALL examples of a particular revision mobo will have issues.

[Paul_s]

Originally Posted by Peter

PS - just out of interest (probably only to me) does anyone know the order in which the A1200 motherboards were released?

taken from amiga-hardware.com

Rev 0 (Engineering Prototype)
Rev 1 (Advanced engineering prototype)
Rev 1A (Initial Release)
Rev 1B
Rev 1D.1
Rev 1D.3 (Budgie Rev -01 Release, E123C and E125C have been removed from motherboard)
Rev 1D.4 (Budgie Rev -02 Release, additionally XR358 470Ohm resistor removed, 470Ohm resistor added to pin 43 of Alice)
Rev 2B (Both budgie Rev -01 and Rev-02 were used. Resistor 118 was changed from 470Ohm to 220Ohm)

[Peter]

That's great - thanks for the information.

[prowler]

Originally Posted by DDNI

I am sure that there may be some configurations out there that have issues.

However it is wrong to say that ALL examples of a particular revision mobo will have issues.

Yes, you are absolutely correct, of course. That is not what I was implying.

The uncertainty about whether a timing fix will be appropriate in each case is caused by component tolerances. For this reason, it is recommended that no action should be taken unless the board appears to be incompatible with accerators based on '040 and faster processors.

@Paul_s,

Thanks for posting that information. I have not seen that before.

@Peter,

FYI, these are the circumstances which gave rise to the timing issues associated with Revision 1-D-4 and 2B A1200 motherboards:

When the A1200 was designed, there were - as is common in designs of that complexity - some "bugs" in the chip designs. Many of these were corrected by the time the main production run started, but some were not, and in fact several bugs still remain in the latest revisions of the AA chip sets.

The main production board issued in the UK - Rev 1-D-1 - had onboard provision for hardware workarounds to fix the main bugs, notably bus timing problems in the Gayle and Budgie chips. In general, these boards work fine with most combination of expansion hardware and were the boards used to test A1200 expansion products before shipment.

The later revisions of motherboards - particularly the Rev 1-D-4 and 2B - are a different story, however. These board revisions were designed to take corrected versions of the Gayle and Budgie chips - and therefore were designed without provision for the earlier hardware workarounds. However, revised versions of the custom chips were never produced and, as a consequence, all but a few of the last AT/Escom-manufactured motherboards were shipped without either bug-free chips or hardware workarounds.

Edit: I have just realized that the last sentence above confirms that the very last A1200 motherboards must, in fact, have been manufactured by AT/Escom.