@punkpie
Here's what Eyetech published in 1999.
'Fortunately, revision 1.D.1 had the most serious timing faults in the Budgie and Gayle chips fixed at the time of manufacture. A few early, pre-main production boards, notably 1.A and 1.B found their way into the shops. As far as we can determine, these also had timing work-arounds fitted.
Commodore always planned to correct these faults and so produced revision 1.D.4 and 2.B without provision for a workaround. Unfortunately, these were shipped with the old chips when the company started getting into financial difficulty. If you have one of these boards, then it will need fixing if you intend to use anything which requires fast bus access. E.g. 030/40 or faster accelerators, PowerPC boards, clock port accessories and IDE Flyer.
(I have a 1.D.4 with all the bells and whistles, and it runs perfectly)
When Amiga Technologies/Escon started manufacturing in 1995, they used the same 1.D.4 and 2.B boards and bugged chipsets. AT began fixing this fairly quickly, but we reckon 15-20% of the new production was shipped without the fix. Unfortunately, there are no records of which serial numbers were fixed. Escom made units had 'Amiga' rather than 'Commodore' at top left of the keyboard and 'Made in France 1995 (or 1996)' on the underside.
Eyetech said that their timing fix was proprietary and not in the public domain. Fixes published on Aminet were different and - in Eyetech's view - and didn't fix the problem.
Though the 1.D.4 and 2.B fixes bring these boards to the 1.D.1 standard, there are other issues which are, for all practical purposes, insoluble. These problems only come to light when using a Zorro 3 card. The necessary redesigned Budgie chip was never produced.'
There are other issues discussed (video tearing, etc) but as far as timing problems are concerned, Eyetech consider revision 1.D.1 to be the best one.
JaX
