Alan,
I specifically ran the code listed below under VFP8SP1 and VFP7SP1 before posting the reply and the correct result 2068.16 is returned.
I'm not sure why you see a different result.
Sometimes KB articles list the specific release where bugs have been fixed, other times they don't, but since VFP7 and VFP8 are not listed in the versions affected by the bug we have two possible choices 1) it's been fixed in them or 2) it's still a bug and the KB article is out of date.
>Using the code produced by MS in the KB to illustrate the bug, the result in version 8-sp1 is the same as in version 6. Since MS acknowledged it as a bug in version 6, I can't see why it wouldn't still be considered a bug in version 8.
>
>>i = 86.173125
>>j = 24
>>? round( i * j, 2 ) && yields 2068.16 for me