>George,
>
>> Further, I don't think I could have found a solution, and you did. It demonstrates your understanding of what was going on, and, in turn, has given me more information than I had before.
>
>It was Hong-Chee Tan at Microsoft that discovered moving the mouse affected doevents while checking a bug report I sent in. I'm only exploiting the discovery *g*

Isn't there some line about good programmers borrowing and great ones stealing?:-) Or is that in reference to one of the other arts? *rbg*

>>Isn't the mouse position sent to the window the "rat's" over via Windows messaging? I think I recall reading something like that. If so, it would explain this behavior. Mouse moved over the form, there's something to process. Mouse not over the form, it enters a brief wait state to see if there's going to be something to process. Of course, this is pure guesswork on my part, but it would explain what's happening.
>
>I believe so, I've not looked at the message queue for a couple of years now, but I think the mouse generates quite a stream of messages while it's moving around.
>
>>INKEY() not eating the keystroke may explain why it works. It would mean that there's an event in the queue. If so, and if the bug gets fixed, then the solution might not work in the future. Again, pure guesswork on my part.
>
>kludge++ *LOL*

Shouldn't that be "Visual kludge++"? *s*

>>I think you've hit upon the solution to my problem with DoEvents. There were only a couple of thousand records being sent by the server, and I was issuing it once per record. Reducing that by a factor of 100 should do the trick, in concert with your solution.
>
>I did go through the effort of writing a doevents wrapper object last year that contained an iteration counter it's on my website. I guess I should update the article now though.

Please let me know when the updated version's available.

Regards,