Can someone shed a little light on how HP's ODBC driver retrieves data
from an Image/SQL database. How does it determine whether to perform
a serial read vs. a chained read on detail set or a serial read vs. a
calculated read on a master set?

My frustration and experience comes from using Microsoft Access 2.0 to
query data from our Image/SQL databases. For the most part it works
well with acceptable performance. But sometimes I feel like I've been
sent to "query hell". Especially if I link Access tables to Image tables.

For example, if I was to create a query to extract a few records out of a
master dataset, where I specified the key item values I was looking for in
the criteria line of an Access Query - I would get the results almost
instantly. On the other hand, if instead of specifying the values  in the
criteria line, I had stored them in an Access table which I then linked to
my Image/SQl table - well it would be time for me to take a coffee break
(and maybe lunch also).

Is there a way to force the query to take a more efficient path to the
Image/SQL data? Or are there any tricks or configuration items I'm
missing?

Eric Popish
Directors Guild of America
[log in to unmask]