Attached is a sample program that creates two Report Sharpshooter reports; one using a SF BBS as the data source and one that uses a dataset. Aside from the data source both reports are identical. The sample program uses the SF Sample database as its data source. What I've found is that when a BBS is used as a data source, it takes RSS longer to render the report and the rendering time increases dramatically as the number of records increases. The SF sample DB only contins around 19K customer records but this is sufficient to demonstrate the difference. On my development system the BBS based report takes about 43 seconds to render while the dataset based report takes about 6 seconds. The tests were run outside of VS. Is this to be expected when using a BBS as a report data source or am I missing something that could improve performance? The sample was created using SF 1.7.0.2 and RSS 4.0.3.5.