Jeff's email got me started investigating the relative performance of VB 6 versus VB.NET. I couldn't find a lot to go on.
Several years ago, Robert Wlodarczyk, who identifies himself as a "Microsoft Student Consultant at the State University of New York" wrote an article titled, "Performance Testing Visual Studio .NET" that reports benchmarks comparing VB 6 against VB.NET, and several other languages as well. In that article, Robert concludes that, "Visual Basic .NET code ran 46.45 times faster, on average, than Visual Basic 6.0 code."
Forty-Six Times As Fast?
C'mon Robert! I'd be willing to believe maybe 100 percent faster. But 4,645 percent faster?
As it turns out, that instinct is correct. After Robert's article was published, it was roundly criticized in the newsgroup, "microsoft.public.dotnet.framework.performance" as inaccurate. Dan Appleman, founder of Desaware Software, co-founder of APress, and author of the legendary "Guide to the Win32 API" for Visual Basic 6 wrote, "The VB.Net benchmarks are all incorrect. Benchmarking is VERY difficult to do correctly." (I contacted Robert directly about this. Unfortunately, he said he had not updated the study.)
But, even given it's defects, Robert's study does show that VB.NET is far from being slow and clunky. It can at least appear to be over four thousand times faster than VB 6!
