To give more down to earth numbers, with a 45 MB CSV file containing 145 fields and 50,000 records, the reader was processing about 30 MB/sec. So all in all, it took 1.5 seconds! The machine specs were P4 3.0 GHz, 1024 MB.
I've started using CSV Parser that is part of the CommonLibrary.NET.
It uses .NET 3.5, has an easy API, and convenient overloads/methods & lamda's for iterations.
I don't have any benchmarks for this one like above, but nice thing about this is that it's just one component of a library similar to Java Commons. So I also get a Command-line parser, Repository implementation among other things.