You will need a very fast disk drive to experiment with it. In effect, only a RAM Drive can expect to keep fast enough steam to feed LZ4, especially when using multi-threading modes.
Block subdivision method was selected, due to its design simplicity and scalability. Compared with previous versions, even single thread mode greatly benefits, since multi-threading requires asynchronous data loading, which means that reading and writing to the disk is done in parallel with compression.
I can't really test beyond 2 threads, since i only have dual-core systems within reach, but the code should prove pretty scalable with quad-core systems and more (up to 32 threads).
You can download and test this new version on the LZ4 homepage.