The low latency might have some positive effect on overall system speed, but as a practical matter it is unlikely enough to be noticeable other than with very sensitive testing software.
If we knew what it was we were doing, it wouldn't be called research, would it?
