Having worked with Dallas 1-Wire^TM and iButtons^TM for a while, I looked at SensorPath^TM a couple of years ago.

I never actually used SensorPath, but these were the things that struck me at the time:

The basic comms concept looks very similar indeed

SensorPath seems to be aimed at one specific application area; viz, on-board sensors, particularly temperature sensors on PC motherboards and the like. This allows the protocol to be a lot simpler than the full-blown 1-Wire, but means there's nothing like the range of available devices.

SensorPath doesn't seem to be aimed at the "heavy-duty" applications like 1-Wire with very long distances and/or very large numbers of connected devices.

It was quite difficult to get the SensorPath protocol definitions; NS seemed to be just using this in their motherboard controller chips - they didn't seem interested in others implementing their own "masters" in microcontrollers, etc.

I don't think SensorPath had an equivalent of the 1-Wire "parasite power" idea (why would it, if it's intended for on-board applications?)

I'm afraid that all sounds like a downer on SensorPath, but it's not supposed to be. I think it's just that SensorPath is more specifically targetted at a narrower application area - so, if it fits your application, it may well be simpler and cheaper!

People here often report difficulties obtaining Maxim parts in anything under telephone-number quatntities - I dunno about NS and/or SensorPath...