Add protocol 01 B5 (Samsung36?) to RM/KM?

[Barf]

At the risk of sounding patronizing: Standards are defined by (authoritative) documents, or in some cases "reference implementations", not by "random" implementations. So, in order to make an authoritative claim that one interpretation is correct, and another one is not, a reference to an authoritative document is compulsory. Studying implementations and measuring is guesswork. In this case, there appears to be no known authoritative description of "Samsung36", everything is based on reverse engineering (another word for guesswork). (That includes the possibly misleading name.)

Having said that, the extra 68 micro seconds gap makes technologically no sense. It extends the preceding gap from 500(1500) micro seconds to 568(1568). This will almost surely pass as a measurement error by a decoding circuit, so from that point of view it hardly matters.

I'd be inclined to doubt either the learns or the OEM remote.

My intuition says that it was likely a glitch in the implementation, for example a computational delay (computing the complement?). Fixing it was not deemed necessary by the manufacturer. Again, guessing....

BTW, finding this kind of difference isn't easy. A widget counts the number of IR modulation edges in 100uS time bins, so it will take many captures to build up the statistics to measure the off time interval.

IrScrutinizer with an Arduino using the AGirsLite firmware based on MikeT's work measures with higher precision. Also the Command Fusion Learner is supposed to be very accurate (but on the expensive side...), and is supported by IrScrutinizer.

My recommendation is to just change the IRP.

+1. I will follow up in IrpMaster/IrScrutinizer.