John, I used GenIRDB (that can't do the parity bit) and just calculated the parity bit with the Excel MOD function and converted it all to 8-bits as you probably suspected.

Since you mentioned it, I actually tested MakeHex and it can do the parity bit with the XOR 1 bit at a time. Here is the irp file.

Device=202
Function=0..63

Protocol=Polycom
Frequency=38000
Time Base=287

One=2,-3
Zero=3,-4

Prefix=9, -9
Suffix=1,-202

Form=*,D:8,F:6,T:1,P:1,_,*,D:8,255:8,_
Define=P as F(0..0)^F(1..1)^F(2..2)^F(3..3)^F(4..4)^F(5..5)^T
First Bit=MSB


But, then again YOU wrote it:) If you ever do want to tackle rewriting it, I'll will be glad to test it and I do have some ideas ...


I completely agree that it should have been "0000 006d 0024 0000" and that it probably WILL be unreliable using "0000 006d 0000 0024"

I will edit the post above with that change in case anybody uses it.

As far as the TSU 3000, I think that the frequency calculation itself is slightly different, but the on and off times are calculated the same way. (In other words 006C may not mean ~38 KHz but you still use the same multiple). But it has been a while and I can't find my calculations. So my memory may be faulty and I only had one data point.

-Jon