Protocol Builder idea: Mid-Frame Bursts

[The Robman]

Hey Mike,
I've been doing some work with mid-frame bursts (MFBs) and I see an opportunity to add some more functionality to Protocol Builder.

See my notes here:
https://www.hifi-remote.com/forums/viewt ... 1&start=15

And a working upgrade here:
https://www.hifi-remote.com/forums/viewtopic.php?t=12226

Basically, the version of the MFB that PB currently supports will give a MFB in the following format:

ONE on time
LEADIN off time

There is a 2nd format that could be supported. If the set the MFB flag (W7.3 in S3C8, $AB.3 in HCS08) and jump to the regular engine you'll get an MFB which is a repeat of the leadin pair.

You can also format a custom pair if you want, either using a different mix of the times in the data block, or if necessary using brand new times.

For S3C8 you would load the timing data into RD0/1/2/3 and for HCS08 you would load it into $9a/b/c/d

I tried to load the RD0 registers directly using LDW but for some reason that didn't work, but when I did it the UEI way (see below) it did work.

BITS W7.3
ld w0, #D0h
ld @w0,pd10
inc w0
ld @w0,pd11
inc w0
ld @w0,pd04
inc w0
ld @w0,pd05
JP 0146h

For the upgrade I created for the Elan v883, I needed to create an MFB with "+800 -800" timings. The -800 was already in the data block for the ONE time, so I needed to add +800, which I put into 2 of the 3 spare bytes between the leadin times (R20/21) and the bit count (R25), but I see no reason why I couldn't have also loaded the times as hard coded values in the code itself.

[mr_d_p_gumby]

It sounds like you've demystified the MFB enough that maybe I can find a way to implement both methods in PB. I'll certainly have a look at it as soon as I stop having to speak Chinese.

Xie xie ni.