Protocol Builder Flags

Post by **vickyg2003** » Sun Sep 05, 2010 4:25 am

I've been working on an Epson EMP710 in protocol builder. SC38+

The approach I was using on the Epson EMP835 didn't work with the Epson EMP710. It crashed the remote, before it actually got to the protocol so I decided to use the pd registers the way that protocol builder would set them up.

Device 4, 8 bit
Command 2, 8 8

My protocol was sending out really really weird signals, so I just did a jp xmitir to see the data in the buffer.

I've got a weird problem.

Instead of my expected 48 bump graph, I've got a 47 bump graph. One of the bumps is double wide. This happens to be the only place in the data where there are two ones in a row, 11. So when a 11 is encountered the time off is ommitted.

So we know what flag this would be?

Post by **The Robman** » Sun Sep 05, 2010 9:48 am

If it's a bi-phase signal, you can't do a double wide bump via the IR engine, the signal would need to be generated manually.

Post by **vickyg2003** » Sun Sep 05, 2010 10:15 am

The Robman wrote:If it's a bi-phase signal, you can't do a double wide bump via the IR engine, the signal would need to be generated manually.

Its not supposed to be a bi-phase signal, its supposed to be a a quad. Things were not going as planned. So I treated it like a regular binary to see what was in the dcbuff, and I see the 11 being shown as a double wide time on.

At this point its academic, since my interface decided it was tired of my uploads (over 100 yesterday), and is no longer talking to my remotes.

I suppose I can switch over to a SC38+ parallel remote.

Post by **The Robman** » Sun Sep 05, 2010 12:20 pm

You can't do quads thru the IR engine, so whatever you would get thru the engine is dependent on how you had things set up in the front page of PB.

Post by **vickyg2003** » Sun Sep 05, 2010 12:43 pm

The Robman wrote:You can't do quads thru the IR engine, so whatever you would get thru the engine is dependent on how you had things set up in the front page of PB.

Yes I know I can't get quads through the S3c8+ engine, the HCS08 handles this kind of quad just fine. With the other Epson Quad putting the pairs in worked fine, but this one seems to have data that screws with the Presetup where the remote loads the dcbuff. So I had to undo it, and then I got this really screwwy output, probably a typo somewhere. So I wanted to see what was in the buffer, so instead of doing my by custom quad routine, I sent it through as just a binary, so I could see that everything was in the buffer the way I expected it to be. Instead of seeing the binary, the double 11's were showing up as one big bump on the graph.

If this doesn't ring any bells with you, then I'll just keep looking. However I am really bummed that my interface decided to take a leave of its senses on a three day weekend, so I can't even get one started in the mail until Tuesday or later. Bummer!

mr_d_p_gumby · Post by **mr_d_p_gumby** » Sun Sep 05, 2010 1:59 pm

vickyg2003 wrote:Its not supposed to be a bi-phase signal, its supposed to be a a quad. Things were not going as planned. So I treated it like a regular binary to see what was in the dcbuff, and I see the 11 being shown as a double wide time on.

I understand that all you wanted was to dump out the DCBUFF contents via JP XMITIR, but it sounds like some other flag in the header has you set in a bi-phase mode. Even so, this still accomplishes what you set out to do, it just makes interpreting the results a little harder because you have to put on your bi-phase decoding hat.