Bosch

eferz · Post by **eferz** » Mon Jan 16, 2012 3:27 pm

3FG wrote:All of these signals have OBC1 = 2, so use the NEC-f16 NEC-f16 3Fixed protocol. Enter 134, 228 and 2 into the 3 device fields. On the function tab, enter in OBC2.

I'll get to the unknown, y1 and y3 signals tonight. BTW, y1 and y3 are just special cases of f16, and regular NEC is an even more special case of f16.

Awesome, thanks! I've already got 99% of the rmdu created so far. It's just those two (maybe three) commands that are holding me back from uploading the complete file set.

Edit, actually, I had it set to official again. So, I've changed it to fixed version per your earlier advice with the memory and key moves issue.

3FG · Post by **3FG** » Mon Jan 16, 2012 11:58 pm

The unknown signal doesn't have enough info to do anything with. The y1 signal is the complement of OBC1 = 02 except for the most significant bit, which is not complemented:
00000010 OBC1 = 2
01111101 OBC2 =7D = 125 Play

the y3 signal doesn't complement the most and least significant bits.
00000010 OBC1 = 2
01111100 OBC2 =7C = 124 Stop

Of course, I also checked the waveforms directly and got the same result.

eferz · Post by **eferz** » Tue Jan 17, 2012 7:12 am

3FG wrote:The unknown signal doesn't have enough info to do anything with. The y1 signal is the complement of OBC1 = 02 except for the most significant bit, which is not complemented:
00000010 OBC1 = 2
01111101 OBC2 =7D = 125 Play

the y3 signal doesn't complement the most and least significant bits.
00000010 OBC1 = 2
01111100 OBC2 =7C = 124 Stop

Of course, I also checked the waveforms directly and got the same result.

Excellent. Thank you. I've created the following files for any Slingbox users which may want to use it.

ICT file: http://www.hifi-remote.com/forums/dload ... e_id=10372
RMDU file: http://www.hifi-remote.com/forums/dload ... e_id=10378
BIN files: http://www.hifi-remote.com/forums/dload ... e_id=10379

BIN file Instructions.

Customization instructions:

http://slingbox.wikia.com/wiki/How_to_c ... te_upgrade

simonsezdothis · Post by **simonsezdothis** » Tue Jan 17, 2012 8:48 pm

Yes, these were learned with an IR Widget.
The decode for button "Right" should be NEC1-f16 too.
I noticed that if I learned with a very short keypress (no repeat) it would decode as NEC-f16 instead.
The "Play" and Stop" keys are different for some reason. I did reproduce this a number of times to check.
The Unknown is a good old fashioned cockup I think! Sorry about that.
(It confuses me that when you open an ict file it adds to the window instead of replacing an open file, probably saved some junk along with it by mistake.)

Struggling to keep up with you guys.
I have non repeating codes that I can fire at the DVR with my USBIRT which work fine now.
Import them into AMX though, and they repeat again. Trying to sort that out now.

3FG · Post by **3FG** » Tue Jan 17, 2012 9:23 pm

The distributions for DecodeIR.dll and RMIR contain the file DecodeIR.html which has information about the various decodes. It explains the meaning of -f16 or -y1.

NEC protocols are all nearly the same signals. The difference between NEC1 and NEC2 is the pattern used with repeated signals. The protocols send the same initial signal but NEC2 simply repeats the initial signal, while NEC1 sends a distinctive "ditto" signal. So DecodeIR can't tell the difference between the two if the button press is short enough so that only the initial frame is sent. As described in DecodeIR.html, if the signal doesn't repeat, DecodeIR simply describes it as NEC or NEC-f16, omitting the 1 or 2. If the signal repeats, then it is possible to distinguish the two protocols, and for the present case, it would be decoded as NEC1-f16.

Of course, since all of the other Bosch signals decode as NEC1, it is reasonable to believe that right arrow is also NEC1.

Barf · Post by **Barf** » Thu May 03, 2012 2:53 pm

IrMaster also won't always work, even with the suitable IRP added to IrpProtocols.ini, because it only handles function numbers up to 32767 rather than 65535.
irp={38.4k,564}<1,-1|1,-3>(16,-8,D:8,S:8,F:16,1,-78,(16,-4,1,-173)*) [D:0..255,S:0..255=255-D,F:0..65535]

Blush. I was recycling code from another project, and using Java's short. As everyone (?) knows, one of Java's sillynesses is the absence of unsigned types, thus 2^15-1. Anyhow, in the new release I use longs, so the limit is lifted to 2^63-1=9223372036854775807, which should be enough for the immediate future.

3FG · Post by **3FG** » Thu May 03, 2012 8:44 pm

Barf,
Thank you. Yep, not having unsigned types can be a real aggravation.