Dreambox and MX-500

[broken]

Hey all, need some help, basically trying to get an MX-500 to learn to control a Dreambox 7000s, it seems to have trouble learning certain keys, here's the players I've got,

Dreambox 7000s
Dreambox V3 Remote ( JP1 but no RDF yet )
MX-500
OFA 8811
JP1 Cable

Have a look at this thread,

http://www.remotecentral.com/cgi-bin/mb ... d.cgi?4443

For example the 8811 learned the Dreambox ON fine from the Dreambox Remote but the MX-500 does not. I learned the Dreambox ON onto the 8811 AUX L1 and then put the MX-500 learned version of the same code onto the 8811 AUX L2 to compare, the resulting IR file is here,

https://www.hifi-remote.com/forums/dload ... le_id=1739

Any help would be great, I'm rather new to JP1 so relatively simple terms would be appreciated

[johnsfine]

Based on the analysis Jon did of what these are supposed to look like, even the one on L1 is bad enough I'm surprised it works. I need to investigate further to understand why that works and to add support in DecodeIr and maybe some other JP1 software for this protocol.

L2 is way off. Are you sure it's the same original (Deambox remote) button you used for L1? The frequency (34188) is surprisingly low. I wouldn't expect that size frequency error from the MX-500, and several of the durations are short by a larger amount than I'd expect. I would start making theories for frequency being low and durations short and thinking of a general work around, except that a couple of the durations are longer rather than shorter. Either that means it's a different button, or it means that the learning error isn't consistently biased in one direction.

What sort of batteries does an MX-500 use and what condition are they in? Some learning problems are due to weak batteries.

I think we'll need multiple tests to see how consistent the errors are. There are two levels of unknow we need to investigate, the errors in learning from the Dreambox to the MX-500 and the errors in learning from the MX-500 to the 8811.

We need to learn the same MX-500 signal to more than one 8811 button to see how consistent those results are, and we need to learn the same Dreambox signal more than once to the MX-500 and relearn each to the 8811 and see how consistent those are.

If there is any consistent bias in the learning by the MX-500 (frequency or duration or both) then we can make the 8811 generate a new signal biased in the opposite direction and hope the MX-500 will mislearn that back to correct.

But if the MX-500 learning is just unreliable (rather than consistently biased) the problem is harder.

Meanwhile, I think some other JP1 expert ought to be able to direct you to the right RDF file.

[broken]

Doesn't seem to be the batteries, meantime, I've compiled further data as suggested,

https://www.hifi-remote.com/forums/dload ... le_id=1740

dreambox.on.8811_mx500_0.ir
Dreambox ON learned onto L1...L4 of 8811 from single MX-500 learn

dreambox.on.8811_mx500_1.ir:
Dreambox ON learned onto L1...L4 of 8811 from second single MX-500 learn

dreambox.on.8811_0.ir:
Dreambox ON learned onto L1...L4 of 8811 from Dreambox ON

[johnsfine]

Doesn't seem to be the batteries, meantime, I've compiled further data as suggested,

You couldn't ask for more consistent than that, but I still haven't figured out what it all means nor why the MX-500 would learn that badly.

All eight samples learned through the MX-500 are basically identical to each other and significantly different from correct.

[broken]

But since they're consistent can't we code an offset version so that'll it'll end up learning correctly like you mentioned previously?

[johnsfine]

Probably, but I don't yet know what the offsets should be.

If every pulse had gotten shorter then I'd try making them all longer. Instead, most got shorter and some got longer. I don't think that behavior could be positional. There needs to be some other rule.

[broken]

For anyone reading, the 8811 is showing,

Code: Select all

Gap-210-2160-8?             10    A
Gap-210-10782-17?  0  1     0     0.1.0
Gap-210-2160-8?             32    20
Gap-210-2160-8?             10    A
Gap-210-10782-17?  0  1     0     0.1.0
Gap-210-2160-8?             32    20

And the MX-500 is showing,

Code: Select all

Gap-202-2078-8?             138   8A
Gap-202-10815-17?  0  1     0     0.1.0
Gap-202-2078-8?             32    20
Gap-202-2078-8?             138   8A
Gap-202-10815-17?  0  1     0     0.1.0
Gap-202-2078-8?             34    22

[damir]

Dreambox 7000s
Dreambox V3 Remote ( JP1 but no RDF yet )

Complete RDF for Dreambox V3 Remote is available in current RDF pack.
https://www.hifi-remote.com/forums/dload ... le_id=1378

[broken]

So it is, thanks. I'd unzipped that set into my rdf directory previously not realizing they unzipped into a sub rdf directory and didn't get picked up.

[johnsfine]

You should use IR.EXE and JP1 to backup the eeprom contents of the Dreambox remote to a file on the PC.

Next, you should try editing the Protocol 016C in your Dreambox remote to generate a higher frequency.

Your MX-500 learned a 38Khz signal from the Dreambox but retransmitted it at roughly 34Khz. That is probably NOT the serious problem. The burst timing changes are probably the serious problem. But the frequency change might be serious and it might be connected to the timing change in some bug in the MX-500's firmware that we won't understand until we test. So I want to see what happens when we make the Dreambox transmit at around 42Khz instead of 38Khz.

After saving an unmodified copy from IR, go to the Protocols tab, where Protocol $016C should be selected.

Click Edit

Select (cursor drag) the 43 8B at the start of the "Upgrade Protocol" dialog.

Type 3D 7D replacing the 43 8B (make sure there is still a space between the 7D you just typed and the 42 that follows it).

Click OK.

Upload to the Dreambox remote.

1) Test the Dreambox remote. I'm curious whether the Dreambox is picky enough about frequency that this change would matter. But whether or not it works continue.

2) Learn one key from the Dreambox to the 8811 (to verify that the change did what I intended it to).

3) Learn a couple keys from the Dreambox to the MX-500, check whether they work, but in any case continue.

4) Learn those keys from the MX-500 to the 8811.

5) Upload 8811 .ir file to the diagnosis folder and reply with its URL.

Note that the 8811 documentation is misleading about learning. You are not limited to the L1 through L4 keys. You can learn onto almost any key of the 8811. You can have a lot more than 4 learned signals in one .ir file (though not unlimited).

[broken]

A'ight, I'll do this first thing when I get home tonight, thanks for all your help.

[johnsfine]

For example the 8811 learned the Dreambox ON fine from the Dreambox Remote but the MX-500 does not. I learned the Dreambox ON onto the 8811 AUX L1 and then put the MX-500 learned version of the same code onto the 8811 AUX L2 to compare, the resulting IR file is here,

Going back to something from the start of this thread: I'm confused.

As my understanding of this protocol gradually increases, there are things I think I understand that indicate the L1 signal in that file is seriously wrong, enough so that it shouldn't work. But you seem to be saying it does work. The L2 signal is much worse, so it isn't surprising that doesn't work, but I'm confused over L1 working.

I think I may understand a consistent distortion that is occuring in the 8811 learning process for all these signals. If I understand that correctly, what we see when you did the double learn is not just the MX-500 learning distortion, but the combination of MX-500 distortion followed by 8811 distortion. If so, I can reason backwards from what the 8811 stored to deduce what the MX-500 really sent, which is a more internally consistent distortion than what was stored, then we can precompensate for that (probably by patching that protocol executor further on the Dreambox).

But why did L1 work?

1) If L1 worked because it's actually a good learn, then my understanding of the protocol is flawed and I can't use a theoretical distortion in the 8811 to simplify the inferred distortion in the MX-500.

2) If L1 worked because the actual device doesn't care that the signal is seriously wrong, then I might be on the right track and the solution might be closer than it looks.

[broken]

Just for a lark I tried getting the MX-500 to learn the code from the 8811, ie,

Dreambox ON > 8811 > MX-500

Didn't work, but I didn't dump the results either since that would've maybe been silly,

Dreambox ON > 8811 > MX-500 > 8811

Hehe, probably would've been useless data at that point.

[broken]

After downloading from the Dreambox remote there's nothing listed under the protocols tab of IR.

https://www.hifi-remote.com/forums/dload ... le_id=1741

[johnsfine]

That's a surprise. I guess it means in this version that protocol is built-in rather than included in eeprom.

So we can add it instead of editing it. Copy the following block from (and including) the word "Upgrade" to (and including) the word "End" and then use the Add button in IR.EXE's protocols tab and paste it in.

Upgrade protocol 0 = 01 6c (S3C80) Manual Settings
3D 7D 42 8B 05 00 00 6A 01 67 6C 03 F6 80 99 E4
07 09 E4 08 0A E4 03 07 E4 04 08 F0 08 56 08 0F
6C 07 F6 80 99 F6 80 38 FB 0D 46 08 80 6C 07 F6
80 99 F6 80 38 7B FB AF 28 03 F6 80 71 28 04 F6
80 71 28 05 F6 80 71 28 06 F6 80 91 C6 F8 19 64
F6 01 45 28 07 F6 80 71 28 08 F6 80 71 28 09 F6
80 71 28 0A F6 80 91 C6 F8 9B 47 F6 01 45 8D 01
0A 48 C2 F0 C2 F6 80 7D 28 C4 8D 80 7D 1C 12 F6
01 39 56 C2 0F 6B 09 C6 F8 00 3D F6 01 45 2A F7
AF F6 80 71 1C 12 8D 01 39 38 C6 C7 03 F0 C0 3E
C7 13 F6 80 C1 3E C7 13 F6 80 C1 3E C7 13 F6 80
C1 60 C0 0E 56 C0 0F C7 26 56 C2 F0 42 20 D7 62
AF 28 C1 F0 C1 02 12 02 01 AF
End

Since the device (SAT/0031) is still a compatible upgrade, we don't need to put in a device upgrade. I hope other things are similar enough between the V3 and older Dreambox remotes for this to work.