Polaroid model PCB FLM 3701 LCD-TV -- JVC protocols question

[ElizabethD]

This file has learned signals, and I'm curious about what I see JVCnitpicking.zip
1. On a DVD of mine, I got used to seeing JVC and JVC{2}, all buttons have "Sent Repeatedly" section.
In the TV signals it's only JVC{2} and only one "Sent Repeatedly" and 0-1 is ambiguous (-1684, -578, +466, where ambiguous=I have no clue what I'm looking at ). Seem to be good learns, all learned buttons worked the TV.

2. Why OEM remote's "Size" signal recorded on TV/Info is so different from all other buttons? Others not held long enough or something?

[ElizabethD]

bump

[The Robman]

I think this is a question for John.

[ElizabethD]

Rob, let's consider it academic now.
Three retries later, I think the missing first part is real, as no matter how long I delay pressing the OEM button, I can't get the start. So it must be the boring truth by definition.
And I can't replicate the issue with one button any more.
Could it have something to do with the JP1.2 remote or different manufacturer signal specs?

In any case, both the leadin-less learned signals and KM upgrade, containing both JVC{1} and {2}, work the TV.
Seeing there are no Polaroid TVs in the collection here, I uploaded the upgrade and supporting files here

[johnsfine]

I missed the first post of this thread when I dug through the big backlog after being out of the country for nine days.

I still don't really understand the question.

I think part of this issue is explained by the JVC intry in the DecodeIr documentation:
http://www.hifi-remote.com/johnsfine/DecodeIR.html#JVC

JVC protocol has a lead-in on just the first frame, unlike most protocols with lead-in that have it on every frame.

As a learning remote tries to distinguish the one-time part from the repeat part, having lead-in on just the first frame can present complications. In many designs, it is important to have the boundary between one-time part and repeat part occur on a relatively large gap. In that case the one-time part must be lead-in plus the basic frame and repeat part would be the basic frame again. But if the learning code doesn't apply that rule, the one-time part may be just the lead-in and the repeat part will be the basic frame. That later form is what DecodeIr calls a pure "JVC" signal. Other patterns are distinguished from a pure JVC signal, but the documentation is intended to tell you to ignore such distinctions for ordinary uses of the decode (such as constructing a JP1 upgrade).

I don't know why that first signal looks like JVC but lacks the lead-in. Maybe it is a different protocol that is just too close to JVC for DecodeIr to distinguish. There are several hard to distinguish protocols similar to JVC. If it is another protocol that close, your results seem to indicate that the device can ignore the lead-in and then also can't distinguish its own protocol from JVC (you can use a JVC upgrade to operate that device).

Those signals are in-between Mitsubishi protocol and JVC protocol in timing, but much closer to JVC. If they were learned correctly, they match Mitsubishi, not JVC, in structure. I suspect the device would also accept Mitsubishi signals with the same numbers, despite the timing being closer to JVC. DecodeIr can barely distinguish JVC{2} from Mitsubishi, so if there is some third protocol in between, distinguishing it would be hopeless.

[MaskedMan]

Some polaroid owners have had success with Mitsubishi TV code:


http://forums.directv.com/pe/action/for ... rtiseCode=


http://www.avsforum.com/avs-vb/showthread.php?p=6599449&&

http://www.remotecentral.com/cgi-bin/mb ... d.cgi?8546

[johnsfine]

I see TV/0150 is quite a good match for those learned signals (though a few of the EFC numbers don't match).

I haven't checked the EFC numbers in the less common versions of that Mitsubishi code set ( TV/0014, TV/0108, and TV/0836) to see if they're a better or worse match than TV/0150. The 8910 doesn't have those other three anyway.

It can't be coincidence that the device number and so many of the EFC numbers match Mitsubishi. So these must not be JVC nor some new protocol. It must be Mitsubishi protocol. So it's even harder than I thought to tell Mitsubishi from JVC{2} in a single sample (DecodeIr always works one sample at a time).

I need to strengthen the hint in the documentation that indicates that you should suspect Mitsubishi rather than JVC if you have a lot of different JVC{2} decodes and no JVC nor JVC{1} decodes.

[ElizabethD]

I still don't really understand the question.

I think part of this issue is explained by the JVC intry in the DecodeIr documentation:
http://john.fine.home.comcast.net/ir/DecodeIr.html#JVC

JVC protocol has a lead-in on just the first frame, unlike most protocols with lead-in that have it on every frame.

You understood it better than my way of asking. Thanks. Neat explanation.
I feel dumb, because your instructions do give an answer. It's just that the only thing I remembered was if it's {1} and {2} and they're the same, it's JVC, ignore {2}. And for {2} only I forgot the Mitsubishi confusion, hence this thread. These things are slooowly sinking in once you look closer.

Some polaroid owners have had success with Mitsubishi TV code

These links are interesting. What a jungle.
BTW, this is a cool TV, all 3 of them. It even has a little blue blinking LED which acknowledges the remote's signals.

Follow up question: Does Mitsubishi decode those repeat bits differently than JVC? I was trying to puzzle it out once as it looked different from the little bit I figured out from my DVD and another real JVC gadget. My math was likely bad, but if not, then KM JVC upgrade working these TVs is puzzling.

Edited 2/8 - never mind the question, I puzzled it out. And thanks again.
I also see that Rob has posted manual protocol upgrade. Missed it before

[ElizabethD]

I suspect the device would also accept Mitsubishi signals with the same numbers, despite the timing being closer to JVC.

It does.

[ElizabethD]

More feedback, in case anyone's interested, 'cause I've been playing
> I tried, on the OEM remote, 2 JVC 5-digit setup codes. They did not work.
> Having found an upgrade of Rob's for 6131, I then made the OEM remote run Mitsubishi setup code 11276. It worked. Mitsubishi 10150 worked few buttons only. All learned signals from 11276 decode as JVC{2}
> Rob's upgrade for 6131 works as well for the most part. Rob's manual setup learned from 6131 to 8910 ended up with identical timing as what the OEM remote put out for me the first time around. It all says JVC{2}.