Help needed to finalize upgrade on JVC RX-5032V

johnsfine · Post by **johnsfine** » Wed Nov 04, 2009 9:40 pm

You decided to make the lead-in first frame only. Great. I think that is what was needed. It's been so long since I've done such things I forget how.

But you shortened the lead-out?? What did I miss? Do we think we know the correct lead-out? You shortened it from what to what?

Long ago the device combiner worked only with built-in protocols, not upgrade protocols. I thought that was fixed. Your symptoms seem to say it wasn't fixed. Is that non working only for certain models?

C9 is a common executor, isn't it? It isn't only for JVC-48 right? If you put in a modified C9 executor you can't have any device upgrades using the unmodified C9 executor.

ylaviolette · Post by **ylaviolette** » Wed Nov 04, 2009 10:48 pm

Thanks Johnsfine and all others,

first don't get me wrong, my comment about another technical person didn't imply that you weren't one yourself

sorry if you felt that way, I was just looking for additional input...

anyways, my learned signals do work correctly however I uploaded the wrong file (...) which was the first of multiple tests. The one provided does indeed provide a non repeating signal. Sorry if you wasted time on it

Since the problem remains, here's a new version with some explanations

https://www.hifi-remote.com/forums/dload ... le_id=7464

Depending on the way I did the learn (long press directly in front of the remote, bringing the already emitting remote in front of the learning one, etc), It generated different results and I have provided three distinct examples

Example 1:
Vol + is "front L+"; functional non repeating signal

Examples 2:

Vol - is "front L-"; perfectly functional signal (same constant speed as the original remote)
Ch + is "front R+"; perfectly functional signal (same constant speed as the original remote)

Example 3:
Ch - is "front R-"; here you have a perfectly functional signal but much faster than the original remote (still constant)

Hopefully this will help pinpoint the problem with the protocol.

If you need further details or testing just let me know,

Thx again.

Post by **vickyg2003** » Thu Nov 05, 2009 6:07 am

johnsfine wrote:You decided to make the lead-in first frame only. Great. I think that is what was needed. It's been so long since I've done such things I forget how.

We make quite a pair, you forget how, and I don't know how.

But you shortened the lead-out?? What did I miss? Do we think we know the correct lead-out? You shortened it from what to what?

When I shot the learned signals, which we know are bad, at my irscope the leadout time was 31500 as opposed to the standard JVC-48 which had a leadout time of 72000. Combined with the sentence where ylaviolette said that the new signal was not repeating as fast as the original signal, it seemed to me that the leadout time was too big.

Long ago the device combiner worked only with built-in protocols, not upgrade protocols. I thought that was fixed. Your symptoms seem to say it wasn't fixed. Is that non working only for certain models?

I really don't know. I have only encountered the device combiner once before and it was for the HCS08 model, and apparently it had been out buggy for almost 1.5 years before someone tried to use it. This one is for the Flash remotes and seems to be considerably different from the SC308+ version, so I don't know if its the built-in thing/versus upgrade thing, or if it is just getting tested on the Flash for the first time.

C9 is a common executor, isn't it? It isn't only for JVC-48 right? If you put in a modified C9 executor you can't have any device upgrades using the unmodified C9 executor.

Whoops, you bet it is!

That's for the panasonic!

Looks like I'd need to start from the ground up since ylaviolette posted working learns after I started working on this. Apparently this time the JVC frame is mising, and these only decode as GAP's.

I'm really at a loss for how proceed. All I ever wanted to do was to learn enough about this, so that I could provide JP1.2 support!

johnsfine · Post by **johnsfine** » Thu Nov 05, 2009 8:18 am

vickyg2003 wrote:ylaviolette posted working learns after I started working on this. Apparently this time the JVC frame is mising, and these only decode as GAP's.

So that tells us what the repeating part looks like. It also tells us the device isn't picky about missing the first frame. It will understand repeat frames even when they aren't preceded by the first frame. Even so, sending the correct first frame is probably a good idea.

I expect that if you learn one of these signals "correctly" you will get a perfect first frame with only part of the repeat frame and the result won't repeat.

But if you start the 8910 learning an instant after the original remote starts sending, you are likely to get a perfect repeat frame with no start frame. Apparently that works better.

Putting the two together tells us what the whole original signal looked like.

Sorry I forgot that earlier. I have told people to do exactly that in previous cases where the start frame was too long to allow the remote to learn the repeat frame. Intentionally do the learn wrong to miss the start frame and you have a good chance of discovering the correct repeat frame. Apparently, Ylaviolette had already done that (maybe by accident, maybe on purpose).

Ylaviolette, can you test whether the upgrade Vicky posted earlier works?

If it doesn't play well with the device combiner, you could still test it without the device combiner to see if it gets the timing good enough. If the C9 id for the executor conflicts with some Panasonic or other upgrade you have for another device, we can change the id later, but you still could test the upgrade first.

ylaviolette · Post by **ylaviolette** » Thu Nov 05, 2009 9:29 am

You are correct, as I mentionned above, when I'm having trouble learning a signal, I always proceed in three steps to see if I get different results:

A: short press directly in front of the learing remote
B: long press directly in front of the learning remote,
C: bringing the already emitting remote in front of the learning one (which seems to supress the first frame?)

I'll be glad to test what Vicky posted earlier however just to be sure I'm testing correctly, could you confirm the exact IR file name/location and which button(s) or signal(s) you want me to test?

Thx again.

ylaviolette · Post by **ylaviolette** » Thu Nov 05, 2009 4:55 pm

By "testing Vicky's upgrade", if you meant trying to add this new protocol:

Upgrade protocol 0 = 00 C9 (S3C8+) PB v4.01
45 91 51 8B 12 8F 48 08 08 00 DA 00 C6 00 DA 02
7B 3E 80 06 D0 03 54 E6 0C E8 60 03 20 01 E4 08
09 B4 03 09 20 10 F6 01 46 F6 01 0A FB 0D 00 0C
EB F4 C6 3E 00 00 B0 4A F6 01 0A AF
End

Yes I tried it without success. Still gives the same results.
However I am not sure about the way I do this; I just basically added this new protocol under the "protocol" tab of IR but do I need to make some kind of association to the device as I still see protocol 144 being matched to my amplifier upgrade?

Keep me posted, thx

Post by **ElizabethD** » Thu Nov 05, 2009 6:57 pm

Associations - See if this helps
Click Devices tab in IR, then Edit. Change first two bytes from whatever it's now to C9 00, yes, reverse the pid.
While this is a Panasonic (I think protocol) in this instance Vicky's thing should do the job.

Post by **vickyg2003** » Thu Nov 05, 2009 8:15 pm

Yes, by testing I did mean to just add the protocol. On MY remote, which is an Atlas, the 144 combiner would use the protocol in the upgrade area, although it would crash right after I pushed any button on the device that used the 144 device combiner protocol. John had mentioned that the 144 had problems with protocols in the upgrade area, so that is why you can't access this protocol. What we really need to do is set up a helper device that uses the JVC-48 protocol. With this protocol, that should send out your signals.

ylaviolette · Post by **ylaviolette** » Thu Nov 19, 2009 2:29 pm

Let's simplify things. Forget about the Combiner. I'm sure it's not related.
I have tested the same commands under the standard (non combined) JVC-48 protocol with the same results. ("flaky" signal instead of smooth ramp up and down)

I did the test with and without the protocol upgrade provided by Vicky

Find below two files I've created : one is the RMDU file used for the 6 JVC-48 commands. The other one is the IR file used. (with Vicky's upgrade)

https://www.hifi-remote.com/forums/dload ... le_id=7500

If you have other ideas, just let me know, I'll be glad to do further testing and hopefully resolve this. I'm sure the solution is not too far.

Thx again!

Yann

Post by **vickyg2003** » Fri Nov 20, 2009 11:21 am

ylaviolette, I don't know how you came up with that IR but your 2005 device was jibberish.

I put your information into a JVC Combo and a helper protocol, and modified this ir. Pleas note because I had to split this into 2 upgrades, I put these JVC-48 keys on the shifted version of the key, because you were running out of keymove space.

So its the Shift Fwd and the Shift Rev that do these troublesome keys.

The upgrades I used to create these are here

ylaviolette · Post by **ylaviolette** » Fri Nov 20, 2009 1:52 pm

Thx Vicky, other than device 2005, you can get rid of everything in my IR file, this is just a test file. All keys except the 6 that have relevant info are valid, all other keys (same irrelevant data repeating) is just remaining code from the old combiner file. If this is an issue I'll reupload a new file clean of other data.

I am not at my PC so I haven't had a chance to test your files. Maybe I'll get my answers later when I try it, but just in case:

Why you are working with the combiner at this point? As I said eariler, the stand-alone JVC-48 si not working either, shouldn't we test the stand alone before the combiner?

Also why did you split it into 2 upgrade?

I'll provide feedback once I test your files, thx much for your assistance.

Post by **vickyg2003** » Fri Nov 20, 2009 3:19 pm

Why you are working with the combiner at this point? As I said eariler, the stand-alone JVC-48 si not working either, shouldn't we test the stand alone before the combiner?

I couldn't find any evidence that the modified jvc-48 was tested in your IR. As I said the 2005 device code was gibberish.

I'm not working with the COMBINER, I'm working with the JVC COMBO. There is a big difference.

The combiner is different for every remote. Since I couldn't get this to work with the combiner for more than one keypress on my own remote and you had reported that the fix didn't work on your remote (although now that I see that your IR had a corrupted upgrade, I'm not 100% sure your test was accurate.) I decided to follow a more universal solution.

Also why did you split it into 2 upgrade?

Your device needs JVC 167, JVC 175 and a modified JVC-48 34.84 signals. We can combine the JVC 163 and 175's using the JVC COMBO. To get the other 5 ignals, we'll use a helper upgrade. That way we can avoid the combiner.

The JVC-Combo is a processor dependent protocol, so it is a lot easier to test than a remote specific dependent protocol. I can test a processor specific protocol, with any remote that has the same processor. I don't have to have the specific remote. So this approach will work on all s3c8, s3c8+ and JP1.3 remotes. If you repote that this works, as it appears to work, I'll add hcs08 support, and then we'll have pretty good remote coverage. Not all of the combiners are thoroughly tested, because the people writing the code don't have access to the remotes. We need people with the remotes to dest the devcombiner.

If you want to concentrate on the jvc-48 just work with the helper device.

Good luck.

ylaviolette · Post by **ylaviolette** » Sat Nov 21, 2009 9:45 am

OK here's where we stand:

unfortunately when I try your IR file, shifted keys you asked me to try still act the same way, the dsp (on swap) doesn't react and the dimmer (tv/vcr) crashes the remote.

I did some further tests below maybe you'll be able to pinpoint a problem somewhere:

1- when I try to open either Keymap files, it states
"The upgrade file you are loading is for the remote URC-9910/8910/UEI HTPro. There is no remote with that exact name. Please choose the best match below:"
So I choose my usual "URC-9910 (New)"

2- The "helper file" has no visible shifted keys and only has two of the keys assigned? (dsp and dimmer), so I have assigned the other 4 "front levels" keys (rmdu attached below)

3- I have slightly rearranged they keys of the combo file to match my original config. (rmdu attached below)

4 - I have cleaned the IR file just to keep the minimum (attached below)

Unfortunately the IR above will work for all protocol 163/175 keys but not the protocol 48 ones. Something is probably wrong with my procedure but I don't know what...

Let me know if I can pursue other tests. Thx

https://www.hifi-remote.com/forums/dload ... le_id=7515

Post by **vickyg2003** » Sat Nov 21, 2009 11:03 am

I went out this morning and purchased batteries for my 8910, to see if I could see what is going on here. Unfortunately I don't have my USB cable here, and my 8910 is not talking with my very-voltage-finicky-simple interface.

Its no surprise that the IR file won't send out the jvc-48 codes. The code isn't assigned to a device button, and there are no keymoves to device 2006. Helper protocols usually have their functions assigned via keymoves.

Hopefully after a night with with full batteries my remote will start conversing with the simple interface.

BTW: I crashed your remote with the dimmer, and after looking at the IR file I posted, I see that I had to hand do the keymove and I picked receiver/2006 instead of vidAcc/2006.

ylaviolette · Post by **ylaviolette** » Sat Nov 21, 2009 12:37 pm

Sorry, I was not familiar with the "helper" concept; from what I understand, when not too many keys are involved, it's basically to use a device upgrade in order to complete another one through keymoves, correct?

Since we are just testing, I have assigned 2006 to another mode (CBL) but still the same flaky results. I have attached the IR file below. Thx

https://www.hifi-remote.com/forums/dload ... le_id=7516