Anyone here got an RC5 device that needs the correct toggle?

[The Robman]

I have just developed a custom executor that can hopefully be used by owners of devices that use the RC-5 protocol who want to use their OFA remote to teach another learning remote.

As you probably already know, there is a toggle bit in the RC-5 signal and while most devices don't care how it's set, there are still plenty of them that do, especially high-end audio devices. The trouble with learning an RC-5 signal is that you can only capture the toggle bit in one state, so when you use the learned signals it's 50/50 whether it's in the right state or not.

My custom exec will always send a quick meaningless signal before sending the proper signal, and the toggle will alternate between the two.

So, I'm looking for someone who has a device that uses the RC-5 protocol and you know that it requires the toggle bit to be set correctly to test my code.

Any volunteers?

[zaphod7501]

Rob, I've got some Hauppage devices (PVR250 and MediaMVP) that are RC-5 and a Sony RM-AV1100A (LCD touch panel, learning, 12 device) that would not learn the RC-5 commands from the original remotes (I didn't know the reason at the time, I just knew that a command would only work every other time I pressed a button after learning it).

Is this the type of test you're looking for? I don't use the Sony remote currently. My 6131's are much more usable than the $365 Sony (received in a parts shipment mixup - cost me $3). I don't know if the Sony can generate the necessary commands regardless of how it is learned but I can try and compare results at least. I've got lots of backup 6131's so I don't have to mess up a working setup to try it. If it works I can use the Sony as a "coffee table" remote. Since it's lighted, it may come in handy when running my new front projector (dark room).

[The Robman]

Yes indeed, that's exactly what I'm looking for. If you've already added upgrades to any of your 6131s to work any of these devices, even better.

All you need to do is drop this protocol upgrade into the IR file...

Upgrade protocol 0 = 00 E8 (S3C8+) RC-5 for learning (PB v3.11)
40 9A 41 8B 0F C5 80 10 08 06 01 C2 01 A4 01 C2
01 A4 DE 2C 08 07 56 C0 03 87 10 03 77 1B 19 03
00 10 F6 01 46 B6 03 20 20 10 46 29 01 8D 01 46
End

The existing upgrades use 3 bytes of fixed data whereas this exec uses 4, the extra byte (which needs to be added after the current 3 bytes) is the meaningless command that will get sent before the real command. To figure out what hex code to use, look at your KM or RM file and look for an OBC that's not used, then enter it to see what the hex code is. Then edit the upgrades, adding this byte after the current 3 bytes of fixed data.

Once this is loaded into the remote, verify that the original upgrades still work, with repeated presses of the same button. Then, try using the 6131 to teach the Sony learning remote and see if that works.

[zaphod7501]

Well, it seems to work. It took a modified learning method with the Sony (single press rather than a hold down); before, learning was very erratic with the Sony. Some commands worked, some repeated, some worked barely with multiple presses. It appears to be learning fairly well right now.

I use RM and didn't see any way to edit the fixed data within RM, so I edited the device within IR, and added the protocol. The fixed data shows the additional byte and the 6131 still works correctly with the MediaMVP. I can't test it with the PVR250 until a new mainboard and CPU comes in. I am running the card in a PC too low in power to run in anything other than command line mode for recording.

My only other learning remote is a URC8910 so you probably need some other people testing this too. I think I have only one upgrade device based on RC-5. Would it be of any benefit to compare the outputs from the remote with the modified Upgrade protocol to the unmodified remote (using the 8910)?

[The Robman]

There isn't a way in either RM or KM to edit the fixed data, so I actually intended for you to edit it in IR.

It sounds like it's working as expected. Am I right that you were able to learn the commands using the Sony remote without any special learning techiniques and that the learned commands worked every time? Am I also right that when you used the 6131 to work the device, the buttons worked every time (including repeated presses of the same button)?

Ws there any noticable delay in the unit responding to the remote? I ask because there is now a single repeat of a meaningless signal before the proper signal starts, so I curious whether he delay is sufficient for you to notice.

[zaphod7501]

Sony learning method: A single press of a 6131 button was needed, holding down continuously didn't work well. I'll try learning with my 8910 tonight to see if there's any difference between "single press" and "hold down" during learning (I haven't used the Sony enough to know what learning method works best and the Sony manual is poor when it comes to advanced functions).

Commands learned on the Sony work every time with no obvious delay including repeated presses of the same button.

6131 operation: appears to work exactly the same as before. I am running the 6131 with the extender so that might change the output timing however.

Still to try
: Adding it to a non-extended 6131 and checking for any 6131 problems and learning ability.
: Adding it to my 8910.
: Learning the Sony output with my 8910 to see if there are any anomalies.

These will have to wait until after work tonight however.

[The Robman]

Using the "short press" learning method probably means that repeating buttons (like VOLUME, etc) won't work properly. I'm surprised that the Sony had a problem learning but then again, I'm not familiar with how the Sony normally performs.

You can also use the URC-8910 to perform this test. Try learning the signals from the URC-6131 using the URC-8910, using the "long press" method and then see if the URC-8910 buttons work properly (ie, they work when held and with repeated presses of the same button).

It shouldn't make a difference to the URC-6131 whether you have an extender installed or not, so there's no need to re-test without the extender.

[zaphod7501]

I'm not familiar with how the Sony normally performs

-- in a word - poorly. Most of the long keypress learns work - notable exception - pause, which will only work with a short press. Short presses work, except for things like the volume as you noted (almost didn't check it since everything goes through the receiver and volume on the MediaMVP is never used - set at max always).

The 8910 learns correctly using the "long keypress" method both repeating buttons like volume and one time like "pause". The menu left and right keys work correctly too. (that's actually significant since they operate as one shot for menu and continuously as volume controls when playing something).

The learns show the meaningless command preceding the actual command so it is working as desired.

It looks like you need some other people to check it with different learning remotes. It looks like it is operating as intended but my Sony remote wasn't the best choice for testing.

The odd thing about the Sony learns is that they don't work very well (pause is most noticed) but when I used the 8910 to learn the Sony output of pause, the 8910 sent that learned command correctly while the Sony sent it incorrectly. Confusing?

6131 -> Sony -> xmit -> poor results
6131 -> Sony -> 8910 -> xmit -> good results

In other words the Sony can learn it but can't send it properly but the 8910 interprets it correctly (from the Sony) and does send it properly. That's why the Sony sits on the shelf and I have eight 6131's. Although I may try again since the lit, labeled, LCD touch panel makes it visitor friendly.

[The Robman]

Thanks alot for your efforts on this Steve, I think this pretty much confirms that my protocol does exactly what I was hoping that it would do, even if the Sony isn't up to the challenge.

The guy that I'm trying to help with this wants to program some MX- remotes, which I believe are pretty good learners.

[BallTongue]

Hi guys,

Got some questions that might seem stupid but im trying to learn as much as i can.
I just purchased a Magnavox TV and i got it to work with one of the codes supplied in the owners manual. (By the way im using a RS 2116 remote). Most of the buttons work but some don't. So i tried learning the buttons and everything seemed to work out ok.
But there are buttons that have 2 functions. Like if i press up on the original remote it will change from 4:3 to 4:3 Stretch and if i press it again it will go back to 4:3. When i tried to learn the button it only learned half of it.
Now if i am reading correctly the code that rob supplied will help my 2116 learn the rc-5 signal correctly? I know you stated how to install it but it seems a little to complicated to me. I think i have to learn all the functions, create my own upgrade and then i can add the code in, right?
Any help from you guys would be a great relief as this is kind of making me mad. Thanks.

[The Robman]

When you learn an RC-5 signal with a JP1 remote, you are indeed only capturing one half of the toggle, but that's enough to get the device code and OBC, which is all you need to build an upgrade using the regular RC-5 protocol, you don't need the custom protocol that I wrote here.

This is for people who want to use their JP1 remote to teach another learning remote.

[johnsfine]

But there are buttons that have 2 functions. Like if i press up on the original remote it will change from 4:3 to 4:3 Stretch and if i press it again it will go back to 4:3. When i tried to learn the button it only learned half of it.

The protocol has a feature that won't let it accept the same learned signal twice in a row. But your learned "up" function still carries the whole meaning. If you use it, it won't work again immediately (to change back) but if you use any other function after using "up" and THEN use "up" again, it will change back.

Now if i am reading correctly the code that rob supplied will help my 2116 learn the rc-5 signal correctly?

No. Rob created a protocol that would let you use your 2116 to teach a signal that would work to some other learning remote that lacks the 2116's ability to handle RC5 signals correctly.

What Rob did does not affect how the 2116 learns from the Magnovox remote. That learning will still have the normal problem of learning RC5 signals.

I think i have to learn all the functions, create my own upgrade and then i can add the code in, right?

It's easiest to just use the setup code you already found and instead of using learned signals for the missing function(s), use KeyMove(s) based on that built-in setup code. You need to find out the EFC number of the required KeyMove, then delete the learned signal and install the keymove.

If you have a JP1 cable, it's quite easy to do all of that in IR.exe.

In IR's learned signal's tab, you can examine your learned signal and see the three possibilities for the EFC number. Deducing which of those three is right may be complicated. Just testing all three is probably simpler.

If you choose to build a whole new upgrade instead, use OBC numbers rather than EFC numbers to avoid that problem.

[BallTongue]

Well guys thank you for your responses. I think i am really getting the hang of this stuff.

This is for people who want to use their JP1 remote to teach another learning remote.

Thanks for clearing this up rob. As you can tell i was up late lastnight trying to figure some stuff out and I misunderstood.

If you have a JP1 cable, it's quite easy to do all of that in IR.exe.

Yes i do have a jp1 cable. Got it from hovisdirect and built it myself. I feel a great sense of joy if i know i built it myself.

It's easiest to just use the setup code you already found and instead of using learned signals for the missing function(s), use KeyMove(s) based on that built-in setup code. You need to find out the EFC number of the required KeyMove, then delete the learned signal and install the keymove.

If you have a JP1 cable, it's quite easy to do all of that in IR.exe.

In IR's learned signal's tab, you can examine your learned signal and see the three possibilities for the EFC number. Deducing which of those three is right may be complicated. Just testing all three is probably simpler.

If you choose to build a whole new upgrade instead, use OBC numbers rather than EFC numbers to avoid that problem.

When you posted this i really didn't understand what i was supposed to do but i followed your directions and it worked perfect. I first tried using the OBC codes (I took the OBC code opened up KM and put them in there so it would spit out the EFC code). But when i put in the EFC code KM gave me, it mapped a totally different button then i wanted it to. (Up and Down were now acting like VOL+ and VOL-) I had to do the latter of the choice (go thru and try each EFC code until I found the real function.) Thank you Robman and johnsfine. Without you guys i would still be kicking my self in the ass trying to figure this stuff out.

[johnsfine]

I took the OBC code opened up KM and put them in there so it would spit out the EFC code

Before that will work reliably, you need to put values on the setup sheet of KM (protocol selection and device numbers) so that it matches values built into the setup code you plan to use for the KeyMoves.

You never mentioned what the Setup code number is, what the device number of the decode is, or what the OBC is, so I can't guess where you got unlucky.

If you followed the typical method ( used the protocol name and device number from the decode for KM's setup sheet ) rather than finding the info on your setup code, I'd still expect it to work.

Most of the Magnavox TV signals are RC5 device 0, and most of the OBC numbers are less than 64, and all (so far as I can find) of the TV setup codes for RC5 device 0, have the 0 in the first device position.

So you should have gotten lucky and had KM show you the right EFC.

If you forgot to select RC5 at all in KM, it would give you the wrong answer. If you have an unusual OBC that is greater than 63 then KM (used simplistically as above) would give you the wrong EFC. Beyond that are other less likely ways you might get the wrong EFC.

[feelfree]

Hi,

I own a Harddisk-Music-Server called Hermstedt Hifidelio which is sold in the US as an OEM under the "Olive" label. The device itself is absolutely fantastic, but its remote is not.
Until yesterday nobody was able to use any remote (OneForAll, Pronto) with the Hifidelio without problems. The RC-5 toggle bit was the problem.
Today I created succesfully a JP1-cable and an upgrade for my URC-7560. I am really amazed of all your work now: The URC works flawlessly with the device!!!

So now, lets come to my question:
There are certain keys (eg. scroll up/scroll down), on which the Hifidelio also needs the correct toggle bit, even with the original remote. That means if you want to scroll down 30 entries in a list, you have to press the arrow-down-button 30 times instead of simply holding it.

After reading this thread, my (maybe stupid) idea is:
Would it be possible to make the URC even better than the original remote if someone would be able to do a "protocol update" for some keys, which let's the Toggle-Bit toggle while the key is pressed down continously?
Of course this would make sense for certain keys only, since eg. "skip-forward" should only skip one song instead of many...

I'm completely new to this whole jp1-thing, but if someone says that my idea should be possible, i am willing to get deeper into this topic and i do appreciate any pointer/link on how to program such protocol updates.