Home AMX User Forum AMXForums Archive Threads Tips and Tricks

Problem learning some IR codes

bobbob Independent ProgrammerPosts: 296
Hi there

I am facing a real problem when trying to learn some of the IR codes from a particular remote.

I am using a pronto and prontoedit to learn the IR codes from the original remote and then cut-and-paste these codes as Discrete IR Hex Codes straight into IREdit. Particular I can not learn the left/right/down/up arrow keys. When learned they do not go one item up or down but several which drives me nutts. I have tried SEND_COMMAND SP with CTON 1, also setting SEND_PULSE_TIME(1) and TO or PULSE without luck. Each time a short press on the viewpoint on the learned UP key produces SEVERAL UP jumps on the TV screen.

The original remote works just fine (this is for a Sigma Designs REALmagic card from inside a dvd/video server computer) and also some other buttons like PLAY or STOP are fine. But other like PREV or NEXT CHAPTER skip several chapters.

Any ideas as to what I may try else?

Many thanks in advance,
Bob.

Comments

  • Thomas HayesThomas Hayes Junior Member Posts: 1,164
    I've had this problem before and somewhere in the forums we discussed about this issue before. I and others ended up changing the pulse time for a few of the commands to work properly. I think I used the 'SP' command to correct the problem. If you take a look in IRedit at the different commands you will probley notice the timing pulse width is different for the commands that are causing the problems vs the ones that work. I'll try to track down the thread and post it for you. :)
  • bobbob Independent Programmer Posts: 296
    Thanks Thomas.

    I know about the SP and the CTON, CTOF commands. I already used them in the past with other remotes to correct this problem. However they doesn't help with this one particular remote!! I am already using

    SEND_COMMAND XCARD, "'CTON',1"
    SEND_COMMAND XCARD, "'CTOF',1"
    ...
    SEND_COMMAND XCARD, "'SP', PUSH_CHANNEL"
    ... or, also tried:
    SET_PULSE_TIME(1)
    PULSE [XCARD, PUSH_CHANNEL] (* also tried the TO command here *)
    SET_PULSE_TIME(5)

    with the same result! I think that the pulse time can not be shorter than 1. I have never tried different values for CTOF though?!

    What also makes me wonder is the fact that most of the time I learn the one particular IR code the hex values for it are different, also the length is from time to time different! But the original remote just works fine!

    Thanks again,
    Bob.
  • Marc ScheibeinMarc Scheibein Junior Member Posts: 688
    With some IR codes with the same symptoms (mostly with the SKIP functions of Kenwood CD/DVD players in the past), I did a workaround which in most cases give me about 95% to work correct:

    PUSH[PANEL,1]
    {
    ON[dvIR,4]
    OFF[dvIR,4]
    }

    It can also be done in a BUTTON_EVENT.

    THis is the shortest "pulse" I could create ever... but as I told above, I only get a rate of about 95% ....
  • DHawthorneDHawthorne Old Timer Posts: 4,584
    This is quite common with remote buttons that repeat when held. I have had limitied success in re-capturing them by the "shutter" method. Hold your button down and block it (I use a large Post-It pad). When your Iris is ready, lift your "shutter" for as near exactly to one second as you can, then drop it back down. Keep the button depressed. When the Iris settles down and is ready to verify, lift the shutter again for one second. If all went well, you should have a ready-to-send condition.

    Arrow keys (and often volume and channel) send a repeating IR pulse for as long as the button is held (some types of remotes do this with every function). The actual pulse for control is very small, and it is extremely easy to learn several pulse in the place of one if you are simply pressing the button. The shutter method allows for better control of the size of the pulse you are capturing. As a rule of thumb, I try to make my capture as short as possible while still being able to verify it.

    Some remotes with a big nav button that you can press the edge for your arrows have the capacity to send a mixed signal to simulate diagonal movement. These I take apart and use a jumper wire to short out the button contact to insure I'm not getting a combination of up-and-right, or whatever.
  • bobbob Independent Programmer Posts: 296

    PUSH[PANEL,1]
    {
    ON[dvIR,4]
    OFF[dvIR,4]
    }

    THis is the shortest "pulse" I could create ever... but as I told above, I only get a rate of about 95% ....
    This one did the trick! I had the same problem with the SKIP and PREV buttons from the same remote! Now those commands work just fine.

    Thanks, Marc! And also thanks to Dave and Thomas!

    Greetings,
    Bob.
  • kstenbackstenbac Junior Member Posts: 16
    DHawthorne wrote:
    I have had limitied success in re-capturing them by the "shutter" method. Hold your button down and block it (I use a large Post-It pad). When your Iris is ready, lift your "shutter" for as near exactly to one second as you can, then drop it back down. Keep the button depressed. When the Iris settles down and is ready to verify, lift the shutter again for one second. If all went well, you should have a ready-to-send condition.
    I always use this method with arrow and skip buttons: I press the button very quickly, so that the green led in Iris will flash only once. Sometimes the verification fails because the second push may be longer than the first one (led flashes more than once). But you can always try again and usually second try is succesfull.

    Bob wrote that he is capturing IR functions with a Pronto and then copies a hex code to IREdit. Unfortunatelly sometimes if you use the method I described, the hex code generated by ProntoEdit has wrong format and it can not be used in IREdit.

    I've also noticed that something that works in Pronto doesn't necessarily work in AMX if it's transferred to IREdit as a hex code (even if IREdit accepts that hex code), but same thing can work correctly if it has been captured with Iris. Strange...

    Marc's code sample looks very interesting, I'll have to keep that in mind.

    -Kari-
  • SpannertechSpannertech Junior Member Posts: 53
    I concur with Kari - if you want super short pulses, you have to learn them super short. It 's pretty instructive actually "looking" at the IR (substitute a visible LED, view it through a digital camera or use an IR->visible gizmo (mine's here: http://www.spannertech.com/spanner/IR/ ) and seeing what's coming out. Results will surprise. Many many library files are learnt far too long. Relearning shorter pulses has usually done more for me than just reducing pulse time. I came across a real gotcha the other day: My jobs nearly always have the HD Tivo (Hughes HR10-250) and I routinely check that the arrows are working by selecting GUIDE and going all directions. It can work OK in the guide, but then still two-step on the menus (Now Playing etc). I've had to relearn the Up/Down arrows extra short to overcome this. This model has a super-sensitive IR window, easily saturated.

    OP
  • DHawthorneDHawthorne Old Timer Posts: 4,584
    We should get this thread moved to the tips sectipn :).
  • bobbob Independent Programmer Posts: 296
    I've had to relearn the Up/Down arrows extra short to overcome this. This model has a super-sensitive IR window, easily saturated.
    OP
    In order to avoid oversaturating the IR receiver, has anyone tried sticking something like a half transparant sort of tape or foil between the IR flasher and IR window on the deck being controlled? Something that would partially absorb the IR light?

    Greetings,
    Bob.
  • SpannertechSpannertech Junior Member Posts: 53
    Yes I've done this in the past, also adding a series resistor.

    OP
  • bobbob Independent Programmer Posts: 296
    I've got an IRIS the other day and tried to relearn the still problematic prev/next chapter buttons. The arrow buttons are fine with the ON/OFF code supplied by Marc, but I still get two-steps with the prev/next buttons. I tried to re-learn them super short again and again, but the still two steps!!

    Has anyone experience with the non-P0 modes on the IRIS? According to the manual they are:

    SP Special Function. Special mode to capture unusual HC functions.
    P3 Sends HC functions to a PC running IRLIB automatically
    after the VERIFY operation.
    P4 Disables the VERIFY operation to capture HC functions.
    Default mode requires the VERIFY operation.
    P5 Loose-timing mode to capture JVC (PQ10956) HC functions.
    P6 Glitch-detection mode to capture Kinderman and some
    Mitsubishi HC functions.
    P7 Disables glitch-detection mode to capture non-carrier HC functions.
    P8 Strict-timing mode
  • kstenbackstenbac Junior Member Posts: 16
    My experience about non-P0 modes is that if you have a remote that can't be captured correctly, you just have to try different modes. Usually it's very hard to define which mode should be used.

    About your actual problem with the prev/next buttons: after you have captured them super-short, try to use the CTON command in your code. I've never had any device which doesn't work correctly after "super-short capturing" and then shortening the pulse with the CTON command, actually I've never had to define the pulse length below 2 (which is equal to 0.2 seconds).

    -Kari-
  • i just wanted to check if you are using the IRS4, external IR card.

    this behaves differently for some products i have to deal with here in Australia, notably Fox Digital TV.

    also, since i describe each IR device in a text file for when the system boots, i tend to specify a CTON/CTOF setting for any device that may give me trouble. (as far as i know, all the IR files have been captured with IRIS at full length).

    one unreliable TV (sharp aquos over IRS4) has been set to CTON=4, CTOF=1.

    and the Fox Digital TV (not over IRS4, wont ever run reliably) is set as CTON=1, CTOF=3.

    yes, i could fiddle with shortening IR codes and recapturing short pulses, but i only have so much time in a day. the system i've mentioned, and being cautious around an IRS4, has worked fine so far.

    i think there are basically two button types, single press (power, pause, play etc) and repeat buttons (numbers, cursors etc), with the latter group causing the trouble.
Sign In or Register to comment.