Page 14 of 17
Posted: Wed Mar 26, 2008 9:07 pm
by unclemiltie
in my zeal to clean things up in the extender, I introduced a bug (that Binky had actually already fixed) in the Multiplex protocol in V0.06. I've fixed it and updated the file.
If you are using V0.06 AND you need device multiplex, you should get it again. If you're not using V0.06 this has no impact on you
Posted: Wed Mar 26, 2008 9:40 pm
by unclemiltie
sriram wrote:I tried $1e $00 and $00 $1e and it still doesn't seem to work :(
OK, I looked at your IR file and am confused....
you have a special protocol for Pause ($1E) bound to Page+
You have a macro F2 that is X_Tv, Shift-Source, Page+, Page+, Shift-Source
you have another Macro Source that is Shift-Source
Source is bound in TV/1000 to a real key
The extender does what is known as Shift cloaking to allow you to get access to real buttons in macros that may be defined on those keys. So when there is nothing bound to shift-key, the extender will then do the unshifted key.
So, in your macro, you will see
X_Tv
Shift-Source -> Source-> Shift-Source -> source
Pause (Page+)
I have no idea if this is going to cause the extender fits,
may I suggest that you bind the macro for testing to something like
X_Tv, Channel+, Pause, Pause, Channel+
and then come back and tell me what happens.
What you should see is the channel go up, wait a bit and then it goes up again. (or you can use something else like guide/exit, or something that is visible in your system to let you know what is really happening)
let us know.
As for the two-byte hex I found a way around that on this extender. I've tested and verified that the hex value does get into the protocol by writing things back into the EEPROM and reading them and timing between blinks, so I think that this is correct. Others have used it and found it to work as well.
-bill
Posted: Thu Mar 27, 2008 1:59 pm
by unclemiltie
Well....
I spoke too soon. Turns out after careful examination the way that the $03 length keymoves work requires that (1) the key value is a valid key value and (2) that the key is in the list of valid keys for that device type. Thus, using this method isn't going to work.
So, this is going to be a restriction in how the extender can be used. To use the Pause, the pause value in hex has to be followed by another (value is ignored) hex value. So, in your example for 3.0 seconds, you would want
$1E $00
(and yes, I know you tried this and it didn't work. still investigating)
Posted: Thu Mar 27, 2008 4:27 pm
by sriram
unclemiltie wrote:
OK, I looked at your IR file and am confused....
you have a special protocol for Pause ($1E) bound to Page+
You have a macro F2 that is X_Tv, Shift-Source, Page+, Page+, Shift-Source
you have another Macro Source that is Shift-Source
Source is bound in TV/1000 to a real key
I wanted the TV/Source to be global accross all device so I assigned it a macro on Source that does X_TV, Shift-Source. It seems to work well, but for debugging purposes I tried replacing the Shift-Source in the pause macro with VOL+ as you suggested. It still doesn't seem to pause, however

(the tv volume only goes up one because, I assume, the second VOL+ is too soon after the first).
Thanks for looking into it.
Posted: Thu Mar 27, 2008 6:33 pm
by binky123
I'm fairly certain the Pause protocol will work if you enter 1E 00 into the Hex box on the Special Protocols tabsheet in IR.exe. Can you post your .IR file where you tried it? I tested using X_TV;PageUp to simplify things and it seemed to delay correctly(at least the LED light stayed on).
Posted: Thu Mar 27, 2008 7:57 pm
by sriram
here's the file I'm using.
After some more testing, pause seems to work if I have it on the TV device before hitting the macro, but it doesn't work if I have it on any other devcie. It's like the X_TV doesn't work....
Posted: Thu Mar 27, 2008 11:48 pm
by greenough1
Sriram,
Your comment about it's like X_TV isn't working is what I was running into. I have a macro that selects TV (X_TV) and changes the input. Then X_RCVR (select AVR) and change it's input. The X_TV + TV input change doesnt' execute, but the X_RCVR + RCVR input change does. Wierd. I haven't had time to work on it, but your comments were too familiar to not chime in on.
Best,
jeff
Posted: Fri Mar 28, 2008 12:18 am
by unclemiltie
re: jeff's comment
hmmm... That'll require a bit more thinking
I can't understand why X_TV would not work and X_other stuff would work. unless I got some of the key definitions off in the RDF or in the extender. I'll have to take a look at that.
Posted: Fri Mar 28, 2008 1:46 am
by sriram
I think it's common to all X_
after some more testing here's what I think is happening:
my "pause" keymove is on TV/MUTE
the macro I'm using to test goes like this: X_TV;VOL+;MUTE;VOL+
If I have the remote on TV mode before hitting the macro everything goes as planned. If, however, I have it on some other mode then when processing the macro the pause keymove is ignored, and the real TV/MUTE is executed (muting the TV instead of doing the pause).
It seems to be the same for all keymoves, so more generally:
if you X_SOMEDEVICE in a macro and call a keymove, the keymove will be ignored and the underlying real key executed unless you are already in SOMEDEVICE mode before hitting the macro.
hope that makes sense
edit: a simple way to test:
have two devices with a menu key (I'll use TV and CBL).
(1) Make a keymove that sets TV/SETTINGS = CBL/SETTINGS.
(2) Make this macro (in addition to the defaults): X_TV;SETTINGS
if you hit "TV" then the macro you'll get CBL/SETTINGS
if you hit "CBL" then the macro you'll get TV/SETTINGS
Posted: Fri Mar 28, 2008 9:38 am
by vickyg2003
It sounds like the active/saved device register confusion wasn't fixed in the x_ device case.
new version posted
Posted: Fri Mar 28, 2008 9:44 am
by unclemiltie
Binky found this one as well. A swap of the R_Dev and R_DevA when searching for an advance code. (and hence you were getting the current device when you were looking for the pause protocol which is really just a sophisticated keymove)
V 0.07 should fix this!
it's here:
https://www.hifi-remote.com/forums/dload ... le_id=5490
Posted: Fri Mar 28, 2008 11:42 am
by sriram
There still seems to be a bit of weirdness in the device selection. Here are the symptoms:
again have two devices with a settings key (I'll use TV and CBL).
(1) Make a keymove that sets CBL/SETTINGS = TV/SETTINGS.
(2) Add pause on TV/Phantom1
(3) Make this macro (in addition to the defaults): X_CBL;SETTINGS;X_TV;Phantom1;SETTINGS
In theory the macro should get you the tv settings, pause, then the tv settings again.
What I get instead is
(1) if I hit CBL first then the macro: settings on the TV and no pause and nothing else
(2) if I hit TV first then the macro: settings on the TV and settings on the CBL and no pause.
Kinda weird; I'll see if I can narrow it down further.
Posted: Tue Apr 08, 2008 10:24 am
by Evan_s
Any more news on this extender? I just got my Direct TV installed so I finally have a good reason to get my OCAP atlases going but I'm definitely going to want to use the extender.
Posted: Tue Apr 08, 2008 12:59 pm
by speaker.guy
Just a social comment, no insult or other bad things intended:
I will never do this. There is not enough time left in my life to both create a extender and use the device it is intended to control!.......
Posted: Tue Apr 08, 2008 1:16 pm
by vickyg2003
There is not enough time left in my life to both create a extender and use the device it is intended to control!.......
Yes, extender writing is a special kind of sickness.

These cable remotes really cry out for an extender. They have very usable layouts, but their native punch through makes them difficult to use without an extender. I'm really fond of the extended atlas although I don't have many applications where I need 5 devices. Its usually 2-3 or 7 or more.