View previous topic :: View next topic |
Author |
Message |
unclemiltie Expert
Joined: 21 Jan 2004 Posts: 1795 Location: Pittsburgh, PA |
Posted: Wed Mar 25, 2020 12:21 pm Post subject: |
|
|
I just did a diff on the code from 3.03 and 3.04/3.05 (same code for both) and see only a few differences. Most of the differences were the add to support the 3165 remote so irrelevant in the 3147 build as that assembly is skipped. There are five changes to the 3147 code, only two of which may impact you:
I swapped the Transport and Channel key definitions to be consistent among all of the remotes. So if you had a configuration that had a Set_Trans in 3.03 it will now be seen as a Set_Chan. I did this to make maintenance of the extender easier when I published the source for the experts in case I wasn't able to continue to work on this thing. (now all of the remotes are the same and use the same order as UEI does on their remotes that have "Home Theater" capability) I don't think this is an issue for you.
I changed the default HT setup table from $00 to $04 to reflect the above change in the HT table order. These bytes are stored in $623 thru $626 in the raw data area. should be no impact
I changed the default device codes for Dev1, Dev 4 and Dev5. Although I think that when the extender is built I put zeros in all of the devices. The defaults are only used for when I'm debugging so I don't have to always change the device codes to find one that is actually in the remote and should not impact you.
For some reason and I don't remember why I changed the first byte pf the remote configuration area. This area is not understood by the experts and changing those bytes can cause the remote to do odd things. This byte is at $61A in the raw data tab. In 3.03 it was $02 and in 3.04/3.05 it is $00. I *think* that byte is the setup lock bit which should have no impact but you could try to edit the raw data and change it to $02 and see if that makes your configuration work.
In the code the instruction to check a keypad press during pause was changed from a JRC (Jump Relative if Carry bit set) to a JRNC (carry bit not set) to fix a bug in the interruptible pause. On the 3147 this instruction is at $08F2 in the raw data. The JRNC is $FB 02, to change it back to the JRC would change it to $7B 02. you could also try that to see if that changes things.
other than that, I see nothing else in the code that is different. If you want to try patching those values and tell me what happens that will help until I can find my remote. _________________ this JP1 stuff is a sickness! |
|
Back to top |
|
|
unclemiltie Expert
Joined: 21 Jan 2004 Posts: 1795 Location: Pittsburgh, PA |
Posted: Wed Mar 25, 2020 12:42 pm Post subject: |
|
|
How the interruptible pause works is this:
The pause value is specified in 100ms intervals. IR loads the number of pause counts into the key move that fires the pause
When the pause is called, it loads a value for 100ms into a register and then counts down to zero.
When the countdown is done it calls a function to scan the keypad to see if any keys were pressed. That function returns a status via the Carry bit
if a key is pressed that JR goes to the end of the pause return and it terminates
if a key was not pressed, the pause decrements the number of 100ms pauses left to do and jumps back and does the whole thing again.
It should not matter if it's inside a ToadTog (which is in essence a macro) or if it is not. All macros are pushed on the stack and replayed back from there, which is why the extenders have almost unlimited macro sizes. I guess if you have an exceptionally long macro you could overflow the stack (which is not checked) and wipe out some portion of memory that would cause the remote to go nuts. But that would be a really big macro _________________ this JP1 stuff is a sickness! |
|
Back to top |
|
|
jeajea
Joined: 24 Feb 2010 Posts: 283 Location: USA |
Posted: Wed Mar 25, 2020 3:53 pm Post subject: |
|
|
I mentioned the toad tog because I don’t know how common a pause called from a toad tog is.
It may be useful for you to take a look at the simpler 3.05 file I uploaded today to see if the problem is caused by my error.
I have also tried a different remote (an AOC 67100) in case the insignia had a hardware problem.
I think I initially used Rmir 2.05 to create 3.03 file that doesn’t have the problem.
I have also used Rmir 2.08 and 2.09 to edit the 3.03 file.
If anyone is interested new AOC 67100 remotes are available on eBay for $13
https://www.ebay.com/itm/AOC-67100BA1-017-R-Remote-Control-Lifetime-Warranty-and-Free-Shipping/202671900342 _________________ Jim Anderson |
|
Back to top |
|
|
jeajea
Joined: 24 Feb 2010 Posts: 283 Location: USA |
Posted: Wed Mar 25, 2020 4:19 pm Post subject: |
|
|
Note: I have no urgent need for 3.05.
As I mentioned before tale care of yourself and family first. _________________ Jim Anderson |
|
Back to top |
|
|
|