Page 3 of 3

Posted: Tue Jan 04, 2005 11:11 am
by mtakahar
Something in common between TV Power and TV/VCR is both of them are learn'able buttons. Still not sure why it would make a difference, though.

Anyway, WaitNoKeys is one of the places that might worth a try.

leres, could you try changing the address of WaitNoKeys to 06BC in the source code (or change "C2" to "BC" at the location 0B3 in the raw image)?

Hal

Posted: Tue Jan 04, 2005 2:06 pm
by leres
The WaitNoKeys change did the trick. With it defined as 06BC, TV/video works correctly as the shift key. I also tested that the shift key times out after 30 seconds. So basically the rom location of the WaitNoKeys routine was incorrectly defined in the extender?

(Prior I tried David's hardwired-shift change which did keep shift in effect at all times.)

Posted: Tue Jan 04, 2005 3:11 pm
by The Robman
This sounds promising. Looking at the raw data in the IR dump, I see an "F6 06 C2" at $0792, is the fix as simple as editing this to "F6 06 BC" ? If so, I'll send a file to the guy I programmed the remote for to see if it works.

Thanks for all your efforts.

Posted: Tue Jan 04, 2005 4:01 pm
by mtakahar
leres wrote:So basically the rom location of the WaitNoKeys routine was incorrectly defined in the extender?
This is kinda yes and no. The original location was equivalent to many other extenders, but they work fine because of combinations with other things, like backlight, lack of learning, slower processor, etc.

I'll upload the fixed extenders repackaged with newer LDKP (LDKP is for ex2 only) later today.

Rob, I don't have the source, etc. in handy right now. get back to you once I get home.


Hal

Posted: Tue Jan 04, 2005 10:14 pm
by mtakahar
The Robman wrote:Looking at the raw data in the IR dump, I see an "F6 06 C2" at $0792, is the fix as simple as editing this to "F6 06 BC" ?
I've just verified - Yup, that's it.

Hal

Posted: Tue Jan 04, 2005 11:42 pm
by mtakahar
mtakahar wrote:I'll upload the fixed extenders repackaged with newer LDKP (LDKP is for ex2 only) later today.
done.
http://groups.yahoo.com/group/jp1/files ... DirecTivo/

Posted: Wed Jan 05, 2005 12:00 am
by leres
What would it take to make the extender boot up into the TiVo device instead of the TV device?

Craig

P.S. Thanks to everybody for their help in solving the shift issue.

Posted: Wed Jan 05, 2005 1:33 am
by mtakahar
leres wrote:What would it take to make the extender boot up into the TiVo device instead of the TV device?
It's just a matter of initializing R_DevO with the value of your choice in the beginning.

For example, you could add "LD R_DevO, 1" after "CLR RD9":

Code: Select all

; ---- Extender

;Extender
        org     Ram
        xorg    ee_main         ;10;

L00:    ;Initialization only needed one time.
        CLR     RD9                     ;Init stack pointer
        LD      R_DevO, 1       ;"Other" device <- TiVo
L01:    LD      R_DevX, WD      ;Turn off temporary DevIndex
L10:    CALL    WaitNoKeys      ;Wait for no keys pressed
        DI
Here, "1" is the device index where 0=TV, 1=TiVo, 2=CBL, and so on.
You can initialize the volume device using R_DevV in the same way.

This is a trade-off of 3 - 6 bytes for saving 2 key strokes per activation.

Hal

Posted: Wed Jan 05, 2005 2:07 am
by leres
I guess another other approach would be to rearrange things so that the TiVo device used device index 0. But then volume device would still be wrong.

The 6131 extender lets you specify the initial O, V (and T) devices. Is that a feature of the 6131's rom that the RM-Y809 lacks or code in the extender?

Obviously this isn't a big deal but it is sort of cool to be able to set the initial devices. For example, if you're doing a volume punch through type config, you can actually save bytes because you never have to use any V_* commands.

Posted: Wed Jan 05, 2005 2:24 am
by leres
Ok, I see the code in the 6131 extender now that does the initial device setup including the role that the rdf plays. I assume it would be possible to do something similar for the RM-Y809 extender although maybe it would make sense to do it in the 2K eeprom version? Maybe I'll play with that once I get one upgraded.

Posted: Wed Jan 05, 2005 7:44 am
by The Robman
Hey Hal and leres, thanks for all your efforts on this. I will patch my file and send it to the user and report back if everything works OK.

Posted: Wed Jan 05, 2005 11:08 am
by mtakahar
leres wrote:Ok, I see the code in the 6131 extender now that does the initial device setup including the role that the rdf plays. I assume it would be possible to do something similar for the RM-Y809 extender although maybe it would make sense to do it in the 2K eeprom version? Maybe I'll play with that once I get one upgraded.
Yup, that'll work.

Hal

Posted: Thu Jan 06, 2005 12:31 am
by leres
Is it possible for the extender to read the state of the TV/sat switch? (Not that I can think of anything interesting to do with it at the moment...)

Also, I assume the RMT-V303 doesn't have such a switch. (And looking at pictures of it it also doesn't have delimiter, TV/video, menu, display or live-TV buttons.)

Posted: Thu Jan 06, 2005 10:48 am
by mtakahar
leres wrote:Is it possible for the extender to read the state of the TV/sat switch? (Not that I can think of anything interesting to do with it at the moment...)
I thought about it, too, but I was just borrowing the remote from someone when I wrote the extender and didn't spend much time trying.
Also, I assume the RMT-V303 doesn't have such a switch. (And looking at pictures of it it also doesn't have delimiter, TV/video, menu, display or live-TV buttons.)
It doesn't have them. There's a less-obvious thing RM-Y809 has but RMT-V303 doesn't. It's a button on the PCB that is not exposed on the key pad. Usually you can't map functions on phantom buttons in the device upgrades, but you can do it on "Phantom0" in RM-Y809.

Hal

Posted: Tue Jan 11, 2005 2:24 pm
by The Robman
Btw, I just got word back today from the guy who I programmed the remote for and he confirms that this fix works.

Thanks again to you guys for helping out with this.