Porting an existing extender to the DCP501-R remote

mtakahar · Post by **mtakahar** » Sat Jul 09, 2005 8:22 pm

Another question. Do you know what the default device is when the remote got reset? Even thought the activation setup code is TV/1800, you should just press a volume key in the default device when you activate the extender.

I was assuming the default was Cable, but the following line may have to be changed if that's not the case:

Code: Select all

RR_DevASetup	=	RR4E	;Setup code of active device

Should be RR50 for Audio, RR52 for TV, etc.

Hal

Post by **The Robman** » Sat Jul 09, 2005 8:35 pm

Seeing as how this is the OEM remote for a DCP501 a/v system, I would assume that the default mode is "audio".

joshville · Post by **joshville** » Sat Jul 09, 2005 10:05 pm

Thanks for the great information, Hal. I only had time tonight to test the blink protocol/upgrade. It worked as advertised, except that "4" blinked 3 times and "3" blinked four times. The first blink on each key was longer than the others, maybe like a double blink with no off time in the middle. A 981 reset gives two sets of two blinks with a short pause between sets.

Thanks!

Gotta go soothe the baby...

-Josh

mtakahar · Post by **mtakahar** » Sat Jul 09, 2005 10:41 pm

joshville wrote:It worked as advertised, except that "4" blinked 3 times and "3" blinked four times.

I just realized 03 and 04 are transposed in my test upgrade, so it's working correctly.

I thought everyting VPT'd to Audio would be enough for an HT system, but you can play with the code if it doesn't work as is. You can use the initial image as the old.txt and run the extinstall against it to install the new image.

Hal

joshville · Post by **joshville** » Sun Jul 10, 2005 11:46 am

I don't think this would work because the remote can't know about the device button settings addresses in the extender RDF.

I didn't set 1800 in IR, I used the setup method on the remote to manually change codes.

The default mode on the remote after a 981 is Cable. It is made by Motorola after all.

I tried a merge and then forced IR to use the unextended RDF as you described by changing the signature. It stilll showed Cable/1800 for all the devices. I switched all the Cables to TV, saved, and then changed the signature back to extended on the merged file. Of course, I got checksum errors, but thankfully IR told me what they should have been. After uploading again, my first volume key press gave me the one long and three short blink pattern I'd gotten before, and the remote is in cable mode. I can't get it to switch devices, though. The LED lights with each press of a key that is in the Cable set, but not when I press any device button or a key that is not in the Cable set. It also seems to be picking up the double key press, because when the remote is unextended I can't double-press a key fast enough to keep it from blinking twice, but when extended a fast double press will light the LED once only. I haven't looked into the double press protocol really at all yet to see what it does, but I thought it might be helpful to know that it (seems like it) is picking up. What would my next debug step be? I thought I would try to assign a macro to one of the keys in the cable set and see what happens. I'm sort of shooting in the dark, though. Most of my programming is in VBA or VBScript, so I use message boxes all the time to tell me what is actually going on. No such luxury here!

Thanks!

-Josh

mtakahar · Post by **mtakahar** » Sun Jul 10, 2005 1:25 pm

When I saw a word "normal", I guess I was automatically thinking it's in IR, even though you explicitly said "on the remote". Perhaps this is a symptom of JP1 addiction.

Based on your observation of device selection not working at all, I found a problem. I forgot to change what the 1st device index was.

Code: Select all

	SUB	RC1, _TV	;Is it a command to change a DevIndex?

This should have been:

Code: Select all

	SUB	RC1, _CATV	;Is it a command to change a DevIndex?

But instead of hardcoding it there, I introduced a new macro _1stDev to make what to change during porting explicit:

Code: Select all

	_1stDev = _CATV
...
	SUB	RC1, _1stDev	;Is it a command to change a DevIndex?

I also double checked the key reading code, but it looks ok. The problem above is bad enough to make other things very random.

I've updated the archive:
https://www.hifi-remote.com/forums/dload ... le_id=1903

Hal

joshville · Post by **joshville** » Sun Jul 10, 2005 2:36 pm

I tried the new file, but still no device selection. I tried assigning a macro for the sequence "1-2-3-4" to the "A" button, but it didn't work either.

Also, there is some strange behavior after the upload to the remote. If I press any of the device keys (before pressing a volume key), the lights under the device keys go through their little sequence show the same as they do when you're manually, on the remote (

), changing the device code for that key.

After initializing with a volume key there is also some strange behavior. If I press only volume keys I get an alternating pattern of no LED response and the 1 long 3 short that I get when initializing the extender. Like:

Press [no blinks]
Press [1 long 3 short blinks]
Press [no blinks]
etc.

I can press any number of non-volume keys and I get a normal single blink of the LED, but then if I press a volume key afterward I get the 1 long 3 short pattern again.

I also have a question about the programming. I've been trying to go through the ASM file line by line to try to deconstruct what's going on. I was trying to check the differences between JR, JRNE, etc, but didn't find JRNE in the processor instruction set referenced by Rob in the previous post. Is it in a newer/different version of the processor? or some undocumented function? or am I just way over my head? (most likely scenario)

Thanks!

-Josh

mtakahar · Post by **mtakahar** » Sun Jul 10, 2005 4:46 pm

Also, there is some strange behavior after the upload to the remote. If I press any of the device keys (before pressing a volume key), the lights under the device keys go through their little sequence show the same as they do when you're manually, on the remote ( ), changing the device code for that key.

After initializing with a volume key there is also some strange behavior. If I press only volume keys I get an alternating pattern of no LED response and the 1 long 3 short that I get when initializing the extender. Like:

Press [no blinks]
Press [1 long 3 short blinks]
Press [no blinks]
etc.

I can press any number of non-volume keys and I get a normal single blink of the LED, but then if I press a volume key afterward I get the 1 long 3 short pattern again.

This sounds like the extender is crashing when you press a volume key. Could try this one with unextended?

Code: Select all

Upgrade code 0 = 2c 50 (TV/1104) (RM v1.26)
 fb 00 f1 
 0a 01 02 04 03 05 06 07 08 09 03 04 05 06 07 01
End
Upgrade protocol 0 = 01 fb (S3C80) pause-DCP501R (RM v1.26)
 00 00 01 28 03 8d 14 50
End

Basically similar to the one I posted before except this one has blinks on a few more buttons. vol+ should blink 3 times, vol-=4, mute=5, ch+=6 and ch-=7.
I notice something unusual around the LED code, but I still can't put my finger on it. Have you noticed anything weird or interesting in terms of LED blinking patterns or something else that are key set specific in the unextended state?

It may be also possible that a volume key sets something in the remote before the extender becomes active.

Code: Select all

; ---- Device/Protocol upgrades for loading extender
tv_1800:                        ;22;
        db      80              ;Protocol = 0x180
        db      0               ;No digit map
        db      41              ;21; volume keys defined

If you modify the last line: "db 41" to "db ff, ff, ff, ff, fe", the extender should get activated with a press of almost any key. Could you try activating it with various different types of keys (power, a digit, a channel, something else) and see if it makes a difference?

If neither the LED code nor the volume keys is the calprit, I'll have to look into it further.

I also have a question about the programming. I've been trying to go through the ASM file line by line to try to deconstruct what's going on. I was trying to check the differences between JR, JRNE, etc, but didn't find JRNE in the processor instruction set referenced by Rob in the previous post. Is it in a newer/different version of the processor? or some undocumented function? or am I just way over my head? (most likely scenario)

This part of the Readme of s3c8asm is explaining it.

The ASM syntax is similar to the various S3C8 dissasemblers around
except:

1) The condition code in a conditional JP or JR are part of the opcode,
rather than being an operand.

What this exactly means is, s3c8asm takes "JRNE <label>" instead of "JR NE, <label>". John did it this way perhaps because this is more common notation for assemblers for many other major chips, such as Intel, MIPS, etc. I feel more comfortable with this style, too.

Hal

joshville · Post by **joshville** » Sun Jul 10, 2005 7:03 pm

Dang, I've been looking at too many different files. Now that you mention about the s3c8asm file, I do recall reading that before. It just didn't sink in to the proper portion of my memory bank. Time for me to RTFRM again.

Your blink testing protocol/upgrade worked again as it was supposed to, with the 3 and 4 transposed. Volume, mute, channel all blinked the appropriate number of times.

I changed the line in the 1800 protocol you suggested. I assembled the asm file and then merged the hex file with the beta 4 extender in IR. After upload the behavior is the same as before, except that any key will activate the extender. The volume buttons still do that alternating 4 blink/no response sequence as before. It's still only the volume or mute that does that, though. As far as funny LED behavior, here goes.

There are 9 LEDs, one under each device button (both real and fake). Unextended, the light under the active device blinks with each key press that is part of that device key set or punch thru. I believe there are several keys in addition to volume that are punched through to Audio by default. The device button lights for both the real device buttons and the fake ones.

Code setting procedure on the remote is holding the desired device button down until all of the programmable device buttons light in sequence:

CATV, AM/FM, TV, VCR1, VCR2, CATV, AM/FM, TV, VCR1, VCR2,

The desired button then remains lit until a code is entered. A good code is acknowledged by 3 blinks. There is a slight delay (0.5s?) between the press of the last key and the acknowledgement.

All of the device buttons (real and fake) remain lit when held down until the sequence begins (real only). The exception to this is the TV button, which only blinks on then back off until the sequence begins.

The only other quirk in unextended mode is that the record button only lights the LED on the second press when in VCR mode. I tried that with a couple of setup codes. Pressing another key after the first press of the record button resets it to a non-pressed state (i.e. you still have to press twice for a blink)

Immediately following an extender upload, pressing any real device button except TV gives the light sequence CATV, AM/FM, TV etc. A code may be entered after the sequence without loading the extender. After either code entry or press of a non-numeric key, the next press of a number, volume, or channel loads the extender. The previously hit device key does the 4 blink pattern and all ensuing key presses in the Cable keyset light that button.

I know this is pretty verbose, but I wanted to give you all the information I had.

Thanks!

-Josh

mtakahar · Post by **mtakahar** » Sun Jul 10, 2005 8:10 pm

joshville wrote:... with the 3 and 4 transposed.

Ok, now I'm confident that it wasn't my fault. The ButtonMaps section in all of those RDF's are wrong. ($25 and $26 are transposed.)

Does the extender look like working if you don't touch any volume buttons or buttons that have macros?

The record button requiring two presses in certain modes is a very commonly seen feature in the UEI remotes.

Hal

mtakahar · Post by **mtakahar** » Sun Jul 10, 2005 9:13 pm

I found and fix at least one place that would cause all sorts of problems. Bad news is the fix made the extender chunk size one byte too many to fit into the $01A - $0FF area (so called "old keymove area", which is not going to be used in this remote without the extender.) So, I had to rearrange the code and as a result, the available EEPROM space became a few dozens smaller than the previous one because the code can't be rearranged just arbitrarily. Once this extender starts working, you could reclaim this space for handling a few more key sets or something else if you are looking for more fun. (hint: see the 6131 extender)

https://www.hifi-remote.com/forums/dload ... le_id=1903

I've corrected the buttonmap section in the RDFs, including non-extended ones in this archive.

Hal

joshville · Post by **joshville** » Mon Jul 11, 2005 4:35 am

It looks like that fixed at least some of what was going on... I still get the four blinks with every volume key press. Device selection doesn't seem to be working completely. However, I assigned macro "1;2;3;4" to the "A" button and it worked. Pressing a real device button also illuminates the LED for the device button that was active at the time of extender initialization, after which no buttons will light it again except another device button.

When the extender loads it works my cable box, regardless of the settings I have in the Device Buttons section of the General tab in IR. After a device button press the remote doesn't respond (no LED blinks and no effect on my equipment) until I hit the CATV device button, after which it controls my cable box again. The exception to the previous sentence is when I hit the "A" button with a macro assigned to it. The LED blinks sort of longish (probably due to the macro length). It sort of seems like it's not finding the device settings?

Thank you again!

-Josh

mtakahar · Post by **mtakahar** » Mon Jul 11, 2005 11:55 am

Could you try commenting out all the occurences of calls to LedOn, LedOff and CheckVolKey and if it makes it any better?
If it seems to work, then put back one thing at a time and see where it breaks again.

Hal

Post by **The Robman** » Mon Jul 11, 2005 1:39 pm

Thanks for catching the RDF bug Hal, I have updated the 3 RDFs that are in the RDF folder.

Nils_Ekberg · Post by **Nils_Ekberg** » Mon Jul 11, 2005 2:03 pm

The Robman wrote:Thanks for catching the RDF bug Hal, I have updated the 3 RDFs that are in the RDF folder.

I also captured them for the next RDF release with the right names