Until I get my PC working again, I'm just using the on-screen programming.

So I added a few devices to my remote, using pre-programmed manufacturer codes as listed in the manual.

But many of the LCD buttons just don't match up at all! For example, on my Sony TV, the button for "Closed Captioning" toggles surround/normal effects. On my Sony CD player, the transport buttons don't work, "7" selects track 6, and "9" selects track 7. And for the TiVo (I have a Philips), the assignments seem just as random.

What's going on? Did I do something really stupid?

I came across that as well. I experienced a similar fate on the SL-9000. It seems all hard buttons are labeled, at start, exactly the same. Changing a device code only changes the preprogrammed signals, not the label... and even then some seem out of sync. Best solution? Re-learn or move the keys.

THAT makes the preprogrammed signals pretty useless.

And if you move the keys, don't you lose the preprogrammed signals? Or is that after you move them, reassigning that device doesn't update moved buttons?

Somebody really didn't think out their use of preprogrammed codes. Especially since even the transport controls on my Sony CD player don't even match up to the manual! (FF skips forward a track and ChUp does nothing...)

Oh well. Time to work on a Pronto -> LDT converter...

I'm not sure how well such a converter would work... it would seem that learned buttons are assigined to a specific coordinate on the MX-1000's screen?

Daniel, I do not think the learned buttons are assigned to a specific coordinate.
A new button, or a default one has a number assigned to it. The DEFAULT for this number is depending on the coordinate. I believe the learned program is assigned
to this number.

So if you learn a button, and then move it, the learned program will still be assigned to it.

You seem to be right there... I learned several buttons on one config, then downloaded another with completely different button layouts. The learned signals seemed to stick -- however I'm still wondering if that's due to what their original "preprogrammed" function was... still have to test further, but not right now! :-)

Hi,

I tested some more with button assignments.
Whenever a button is created it is assigned a "button number". This initial number is a value from 1 to 36.
The value is initialy a grid position, where upper left corner is 1 and lower right is 36 (the previous
page button).

If you copy a button, the same "button number" is assigned. So if you learn something to one of those, the other will get the same learned command.

If you copy a button to another page in the same device, the button value will be the Button number + 40*page number. So button 127 on page one is the same as button 7 on page 3.

If you move a button the value stays the same.

So the easiest way to create a conflict, is to create
a button somewhere in the grid, move it to another location, and create a new button in the same place as the first. They now have the same "button number", and
learning one of them will also learn the other.

The preprogrammed buttons mostly have values 7 9 11 13 .... (the 3*4 button layout). It seems that
the preprogrammed buttons are connected to odd and even
pair "button numbers". So button 7 and 8 have the same preprogrammed values. I do not know yet if this is the same for learned commands, but I will test it.

See my post in "MX-1000 and Discrete codes" for more info on the LDT files. Making a Pronto to LDT converter wouldn't be too hard, but would be most useful if you had a complete tool that mimics the current PC design tool, but also lets you directly modify the device assignments, etc. For example Pronto code XY should be assigned to SAT, Page 1, Button 3 (C6133).

At a minimum, you probably could have a tool that took the Pronto files, ask you which Pronto button should go to which MX1000 button,assign it to that button, and then translate the IR CODE. I took a quick look at some pronto information and didn't see an easy mapping betwen pronto and MX1000 IR codes.

I have to figure out if there is any special considerations for addressing that has to be taken into consideration. I suspect there may also be either a checksum or identifier for the length of the IR commands, because there are two sets of 4 bytes that I cannot identify their purpose yet.

Rene,

Did you get my spreadsheet? Did you see anything that was inaccurate or do you have more info?

After some more tests, it seems that learned commands are also connected to the odd-even pair of button numbers.
Also I was unable to use numbers 1-6.

Jason, Yes I did, but I must confess I didn't really
looked at it (yet).
I also found out (but you probably already know) that text in table BUTTONTEXTTB will only show on the remote
iff the bx1 and by1 are exactly the same as corresponding fields x1 and y1 in BUTTONTB.
(yes I found out the hard way, changing x and y in
BUTTONTB to align my buttons ==> no more texts on remote (in PC software they did show !!!).