IR: replace data message during download

[ElizabethD]

Replay remote. Keymoves at $1C. One sticker doesn't match RDF specs.

#1. When IR issues a message that the checksum or signature doesn't match the RDF and when it asks whether to replace - WHEN is the replacement taking place? At which point is the RawData not longer an image?

#2. When RDF name contains 7-character signature, does IR make up the 8th character when you download or is the 8th character in RawData signature exactly what's in the remote? What's the impact of Yes or No answer to #1? The value seems to be OEM3OEM0 <-- $30 zero, not binary. But I don't know what I answered to #1.

#3. In IR there is no macros tab. There are no macros in RawData that I can see.

#4. There are 12-14 keymoves from $7C setup code and one other, most on transport keys, 2 sets. Is the keymoves tab missing because I've used RDF for unknown Replay rather than v1 or v2? Keymove tab does reappear when v1 or v2 RDF is selected. It looks better with v1 than v2 but still not the same as what Raw data says.

Just trying to understand how downloads work. And how this group ever figured out sooooooo many remotes.

Unrelated to the download question: there are many, mostly Panasonic, devices and protocols plus 2 stray bytes in odd places - does this indicate a jp1-used remote? Kinda strange, because it still has the little plastic protector over the IR part.

[The Robman]

Replay remote. Keymoves at $1C. One sticker doesn't match RDF specs.

So, what do the stickers say?

#1. When IR issues a message that the checksum or signature doesn't match the RDF and when it asks whether to replace - WHEN is the replacement taking place? At which point is the RawData not longer an image?

Any time that IR selects an RDF you can't be 100% sure that the data in the Raw Data panel is a true reflection of what's in the remote. If you want to be 100% sure that the data is what's in the remote, do a RAW DOWNLOAD.

#2. When RDF name contains 7-character signature, does IR make up the 8th character when you download or is the 8th character in RawData signature exactly what's in the remote? What's the impact of Yes or No answer to #1? The value seems to be OEM3OEM0 <-- $30 zero, not binary. But I don't know what I answered to #1.

Whenever we use an incomplete sig in the RDF name, the missing trailing positions are treated as wild characters. So "OEM3OEM" would match 'OEM3OEM0", "OEM3OEM1" and "OEM3OEM2".

Just FYI, all versions of this remote natively use the "OEM3OEM0" sig, the "1" and "2" sigs were created by me. If you initially select the "unknown version" RDF and then select the correct version of the remote using the dropdown, IR will replace the 8th byte of the sig with the value you selected. If you select the correct RDF during the initial download, it will also fix the 8th byte of the sig for you.

#3. In IR there is no macros tab. There are no macros in RawData that I can see.

That's because this remote doesn't support macros!

#4. There are 12-14 keymoves from $7C setup code and one other, most on transport keys, 2 sets. Is the keymoves tab missing because I've used RDF for unknown Replay rather than v1 or v2? Keymove tab does reappear when v1 or v2 RDF is selected. It looks better with v1 than v2 but still not the same as what Raw data says.

There's no Keymoves tab when you select the "Unknown Version" RDF, because IR doesn't know where the keymove section starts yet. Once you have selected the correct RDF, you will be able to see the keymoves.

Just trying to understand how downloads work. And how this group ever figured out sooooooo many remotes.

We've been doing this since late 2000, so that's over 5 years of work already, and the Replay remote was one of the first 3 remotes we decoded (the others being the 15-1994 and the URC-7800).

Unrelated to the download question: there are many, mostly Panasonic, devices and protocols plus 2 stray bytes in odd places - does this indicate a jp1-used remote? Kinda strange, because it still has the little plastic protector over the IR part.

This is a factory image that you are looking at. The keymoves are there to fix mistakes in the upgrade that was added for ReplayTV, and to populate the ReplayTV-specific buttons in all the other device modes.

UEI made a small change to the $00C9 and $001F executors before releasing this remote, and they wanted the changes included in the remote. Due to the fact that built-in codes never look at the EEPROM for protocol upgrades, in order to make the built-in Panasonic codes use the updated executors, they had to load new copies of all the Panasonic codes.

[ElizabethD]

So, what do the stickers say?

Sorry, I didn't have'm at work. One sticker inside matches RDF v1: 3401B02. One sticker doesn't match v1 nor v2: G042304.
Batteries are nearly impossible to get out. Tight springs in there.

If you want to be 100% sure that the data is what's in the remote, do a RAW DOWNLOAD.

I just found and did RAW DOWNLOAD. I wish I knew about it before.

Just FYI, all versions of this remote natively use the "OEM3OEM0" sig, the "1" and "2" sigs were created by me. If you initially select the "unknown version" RDF and then select the correct version of the remote using the dropdown, IR will replace the 8th byte of the sig with the value you selected. If you select the correct RDF during the initial download, it will also fix the 8th byte of the sig for you.

That's the key to my puzzlement all along which triggered this thread. Ok, so the signature is OEM3OEM0, as displayed, and if I were to use the remote, v1 RDF is it.

All the rest is clear now. Thanks so much for clarifications

Just couple more followup things, please:
1. The 2 stray bytes in a section filled with $FF: at $007F is $0F, and at $0096 is $11. What do they use it for?
2. I tried asking for checksum calculation. Cool tool, but IR said "invalid class typecast". In English?
3. I tried filling the whole thing with FF, but IR said, "priviledged instruction". Windows issue? Rights issue?

[The Robman]

One sticker inside matches RDF v1: 3401B02.

That's the sticker you want to look at, you have a version 1 remote.

One sticker doesn't match v1 nor v2: G042304.

That sticker is in every UEI remote, including the "off the shelf" OFA remotes, I think it's a batch number.

Batteries are nearly impossible to get out. Tight springs in there.

Oh yeah, they're a real bitch!

1. The 2 stray bytes in a section filled with $FF: at $007F is $0F, and at $0096 is $11. What do they use it for?

Nothing, that sounds like spurious data.

2. I tried asking for checksum calculation. Cool tool, but IR said "invalid class typecast". In English?

Never seen that one, what did you do to make it happen?

3. I tried filling the whole thing with FF, but IR said, "priviledged instruction". Windows issue? Rights issue?

Dunno. If you can use IR to load to the remote, you should be able to init to $FF.

[whompus]

While playing with those replay remotes, Take a look at the video code in them 0059. It has many pioneer service remote functions in it. Makes me wonder if that's the reason the code 0059 is not in the newer uei remotes.

[ElizabethD]

2. I tried asking for checksum calculation. Cool tool, but IR said "invalid class typecast". In English?

Never seen that one, what did you do to make it happen?

Don't waste any time on it - no point, it's totally harmless. I don't know what made it happen. But for the record, here are two ways to reproduce it:
1. Get the checksum calculator window and click where it makes no sense to click - in the empty right box.
Or, follow this to watch a focus delay as well as the subject message:
2. Type in a range to use, then range to exclude, click Add, the button gets focus and the range goes up
> Click on the entry in that upper text area to select it. Now the error hits. Read it, click OK
> Note that the entry is now selected
> Click Delete - nothing happens and Delete has no focus
> Click it louder, the entry goes away, and Delete gets focus, in that order
> Checksum gets recalculated, I think correctly. Fabulous little tool

3. I tried filling the whole thing with FF, but IR said, "priviledged instruction". Windows issue? Rights issue?

Dunno. If you can use IR to load to the remote, you should be able to init to $FF.

Normally it's not a problem. Rough to reproduce. Just happens every once in a while and I don't see a pattern. It might help to know whether it's priviledged on the remote side or the Windows side.

Take a look at the video code in them 0059

I suppose that's some built-in protocol or another upgrade of the sort Rob described? All that's visible here is Panasonic, Panasonic VCR combo and Dishplayer. Or something with setup code 0124 if I select another RDF. How do you take a look?

[The Robman]

I think he's talking about the built in LDP/0059 code (which is a VCR code in the Replay remote).

I'm guessing that it's not in the more recent remotes because they probably figured it's time to start dropping the laserdisc codes, right along with the cassette tape codes. I doubt that the fact that laser disc codes can be used as service codes for Pioneer gear even factored into it.

[whompus]

Yeah... That's the code I was talking about.

Sorry to have swayed off the topic of the post. That code just intrigued me with the lack of it in the newer remotes.