Page 1 of 1

RDF definition for memory?

Posted: Wed Apr 05, 2006 4:09 am
by xgamer123
I did some search in this great site, but failed in RDF definition for memory. Hope experts can make a document to describe more detail about definition in RDF.

There is no any problem when I use regular RDF to make my own device upgrades/KeyMoves/Macros. The problem that I need help is I make some changes of memory locate definition like below:

[General]
Name=URC-881x/801x/601x MoreKey <-... My own name
EepromSize=$800
FavKey=$30, $022, 15, 5
AdvCodeAddr=$024..$1FF <-... Need more key moves space
UpgradeAddr=$200..$3FF <-... don't need many Upgrades
LearnedAddr=$400..$7FD

When I run IR with new RDF file. It looks everything OK in raw data area. I did check pointer after I put some device upgrades and key moves. It all looks good. But TV setting in general code change back to 0047 after upload to and down from OFA 8810. This happens on devices upgrades. It is not allow me to change to devices upgrades code, but all data remain in raw data. What and where I should do it? Where I can download a good and fully describe about RDF definition?

Posted: Wed Apr 05, 2006 4:27 am
by Mark Pierson
You can't increase or change EEPROM space by simply editing the RDF. You'll need to look into using the extender if you need more key move space.

New Ext RDF?

Posted: Wed Apr 05, 2006 10:04 pm
by xgamer123
Thanks for your quick reply.

I've read extInstall readme and 6_80_XxX ext RDF files. I need to keep learning code support to parse some unknown IR, and don't need large space for fancy upgrades. Can you experts make a new 2K 6_806_80 (URC-881x_801x_601x family) ext RDF file that allow user change EEPROM space and still keep original functions?

Posted: Wed Apr 05, 2006 10:15 pm
by The Robman
You need to understand that the RDFs don't actually "do" anything, they merely "reflect" what already exists. When we create an RDF, we don't arbitrarily decide how much memory to give each of the sections, we determine how much memory each of the sections has been given in the remote's firmware (which is in the un-changeable ROM).

Even with the extenders, it's not the RDF that dictates how much memory each section will get, it's the actual assembler code that the extender is written in that decides that, the RDF just reflects those limits. So what you're really asking for is for someone to write a brand new extender that supports learning and uses different sizes for some of the sections. That's an awful lot of work that you're asking for, so it's very unlikely to happen.

A much better approach would be to post an IR file that contains all the learned signals that you have been unable to convert into upgrades so that one of us can have a crack at writing an executor for them. Then you'll be able to use the extender.

Why do you think you need so much keymove memory anyway?

Posted: Wed Apr 05, 2006 10:55 pm
by xgamer123
Thanks Robman,

I'm sorry for this topic involve such things. Because new model of remote controllers are not typically - some bottoms with new functions like color bottoms. I assign these keys to VCR/DVD... control set that cause lot of key-moves. And I put some key press sequences in groups to change A/V sets that consume another macros space. This is the reason which I'm trying to get more key space.

Posted: Thu Apr 06, 2006 6:10 am
by The Robman
If you are using keymoves to program buttons like the TV/VCR button, you should be creating upgrades instead. There's plenty of upgrade memory but limited keymove/macro memory, as you've discovered.