| View previous topic :: View next topic |
| Author |
Message |
jon_armstrong
Expert
Joined: 03 Aug 2003
Posts: 1238
Location: R.I.P. 3/25/2005
|
 Posted: Thu Sep 09, 2004 8:47 pm Post subject: NEC 2DEV Combo protocols.ini entry PID 01 1A |
 |
|
At the end of the second page of this thread I posted a protocols.ini entry that fixes the NEC 2dev combo. It also has a description of what bits call certain features if Mark or Mike are going to tackle KM Master that is also broken for this protocol. I tested it with RM and it all seems to now work as expected.
It is a good protocol because you can mix 2 deivce/sub-device combinations and call NEC1, NEC2, NECx1, or NECx2 style for each command.
Edit: I finally found the old protocols ini entry for this protocol in RM which will explain my previous confusion and edited this message to make more sense.
_________________
-Jon |
|
| Back to top |
|
 |
garypen
Joined: 03 Apr 2004
Posts: 145
|
| Posted: Fri Sep 10, 2004 11:22 am Post subject: |
|
|
I pasted the text into the appropriate place in the protocol.ini file. However, I don't think it fixes the protocol. It keeps changing the entries that I make on the functions page. For example:
When I enter the num0 EFC of 164, the device column automatically changes to 110 (device2), and protocol NECx2. When I correct it to 45, it changes the EFC to 168. If I try and correct the protocol to NECx1, it changes the EFC to 172 and device 110. When I change the device back to 45, it changes the EFC to 176!
All during this goofiness, the OBC does not change at all. Isn't that supposed to automatically change based on some formula? |
|
| Back to top |
|
|
johnsfine
Site Admin
Joined: 10 Aug 2003
Posts: 4766
Location: Bedford, MA
|
| Posted: Fri Sep 10, 2004 12:16 pm Post subject: |
|
|
| I don't know why Jon (apparently) changed the CmdIndex from 0 to 1. I'm pretty sure 0 was correct. If the CmdIndex is wrong then the EFC will be garbage both for input and output. You could still make it work by entering commands via the OBC and other fields and ignoring the EFC. But it's better to fix protocols.ini to put the CmdIndex back to 0. |
|
| Back to top |
|
|
garypen
Joined: 03 Apr 2004
Posts: 145
|
| Posted: Fri Sep 10, 2004 12:53 pm Post subject: |
|
|
I changed it to 0. That worked in terms of correcting the strange changing of the values. However, I think the hex code is now incorrect.
When I enter 145, 110, NEC1, I get "70 a0". Shouldn't that be "70 80"? (I've already proven that 80 works as NEC1, 110 by trying it with my device.)
Also, when I enter 164, 45, and NECx1 as EFC, device, and protocol, I get "dd 30" as the hex code.
I will take your guys word that "30" is correct for NECx1, and "24" was NEC1, but accepted by my device anyway. (Although, I don't understand how 24 could be NEC1, if 80 is NEC1, unless 24 is NEC1, 45, 45, and 80 is NEC1, 110.) I won't really know until I try it. |
|
| Back to top |
|
|
johnsfine
Site Admin
Joined: 10 Aug 2003
Posts: 4766
Location: Bedford, MA
|
| Posted: Fri Sep 10, 2004 1:11 pm Post subject: |
|
|
| garypen wrote: | I changed it to 0. That worked in terms of correcting the strange changing of the values. However, I think the hex code is now incorrect.
When I enter 145, 110, NEC1, I get "70 a0". Shouldn't that be "70 80"? |
"70 80" was sort of an error (my error that Jon caught), but an error that didn't matter. "70 A0" is correct.
"70 A0" tells the remote to make an NEC1 signal using the second device.subdevice pair from the setup sheet.
"70 80" tells the remote to make an NEC1 signal using the second device from the setup sheet, but ignore the subdevice from the setup sheet and force the signal to have no subdevice.
Since the second subdevice on the setup sheet is blank, there is no difference in the actual IR generated between using that and not using it. |
|
| Back to top |
|
|
garypen
Joined: 03 Apr 2004
Posts: 145
|
| Posted: Fri Sep 10, 2004 1:18 pm Post subject: |
|
|
I see. Then, it sounds like it will work!
BTW, in addition to changing CmdIndex back to "0", I also got rid of the "*" in the protocol name. It now reads "NEC 2DEV Combo", instead of "NEC 2DEV Combo (*)" to better match the protocol name in KM, reducing error when opening a KM file. |
|
| Back to top |
|
|
jon_armstrong
Expert
Joined: 03 Aug 2003
Posts: 1238
Location: R.I.P. 3/25/2005
|
| Posted: Fri Sep 10, 2004 1:32 pm Post subject: |
|
|
| johnsfine wrote: | | I don't know why Jon (apparently) changed the CmdIndex from 0 to 1. I'm pretty sure 0 was correct. If the CmdIndex is wrong then the EFC will be garbage both for input and output. You could still make it work by entering commands via the OBC and other fields and ignoring the EFC. But it's better to fix protocols.ini to put the CmdIndex back to 0. |
I started out with the wrong protocols.ini entry to begin with and modified it. I used the NEC1 2DEV Combo (since NEC 2DEV Combo[*] was before NEC1 and all the other NEC combo protocols were at the end of the NEC section.) It seemed to work (and I didn't even think about EFC's) I'm gaining competence with protocols.ini, but I'll confess until now, I didn't know what CmdIndex did. Now I understand that it tells RM which byte of variable data maps to the EFC and in this case that's the first byte. If omitted does it default to 0?
BTW, I think the NEC1 2DEV Combo is broken(PID $0141). The protocols.ini entry calls for 6 bytes of fixed data and I'm pretty sure that there are only two. The command translator is lsb and not lsb,comp etc. Needless to say when I first started testing, I became very confused. But what else is new ...
_________________
-Jon |
|
| Back to top |
|
|
The Robman
Site Owner
Joined: 01 Aug 2003
Posts: 21234
Location: Chicago, IL
|
| Posted: Fri Sep 10, 2004 1:51 pm Post subject: |
|
|
| jon_armstrong wrote: | | BTW, I think the NEC1 2DEV Combo is broken(PID $0141). |
I'm pretty sure that this is a protocol that I wrote and I'm not even sure that it works. Regardless, whenever an official protocol is released that matches a hacked one that we were using, we should modify our tools to support the official version and drop support of the old hacked version. That way, when the user has a remote that has the protocols in the ROM, they won't need a protocol upgrade, whereas a hacked protocol will always require an upgrade.
_________________
Rob
www.hifi-remote.com
Please don't PM me with remote questions, post them in the forums so all the experts can help! |
|
| Back to top |
|
|
jon_armstrong
Expert
Joined: 03 Aug 2003
Posts: 1238
Location: R.I.P. 3/25/2005
|
| Posted: Fri Sep 10, 2004 2:03 pm Post subject: |
|
|
I didn't find PID $0141 in KM so it apparently no longer supports it. I am all for eliminating it in RM. Of course, Senator Fine and President Bush are in charge of RM and what I think, occasionally fails to persuade them 
_________________
-Jon |
|
| Back to top |
|
|
johnsfine
Site Admin
Joined: 10 Aug 2003
Posts: 4766
Location: Bedford, MA
|
| Posted: Fri Sep 10, 2004 2:28 pm Post subject: |
|
|
| jon_armstrong wrote: | | I didn't find PID $0141 in KM so it apparently no longer supports it. I am all for eliminating it in RM. Of course, Senator Fine and President Bush are in charge of RM and what I think, occasionally fails to persuade them |
I'm OK with your suggestion for RM. I tend not to take insult easily, so I'll let that apparent comparison go. This is not a political discussion area, so I expect I've already shown my political views too much, but as an extra hint: I live in Massachusets, but I'm NOT going to vote for the candidate that all the dead voters (and most likely the majority of live ones) will vote for. |
|
| Back to top |
|
|
The Robman
Site Owner
Joined: 01 Aug 2003
Posts: 21234
Location: Chicago, IL
|
| Posted: Fri Sep 10, 2004 4:14 pm Post subject: |
|
|
| johnsfine wrote: | | [...the candidate that all the dead voters (and most likely the majority of live ones) will vote for. |
We can only hope!!! ABB
_________________
Rob
www.hifi-remote.com
Please don't PM me with remote questions, post them in the forums so all the experts can help! |
|
| Back to top |
|
|
garypen
Joined: 03 Apr 2004
Posts: 145
|
| Posted: Fri Sep 10, 2004 4:33 pm Post subject: |
|
|
| Well. We all know which one the brain-dead voters will be choosing. |
|
| Back to top |
|
|
The Robman
Site Owner
Joined: 01 Aug 2003
Posts: 21234
Location: Chicago, IL
|
| Posted: Fri Sep 10, 2004 6:43 pm Post subject: |
|
|
| garypen wrote: | | Well. We all know which one the brain-dead voters will be choosing. |
You'll get nowhere around here by insulting John like that!!! 
_________________
Rob
www.hifi-remote.com
Please don't PM me with remote questions, post them in the forums so all the experts can help! |
|
| Back to top |
|
|
johnsfine
Site Admin
Joined: 10 Aug 2003
Posts: 4766
Location: Bedford, MA
|
| Posted: Fri Sep 10, 2004 7:02 pm Post subject: |
|
|
| garypen wrote: | | Well. We all know which one the brain-dead voters will be choosing. |
I didn't know Buchanan was running again  If you're stupid enough to WANT to vote for Kerry you might be too stupid to find his name on the ballot.
(I'll try to stop now). |
|
| Back to top |
|
|
The Robman
Site Owner
Joined: 01 Aug 2003
Posts: 21234
Location: Chicago, IL
|
| Posted: Fri Sep 10, 2004 7:14 pm Post subject: |
|
|
| johnsfine wrote: | | If you're stupid enough to WANT to vote for Kerry you might be too stupid to find his name on the ballot. |
And to all those people who were too stupid to find Gore's name on the ballot last time around (and gave the nation "W"), I curse you to not 1, not 2, but THREE hurricanes!!!
_________________
Rob
www.hifi-remote.com
Please don't PM me with remote questions, post them in the forums so all the experts can help! |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
File Section
Wiki
