Im trying to import a CCF file into the 850 universal browser. (Discrete power for a 942 dish Netowrk) I have the CCF file saved on my drive. I open universal browser, and browse to the ccf file. The same panel comes up that was dsplayed in Pront Edit. If i highlight the desired button in univeraal browser, "learned data" appears under "Action list" If I drag the "learned data" box OR the button itself to the desired location on hte 850, it appears to work, as the button adds the "L" for "learned"

Problem is the commands dont work after downloaded. Adress code is 1, and I used the "1" file from Pronto. The hex codes start w 0, not 5, 6, etc as mentioned elsewhere.

Power Off

0000 0048 0001 0011 0017 0163 0017 0061 0017 00a3 0017 00a3 0017 0061 0017 0061 0017 0061 0017 00a3 0017 00a3 0017 00a3 0017 00a3 0017 00a3 0017 00a3 0017 00a3 0017 00a3 0017 00a3 0017 00a3 0017 0163



Power On

0000 0048 0001 0011 0017 0163 0017 00a3 0017 00a3 0017 00a3 0017 00a3 0017 00a3 0017 0061 0017 00a3 0017 00a3 0017 00a3 0017 00a3 0017 00a3 0017 00a3 0017 00a3 0017 00a3 0017 00a3 0017 00a3 0017 0163

What am I doing wrong? Is there anyway to CONFIRM the DATA is actually on the button Im dragging, and that Im not just dragging the button TEXT?


Frank

www.customdigitalinc.com

Most likely you used the universal browser correctly and really have that learned data on the desired mx850 buttons.

That would mean those commands don't do the action you expect on your 942.

I don't have a good guess of whether that means your 942 is set to a different address, or doesn't support discrete power commands, or uses another code set entirely.

The Pronto Hex you posted has a zero as its address part. The dish info I have from the JP1 groups says that would be correct: The address part of the signal is one less than the user visible address, so it should be a zero for address one. But maybe the source you have for saying "Address code is 1" doesn't obey that rule (actually encodes a 1 in the signal, which would be address 2 by the usual rules).

Do you have an mxd file for any commands that do work with your 942? I know enough about manually decoding mxd files that I could look at it and determine whether it is using the same code set and if so, what the address part is.

I have the current 850 file that I am using, but the dicrete power commands that are on RC under the dish net files dont work (at least on the 942)

I DID try the CCF file on a 311 I have in the showroom as well. The command didnt work for the 311 either.

As far as my "source" for the remote adress, it is from the system info screen.

I will mail you a copy of the 850 file to see if thats what your looking for.

Thanks

BTT.

Anyone have any other ideas?

I never got any such email.

Just realized I forgot the period between your first and last name. I just re-sent it. Let me know if it still didnt come through.

I just investigated and emailed back a CCF file. The results were quite strange (items 1 and 3 below are not part of that strange result, they're just about my incompetence in the MX-850 software):

1) I don't know the MX-850 editor software well enough for some basic operations (that I thought I knew how to do until I tried this). I wanted an MXD file with just the relevent signals (the power toggle that works and the power discretes that don't). I received an MXF file with many devices including the DISH device with many functions, including the three I wanted. I should have been able to construct a new device with just what I wanted to export to mxd, but I got lost every time I tried to edit anything in this mxf file.

2) Fortunately, someone had loaned me an MX-850 (fairly long term) and I have a very good IR capture and decode system, so I loaded that mxf into the MX-850 and tested those three buttons.

The power toggle gave a bunch of IR garbage with a correct Dish_Network signal buried in the middle.

The two discretes also gave a bunch of IR garbage with a Dish_Network signal in the middle and those signals were commands 1 and 39, just like the discretes in the attached (same email) CCF file, BUT the signals and surrounding garbage were all one third faster than correct Dish_Network signals. In the CCF they were correct speed and the learned power toggle was correct speed.

3) I tried using the Universal browser myself to redrag those signals from the CCF. I put them in reversed positions as a detector for exactly the sort of failure I then got (but still don't understand) and downloaded to the remote and nothing had changed. The commands weren't even reversed. In a bunch more tests, everything I did looked like it was saving its effects in the PC image of the MX-850 and looked like downloads worked, but nothing changed what the MX-850 actually sent.

4) Starting with an empty configuration, I had no trouble using Universal browser to load the two signals from the CCF onto the desired buttons and exactly duplicate that one third speedup.

5) I took the appropriate .irp file from my MakeHex program and edited it to produce Dish_Network signals that were 50% slower than correct signals (1.5 times correct duration). I used the Universal Browser to drag those into my clean MX-850 config and downloaded and tested and got something that looked structurally just like the learned power toggle: IR garbage at the speed of a correct Dish_network signal with a correct (speed and content) signal in the middle of it. So as I hoped these signals that I slowed down by 50% were sped up by one third (of the new duration) getting back to correct timing.

I emailed that CCF file.

6) I tried the Universal Browser on some other Protocols and had no similar problems. It neither changed the speed of the signals from what the Pronto Hex specified, nor surrounded the specified signal with IR garbage. I don't know what is special about these Dish signals that makes the Universal Browser malfunction this way. I hope the slowed signals give correct results. To my IR capture and decode system those signals look just as good as the learned IR toggle. But I don't have any actual Dish hardware. How the signal looks to Dish hardware is what actually matters.

Wow! So much above my head! (slowing down speeds, extra ir garbage, etc)

Thanks a bunch for the time spent. I certainly didnt expect you to go to all that trouble.

Ill check them out when I get back in the office next week,

Thanks again

I sure am glad I came across this thread. I was going crazy with these dish codes. I was using makehex to generate my codes (nice program John), but they wouldn't work. I used my slink-e with the ez-learn program and noticed the same thing John noticed, that the signal was about a 1/3 too fast. Well I exported the faulty dish codes to an mxd file and looked at it in a hex editor. It turns out the the ir format in an mxd file is essentially the same as the format for an mx500 which is fairly well documented over at the old irclone site. To summerize, the specified frequency was correct, but the 4 timebase parameters (400 us, 1600 us, 2800 us and 6000 us) were all off by 1/3. Bumping the time base in the irp file for makehex from 400 to 600 fixes the problem.

This is a bizarre bug in the universal browser. It makes me wonder what other devices are effected. Perhaps I will post a new thread in an effort to get URC to look at this.

I have a pioneer receiver that may be affected as well. I use the codes database in the mx3000 software for most functions, but there are a few that are unavailable (Surround: Channel Select, Adjust up & Adjust Down).

I've tried learning the code several different ways which works, as per the editor, successfully but does not work with the receiver.

I own a pronto 1000 that I used for years pre-mx3000. I learned and tested the codes with the pronto (IE; the pronto controls the receiver properly with learned codes) but importing the codes through the universal browser leads tho the same end as above. The import goes fine through the browser but the codes don't work.

I've never got into the codes before but it looks like I may have to.

How are you looking at the codes that the mx editor creates after import.?

If I need utility programs, where can I get them?

Is there a source for easy to understand explanations of the the codes, codes structure etc?

Thanks
Lane Bailey

I don't think those could be affected by the same bug. But it might be a different bug.


Pioneer has CCF files on their web site. They used generated commands in those CCF files (not learned commands), so they should be imune to all the variety of effect you might see with learned signals.

Try using the Universal browser on their CCF file. I think the universal browser can read perfect Pioneer signals.

I have some guesses about what sort of imperfect learns of Pioneer signals might work in a Pronto but fail in the Universal browser. I'd rather not speculate blindly about that. If you want to give me the ccf file with your learned signals I can check directly. It's better if you can put it up in some web space and post a URL, but emailing the file is OK if you don't have any web space under your control.


Save to an mxd file, then open in a hex editor, and use general knowledge about IR signals to find the understandable IR signal within the largely non understandable (by us) contents of the mxd file.