This may fall in the “You ain’t gonna believe this!” category, but it IS what it is. I hope someone can shed some light on this.
I have had quite some difficulty getting my MOXI to respond correctly using TouchSquid with my new WF2IR device. The problem manifests itself by frequently duplicating DIGIT keys. Testing shows this to be frequent, but unpredictable. On average it happens probably 75% of the time. I’ll type ‘112’ and get ‘1112’ or ‘11112’ or ‘1122’ response from the Moxi. Makes it very difficult to use the TouchSquid app.
In my effort to resolve this I eventually turned to the iLearn utility that came with the WF2IR unit. After futzing (very descriptive word for my test procedure) around for a couple of hours I stumbled upon what appears to be the problem. If I can describe it sufficiently, perhaps you can make better sense of it. Let me describe as best as I can the most trivial case I found:
1st I simply setup iLearn to capture a test key from my Moxi remote. I chose to test the ‘1’ digit key. I set iLearn Connector to 3 and left other settings alone. The iLearn program captured the ‘1’ digit key as:
sendir,1:3,1,36443,1,1,
98,32,17,16,17,15,17,32,17,32,49,33,16,16,17,16,17,15,17,16,17,15,17,16,17,16,16,16,33,16,17,16,16,32,17,16,33,16,17,32,17,15,17,16,33,32,17,16,17,15,17,16,17,15,17,16,17,16,17,15,17,16,17,15,33,2510,
98,32,17,16,16,16,17,32,17,32,49,32,17,16,17,16,16,16,17,16,17,15,17,16,17,16,16,16,33,16,17,16,16,32,17,16,33,16,17,32,17,15,17,16,33,32,17,16,16,16,17,16,17,15,17,16,17,16,16,16,17,16,17,15,33,2510,
98,32,17,16,17,16,16,32,17,32,49,33,16,16,17,16,17,15,17,16,17,15,17,16,17,16,16,16,33,16,17,16,16,33,16,16,33,16,17,32,17,16,16,16,33,32,17,16,17,15,17,16,17,16,16,16,17,16,17,15,17,16,17,15,34,4664
Now it gets interesting. Using the above captured code I get the same frequently duplicated keys (e.g., I press to test the IR data once and I get either ‘1’ or ‘11’. If I press it twice I often get ‘111’ and sometimes ‘1111’. This was a good thing as it tells me that the problem is probably not rooted in TouchSquid. But rather something else. But what? So I experimented with the iLearn settings at hand…
I randomly changed Repeat, Offset, Gaps, Gap (pulses), and Separation (ms). As I was testing these settings, the ‘captured data’ showing in the window inexplicably changed. What do I mean? Well the code listed above, suddenly, and without explanation became:
sendir,1:3,1,36443,1,1,
98,32,17,16,17,15,17,32,17,32,49,33,16,16,17,16,17,15,17,16,17,15,17,16,17,16,16,16,33,16,17,16,16,32,17,16,33,16,17,32,17,15,17,16,33,32,17,16,17,15,17,16,17,15,17,16,17,16,17,15,17,16,17,15,33,3644
It did this on its own. And I soon found that this ‘change’ was repeatable in iLearn. It usually occurred right after I changed the Gaps setting to something other than 1 and again pressed the learned IR test key. And I will stress that pressing the TRIM button in iLearn did not do this. In fact I tried the TRIM button multiple times hoping to get this change but it did not work. Only by changing the Gaps (or perhaps other settings?) did it change the displayed 'captured IR code'.
NOW THE MOST INTERESTING PART. With the 'changed and abbreviated' IR code, the Moxi responds perfectly. In fact I changed a variety of settings (including returning the Gaps to 1) and the Moxi continued to respond correctly by always displaying a ‘1’ when I sent the shorter learned IR code. If I sent it three times, I would get ‘111’ just as it should.
So what does this all mean? As you can see the end data grouping value changed between the longer code and the shortened code. It did this w/o recapturing anything from the remote. I think this clue holds the key to making my Moxi work with TouchSquid. Just not sure how to most easily implement this into the TouchSquid. I hope I don't have to go through such a complicated 'rain dance' with every button on my remote to get everything working. I will if I have to - but surely there is an easier solution.
If you have some helpful input I'd love to hear it. I love the TouchSquid gui and its flexible, user definable setup. But the duplicating key entries are making it impossible to use just yet. I will get it working. Just need some help understanding what is going on.
What I _think_ I need is some way to capture these codes (maybe using iLearn if necessary) and then capture the ‘shortened’ version and put them into a TouchSquid compatible format for use. Beyond the iLearn process I have no clue yet on how to migrate these into a TouchSquid usable format, nor how to import them into TouchSquid. But I am hopeful it isn’t too terribly complex.
Can someone with Global Cache explain the 'change' that occurs in the displayed IR Capture data of the iLearn screen? It isn't anything as simple as truncating part of the codes as the final data group changes totally. Help me understand just that part if you can.
