Sean Meighan
Welcome => Do You Need Help? Post it here => Topic started by: dslowik on November 03, 2015, 05:34:19 PM
-
For this year, I am creating a sequence in Xlights 4 for my pixel megatree. After getting it like I want, I do a render and export model to a LOR clipboard. I load the clipboard into LOR and paste it in with the Paste Mode set to Paste by Time. I have done this for 8 songs without a problem, but the 9th one is causing me problems. When I paste it, the timing does not line up where it should. It plays the effects to quickly after the paste in LOR. The effects look like they have the correct timing when I play it in Xlights.
I verified that Xl says the sequence is 128.967 seconds long and LOR says 2:08.97. I re-exported the audio file as a CBR file (just in case it wasn't) and re-imported it back into both programs.
I'm at a loss at what else to look at. Ideas?
-
Started recreating my Xlights sequence from scratch and that seems to be working.
-
Is there a trick to this. I created a new xLights sequence and add some event. Created a new sequence in LOR based off the same audio file. Exported two different single channel models. Loaded the clipboard in LOR and one is off by about 1/2 second and the other about 3.
Is there a trick to this Something I need to do in xLights or LOR when creating the sequences?
-
I found the the answer to my question. I was pasting by "Cell" instead of by "Time". Once I switched it to time it worked properly.
-
Spoke too soon. Pasting by time got it to start at the correct time but the data is off after that. :(
-
I just tested and it looks fine. Do I need to post screenshots? Also be aware when you export a model you only get the effects placed on that model. If you have effects elsewhere on a model group that data doesn't make it into the export.
-
Here's a screenshot. Data in LOR after pasting clipboard is on top. Data in xLights is on bottom.
Are you sure you're selecting the leftmost cell in LOR when you do the paste?
-
Yours looks good. Here is mine. Mine is just a single AC channel if that makes a difference. Thanks for taking a look.
-
Attached is a snipit of the xml from the sequence and the LCB file. This might provide somthing more to go on.
-
I don't know what you have happening. I tried a single channel white and it worked as expected.
-
Any chance you could post the directory of the files you used for your test. I have tried multiple machines, directories, exporting to lcb using the convert tab and a hundred other tests and every time the timings come up short. I just don't get what I could be doing that others should not have seen.
-
I think it would be more productive for you to post your files with a description of what steps to do to recreate it...i.e. which model are you exporting.
-
That it would. Attached is a clean directory with one sequence, one model and 5 fades coming from a clean computer.
The image of the xml is the portion from the sequence and the lcb file pasted together so the timings can be compared. The two in yellow are just the first to showing the offset difference.
Thanks for taking the time to look at this.
-
Well it was pretty obvious that all the efffects are coming in exactly half the size. Didn't you notice that? So first thing I checked was did you use 25ms for your sequence and sure enough that's what I found. Create 50ms sequences in xLights and it will work.
-
I'm seeing it off by about 20% by looking at the images attached. I was using 25ms. I just tried modifying the xml to 50ms, opened, saved and I'm seeing basically the same thing. One timing has a slightly different start time but they are still off by the same amount overall.
-
I'm seeing it off by about 20% by looking at the images attached. I was using 25ms. I just tried modifying the xml to 50ms, opened, saved and I'm seeing basically the same thing. One timing has a slightly different start time but they are still off by the same amount overall.
Delete the fseq file.
-
Ok yeah I should have tested more. There does appear to be something wrong I'm just not sure why it didn't show up with my file yet. I tried with 50ms file and then they were only like 30% of what they should be.
-
Ok scratch my last comment. It is working correctly for 50ms sequences. The reason it isn't working for 25ms is we calculate an interval by dividing by 10 so 25 becomes 2.5 which turns into 2 when assigned to an integer so it causes each cell to increment by 20ms instead of 25. But 50ms is working. I just had the wrong number in my head where the first effect should have ended so I double checked and its good. I usually use the Render then Export option.
-
Mine are spot on for both 25 ms and 50 ms. I did the test where I set the MS, deleted the .fseq file, opened sequence, saved and exported. This time I set each cell to 1 second so it could be seen more clearly. AutoBeat1 is 25ms and AutoBeat2 is 50ms.
Is there anything else variable wise in the code that could affect it that I could be missing?
-
Your doing something different there is no way the results can be the same for 25ms and 50ms.
Using your original test file:
50ms will begin like this in the lcb file:
<effect type="intensity" startCentisecond="0" endCentisecond="5" intensity="100"/>
<effect type="intensity" startCentisecond="5" endCentisecond="10" intensity="80"/>
<effect type="intensity" startCentisecond="10" endCentisecond="15" intensity="60"/>
<effect type="intensity" startCentisecond="15" endCentisecond="20" intensity="40"/>
<effect type="intensity" startCentisecond="20" endCentisecond="25" intensity="20"/>
<effect type="intensity" startCentisecond="30" endCentisecond="35" intensity="100"/>
25ms will begin like this:
<effect type="intensity" startCentisecond="0" endCentisecond="2" intensity="100"/>
<effect type="intensity" startCentisecond="2" endCentisecond="4" intensity="89"/>
<effect type="intensity" startCentisecond="4" endCentisecond="6" intensity="80"/>
<effect type="intensity" startCentisecond="6" endCentisecond="8" intensity="69"/>
<effect type="intensity" startCentisecond="8" endCentisecond="10" intensity="60"/>
<effect type="intensity" startCentisecond="10" endCentisecond="12" intensity="49"/>
<effect type="intensity" startCentisecond="12" endCentisecond="14" intensity="40"/>
<effect type="intensity" startCentisecond="14" endCentisecond="16" intensity="29"/>
<effect type="intensity" startCentisecond="16" endCentisecond="18" intensity="20"/>
<effect type="intensity" startCentisecond="18" endCentisecond="20" intensity="9"/>
<effect type="intensity" startCentisecond="22" endCentisecond="24" intensity="100"/>
-
It is my setup. :o
I loaded up my wifes computer that had an old copy of LOR, installed xLights and ran the tests again from scratch and 50ms is spot on and 25ms is off as expected. My previous test were on my desktop and laptop which I use xLights daily.
As usual. You win.
Now the next question. Are there other setting files that are stored outside of the xLights directory that could cause something like this to go off? How can I exterminate my machine and start over.
Send me your PayPal address. I owe you.
-
I don't know anything that could cause that. Did you look at the lcb file and is it exporting incorrectly on that one machine?
I don't want any money for helping. I chose to help and it did reveal the 25ms issue.
-
I'm starting to wonder if I should have shutdown xLights between tests. So many tests now I'm not sure. Now that I know better what the XML should look like I'll keep an eye on it and report back if I see something crazy.