Sean Meighan
Software => Bugs in xLights/Nutcracker => Bugs (Please dont post here, soon to be deprecated) => Topic started by: RoboMello on September 08, 2015, 06:39:38 PM
-
The title explain the problem.
Since 4.2.7 when you Zoom In past a certain point you no longer will have the option to separate time marks with the hand indication.
Sorry for the poor information, but if you used before, you will know what I am talking about.
If needed I can provide a video :-)
Thanks
RoboMello
-
It's there but the selection zone is only a few pixels wide.
You need to zoom out a bit to get the hand to un-magnetize the effect from the adjacent effect.
-
It works on 2.4.6, and is gone on 2.4.7
And when you have time marks that are 1 second apart is hard to get the correct one and so on.
-
The selection zones are a fixed number of pixels. The edge cursor should be when it's 6 pixels from the edge and the hands should be the next 4 pixels. I know Dan made a slight change to the selection of small effects. It used to be that anything under 22 pixels and all you could do was a center selection and move the effect. He changed it so that now anything under 8 pixels and it will only grab an edge to resize it. I thought it seemed ok when I tested the changes.
I wouldn't mind seeing a video so I can see the size of your effects and which zoom level is having issues.
-
Hopefully this Video helps
https://youtu.be/6cPBms9r3qA
Thank you
RoboMello
-
I did some sequencing today and some effects would have the hand, but didn't have the ability to grab from each end, matter of fact I could usually grab the left side but not the right. So, it was always reversed, if I could grab the right side, the hand wouldn't work. If the hand would work, I couldn't grab the side.
-
Ok I checked in a fix. The changes that were done in 4.2.7 were converting the mouse position to time and then back to a position for hit testing but the conversion to time was calling the function that rounds the time to the nearest frequency interval. So basically if you had a 50ms sequence you were only getting mouse positions rounded to the 50ms intervals so several pixel positions were not achievable causing it to skip over the positions that gave you the hand. This did actually get worse the more you zoom in so I'm sure when I tested Dan's original changes I wasn't zooming in enough to see this problem.
-
released 4.2.9 with this fix
-
Ok I checked in a fix. The changes that were done in 4.2.7 were converting the mouse position to time and then back to a position for hit testing but the conversion to time was calling the function that rounds the time to the nearest frequency interval. So basically if you had a 50ms sequence you were only getting mouse positions rounded to the 50ms intervals so several pixel positions were not achievable causing it to skip over the positions that gave you the hand. This did actually get worse the more you zoom in so I'm sure when I tested Dan's original changes I wasn't zooming in enough to see this problem.
I got lost on the 1st conversion to conversion :-)
Thank you for fixing it, sorry for being annoying.
RoboMello
-
released 4.2.9 with this fix
Thank you, works fine now :-)
RoboMello
-
Ok I checked in a fix. The changes that were done in 4.2.7 were converting the mouse position to time and then back to a position for hit testing but the conversion to time was calling the function that rounds the time to the nearest frequency interval. So basically if you had a 50ms sequence you were only getting mouse positions rounded to the 50ms intervals so several pixel positions were not achievable causing it to skip over the positions that gave you the hand. This did actually get worse the more you zoom in so I'm sure when I tested Dan's original changes I wasn't zooming in enough to see this problem.
I got lost on the 1st conversion to conversion :-)
Thank you for fixing it, sorry for being annoying.
RoboMello
Yeah sometimes I go into too much detail. Basically if you zoom way in and drag and effect edge you will see how it snaps to the 50ms lines. Well just imagine the mouse position was doing the same thing internally so it never could respond to any of the inbetween positions. That's why the further you zoomed in the worse it got because you were expanding the dead zones.
-
Just curious if anyone is having issues with the copy/paste functionality using the 50ms timings? I updated to 4.2.9 and that definitely helped the selection and highlighting cells, but I am unable to highlight a fade up and down effect across multiple cells and do a copy/paste. I can select one effect and copy/paste find but joining fade up and down effects, I can not do that. I was doing this fine on 4.2.5 until i started had a lot of XL crashes which I posted yesterday. I appreciate this selection fix and can work on my sequences; just not as fast.
-
The copy/paste logic doesn't care what type of effect it's copying. Are you talking about the ON effect with ramp intensities? I really don't know what a fade up and down effect is because you could be talking about using the fade text boxes on any effect. Why don't you just click on one of the effects, copy it, and paste it into this thread?
-
Yes Gil - what I call fade up and down is the ON effect where it goes from 0 intensity to 100 (or some value) and fade down is the ON effect where intensity starts at 100 (or some value and goes down to lower value. I have a increasing ramp and decreasing ramps back to back so I select both and try to copy / paste. Does not work for me. I can select one and copy/paste fine. Used to select a range of effects and copy /paste but that is not working for me. Note: this is at the 50ms timing setting using windows.
-
Well I'm not able to look at this for several hours so the more investigating you can do the better. Try it with two other effects because I suspect it has nothing to do with the effect type. Try pasting in a completely open area with nothing interfering on both sides of where you are pasting. It sounds like its only a problem when you are copying multiple effects. Does it paste any of them? It's possible the updated logic using time instead of position is incorrectly thinking that adjoining effects are blocking each other since the end marker and start marker of the next effect share the same spot. Try pasting a single effect and then pasting again in a cell directly beside it. You should be able to determine a pattern of exactly what isn't working.
-
I noticed something last night which might be part of the same issue. If I create a block of random effects and then select the bottom right 25% of them and alt drag them to the right I'm not able to drag them back to the left unless I move the entire block to the right slightly so the code is thinking there is interference when there isn't. I warned the guys that selection and position logic was a house of cards. We janked a card out and now I gotta rebuild it.
-
Gil,
I did a screen capture (windows) of some copy/paste for my megatree. I was adding effects on each leg of the tree. Seems when I copy an effect that it is not overwriting what I copied previously so when I paste, it is the prior effect that was copied. Happens on different effects. A single copy / paste will sometimes work but if I try to copy/paste multiple effects, it only pastes a single effect. The recording was not done at the 50ms timing grid.
[attachment deleted by admin]
-
I don't know what the heck that filetype is you attached. I did a bunch of copy/pasting and I can't get it to produce any issues. If anyone is really having a problem they can duplicate then provide the necessary files and instructions to reproduce.
-
It opens and plays just fine with VLC media player.
-
Rando, when you copied 3 effects and pasted into one cell, and it appears only the last effect is there.. I bet if you dragged the last effect to the right, you would see the first two effects are actually there under the 3rd one.
Are you using 4.2.8 ?
-
I was able to open the file up in VLC and play it too.
What I'm seeing is when the first cell is being selected, then dragged to highlight a second cell, the effects themselves aren't being highlighted in magenta to indicate they are selected for copy and paste. So when copy is being selected, nothing happens. Pasting then just pastes the last effect on the clipboard.
I can't recreate this under Mac OS X. As soon as I select a cell it highlights the effect automatically.
-
Well I've never heard of VLC and I don't open strange files.
-
I was using Microsoft Expression Encoder as the screen capture. I will try another product.
Steve - used 4.2.9, but this was doing same thing in 4.2.8. I have tried to extend the effects out but it appears to me that the copy memory (if that makes sense) is not clearing out when I try another copy. Also, I can not select several cells and delete the effects; I have to select each one individually. As I mentioned in another post, I had several crashes yesterday (using 4.2.5), maybe as many as 15 and right after that this started showing up. I do not know if anything has been corrupted in my installation so I upgraded to 4.2.8 and then 4.2.9.
-
Did another video with Screencast-o-matic. Hope this is better.
[attachment deleted by admin]
-
I just watched you new video and unless the effects are highlighted in magenta, they aren't going to copy. When you're doing multiple cells none of the effects lines (between timing marks) are turning color. When you do a single selection you can clearly see them turning magenta indicating they are indeed selected.
As for paste, it's always pasting the last effect that's in memory which is your single selections.
-
Ok I get that; then tell me why I highlight multiple cells that it is not turning magenta. It used to do that fine. Same thing when selecting multiple cells to delete. Nothing deletes unless I select one individually (and it is magenta). If others can select multiple cells and it turns magenta, then why is mine all of a sudden stopped working.
Fyi - I just uninstalled XL and reinstalled 4.2.9. I was thinking maybe something had gotten corrupted. Didn't help. Thanks for the feedback.
-
Kevin - taking your comments and others, I did see where the magenta highlight was working and where it was not. For some reason, on one model that I have been working on over the past 2 days, especially when I was getting all my crashes, is the only model that the copy / paste is not working. I can highlight several cells and some or none will be magenta. Not consistent. On other other models, copy/paste is working fine. Oh well, let me finish on this one model and move on.
-
then tell me why I highlight multiple cells that it is not turning magenta.
I wish I could answer that. Gil or Dan will have to investigate. I suggest you post your xml files so that either one of them can recreate your set up and see why there is a issue with your one model.
-
I noticed in the video you are trying to copy at the strand level. Can you put the effects at the model level and try it? That way I'll know if its just not working at the strand/node level.
-
Yes it is happening at strand and model level. Funny - when I first started sequence, I could copy multiple effects and paste, but after the first time, it would only copy a single effect. Could highlight multiple cells, but nothing had the magenta designation that it was copying unless I specifically selected one effect.
-
Well the easiest way for us to try to reproduce a user's issue is for you to ZIP up your xlights_rgbeffects.xml and the xml file for the sequence giving you trouble. Those files should be in your current show directory.
-
Sorry - I went out of town this past weekend and no computer. I installed 4.2.10 just now and it appears all the copy/paste issues I had are corrected. I tried at model level and node level and everything copies/pastes as expected. Thanks Gil for the fix.
-
Sorry - I went out of town this past weekend and no computer. I installed 4.2.10 just now and it appears all the copy/paste issues I had are corrected. I tried at model level and node level and everything copies/pastes as expected. Thanks Gil for the fix.
Ah so you were the one with this issue. I couldn't find what thread this was in so thanks for replying. The problem you were having was a bug that was fixed. Someone else on FB ran into the same thing and I finally stumbled on the cause. What nobody told me and I had to discover on my own was that the problem you were having only happens when you had moved the vertical scrollbar because it wasn't subtracting off the number of rows that had scrolled off screen when it was doing the selection calculation. That's the kind of thing to keep an eye out for when reporting problems cause obviously selections were working for most of us. Unless you had a lot of models on the screen or were expanding strands and nodes and then scrolling down you would never see it. I noticed the selection was always selecting effects offset by the number of rows I had scrolled off the page.