Sunday, July 24, 2016

Jiggle & Grow 2.2.2

Version 2.2.2 has been uploaded for anyone to download - What's changed?

  • Simplified Mask generation is now included - script will ask if you want the cutout mask simplified so you don't have to work with lots of anchor points when creating the expansion mask. It will only follow through if there are enough points to work with. Working with the new simplified mask is a lot easier, but the mask won't line up 100%. You are free to add more points if needed.
  • Changed much of the code to compare points differently along the paths instead of basing it on anchor points. This should hopefully, finally, kill the mismatch of anchor points. (Error at line 244)
  • The script tries to auto 'reverse' rotation for you.
  • A new smear point has been added. Instead of having a single center smear point on each breast, there are now two in order to add a bit of wiggle to the movement. Currently very subtle, might get turned up later.
  • Toned down the number of smear points around the edges the script gives. It used to try for around 24-28. Now tries for around 10-20.
This version is much larger than the others, 90kb compared to 130kb. Most of it is the new simplify mask function, which is really cool by the way. It really helps out so you can cutout your image really tight and not have to worry about moving all those points later. You can even add more points to the simplified mask if you need more control, very useful. I spent the day testing it and finding bugs here and there, so I hope I got them all, but I'm not confident based on the number of cases I have to try. If you run into it crashing, please let me know the error and what you were doing, if you can send me the .aep file I can figure it out much quicker. Thank you to those of you that do this, I'm grateful you don't just give up on the script. Main J&G Page

-sp

Monday, July 18, 2016

July Update

I am doing things, I promise. Right now I'm working on a large update to the Jiggle and Grow script. The last update left a lot to be desired in terms of actual results. It actually is a slight set back in terms of quality, but as the trade off was less restrictive on the masks. I'm not a fan of that quality drop.

I've been working on getting the script to simplify the mask, meaning you can use a ton of anchor points cutting out the mask without having to move all those points in the next step. The mask you work with when expanding will be a simplified version. Simplifying bezier mask paths isn't exactly easy - it is it's own fancy script 300 lines long.

So what's taking so long if that parts done? The current J&G script relies a ton of comparing two masks... now there's a third. So practically every section of code has to be adjusted and now take into account the 3 masks. Certain sections (SmearPoints, cough cough) are not cooperating and taking forever.

I hope to have it done by the end of the week so I can start on other things.

Update (7/22): Major progress has been made on the script. I'm going to be testing this one pretty heavily to try and find all the errors this time before pushing it out. So far, it's looking good. The simplify mask option is really helpful and caused significant overhaul of the comparison code, so I hope it will work out. There's a couple other goodies I added that will get added to the patch notes.

-sp

Sunday, July 3, 2016

Jiggle & Grow script 2.1

Version 2.1 brings two new things that are reactionary to some errors people were running into.

  • The expansion mask no longer requires the same number of anchor points as the original mask. Quite a few people were running into "Error at line 244", this solves that error. So rejoice in that!
  • How the script selects its smear points is much more accurate now. It rides the bezier curve instead of picking points directly in a line to the next anchor point. Because of the improved accuracy, the script no longer needs to ask you for the number of smear points.
Fixing the first one required me to rework how the Smear Effect Points were calculated because they were based on pairing up the anchor points on the original mask with the expansion mask. Which, if they have a different number of anchor points becomes impossible to calculate. The script had a section in it called 'magic' because it was an absolute nightmare to code and keep straight in my head what it was doing, and that was the section that needed to be redone.
Thankfully I found out bezier curves are actually easy to calculate, so I reworked the Smear Point process to run off of the bezier curve points instead of selecting them in a line. Using this in conjunction with a different way of calculating what is a 'stuck' point gets rid of the need to have the same number of anchor points. Whew! I will say, though, when the number of anchor points do match up, you'll get a better calculation. If they don't match, the script has to guess at certain things, which could lead to some remote cases of error. When in doubt, use the same number of anchor points on the masks.

Things on the to-do list: (in no order of completion)

  • Adding the ability to use the script with a video (keyframes from cutout mask). Will need to build in my UberKey script - which still needs testing.
  • Generate a 'simplified' expansion mask instead of a duplicate now that the number of anchor points don't need to match up.
  • Add highlight/shadow to the bulge that runs across TopClothingLayers.
  • Add dimples to the heftiness effect.