- January 26, 2021 at 7:59 pm
I know that the scripts offer features beyond what you’re looking for, but they’re still perfect for what you need.
You can build the template in Ae once, set both layers to be replaced with footage, then use CompsFromSpreadsheets with a list of the filenames and output comp names. After you’ve built the Ae template, you could get your 70 outputs with literally just a couple minutes of work.
- January 26, 2021 at 8:37 pm
Swing and a miss.
Bridge DOES allow original file names to be saved in the XML raw data but neither AE nor Premiere nor PS have the ability to read “preserved file names”. Leastways I didn’t see it as an option
If the CC apps had the option to read “Preserved file names” in the XMP Metadata window I *think* that would have been a viable solution.
- January 26, 2021 at 8:45 pm
Thanks Walter and I know you are right.
There is just something infuriatingly maddening that a task so simple requires a $50 plugin.
It’s almost a matter of pride at this point. Plus I am learning a lot of stuff in the process.
If I can’t make it work it’s good to know I can fall back on the script.
- January 27, 2021 at 12:06 am
Here’s a script that shows how to automate this kind of work. After you unzip the JSX file, you can install it as a script or run it from the the File > Script menu.
You’ll have to import all the images you want to process and select them in the project panel before running the script.
Note that there are a few variables up at the top that you can change to modify the behavior of the final animations.
I’ve commented throughout so you can get a sense of what’s going on. The only thing that’s probably a little wonky-looking is my use of “match names” for the effects and their parameters; we’re using those so that they work irrespective of localization (language) of the user.
I’ve done this more to provide some education than to do your homework for you, so please look it over and feel free to ask questions so you can understand how it’s all working!
(I should add — there’s zero error checking in here. It will not fail gracefully if you use it any way other than as-expected. That’s ok for a quick demo, but it’s not a great development practice for real-world work. Normally I’d check for errors throughout and provide user feedback in case anything unexpected occured!)
- January 27, 2021 at 12:51 am
That’s great! Thank you! I’ll open this up tomorrow and see if I can’t waddle through it.
- January 27, 2021 at 2:14 am
OK, Holy Cow, Walter. That’s insane. I couldn’t wait until tomorrow to try it out. And in literally less than a minute it made the 70 files into the comps I needed.
And after I ran the script I was gonna wait until tomorrow to look at the code but nope, had to have that immediate gratification.
The script looks so clean and I understand a lot of it. I’ll need to drill down a bit on the math formulas for the blur, drop shadow etc. I am assuming the formulas are intended to have those effects come in as a ratio of the actual image and not a hard number that results in some attributes being too blurry, shadows too big/small, etc.
And thanks for the // notes. Very helpful.
- January 27, 2021 at 7:54 am
Adobe’s extendscript editor is deprecated, very buggy and infuriatingly slow. So you can ignore it 🙂
VSCode is the official new platform for script development
- January 27, 2021 at 6:24 pm
I’ll need to drill down a bit on the math formulas for the blur, drop shadow etc. I am assuming the formulas are intended to have those effects come in as a ratio of the actual image and not a hard number that results in some attributes being too blurry, shadows too big/small, etc.
In Premiere, you’re effectively normalizing the values in all your effects with “Scale to Frame Size” set; that fits the image to the frame and then says “ok, this is the new 100%.” In Ae, this would be like scaling to fit, then precomping without “collapse transformations” enabled.
Back in the script, since we can easily do the math, we’re concatenating transforms. Rather than transforming to comp size, rasterizing at comp size, and then transforming again, we’re combining the two transforms with no rasterization in between. This gives us less loss of quality, but also less consistency.
Because the scale of each image could be different, we need to make sure that all values for effect properties which are expressed in pixels are scaled according to the image itself. This will make them affect all images equally, no matter what size the image itself comes in.
Log in to reply.