3d modeled in Modo, assembled in Houdini, rendered with Redshift and post in Nuke.
When I need to create gifs I ususally use After Effects and plugin GifGun.
But sometimes it doesn’t play nice with colors. They have an “experimental engine” inside as an option but that result in huge size gifs.
Some people use Ezgif. But it has some limitations and you need to upload to web.
So this weekend I tried ImageMagick.
It also installs FFMPEG. So you use it first to convert videos or image sequeces to gif first. And then compress it with ImageMagick. There is no GUI on Windows so you use Command Prompt (it’s like Terminal on Mac).
First you change your current location to the directory where you files are with. Just type cd, drag and drop directory in the “terminal” and hit enter.
I rendered 720x1280px image sequece in houdini with names like this:
So to convert it to 24 fps gif with half-resolution and custom color pattern (for better colors representation) I used this line:
ffmpeg -r 24 -i CAM_top.%04d.tif -filter_complex “[0:v] scale=360:-1,split [a][b];[a] palettegen [p];[b][p] paletteuse” output_360px.gif
And then compressed it with ImageMagick with this:
magick convert output_360px.gif -fuzz 1% -layers Optimize output_360px_fuzz1.gif
fuzz1 – controls the amount of compresion. For example in some other tests fuzz5 gave ok results and smaller files size.
Here is another example of creating gif from mp4 – with rescaling, changing fps and single pattern generation. Split just means that we split this video into 2 and one use for pattern generation and other for convetion to gif after that:
ffmpeg -i video_source.mp4 -filter_complex “[0:v] fps=12,scale=540:-1,split [a][b];[a] palettegen [p];[b][p] paletteuse” output_name.gif
Creating gif from mp4 – with rescaling, changing fps. Pattern generated for every frame – this helps if there are lots of color variations in gif but will increase the size:
ffmpeg -i video_source.mp4 -filter_complex “[0:v] fps=12,scale=w=540:h=-1,split [a][b];[a] palettegen=stats_mode=single [p];[b][p] paletteuse=new=1” output_name.gif
Here is single camera view (6 MB):
And here is 15 seconds version with lots of motion (21 MB):
I’ve been using Houdini on and off for a couple of years. And some concepts are difficult to grasp. Especially if you coming from another 3d application.
There is short course that I can now recommend to anyone who want to start having fun with Houdini:
Houdini Made Easy 01 – The Core Essentials from Hipflask. It’s completely free.
They also have 50% off until April 30th, 2020 on 2 other courses. I finished them and can’t recommend enough. One of the best teachers and flow of presenting information.
Random short animation:
I made a small Telegram group for 3d artist in Panama to share tips and tricks. Here is link if you want to join:
Some experiments that I made figuring out the best way to pass geo from Houdini to Modo to render with V-Ray:
I’m in love with SideFx Houdini. It’s the most fun and liberating app for 3d that I’ve ever seen. It’s procedural, node-based and the developers are moving to the future (looking at you, Modo, with sad eyes). Around a month ago I also bite the bullet and bought Redshift rendering engine.
There are tons of tutorials for 3d artist of every level. But this one gives a really strong foundation of poly-modeling:
Vehicle Modeling in Houdini 16.5
I made a short animation:
And here is how it was made:
There is still a lot to learn. For example, I don’t know how to get motion blur to work when the amount of geometry is changing from frame to frame. I guess I need to calculate velocity on birth and pass it to final geo but don’t know how exactly to do that.