Animation Blender 3D Tutorials and Resources

Blender Network Rendering: Revisited – Prism

Almost 10 years ago I built a small basement render farm with a few spare PCs to speed up renders. However, planning and pre-production ended up being far more difficult than rendering. Since I’ve returned to animation I’ve mostly been working in 2D, which is far less render intensive, but a recent 3D project revived my research into ways to use multiple systems to speed up Blender output. The first option I tried, Prism, ended up not being a system manager, as Blender’s previous Network Render panel was, but an entire pipeline manager with some helpful render options built in.

Installing Prism on Windows 10 was mostly straightforward, download from the site, run the .exe install, add your name when it asks, start a project, and in Prism Settings -> DCC apps add Blender. (Reading the manual is definitely recommended.) A few important notes:

  1. Read the manual, saying it again. This is an entire pipeline manager, by it’s nature, it can’t be simple.
  2. When setting up a project, put it in a syncing app such as Dropbox, Creative Cloud Files, or a Network Drive. You probably already have project files syncing but having the pipelines open right away on 2 systems was immediately satisfying.
  3. I had trouble with the Blender 2.8.3 connection not installing a Python Library. I installed Prism to Program Files (x86) instead of the default /prism directory which may cause permissions issues. To remedy this, I manually downloaded the install .zip and copied the Python 3.7 libs into the /prism/pythonlibs/python37. It will still probably bring up the install prompt in Blender, let it run, after that should work. (The auto-install prompted in Blender would present a permissions error.)
  4. Speaking of permissions, Windows smartshield complains from install to each run. Not much to do here but be aware.

After adding a project, the power comes in when going into the assets, or shots, adding a scene, then creating multiple render or playblast tasks. From there the render tasks can run on the machine of choice. While it’s helpful to have multiple CPUs or GPUs tied in together handing off tasks, this management immediately can help even single shot projects by allowing for multiple output settings and a tracker for those tasks.

Oh, and the same folks have a Renderfarm Manager named Pandora. I think that’s what I meant to download.

Tutorials and Resources Web Development

Want to learn Laravel? Tear apart Illuminate

There are any number of ways to learn Laravel. From the documentation right on to the very complete tutorials over at but there is also great way to learn the basics bundled in the install. Navigate to /vendors/laravel/src/framework/illuminate and dig into the modules that form the key functionality included in the original configuration.

Each one of these modules is registered in the app/config and, since Laravel 5, they now reside in the vendors directory leaving the app directory fairly spartan. Spartan is good but figuring out how the original functionality by looking through that directory is a bit tricky.

The vendors directory holds all packages added to the base install as well as holding the core packages in /vendors/laravel/src/framework/illuminate. By opening up /illuminate you can look through, and learn from, a number of packages including:

  1. Hashing, great for learning simple service provider setup.
  2. View, which has some pieces in the app directory as well and covers middleware.
  3. Authentication was added as core functionality and has a number of working parts good to explore.


Tutorials and Resources

PhotoShop CS6 and 30 Bit Color

Thank you Ars Technica for pointing this out.

Tutorials and Resources

Dynamically generated HTML5 manifests

In the specification for HTML5 several methods for storing data locally are outlined including localStorage and manifests. While building out the offline storage for Animatic Builder, I attempted to keep the stored data dead simple; as in the case of the shot information which is stored as one long JSON string. In this way the shot data can be pulled into any other use by reading the string. Keeping the images stored proved more difficult due to the number, potentially hundreds, and their format as many separate files. As well as making sure the storage is universal on mobile and full client systems.

%d bloggers like this: