 In this video I want to give you a bit of an overview of geometry nodes in Blender. What are they, how do they work, why are they great and what can I do with them. So geometry nodes are a system in Blender that gives you a huge amount of tools for procedural workflows all around modifying geometry, making use of a node system. What it means that these tools are procedural is basically that you can create your own fully custom geometry operations that you can run automatically on any object to very easily create complex results from a simple input. This means that you can build your own modifiers with settings and parameters that you can integrate into your workflow, share with others or simply have an accessible way to interface with geometry data on a more fundamental level. And with geometry nodes, all of that is made possible using a node system that is directly integrated into Blender. Geometry nodes directly hook up to the familiar modifier stack that you can use to procedurally modify an object's geometry. You can simply add a geometry nodes modifier just like you would any other. That modifier can then be instructed to do anything you want by selecting a node tree that defines exactly what it does. Just like other modifiers, it takes the geometry from the previous one in the stack. But now you get full control over what happens with that geometry using the node tree before it is passed back out to the next one. Once you have picked what node tree the modifier uses, you can also get a bunch of custom parameters which define settings of the modifier that you can adjust from the outside. But when you want to actually get full control over the modifier, you have to go to the node editor. And this is where the real magic happens. Here you can, by making node links, pass around geometry data between the nodes, process and change the data, split off branches and merge them back together up to the final point where you pass the geometry through the output to the next modifier in the stack. With that, you are entirely free in what you do with the geometry and only limited by the nodes that are available and your creativity of how to use them. You can even make use of the benefits of different types of geometry and convert them on the fly. For example, you can draw a curve as an input, create a mesh from that, scatter a bunch of points on the mesh, and turn that point cloud into a volume to get a simple setup that creates nice little puffy cartoon clouds just from a simple doodled curve and a handful of nodes. And that's the power of procedural systems. You can take very simple inputs that you can iterate over very quickly and let the node system do the rest to generate a more complex result on top of it. Another strength of procedural systems comes into play once you introduce randomization. With random elements in your procedural generation, you can very easily and quickly generate a large number of random variations simply by changing the seed value of your random input. This is great for assets that you need a lot of without caring enough to handcraft them individually or to generate a bunch of random variations as an inspiration in your design process. You never know what you might get. I want to encourage you to take the ideas that I presented in this video as an inspiration and pick up geometry nodes just to play around and try to achieve something that peaks your interest. Some examples would be a system to scatter objects on a surface, a more advanced custom array modifier, or a little random spaceship generator. I brushed over some of the more complex concepts of geometry nodes very quickly in this video. If you want to get a deeper understanding of how these things work, you can head over to studio.blender.org and check out our most recent course, Geometry Nodes from Scratch, where I walk you through all the concepts you will need to know with both a theoretical and a more hands-on approach. Take care.