played with d3 force layouts on a similar graph project. the collision force is usually what people undersize, causing overlap on dense clusters. what values did you land on for collision radius and alpha decay
This is a great visualization. There's something deeply satisfying about seeing a network topology laid out like this.
I've been working on some Markov chain visualizations for a lottery variance engine recently, and getting the node-edge density right without it becoming a 'hairball' is a constant struggle. What did you use for the layout engine? Is it custom D3, or are you using something like Sigma.js for the heavy lifting?
Thank you! It uses D3js as-is for no other reason than being the one I'm more familiarized with.
I also had issues with node density. The way I solved them was by adjusting the forces between the nodes, particularly the repulsion and collision forces so that there is is no superposition. Values too high or too low can make the simulation chaotic or too dense, while others make it more fluid or faster to stabilize from an initial seed.
That being said, if it was a bigger simulation I probably would have had issues keeping the performance, and it's possible I would have considered a different engine.
played with d3 force layouts on a similar graph project. the collision force is usually what people undersize, causing overlap on dense clusters. what values did you land on for collision radius and alpha decay
This is a great visualization. There's something deeply satisfying about seeing a network topology laid out like this. I've been working on some Markov chain visualizations for a lottery variance engine recently, and getting the node-edge density right without it becoming a 'hairball' is a constant struggle. What did you use for the layout engine? Is it custom D3, or are you using something like Sigma.js for the heavy lifting?
Thank you! It uses D3js as-is for no other reason than being the one I'm more familiarized with.
I also had issues with node density. The way I solved them was by adjusting the forces between the nodes, particularly the repulsion and collision forces so that there is is no superposition. Values too high or too low can make the simulation chaotic or too dense, while others make it more fluid or faster to stabilize from an initial seed.
That being said, if it was a bigger simulation I probably would have had issues keeping the performance, and it's possible I would have considered a different engine.