Unfortunately my macbook is old and does not support WebGL.
I am using a lot of WebWorkers for the rendering. The code is using a mix of SVG and Canvas.
I don't know if I can explain it here in a comment. Roughly in the final passes I am preprocessing SVG paths into Path2D's[0] that then get rendered into an offscreen canvas. (they get cached)
The rendering code keeps 4 offscreen canvases, one for the layers bellow the current layer, one for the current layer, one for the layers above the current layer, and finally one for the HUD (the grid lines and zoom info).
These offscreen canvases are then rendered into many final canvases on the DOM in different ways (to do splitting animations, like row splitting or column splitting that happen to allow the UI to react to some specific user interactions like editing a grid element).
This way the rendering is smooth with simultaneous users on the same project on my old hardware.
I am planning to package this specific rendering part into npm and release it soon.
Illustrator has a similar feature called shape builder. You can create new objects from intersections and such.
Illustrator is an amazing tool, I have seen a few use cases of that feature in logo drawing tutorials. My main differentiation points with Illustrator (and sketch) are:
* Web based
* Grid based (there is always a grid bellow that you can use to create patterns)
* Free (right now the export to svg is the only planned payed feature, I hope to make it enough to cover my cost of living, maybe I will add pro accounts to those that need to produce non CC0 stuff if there is audience for it, but I will need to pay a lawyer to help me with that feature if it comes to it)
I have since then managed to get some funding and have been working on it more seriously for the past 4 months (since may 2017). In two weeks I will have another closed alpha-demo that is then going to be used to produce a video for the landing page, and opened to the public if there is no big bashing from my zen target audience friends.
It does seem like a huge desert to cross. Motivation comes and goes. In the meantime I managed to kill my desire to feature creep it to death and learned a lot about the fine art of listening/ignoring the target audience requests. I know for sure that my approach for my next project is going to be completely different than my current one. But in the meantime the rent is due :)
The landing page: https://www.pixnit.com
My blog with some thoughts on the bootstrapping experience: http://www.hugodaniel.pt