Open ambrosejcarr opened 6 years ago
the precomputed format is essential for achieving interactive z-scrolling and zooming in 3D that you see in neuroglancer demos. in our hands (with lightsheet data) render just can't serve up multiresolution volumetric data fast enough for it to really be interactive. this isn't a surprise- the precomputed format is chunked in 3D but render is not.
This is a data format that @berl suggested we look at.
cloud-volume
It has a precomputed data format:
https://github.com/google/neuroglancer/tree/master/src/neuroglancer/datasource/precomputed