Philipp Otto

> I’m not entirely sure which route this concerns. Is it about segmentation data loading? If so, I’d suggest to use the datastore indeed, as this will save a hop,...

@MichaelBuessemeyer As discussed, feel free to do a rough review of this rough PR. I managed to finish a first version of cuckoo hashing for applying the mapping on the...

A note about performance: Before I introduced the cuckoo hashing for mapping the data, the FPS value was at ~22 when flying through a certain DS. With the cuckoo hashing...

@daniel-wer I just finished testing and fixing the 53 bit support. I would love to also get some PR feedback from you :pray: @MichaelBuessemeyer I incorporated most of your feedback,...

Thank you for the review and the testing report :pray: > * [x] If an annotation has an editable mapping, the supervoxels within the active mesh are always highlighted when...

> If an annotation has an editable mapping, the supervoxels within the active mesh are always highlighted when hovering in the 3d view, even if the proofreading tool is not...

With [7588](https://github.com/scalableminds/webknossos/pull/7588), one can annotate transformed datasets (skeletons and volume annotations). Proper systems would still be desirable, but I suggest to defer this until the actual need arises.

Potential solutions: - defer olvy initialization and - fetch the "unread" count by proxying via back-end - OR use another olvy api route to fetch the count - switch from...

The behavior is somewhat expected. Simple example: - user annotates one section in mag 1 --> z=0 is annotated - user zooms out to mag 8-8-8 and extrudes to z=26...

While the supervoxel proofreading performance has been improved by #6909, there is still some room for improvement. We should investigate the current bottlenecks for proofreadingmedium-sized agglomerate files (splits and merge...