You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This is not optimised for running over a vpn or slow network. This is a list of ideas of things that could be investigated to improve performance:
in companion generate checksum for bitmaps and only send if not already 'known' by this remote. Instead of being sent every bitmap when the page changes, we could be more intelligent and only be sent new images if the checksum from above hasnt changed
compression of the bitmaps? (rle, png, gz)
The text was updated successfully, but these errors were encountered:
look at: https://facebook.github.io/zstd/ for compression. Very fast, should be good enough and wide support for it.
Also these assets of 72x72 pixels rgb-values ?? Might be good to compress in one go, all images in one bulk to get better overall compression ratio... a good thing about zstd is that you can train it on a set of representative data for compression yielding better rates even for small sets like 72x72x3 bytes per icon (if not batched together).
No matter what I would add a small hook as soon as possible to the icon-pipeline, this hook should hash the data and create some statistics over unique hashes over time.
My guess is that you see many new hashes early on but most of them will be repeated moving forward in time.
This is not optimised for running over a vpn or slow network. This is a list of ideas of things that could be investigated to improve performance:
The text was updated successfully, but these errors were encountered: