mikemccand commented on PR #13767: URL: https://github.com/apache/lucene/pull/13767#issuecomment-2344025785
I also wonder if we still need to give the `compress` choice to users? I know originaly @benwtrent was concerned about the added cost of decoding the packed/compressed form (two dimensions in one byte), but I think I also saw a change to SIMD-optimize that case? Maybe the performance penalty is not so bad and we could always do `compress=true` when `bits=4`? I'll open a spinoff issue, and try to benchmark this using luceneutil's `knnPerfTest.py`. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
