Charlie Zender

I will try to implement a straightforward bitround algorithm in the January timeframe. This would take as input the integer number of keepbits (for example as output by the 99%...

Thank you for that analysis, @milankl. A surprising (to me) result. If true for multiple lossless codecs then provides a way to gain a few more percent compression. It will...

@rsignell-usgs The BitRound quantization algorithm is now in the current snapshot of NCO as `baa=8`. If all goes well the then-current snapshot will be released as NCO 5.0.5 this Friday,...

There was a two-week holdup for unrelated reasons, and now NCO 5.0.5 with BitRound is finally released.

I learned something new from your careful reporting, thank you. You have found a good work-around for this corner case, which we will try to address in due course. BTW,...

Yes, Dennis, all of this is done with netCDF 4.9.X, not with CCR. The files are both quite small...[in.nc](https://github.com/Unidata/netcdf-c/files/9074918/in.nc.txt) and [foo.nc](https://github.com/Unidata/netcdf-c/files/9074923/foo.nc.txt). It would be great to hear what you think...

Thanks much for looking into this, Dennis. Always turning on Blosc-shuffle when any Blosc is required is an easy change to make in NCO (though ideally it would not be...

Great. Looking forward to it. I will re-test for robustness across datasets and sub-compressors when it lands in main branch.

Ahhh. Quite relevant. This suggests that the calling application skip invoking the Blosc codec for block sizes < 4 kb. Sound reasonable?

NCO uses an optimization called [USE_NC4_SRD_WORKAROUND](https://github.com/nco/nco/blob/master/src/nco/nco_msa.c) that is only invoked when there is a single strided dimension in a multi-dimensional variable and it is the first dimension. Who knows how...