2,652
edits
Eiger: Difference between revisions
→A benchmark
(→A benchmark: add CBF timings) |
|||
| Line 33: | Line 33: | ||
so there's still some room for improvement. | so there's still some room for improvement. | ||
With program versions as of 2016-03-10, miniCBF is practically as fast as H5ToXds binary; hdf2mini-cbf is somewhat slower. | With program versions as of 2016-03-10, miniCBF-Linux is practically as fast as the H5ToXds binary; hdf2mini-cbf is somewhat slower. | ||
When unpacking the .h5 files to .cbf files and processing those, I get (on the same machine and with same processing parameters) on three successive runs: | When unpacking the .h5 files to .cbf files and processing those, I get (on the same machine and with same processing parameters) on three successive runs: | ||
| Line 39: | Line 39: | ||
Total elapsed wall-clock time for XDS 110.7 sec | Total elapsed wall-clock time for XDS 110.7 sec | ||
Total elapsed wall-clock time for XDS 111.1 sec | Total elapsed wall-clock time for XDS 111.1 sec | ||
which indicates a 8% overhead due to the HDF5-to-CBF conversion. | which indicates a 8% overhead due to the HDF5-to-CBF conversion. The latter 3 timings do not include the time for the HDF5-to-CBF conversion. | ||
== Troubleshooting == | == Troubleshooting == | ||