diff options
Diffstat (limited to 'relnotes-0.6.0')
| -rw-r--r-- | relnotes-0.6.0 | 257 |
1 files changed, 257 insertions, 0 deletions
diff --git a/relnotes-0.6.0 b/relnotes-0.6.0 new file mode 100644 index 00000000..22efa56d --- /dev/null +++ b/relnotes-0.6.0 @@ -0,0 +1,257 @@ +CrystFEL - Crystallography with a FEL +------------------------------------- + +Release notes for version 0.6.0 + +Copyright © 2012-2015 Deutsches Elektronen-Synchrotron DESY, + a research centre of the Helmholtz Association. + +Authors: + Thomas White <taw@physics.org> + Richard Kirian <rkirian@asu.edu> + Kenneth Beyerlein <kenneth.beyerlein@desy.de> + Andrew Aquila <andrew.aquila@cfel.de> + Andrew Martin <andrew.martin@desy.de> + Lorenzo Galli <lorenzo.galli@desy.de> + Chun Hong Yoon <chun.hong.yoon@desy.de> + Kenneth Beyerlein <kenneth.beyerlein@desy.de> + Karol Nass <karol.nass@desy.de> + Nadia Zatsepin <nadia.zatsepin@asu.edu> + Anton Barty <anton.barty@desy.de> + Cornelius Gati <cornelius.gati@desy.de> + Fedor Chervinskii <fedor.chervinskii@gmail.com> + Alexandra Tolstikova <alexandra.tolstikova@desy.de> + Wolfgang Brehm <wolfgang.brehm@gmail.com> + Valerio Mariani <valerio.mariani@desy.de> + Parker de Waal <Parker.deWaal@vai.org> + Takanori Nakane <nakane.t@gmail.com> + Keitaro Yamashita <k.yamashita@spring8.or.jp> + Oleksandr Yefanov <oleksandr.yefanov@cfel.de> + +CrystFEL is free software: you can redistribute it and/or modify it under the +terms of the GNU General Public License as published by the Free Software +Foundation, either version 3 of the License, or (at your option) any later +version. + +CrystFEL is distributed in the hope that it will be useful, but WITHOUT ANY +WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A +PARTICULAR PURPOSE. See the GNU General Public License for more details. + +You should have received a copy of the GNU General Public License along with +CrystFEL. If not, see <http://www.gnu.org/licenses/>. + + +Overview +-------- + +The most important new features in this version of CrystFEL are: + +- Support for multi-event HDF5 files (e.g. CXI format as used by CXIDB) + +- geoptimiser: a new tool for precisely optimising the detector geometry + +- Removal of beam files / automatic determination of spot prediction parameters + +- whirligig: a new tool for finding clusters of similarly-oriented + crystal snapshots, e.g. for finding "mini rotation series" in data + from slow extrusion sample delivery methods. + +- Introduction of "CrystFEL unit cell files". + +These new features have individual sections below. In addition, there are many +other new developments. See the ChangeLog or the changes page on the website +for more details. There were, of course, the usual large number of smaller +refinements and bug fixes. + + +Support for multi-event HDF5 files +---------------------------------- + +CrystFEL's handling of HDF5 files has been made much more flexible. Most +importantly, it now offers the ability to handle HDF5 files which contain more +than one event (frame). Until now, CrystFEL has required that each event be +contained in its own file, which can place a lot of unnecessary strain on +filesystems. Recent versions of Cheetah allow you to create larger files which +contain many events each using the CXI format (see http://www.cxidb.org for more +details). CrystFEL now supports this as well as almost any reasonable +multi-event HDF5 layout. + +If you choose to continue using many small files, you should not notice much +difference. However, some small updates will be required. Firstly, +indexamajig's "-e" and "--image" command-line parameters have been removed. +Instead, you need to edit your geometry file and add a line such as this near +the top: + + data = /data/rawdata + +Geometry files now contain all the information needed to interpret the contents +of the HDF5 files as a physical setup, including the photon energy. Beam +parameter files have been removed. See below for more information about this. + +When using multi-event files in CXI format, the peak lists are read differently. +Use --peaks=cxi to retrieve peak lists from CXI files. + +hdfsee has been extended to support multi-event files, including navigation +between events. You should always provide a geometry file on the command line, +otherwise it won't know how to interpret the contents of your HDF5 file. + +A new tool in CrystFEL 0.6.0, list_events, is provided to simplify the process +of creating lists of individual events. However, most users will simply be +able to list the multi-event filenames themselves in the input to indexamajig, +which will then process all of the events in the file. See "man indexamajig" +and "man list_events" for more details. + +Finally, it may be necessary to update your check-near-bragg and +check-peak-detection scripts. Simply make a fresh copy from the CrystFEL +"scripts" folder or download them from the website. + + +geoptimiser: a tool for optimising the detector geometry +-------------------------------------------------------- + +CrystFEL 0.6.0 introduces a new tool, geoptimiser, for optimising the detector +geometry. You simply give it a stream containing indexing results obtained with +the approximate geometry, and it gives you a refined geometry. It can refine +panel translations, rotations and camera lengths. + +To use geoptimiser, you need to add some extra information about the +mechanical construction of your detector. This is used to constrain the +refinement appropriately, for example to ensure that panels which share sensor +silicon do not move relative to one another. If you're processing CSPAD data, +you can simply copy all the rigid_group lines from one of the CSPAD geometry +file examples (folder doc/examples) into your own geometry file. + +Refer to "man geoptimiser" and "man crystfel_geometry" for more information. + + +Removal of beam files / automatic determination of spot prediction parameters +----------------------------------------------------------------------------- + +As of version 0.6.0, "beam files" have been removed completely. This should +simplify usage for most people and remove a lot of ambiguity. Programs which +need X-ray beam parameters, such as pattern_sim, now have additional command- +-line arguments to provide them. indexamajig now determines these parameters +automatically for the purposes of spot prediction. + +We are interested in feedback about the automatic spot prediction parameter +determination. If it appears not to work well for you, you can restore the old +behaviour by using the new command-line options for indexamajig: +--fix-profile-radius, --fix-bandwidth and --fix-divergence. Simply set these +parameters to the values you had in your old beam file. + +The photon energy, or information about where to get it, now needs to be in the +geometry file. This can be done with a line like this: + + photon_energy = /LCLS/photon_energy_eV + +or, for a fixed value: + + photon_energy = 8300 + + +whirligig: a tool for finding "mini rotation series" +---------------------------------------------------- + +CrystFEL 0.6.0 introduces yet another tool, whirligig, which can be used for +locating "mini rotation series" in the output from indexamajig. This might be +used to perform an experiment similar to that described by Gati et al., IUCrJ +1 (2014) p87. In this initial version, whirligig finds runs of consecutive +frames which contain crystals in similar orientations. It writes the +corresponding filenames and event/crystal identifiers to log files, which might +be useful for further analysis. + +This is the program described by Nogly et al., IUCrJ 2 (2015). Refer to "man +whirligig" for usage information. + + +Introduction of CrystFEL unit cell files +---------------------------------------- + +CrystFEL offers you the ability to index patterns using Bravais lattice +information but without unit cell parameters, for example: "index these using +only tetragonal primitive lattices, but any parameters". However, it's awkward +to give this information using a PDB file: you essentially have to "trick" it +into interpreting the parameters correctly, then throw away the parameter +information. + +Version 0.6.0 of CrystFEL introduces a new way of specifying unit cell +information. These new unit cell files look like this: + + CrystFEL unit cell file version 1.0 + + lattice_type = cubic + centering = I + + a = 66.2 A + b = 66.2 A + c = 66.2 A + + al = 90.0 deg + be = 90.0 deg + ga = 90.0 deg + +In the event that you want to specify the lattice type information alone, you +can simply omit the cell parameters: + + CrystFEL unit cell file version 1.0 + + lattice_type = tetragonal + centering = P + unique_axis = c + +Note that a unique axis must be specified for all types of cell where this makes +sense, and can be omitted otherwise. This was done in the first example above, +which is cubic and therefore has no unique axis. + +You can, of course, continue to use PDB files just like before. + + +API changes +----------- + +The following changes have been made to the libcrystfel API: + +New functions: + - The event API (see libcrystfel/src/events.h) + - load_cell_from_file() + - cell_has_parameters() + - find_orig_panel() + - crystal_{get,set}_num_implausible_reflections() were added + - panel_is_in_rigid_group() + - rigid_group_is_in_collection() + - single_panel_data_source() + - find_rigid_group_collection_by_name() + - write_detector_geometry_2() + - find_intersections_to_res() + - sphere_fraction() + - gaussian_fraction() + - hdf5_read2() + - check_path_existence() + - get_peaks_cxi() + - hdfile_get_value() + - fill_event_list() + - image_reflection_closest() + - intmat_identity() + - extract_f_from_stuff() + +Removed functions: + - cell_set_cartesian_{a,b,c}() + - cell_{get,set}_pointgroup() + - twod_mapping() + - get_value() + +Changed function prototypes: + - record_image() + - get_detector_geometry() + - fill_in_values() + - write_detector_geometry() + - hdf5_write_image() + - hdf5_read() + - hdfile_set_image() + - hdfile_get_string_value() + - {get,set}_partial() + - write_chunk() + - prepare_indexing() and {dirax,mosflm,reax,grainspotter,xds}_prepare() + +beam-parameters.h was removed, and "struct beam_params" is now defined in +image.h. |