Contents

• List of Tables
• List of Figures
• Getting started
  • Updating this document
  • General remarks
  • Complete programs, and subroutine libraries
  • Subroutine library coding conventions
    • The file dm.h
    • #defines: system-level, and our own ``invented'' ones
    • Array memory allocation
    • Array ordering and centering conventions
    • Complex arrays
    • Using the predefined array structures
    • Testing your implementation of the subroutine libraries
  
  
• Data files
  • Reading and writing ALS 9.0.1 .nc files
    • Routines for reading and writing .nc files in C/C++
    • Routines for reading and writing .nc files in IDL
  • HDF 5 files involved in iterative reconstructions
    • General structure of our HDF 5 files
    • The assembled diffraction intensity group "/adi"
    • Data error factor in the "/adi" group
    • Assembled diffraction sequence "/ads" group
    • Assembly info "/ainfo" group
    • Support mask "/spt" group
    • Iterate amplitude "/itn" group
  • 3D data assembly .3da files
  • The subroutine library dm_fileio.c
    • Creating an .h5 file and accessing an existing file
    • Reading and writing/updating the "/comments" group in C
    • Reading and writing the "/adi" group in C
    • Reading and writing the "/comments" group in C
    • Reading and writing the "/ainfo" group in C
    • Reading and writing the "/spt" group in C
    • Reading and writing the "/itn" group in C
  • IDL routines for file I/O
    • Reading and writing the "/ainfo" group in IDL
  
  
• Procedures for diffraction data analysis
  • Cataloging files: the routine dt_list_series.pro
  • Merging individual ALS 9.0.1 exposures for one 2D view: Henry Chapman's merger
    • Using merger_manager
    • Wish list for enhancements to merger
  • Merging individual ALS 9.0.1 exposures for one 2D view: commie
    • A word on the commie distribution
    • Quick start: Assembly file
    • Quick start: Command line arguments (CLAs)
    • Quick start: Program flow
    • Further information
  • From 2D to 3D data: Anton Barty's program Xewald
    • Execution
    • Using
    • Output
    • Wish list for enhancements to Xewald
  • Ideas for programs we should have
    • Cewald: a C program for 3D data assembly
    • Estimating the center of a diffraction pattern
    • Estimating information content in a 2D view
    • Characterizing noise
    • Masking the beamstop and detecting saturated pixels
    • Displaying autocorrelation function from subregions of the diffraction data array
    • Estimating support from the autocorrelation
  
  
• Iterative phasing programs
  • Pierre Thibault's retriever
  • Anton Barty's 3dmpi
    • Wishlist for 3dmpi
  • Wishlist for reconstruction programs
    • Application of Fourier modulus constraint with error factor per pixel
    • Application of support constraint
    • Calculation of the next iterate
    • Calculation $I_{\mbox{recon}}(f)/I_{\mbox{data}}(f)$ at a particular iterate
    • Calculation of error metric at a particular iterate
    • Script language for running iterations
    • Structure outside the support
  
  
• Visualization of results
  • Routines for making movies of iterates
  • Routines for viewing 3D data
  
  
• The array calculation subroutine library dm_array.c
  • MPI-enabled FFTs
    • Working with FFTW 2.1.5 MPI
    • Working with dist_fft
  • FFTs using dm_array_fft()
  
  
• A roadmap for work at Stony Brook
• Running MPI code
  • Running code using LAM MPI on Stony Brook's Apple cluster
    • Things to set up just once
    • Running your code
  • Running dist_fft_test using MPICH
    • Running MPICH code on a subset of the cluster
  
  
• Running MPI DIST_FFT
  • Compiler
  • Header file modification