Crystallography Online: the website of the International Union of Crystallography

If the axes have been set up correctly, the beam center information
is redundant, since the beam center is implied as the intersection
of the Z-axis with detector face.  You are right to cross-check
with the set and get routines.  You should look at both the
mm and pixel values.  Let me know if you have problems.

   Regards,
     Herbert

At 5:26 PM -0700 10/15/07, Chris Nielsen wrote:
>	Hi Herbert,
>
>	I have discussed Andy's question (which seems to be the genesis
>of the
>current discussion) with him over the phone.  We are going to see how
>consistent
>the mm definitions of the SLS header are with their idea of the pixel
>origin
>assuming the array is stored as described in the CBF header using the
>cbf_simple.c
>beam center get/put functions.  Typically, the detector's beam center in
>terms
>of pixels is known, and the idea is to convert to mm as per the standard
>CBF definition.
>If the "set" followed by the "get" from cbf_simple is consistent, we
>have it.
>I would like to encourage the use of standard library functions to make
>the process of
>storing a useful beam center painless.
>
>
>	Personally, I think the "pixel" definition is easiest for
>everyone to understand, and fastest to get something useful from just
>from "more"ing
>the header, and for that reason is desirable:
>
>>>   # _diffrn_data_frame.center_fast
>>>   # _diffrn_data_frame.center_slow
>>>   # _diffrn_data_frame.center_units
>
>
>	Was there a consensus in the May meeting regarding the pixel
>origin question?
>I may have left before the discussion was finished.
>
>	Chris
>
>
>
>-----Original Message-----
>From: imgcif-l-bounces at iucr.org [mailto:imgcif-l-bounces at iucr.org] On
>Behalf Of Herbert J. Bernstein
>Sent: Monday, October 15, 2007 4:35 PM
>To: The Crystallographic Binary File and its imgCIF application to image
>data
>Cc: Miroslav Kobas
>Subject: Re: [Imgcif-l] question on beam center
>
>As long as there is a clear and consistent definition, either will work,
>i.e. either using the pixel origin for the bins or defining a new origin
>for bins.  I would appreciate hearing from others on the list as to
>which one would be easier for them to work with. -- Herbert
>
>
>At 12:41 AM +0200 10/16/07, Jon Wright wrote:
>>Dear Herbert,
>>
>>The fast/slow/units seems like a good solution. Is there some reason
>>for this choice of definition?
>>
>>     ... In all cases the
>>     center distances are measured from the center of the
>>     first pixel, i.e. in a 2x2 binning, the measuring origin
>>     is offset from the centers of the bins by one half pixel
>>     towards the first pixel.
>>
>>Why not measure from the centre of the bins in the case of bins?
>>
>>Best,
>>
>>Jon
>>
>>
>>
>>
>>
>>Herbert J. Bernstein wrote:
>>>   Dear Miro,
>>>
>>>       The problem is not what I like, but what processing programs can
>
>>>  deal with.  In the last round of proposed dictionary changes, we
>>>   suggested:
>>>
>>>   # _diffrn_data_frame.center_fast
>>>   # _diffrn_data_frame.center_slow
>>>   # _diffrn_data_frame.center_units
>>>
>>>   which would allow you to specify the center in mm, pixels or bins.
>>>   What we need now is general agreement on this approach.
>>>
>>>       I am forwarding this message to the imgcif list to see if anyone
>
>>>  objects.
>>>
>>>       Regards,
>>>         Herbert
>>>
>>>
>>>
>>>   At 6:54 PM +0200 10/15/07, Miroslav Kobas wrote:
>  >>>  Dear Herbert,
>>>>
>>>>   I think that Andy Arvai has already contacted you concerning this 
>>>>  topic. It is about the fullCBF header. I have made a recent dataset
>
>>>>  available on our webpage (<http://pilatus.web.psi.ch/datasets.htm>
>>>>   http://pilatus.web.psi.ch/datasets.htm) with the miniCBF and the 
>>>>  fullCBF header (conversion done with convert_minicbf). After 
>>>>  experiencing some problems with adxv, I asked Andy, Harry Powell, 
>>>>  Jim Pflugrath to adapt their programs such that they take all the 
>  >>> necessary information from the CBFheader.
>>>>
>>>>   Andy had problems with the definition of coordinate systems and how
>
>>>>  to get the beam center (units of mm) into a know coordinate system 
>>>>  (in units of pixels).
>>>>
>>>>   These are the 3 lines defining the beam center in units of mm in
>>>>  the  full CBF header:
>>>>   _diffrn_data_frame.center_slow 225.406000 
>>>>  _diffrn_data_frame.center_fast -211.577200 
>>>>  _diffrn_data_frame.center_units mm
>>>>
>>>>   For our detector, they correspond to the following beam center in 
>>>>  units of pixels:
>>>>   slow 225.406000 corresponds to y at 1310.50  fast -211.577200
>>>>  corresponds to x at 1230.10
>>>>
>>>>   This conversion can only be done, if one understands the definition
>
>>>>  of coordinate systems properly. I have asked several people who have
>
>>>>  experience with CBF if they understand how the coordinate systems 
>>>>  are defined, but they could no give me a unified solution.
>>>>
>>>>   Andy has asked me to include the following lines in our header:
>>>>   _diffrn_data_frame.adxv_beam_center_x 2463 
>>>>  _diffrn_data_frame.adxv_beam_center_y 2527
>>>>
>>>>   But since we really head now for the standard, full CBF-header
>>>>  (Eric  is implementing the full CBF-header into our dataformat), we
>>>>  only  want to use items, which are defined therein. Therefor I have
>
>>>>  suggested Andy to ask you, if it would be possible to add 2 more 
>>>>  items in the CBF header giving the beam center in x and y. Something
>>>>   like:
>>>>   _diffrn_data_frame.beam_center_x 2463 
>>>>  _diffrn_data_frame.beam_center_y 2527
>>>>
>>>>   What do you think about that? The consequence would be that there 
>>>>  would also be changes in convert_minicbf and and in our header 
>>>>  structure. Eric has told me that it is not a problem for him to 
>>>>  include additional items in our header. But I would easily 
>>>>  understand that you would not like to include these items in the 
>>>>  full CBF header, since this information is already there in a
>>   >> slightly different form.
>>>>
>>>>   If you want to check the above values, I have attached you one of 
>>>>  our recent images with the full CBF-header.
>>>>
>>>>   Thanks for everything and have a nice day.
>>>>   kind regards
>>>>   miro
>>>>
>>>>   --
>>>>   ________________________________________
>>>>   Dr. Miroslav Kobas
>>>>   DECTRIS Ltd.
>>>>   OSRA 007
>>>>   5232 Villigen PSI
>>>>   Switzerland
>>>>
>>>>   <mailto:miroslav.kobas at dectris.com>miroslav.kobas at dectris.com
>>>>   <http://www.dectris.com>http://www.dectris.com
>>>>   <http://pilatus.web.psi.ch>http://pilatus.web.psi.ch
>>>>   +41 56 310 3269 phone
>>>>   +41 56 310 5454 fax
>>>>   ________________________________________
>>>>
>>>>   Content-Type: application/octet-stream; name="run2_1_FH_00001.cbf"
>>>>   Content-Disposition: attachment; filename="run2_1_FH_00001.cbf"; 
>>>>  filename="run2_1_FH_00001.cbf"; filename="run2_1_FH_00001.cbf"; 
>>>>  filename="run2_1_FH_00001.cbf"; filename="run2_1_FH_00001.cbf"; 
>>>>  filename="run2_1_FH_00001.cbf"; filename="run2_1_FH_00001.cbf"; 
>>>>  filename="run2_1_FH_00001.cbf"; filename="run2_1_FH_00001.cbf"; 
>>>>  filename="run2_1_FH_00001.cbf"; filename="run2_1_FH_00001.cbf"; 
>>>>  filename="run2_1_FH_00001.cbf"; filename="run2_1_FH_00001.cbf"; 
>>>>  filename="run2_1_FH_00001.cbf"; filename="run2_1_FH_00001.cbf"; 
>>>>  filename="run2_1_FH_00001.cbf"; filename="run2_1_FH_00001.cbf"; 
>>>>  filename="run2_1_FH_00001.cbf";
>>>>    filename="run2_1_FH_00001.cbf"
>>>>   X-Attachment-Id: f_f7t7fn8h
>>>>
>>>>   Attachment converted: Macintosh HD:run2_1_FH_00001.cbf (    /
>>>>) (001C5C25)
>>>
>>>
>>
>  >_______________________________________________
>>imgcif-l mailing list
>>imgcif-l at iucr.org
>>http://scripts.iucr.org/mailman/listinfo/imgcif-l
>
>
>--
>=====================================================
>   Herbert J. Bernstein, Professor of Computer Science
>     Dowling College, Kramer Science Center, KSC 121
>          Idle Hour Blvd, Oakdale, NY, 11769
>
>                   +1-631-244-3035
>                   yaya at dowling.edu
>=====================================================
>_______________________________________________
>imgcif-l mailing list
>imgcif-l at iucr.org
>http://scripts.iucr.org/mailman/listinfo/imgcif-l
>_______________________________________________
>imgcif-l mailing list
>imgcif-l at iucr.org
>http://scripts.iucr.org/mailman/listinfo/imgcif-l


-- 
=====================================================
  Herbert J. Bernstein, Professor of Computer Science
    Dowling College, Kramer Science Center, KSC 121
         Idle Hour Blvd, Oakdale, NY, 11769

                  +1-631-244-3035
                  yaya at dowling.edu
=====================================================