Minutes
              Einstein Observatory Users Committee Meeting
                      September 25. 1984

Attendees :

C. Canizares, MIT         P. Charles, GO
G. Fabbiano, SAO          D. Harris, SAO/GO
F. Marshall, GSFC         F. Primini, SAO
F. Seward, SAO/GO         H. Tananbaum, SAO
S. Murray, SAO            F. Walter, GO


Since the Boston airport was unexpectedly closed by fog the morning  
of the meeting. D. Helfand. CAL. and L. Ka1uzienski, NASA. were  
unable to attend the meeting. They were consulted after the meeting  
and concurred with the
committee's recommendations.

I. Status of MPC Data Reprocessing

These data are being reprocessed on the M600 to derive a more  
accurate determination of the background which changes with time and  
position over the Earth's surface. Six hundred thirteen days  
(approximately 70%) have been processed so far. Two hundred seventy- 
eight days remain to complete. At the present average speed of 1.7  
real time, it will take 5.5 months to complete. By ~ March 1985 all  
MPC data should be in the Data Bank.

Since reprocessing started, a question has arisen concerning accuracy  
of background subtraction for the higher energy channels. Should our  
estimate of the most suitable background change, a change would be  
made in a calculation applied after the reprocessing. No additional  
processing of the data base will be necessary.

II. Reprocessing of IPC Data

Since January 1, 1984, two major new routines have been added to the  
Rev 1 processing. MDETECT (an algorithm which detects significant  
deviations from a calculated background map) was on line January 16.  
Timing FFT (a fast Fourier transform which searches for regular  
periodicity in X-ray flux from moderately strong sources) was on line  
February 28. These new features are time consuming. Last year IPC  
data were reprocessed at a rate of approximately 1.5 real time, the  
present rate is 0.55 real time. To process 1 day of IPC data, 1.8  
days are required. We find that the minimum Rev 1 processing accounts  
for 50% of processing time.  MDETECT adds another 25% and Timing FFT,  
25%. At the present rate, the Rev 1 processing on the S230 will need  
an additional 28 months for completion. After the MPC processing is  
completed, the M600 will be available nights and weekends for IPC  
processing. Using both computers approximately 21 months will be  
needed to complete the IPC processing. Since the FFT is of minimal  
interest for most of the data, CFA recommended that all data be  
reprocessed first without the FFT. Approximately 16 months using both  
computers would be required for completion. We could then go back and  
do FFT for all IPC data or for just those observations where this  
analysis seems most promising.

A troublesome bug was found in the reprocessed IPC data. Exposure  
maps being applied to the data were not positioned correctly,  
resulting in incorrect exposure times at the boundaries of ribs and  
edges and causing false sources to appear. This problem is only acute  
in a few fields, but most fields are affected slightly. The false  
sources appear just above threshold. Strong sources are not affected.

Appendices concerning IPC image quality, error management in Rev 1  
IPC, and a description of MDETECT are attached.

III. Early Release of IPC Data

Because the IPC reprocessing is proceeding more slowly than  
anticipated, it was proppsed that all IPC data be released to the  
Data Bank January 1, 1985. Thus all Rev 1 data at this time would go  
into the Data Bank, and any time remaining in the 6-month original  
observer exclusive use restriction would go to zero. Newly  
reprocessed data would go both to the Data Bank and to the "owner" at  
the same time. If Data Bank requests were received for Rev 0 data,  
these data would be reprocessed out of sequence and the Rev 1 results  
sent to both original owner and Data Bank requestor at the same time.

There was unanimous agreement that this should be done. Therefore,  
starting January 1, all consortium data will be handled in this  
manner. A letter will be written to all Guest Observers advising them  
of the committee's decision and giving them the option of retaining  
the 6-month exclusive use of their data if they so desire. A Guest  
Observer expecting the 6-month grace period would not be deprived of  
it without his consent. Response to this letter will be less than  
100%, so the letter will state that the 6-month restriction will be  
waived unless the data owner communicates to us his desire to retain it.

Since there might be some confusion as to what data have been  
reprocessed to date, all Guest Observers will be mailed a summary of  
reprocessing status of their projects. CFA will institute immediately  
a small programming effort so the computer can produce these guest  
observer summaries for mailing. This will enable observers waiting  
for all of their observations to be reprocessed to know the current  
situation of the project and to request immediate reprocessing of  
fields remaining. CFA does not anticipate that these requests will  
create an unreasonable burden on the system.

IV. Revised Users Manual

The Revised Einstein Users Manual is now available in loose-leaf  
form. This has been distributed among consortium institutions, and  
all Guest Observers have been informed. Rather than mail the Manual  
to all Guest Observers, we have asked that those who want it return a  
request to us and it will be then mailed.

The Revised Users Manual contains both a description of the Rev 1  
processing and the latest instructions for running off-line analysis  
software. We anticipate our supply of these Manuals will be greater  
than the demand for them, so there are plenty available for  
distribution.

A second document of interest to a limited number of observers is  
SAO's Special Report 393, specifications of the IPC data analysis  
system, available on request.

Total number of Data Bank users is now 159. Requests are currently  
being received at the rate of 10 per month, an average rate double  
that of a year ago. A list of Data Bank users and the sequence  
numbers requested is available to anyone requesting it. It is also  
possible to determine the Data Bank users who have requested any  
particular sequence number. The committee felt that this information  
should be available as a matter of course to those making Data Bank  
requests. Henceforth, when a new request is received, CFA will  
produce a list of previous requests for each sequence number and send  
this to the requestor, who can then communicate with the other data  
users if he or she. so desires.

The committee felt that a yearly newsletter to Einstein data users  
would be of value. This letter should point out the existence of the  
data bank and rules for its use and might incorporate other  
information which has become available or emphasized since the last  
communication.

VI. A Catalog of Einstein Spectra

The committee discussed the desirability of having a catalog of  
Einstein SSS, OGS, and FPCS measurements. These data would be in the  
form of pulse height or other basic spectra available without too  
much processing or analysis. It is desirable to present all data  
obtained in a form so its basic properties could be ascertained by  
potential data users or people contemplating observations with  
similar instruments. Thus an outsider with such a catalog might  
observe the relative prominence of spectral lines, unusually hard or  
soft spectra, and whether or not data for that particular observation  
were statistically significant. GSFC and CFA said they might  
investigate this for the SSS and OGS data respectively and would  
present a proposal for-so doing at the December 1984 NASA review of  
the Einstein program.

VII. Next Meeting

Most members of the committee thought another meeting would be  
desirable and that it should be scheduled approximately one year from  
now.

Frederick D. Seward

10/3/84