Here are our collective thoughts on this document and related issues
that were brought up:

The following comments regarding the document were raised during our
recent group meeting.  We've slightly expanded upon them here taking into
account the new version Malcolm sent out yesterday afternoon.

Specific points:

i   There seems to be no provision for intrapixel sensitivity tests, which
    as these are new detectors would seem to us appropriate to include.

ii  Categorising possible adjacency effects with spot projection optics
    seemed to be in the first draft but is not in the current one.

iii It is unclear whether the tests indicated would pick up any
    significant reset anomalies.

iv  Persistence and cross-talk measurements really need doing on all
    detectors and also in the final assembled camera (this is true of most
    of these measruements of course).

v   How is the (non)linearity going to be measured (fig. 8 setup ?).
    Stability of the illumination source is often an issue.

vi  A cross-reference to the requirements is needed and it would be
    useful inter alia to know what these are.  [These may be in the
    applicable/reference documents - but not clear if we have access
    to these ?]


General points:

a.  Overall we felt that the document just listed a set of entities to
    measure with no suggestions on what to do if any of these parameters
    were problematic or out of spec.  This may be outside the scope of
    this document but could be an issue.

b.  It was stressed that there was a need for doing these tests in the
    laboratory. Carrying them out on the telescope was felt to be
    unacceptable.  Concern was expressed about the possibility that
    laboratory tests like these could be cancelled or not fully
    completed in order to save money.
c.  In VISTA is there any provision for measuring the repeatability
    of the filter placement in the lab (impacts on flatfielding), using
    the real detector system ?

d.  We want to confirm that the reset-correction is being done in the
    aquisition computer together with desired coadditions (ie. is
    transparent to the user).

    

CASU