Re: Bits and pieces

[Tom Droege <tdroege2@earthlink.net>]

Yes, that is the first approximation.  Yes, the difference is small 
compared to the usual noise.   I admit that I have not yet done a really 
good experiment.  Mostly because I have not yet figured out how to do 
it.   I need to learn to use some of the tools on linux.  So far I do such 
things with "Image Scientist" on a Windows machine.  This is one of the 
problems of working solo here.  I don't have a second observer to grab some 
interesting thing and go off and do it.  I put out a fire a day.  Sometimes 
two or three.  Today TOM3 is dead.  It was not something easy.  So tonight 
when Dan arrives we will go after it.

So far, I change flats when something big happens.  Like changing a 
lens.  My long term plan is to take flats once a month when the moon is out 
whether they make any difference or not.  Because of the way TOM2 and TOM3 
are mounted, I need Dan here to do flats.  Darks are easier to take. Darks 
are pretty constant.  It depends on keeping the temperature control system 
working well.  The dark column correction stabilizes darks well within the 
noise.  They are already well within the sky noise.

My general concept is to keep taking data while improving what we can.  We 
will then attach a quality figure to time.  Data taken before some date has 
quality level q1.  Data taken between some dates has quality level q2 ...

If I don't take this approach, I will be dead before the data gets good 
enough.  The worst of the data is good enough to see a lot of interesting 
things.

Tom Droege


At 09:23 AM 6/3/03 -0700, you wrote:
>I think if the above were true then you could subtract two random
>flats taken weeks apart and the differentce would look like noise.
>In other words if flats really don't matter then all master flats
>are randomly interchangable.
>
>Do you really get the same result if you use the "wrong" flat?