[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: The Saga of the Noise Floor



On Sat, 24 Mar 2001 17:46:19 -0500, you wrote:

>
>  In TN 76, I showed that repeated measurements of bright stars
>from Mark IV images exhibited a "noise floor" of about 0.006 mag.
> ...
>Rats.
>
>  I still haven't found the major contributor.

Me too.
I'm still getting nearer 0.01 mags for data set T.

I am still not getting much improvement out of
magnitude-dependent and colour-dependent corrections.
There is definitely a bug in my code: the statistics
along the way say I am eating away at the chi-square
as I add correction terms but the final plots look 
progressively worse! I resorted to serious
bug removal techniques: a full evening staring at the
code and then to bed. A horribly broken night with a
variety of dreams in one of which I was presented with
a clear image of the faulty routine with the mistake
(an extra instruction, effectively applying the error
correction twice) neatly flagged. On waking, I went
straight to the code ... but the extra line wasn't there.

But I did have another dream that completely explains the
problem. I think I was the CCD, looking out through the
lens. Tom: did you know that you have polar bears coming
round and peering into the telescopes? Damn big ones too,
or maybe that's the optical effect of being inside the lens. 
Not at all friendly. Try to get them to stay away.

Andrew Bennett, Avondale Vineyard