-------- Original Message --------
From: Bill J Gray <
pluto@...>
Apparently from: sentto-1412422-6786-1218849082-bydra=
Safe-mail.net@...
To:
guide-user@yahoogroups.com
Subject: Re: [guide-user] Hipparcos 2 TDF file for later update to Guide 8
Date: Fri, 15 Aug 2008 21:11:32 -0400
> Hi John,
>
> "...For no entirely apparent reason I've done a TDF [for]
> Hipparcos 2..."
>
> This may prove quite useful, since I'm now hearing reports
> (from you and other sources) of issues with Hipparcos 2.
> Brian Skiff has mentioned some problem cases, at
>
> http://tech.groups.yahoo.com/group/mpml/message/20890
>
> Thus far, I've gotten to the point of having Guide be
> able to show both the "old" HIP and (some of) the new Hip-2
> data, as described at
>
> http://www.projectpluto.com/new.htm#new_hipp
>
> Being able to actually show on-screen both HIPs may be
> helpful for doing comparisons.
>
> -- Bill
Hi Bill
Ah, sorry Bill, I hadn't even noticed you'd already done it.
If you didn't do so much we'd all be able to keep track better :-รพ
Anyway, it doesn't mean I've done it right. I originally set 'epoch' in the tdf to 1991.25, for instance, and that was totally wrong, so it's the epoch of the equinox of the dataset being investigated's coordinate system that that bit means, then (I nearly always have it 2000 so that should have been obvious to me).
Around 60% of any TDF I do is still achieved via trial and error, ie plug at it till it works.
I still don't know why that 2xmm one crashes Guide occasionally when it's used, because it can't be the immense line length after all, as I found I have a download dataset that's got even longer lines, and it never crashes it. It'll probably be one single weird character hidden in umpteen megabytes of ascii like it was for something else once, hence why it only crashes the setup occasionally.
Cheers
John
PS yep, you find out about catalogues and their quality by using them for one offs and certain cases. Global appraisals seem to miss these little weirdities in catalogues. However, I suppose if global statistic checking of data finds no problems, the problems that are there are probably relatively quite rare. But not too rare that they don't get tripped over from time to time.
PPS Just an FYI: I don't think the TDF documentation explicitly states that radians are covered by unitsn -1, where n is a number from 0 to 5 but is mostly always going to be 0 or 3 for stuff like this, I suppose. It mentions you allow for radians and that you allow for decimal degrees as input source, and that decimal degrees of RA is achieved via RA H etc, and then using units0 -2, and I kind of guessed it from there. It was a while before my brain caught up, though, and told me to put units3 -1 for the de d bit, instead of the units0 -1 I'd copied from RA H.