Requesting a clear criteria for Bad Focus

Discuss your experiences with and ideas about Stardust@home here.

Moderators: Stardust@home Team, DustMods

Post Reply
Vacuity
Posts: 2
Joined: Tue Mar 23, 2010 8:47 am

Requesting a clear criteria for Bad Focus

Post by Vacuity »

I stumbled across a calibration movie that made me question my previous criteria for hitting Bad Focus. Looking trough the FAQ and tutorial I have not been able to find a clear answer for what what should be considered Bad Focus. Both are vague, talking about 'deep enough' but not mentioning what 'enough' is, and also some of the example movies in the FAQ do not work.
I tried to search the forums but I failed finding an answer, so here goes a new thread.

The movie in question is this one. Notice how the track is at the very bottom, with the surface just coming into focus at the second bar. For this single movie having the surface in focus in the top 20 % (24 microns) would be an absolute minimum! The surface above the bottom 25 % is the rule me and many other have gone by, but for this movie that is no where near adequate.
Now you may well be able to find the top of the track in another movie and trace it to the next, but it still makes me wonder what we should be aiming for. How deep would we have to go in that movie to find the track leading into this one?

So, how deep must we be able to see for you to feel confident that we do not miss tracks? How many microns below the the lowest point of the surface must be in focus for the movie itslef to be considered In Focus?
Chuck Crisler
Posts: 30
Joined: Wed Aug 02, 2006 9:44 am
Location: Windham, NH

Re: Requesting a clear criteria for Bad Focus

Post by Chuck Crisler »

I can't give any official answer for this, just my opinion. Several years ago I read that if the surface isn't in focus, then reject the movie as a 'bad focus'. However, you have to be able to go at least a bit lower than the surface. Your example move doesn't seem to bad to me, the track is down at 307, 304 (the coordinates could be reversed). As an alternative to your question but in the same thread, what about movies in which you can't see *ANY* change, except maybe some background shading? I usually mark those as bad focus even though the tutorial implies/states that you shouldn't. I would like my threshold to be that if I can't see Orion, then it is a bad focus movie. However, that would mean 15-20% would be bad, which probably isn't right. Orion is pretty hard to see. If I can see anything that changes in the focus movie (at least from in focus to out of focus), then I consider the movie acceptable.
DanZ
Site Admin
Posts: 777
Joined: Fri Feb 27, 2009 2:44 pm
Location: Berkeley, CA
Contact:

Re: Requesting a clear criteria for Bad Focus

Post by DanZ »

I pretty much agree with Chuck and have been (subconsciously) doing basically the same. However, that's not a great answer so I'll try and find out if anything more official was ever developed.

Dan
DanZ
Site Admin
Posts: 777
Joined: Fri Feb 27, 2009 2:44 pm
Location: Berkeley, CA
Contact:

Re: Requesting a clear criteria for Bad Focus

Post by DanZ »

After a little searching of my own, it appears these guidelines were as official as we ever got (and probably ever will be). Otherwise, for some (including myself) it will always be a gestalt type of thing I imagine.

See:

http://stardustathome.ssl.berkeley.edu/ ... .php?t=645

and Q1 here:

http://stardustathome.ssl.berkeley.edu/vmhelp.php

Dan
Post Reply