"Unfair" calibration movies??!

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

Moderators: Stardust@home Team, DustMods

stephan.1969
Posts: 8
Joined: Sun Sep 09, 2007 5:40 am
Location: Germany

"Unfair" calibration movies??!

Post by stephan.1969 »

Hi everybody,

sorry, but I just want to state here, that I feel some of the calibration movies
are kind of "unfair".

Unfortunately during logging into forum I lost the copied movie ID and now i cannot find the cal. movie any more. But the certain mark I was proposed to click on, was a tiny black dot, which was a sharp distinct dot even focussed far above(!) the aerogel surface. So I decided to mark it as no track found. So I think of those calibration movies being quite unfair, as possible idp's are not clearly seen even above aerogel surface. Digitally inserted tracks should show a natural structure and correct focus depth - below aerogel surface only - there seem to be some discrepances in that.

I mean I surely look quite carefully and try to keep up my sensitivity, but unreal looking dots I will not click on.

jsmaje
Posts: 616
Joined: Tue Aug 15, 2006 8:39 am
Location: Manchester UK

Re: "Unfair" calibration movies??!

Post by jsmaje »

stephan.1969 wrote:I feel some of the calibration movies are kind of "unfair"
Couldn't agree more, not to mention "careless".
This 'mid-air track'is more than a "tiny black dot", so is unlikely to be the one you're talking about.
Looks like there could well be more of them out there.

speck
Posts: 16
Joined: Wed Aug 09, 2006 6:14 am
Location: Seattle, Wahington, USA, Earth

Post by speck »

Have to add my frustration too.

I have encountered several CM with the purported track in focus from at or ABOVE the surface of the gel.

I have also encountered a couple where the surface of the gel is slanted and the supposed track appears to actually be a surface feature coming into focus with the surface of a lower section of the gel.

I too examine the slides very carefully and am feeling a bit frustrated both because some of the CMs look like poor examples to me and because I have no avenue to study more CMs with explanations as to the features being identified with tracks, which I would be happy to do if it would improve my effectiveness.

All that said, I'm still having a great time and I feel fortunate to be able to participate in an effort like this.

Thanks for all the hard work on the part of the Berkley group and all the dedicated BB moderators. You guys rock!

speck
- Starlight, star bright, first star I see tonite - I wish I may, I wish I might, have the wish I wish tonite. - Anonymous

fflo
Posts: 63
Joined: Tue Aug 08, 2006 7:37 am
Location: Eindhoven, Netherlands

Post by fflo »

speck wrote:Have to add my frustration too.
..........................................................................................................
............................................................................
All that said, I'm still having a great time and I feel fortunate to be able to participate in an effort like this.
Thanks for all the hard work on the part of the Berkley group and all the dedicated BB moderators. You guys rock!
speck
Hi you frustrated Dusters,

I very much appreciate speck's attitude as quoted. So:
Firstly, let me remember you that all of us are VOLONTEERS;
Secondly, that every time you start using the VM, you encounter the following advise:
This is Stardust@home Phase 2 . It will be even more challenging than Phase 1. Phase 1 dusters, please review the new training movies, starting on screen 13. Please expect it to be much more difficult to get a high score, and expect your sensitivity and specificity to be lower than in Phase 1. This is on purpose.
Remember
If you're not having fun, you're not doing it right
Expect the Unexpected
This is research: the outcome of this project is highly uncertain
Keep smiling and dusting!
fflo

From dust we come - for dust we search - to dust we go :) :D

KevinLMoore
Posts: 4
Joined: Mon Jul 16, 2007 9:28 am
Location: Roswell, New Mexico
Contact:

Unrealistic calibration movies

Post by KevinLMoore »

I too have to add that some of these CM's aren't quite as they should be. I will try and remember to write down the id of the next movie I see showing this. I agree with Speck, that I will not click on a spot that does not seem natural, ascribing it to a contaminate on the microscope lens or something else, perhaps mechanical in nature. I have been marked incorrect a number of times and it is VERY frustrating. I too wish that if I get a movie wrong, that there was some way to show me why I go it wrong so that I can become more proficient at finding possible tracks. I don't mind the challenge, but it seems to me that we are being "punished" for erring on the side of caution.
Hope we all can find a particle! Good hunting!

Kevin

startrak
Posts: 52
Joined: Tue Jun 05, 2007 3:53 pm
Location: Kennebunk, ME

Personal best

Post by startrak »

I was going for another personal best of continuous correct calibration movies when this one caught me dead in my tracks. Pun intended. I'm assuming coors. 300,348 is the intended track.

http://stardustathome.ssl.berkeley.edu/ ... =9958181V1

BruceHull
Posts: 35
Joined: Mon Mar 05, 2007 11:18 am
Location: Silver Spring, MD USA

Personal best

Post by BruceHull »

I was going for another personal best of continuous correct calibration movies when this one caught me dead in my tracks. Pun intended. I'm assuming coors. 300,348 is the intended track.

http://stardustathome.ssl.berkeley.edu/ ... =9958181V1
In my opinion, CM 9958181V1 should be marked Bad Focus. Much of the surface comes into focus on the final frame.

jsmaje
Posts: 616
Joined: Tue Aug 15, 2006 8:39 am
Location: Manchester UK

Re: Personal best

Post by jsmaje »

BruceHull wrote:In my opinion, CM 9958181V1 should be marked Bad Focus. Much of the surface comes into focus on the final frame.
At least it's typical of those surface 'blurs' of the sort one can see turn into a track below-surface, and not actually a mid-air track (see above).
I guess it's another of those deliberately tricky ones they're slipping in to really test us, and that it would be legitimate to either click it or say 'bad focus'. Either way, if it were a 'real' for example, it would get looked at.
Being a CM however, you'll miss gaining a point by saying 'bad focus', but at least won't lose one by saying 'no track'.
It's clearly such subtleties that they're wanting us to look out for.

startrak
Posts: 52
Joined: Tue Jun 05, 2007 3:53 pm
Location: Kennebunk, ME

Re. CM9958181V1

Post by startrak »

Bad Focus came to mind but I recall reading somewhere tracks in CMs will always be in focus. Guess I made the wrong decision on this one.

But I really do enjoy and feel honored to be able to participate in this great, unbelievable adventure.

Malvineous
Posts: 1
Joined: Sat Jun 10, 2006 5:00 pm
Location: Brisbane, Australia
Contact:

Post by Malvineous »

I agree - phase two is a lot less forgiving than phase one.

9285810V1 got me, I can see three tracks in there but either I clicked on the wrong one, or there aren't any at all... I suppose that's the problem when you're just told you're wrong, it would be nice to have an explanation or hint to learn from.

jsmaje
Posts: 616
Joined: Tue Aug 15, 2006 8:39 am
Location: Manchester UK

Post by jsmaje »

Malvineous wrote:9285810V1 got me, I can see three tracks in there but either I clicked on the wrong one, or there aren't any at all...
It's definitely a +ve CM with a small track at 87,47 (top left). Though pasted perhaps slightly high, it shows an entry blur as the surface comes into focus, and remains clear-ish all the way to the bottom.

Compare that with the other specks, i.e. inclusions, which only focus below the surface over 2 or 3 bars (don't be mislead by the circular focus artefacts above and below them - see here).

Hermax
Posts: 9
Joined: Fri Aug 04, 2006 8:21 am
Location: Belgium
Contact:

Post by Hermax »

I have exactly the same frustration. I find it very hard to accept that I'm being "punished" for not finding a track in a CM, even though I DID see it but had serious doubts about it.
This frustration started for me at the end of phase 1, where I had a specificity of 100%. But then I clicked on something that looked like a track but turned out to be a CM without a track. Bang, there went my long preserved score. :(
My reaction was "So, you don't want me to report everything that looks like a track?" Sorry, but I'm not good enough to determine the difference between a possible inclusion or a possible track, when they both are very small.

All this must be seen as a very little detail of course. I'm having fun most of the time, especially with this phase. The enhanced new movies are a pleasure to look at.

stephan.1969
Posts: 8
Joined: Sun Sep 09, 2007 5:40 am
Location: Germany

Post by stephan.1969 »

Alright, I got one http://stardustathome.ssl.berkeley.edu/ ... rds=?-1,-1
as example...

Probably I am suposed to klick on thjat one, but normally I wouldn't, 'cause it does not look natural and is sharp and distinct even above aerogel surface..... never would be one in real...

.



[/img]

stephan.1969
Posts: 8
Joined: Sun Sep 09, 2007 5:40 am
Location: Germany

Post by stephan.1969 »

Yeah, it WAS a cm with a Track inserted.... would You have clicked on that, or not??

stephan.1969
Posts: 8
Joined: Sun Sep 09, 2007 5:40 am
Location: Germany

Re: "Unfair" calibration movies??!

Post by stephan.1969 »

Gee, just one after the other rigth now.... all the same- unnatural look, sharp over almost the entire range of focus... but supposed to be marked as track... :roll:


http://stardustathome.ssl.berkeley.edu/ ... s=?295,121[/url]

Post Reply