"There's a problem with this focus movie"

Post here if you are having any kind of problem with the Stardust@home website.

Moderators: Stardust@home Team, DustMods

fjgiie
DustMod
Posts: 1253
Joined: Sat May 20, 2006 8:47 am
Location: Hampton, SC, US

Re: "There's a problem with this focus movie"

Post by fjgiie »

jsmaje wrote: ... I'm also a little puzzled: while 7487044V1 certainly does show the same feature plus companion at a similar size but different orientation around 420/300, why indicate that dot in the first (& which I don't see in the second) rather than the shared 'H-type' track?

John
The dot shows up better near bottom focus because the dot is lower down into the Aerogel in 7487044V1.

In 9357982V1, the dot is below the bottom edge of the movie until bottom focus is reached.

[edited]      to add
This movie shows that track in a larger form: 2866673V1 [/edit]

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

Re: "There's a problem with this focus movie"

Post by jsmaje »

Thanks fjgiie & greuti - I think I see what you mean now.
But that really would be a cheeky one, and I'd still like to know what that little out-rider parallel to my-so-called 'H-type' could be in both movies which you consider to be the entry region. Can Dan enlighten?

If, as I believe, I nevertheless did my job of identifying a potential track (albeit a test) at whichever end of it, I think the continuing 30-pixel radius tolerance for 'calibration' tracks first applied in phase I for the small range of purely circular example 'carrot-tracks' is now completely outmoded, given the greater diversity of test-track morphologies and dimensions we are now asked to look out for in phase 4 (let alone all the coordinate & red-bar errors), and that this issue should become part of the current consultation regarding the best 'Skill' formula.

John

laserphil
Posts: 81
Joined: Sat Dec 02, 2006 8:28 pm
Location: Sydney Australia

Re: "There's a problem with this focus movie"

Post by laserphil »

2910721V1

There is a problem with the co-ordinates for correct ID of the track/particle in this movie. After successfully completing a number of sessions without missing a track (including a 70 pointer) this one caught me and I believe that I have reported it previously. It also highlights my thoughts on John's comment
jsmaje wrote:I think the continuing 30-pixel radius tolerance for 'calibration' tracks first applied in phase I for the small range of purely circular example 'carrot-tracks' is now completely outmoded, given the greater diversity of test-track morphologies and dimensions we are now asked to look out for in phase 4
From my perspective I think the thirty pixel limit is still valid. The intent of the project was and still is to find IDP's not just the tracks that they have formed. It is for this reason that the tutorial emphasises that the deepest point of the track should be identified(clicked) as this is where the particles are most likely to be found. It is also my gripe with the above movie as the co-ordinates indicate the top-most part of the track. The deepest particle in this case is around 50,375. Someone please tell me if I am wrong as I am becoming paranoid about where to click on any HAT/LAT's!

Thanks Fjgiie for the info on the url - looks like it worked!

greuti
Posts: 200
Joined: Fri Aug 11, 2006 10:11 am
Location: Switzerland

Re: "There's a problem with this focus movie"

Post by greuti »

8744030V1 has the same track and the same problem as Fred posted here (8203123V1)

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

Re: "There's a problem with this focus movie"

Post by DanZ »

jsmaje wrote:My suggestion is that the team make it official that everyone should use the only-so-slightly-longer option of posting a movie link as full web address[/i]]movie number alone as used above, which results in a more compact, explicit, and I suspect searchable reference. It might also help cut down on any personal angst that one may be alone with a particular movie issue, stem redundant postings and possible overload of moans to the team such as this! John
It's a good suggestion John. Let's just say it's preferred. Otherwise, bad PM reported.

Dan

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

Re: "There's a problem with this focus movie"

Post by DanZ »

fjgiie wrote:Hi John,
There were no coordinates listed in your post for movie 9357982V1 so I am taking a chance here.
The database has the Y for this PM at 384, which is off the scale, so I've reported it.

Dan

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

Re: "There's a problem with this focus movie"

Post by DanZ »

greuti wrote:8569028V1 That's impossible, isn't it. I guess this track looks mostly misplaced.
Why do you say that? It seems to "jump" a little and come into focus only at the very end of the available focus bars, but we've seen those before. What I can't understand is why it's only worth 6 points, so I'm bringing it up at a meeting tomorrow.

Dan

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

Re: "There's a problem with this focus movie"

Post by DanZ »

jsmaje wrote:Thanks fjgiie & greuti - I think I see what you mean now.
But that really would be a cheeky one, and I'd still like to know what that little out-rider parallel to my-so-called 'H-type' could be in both movies which you consider to be the entry region. Can Dan enlighten?

If, as I believe, I nevertheless did my job of identifying a potential track (albeit a test) at whichever end of it, I think the continuing 30-pixel radius tolerance for 'calibration' tracks first applied in phase I for the small range of purely circular example 'carrot-tracks' is now completely outmoded, given the greater diversity of test-track morphologies and dimensions we are now asked to look out for in phase 4 (let alone all the coordinate & red-bar errors), and that this issue should become part of the current consultation regarding the best 'Skill' formula.

John
I'll ask! Though your "out-rider" may just be an artifact of cut and paste for PM purposes.

Dan

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

Re: "There's a problem with this focus movie"

Post by DanZ »

fjgiie wrote:
jsmaje wrote: ... I'm also a little puzzled: while 7487044V1 certainly does show the same feature plus companion at a similar size but different orientation around 420/300, why indicate that dot in the first (& which I don't see in the second) rather than the shared 'H-type' track?

John
The dot shows up better near bottom focus because the dot is lower down into the Aerogel in 7487044V1.

In 9357982V1, the dot is below the bottom edge of the movie until bottom focus is reached.

[edited]      to add
This movie shows that track in a larger form: 2866673V1 [/edit]
I was with John there until you came up with 2866673V1. Very interesting! But now I have to agree. However, 9357982V1 still has bad actual coordinates that have been reported.

Dan

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

Re: "There's a problem with this focus movie"

Post by DanZ »

laserphil wrote:2910721V1

There is a problem with the co-ordinates for correct ID of the track/particle in this movie. After successfully completing a number of sessions without missing a track (including a 70 pointer) this one caught me and I believe that I have reported it previously. It also highlights my thoughts on John's comment
jsmaje wrote:I think the continuing 30-pixel radius tolerance for 'calibration' tracks first applied in phase I for the small range of purely circular example 'carrot-tracks' is now completely outmoded, given the greater diversity of test-track morphologies and dimensions we are now asked to look out for in phase 4
From my perspective I think the thirty pixel limit is still valid. The intent of the project was and still is to find IDP's not just the tracks that they have formed. It is for this reason that the tutorial emphasises that the deepest point of the track should be identified(clicked) as this is where the particles are most likely to be found. It is also my gripe with the above movie as the co-ordinates indicate the top-most part of the track. The deepest particle in this case is around 50,375. Someone please tell me if I am wrong as I am becoming paranoid about where to click on any HAT/LAT's!
It does look familiar! I'm bringing it to a meeting tomorrow to get a face-to-face answer.

Dan

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

Re: "There's a problem with this focus movie"

Post by DanZ »

greuti wrote:8744030V1 has the same track and the same problem as Fred posted here (8203123V1)
The answer I got before was that they are just very "shallow" tracks, but let me bring it up again.

Dan

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

Re: "There's a problem with this focus movie"

Post by DanZ »

DanZ wrote:
jsmaje wrote:Thanks fjgiie & greuti - I think I see what you mean now.
But that really would be a cheeky one, and I'd still like to know what that little out-rider parallel to my-so-called 'H-type' could be in both movies which you consider to be the entry region. Can Dan enlighten?

If, as I believe, I nevertheless did my job of identifying a potential track (albeit a test) at whichever end of it, I think the continuing 30-pixel radius tolerance for 'calibration' tracks first applied in phase I for the small range of purely circular example 'carrot-tracks' is now completely outmoded, given the greater diversity of test-track morphologies and dimensions we are now asked to look out for in phase 4 (let alone all the coordinate & red-bar errors), and that this issue should become part of the current consultation regarding the best 'Skill' formula.

John
I'll ask! Though your "out-rider" may just be an artifact of cut and paste for PM purposes.

Dan
I talked to Dr. Westphal today John and he has no idea what that little "out-rider" is. His guess is just a surface feature. Notice that, although it does change when you focus down, it seems to change all at once vs. a more gradual "moving" change that we see in most tracks.

But the bigger news - we're increasing the click radius significantly! When I get the exact pixels, I'll let you know.

Dan

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

Re: "There's a problem with this focus movie"

Post by DanZ »

DanZ wrote:
greuti wrote:8744030V1 has the same track and the same problem as Fred posted here (8203123V1)
The answer I got before was that they are just very "shallow" tracks, but let me bring it up again.

Dan
Consider this the "final" answer on these sorts of tracks:

Yes they do seem to be on the surface. But as we've always said, expect the unexpected. We certainly didn't expect to see behavior like this, but this track was found, and it looked/behaved just like this. Therefore, we need dusters to be on the lookout for them too. Actually, that's how we found it in the first place - dusters clicking on it! Now it's a PM to help find possible others like it. Why does it look/behave like this? Seems to be a very shallow or just scrapping the surface track, i.e., doesn't go very deep into the aerogel at all. But bottom line, they will remain PMs.

Best I can do!

Dan

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

Re: "There's a problem with this focus movie"

Post by DanZ »

DanZ wrote:
laserphil wrote:2910721V1

There is a problem with the co-ordinates for correct ID of the track/particle in this movie. After successfully completing a number of sessions without missing a track (including a 70 pointer) this one caught me and I believe that I have reported it previously. It also highlights my thoughts on John's comment
jsmaje wrote:I think the continuing 30-pixel radius tolerance for 'calibration' tracks first applied in phase I for the small range of purely circular example 'carrot-tracks' is now completely outmoded, given the greater diversity of test-track morphologies and dimensions we are now asked to look out for in phase 4
From my perspective I think the thirty pixel limit is still valid. The intent of the project was and still is to find IDP's not just the tracks that they have formed. It is for this reason that the tutorial emphasises that the deepest point of the track should be identified(clicked) as this is where the particles are most likely to be found. It is also my gripe with the above movie as the co-ordinates indicate the top-most part of the track. The deepest particle in this case is around 50,375. Someone please tell me if I am wrong as I am becoming paranoid about where to click on any HAT/LAT's!
It does look familiar! I'm bringing it to a meeting tomorrow to get a face-to-face answer.

Dan
Okay, don't be paranoid. This is a "track," but it came up through the aerogel from the OTHER SIDE! (just please don't ask what it is, because right now, we're not sure) Anyway, that's why the terminal end (which is where the "where to click" got placed) is not actually the deepest focus. But of course that causes the confusion you've experienced, which is not really fair and it probably should have never been turned into a PM. Thus we're working on removing it as PM. But there are "cousins" (other PMs made from the original), so if you spot them, please report them here too. Thank you!

Dan

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

Re: "There's a problem with this focus movie"

Post by DanZ »

jsmaje wrote:I’ve seen this strange double-track power movie in different orientations many times and always, as today, been stung by clicking in the ’wrong’ place or on 'No Track'. I’m sure another version of it has been reported before, though can’t find reference via the forum search facility to this particular VM number: 3758179V1

According to the ‘missed power movie’ red-bar coordinates I was presented with at the time, the team's coordinates were half-way between the supposed tracks, where in fact I can see nothing (url provided: http://stardustathome.ssl.berkeley.edu/ ... 3&theY=132):

Image

What do other dusters see using those links?

It was received wisdom in previous phases that the chances of two tracks appearing in the same frame were virtually zero anyway, let alone two identical-looking tracks like this in alignment, also atypical regarding their 'bow-shock' features more commonly seen with ?micro-draught surface artefacts.

It seems that several things have gone wrong here, and I suggest all versions of this movie be withdrawn immediately and the issues investigated.
John
Okay, it's really two tracks and this is how they appeared when they were discovered (in line like this). Expect the unexpected!

But when they got turned into PMs, lots of unintended coordinate misplacements, etc. turned up. We're correcting the bad ones, i.e., coordinates that are just nowhere near the track (and continue to report these please if you find them in the future). But "where to click" in the future will be (should be!) anywhere along the track(s). Why will that work? We're increasing the click radius so that it does! Should be interesting I know. Let's see how it goes and thanks for your patience. [Note: "Anywhere along the track" won't work for all tracks, so please still try and click on the deepest part of the track, especially for really long ones like this, since the increased click radius won't ever be that large.]

Dan

Post Reply