Variable speed auto-scan?

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

Moderators: Stardust@home Team, DustMods

Post Reply
jdrockster
Posts: 27
Joined: Wed Aug 16, 2006 11:14 am
Location: Northern California

Variable speed auto-scan?

Post by jdrockster » Mon Jan 24, 2011 3:09 am

I like the concept of the auto-scan but rarely use it. I would use the feature more if I could control the speed of the device. I don't know how difficult it is to add such a feature; (variable speed), but it would really help me if I could match the focus speed with my eyes and brain on any given day.

It would also be cool if new movies were added, (just a few) on a regular basis. It would encourage old "DUSTERS" to come back. It can be discouraging to click on a track and see that it has been viewed by 250 others. If we knew that new movies were added on a regular basis, I, and others, would play this scientific lottery game more often. After viewing many of the same old movies many times(not calibration movies), their distinctive features are quite obvious. It would be more interesting if a random generated number would start us at tile x,y,z quadrant a,b,c, etc. at each login. It was very exciting at the beginning of phase 3 to find a potential track, click it, and see that it was only viewed by a few people, and I was the first to see it. Many of those movies have still only been viewed by a few "DUSTERS". Did you remove or change those movies?

I hope I am not perceived as a disgruntled "DUSTER", I am not. I am concerned that interest seems to be waning in the project by many people on this side of the lab. I have donated a lot of time in the interest of science, ie; Intel international science fair judge volunteer and many other science projects. Retirement has blessed me with the time to play with the things I like to do such as this. I know you must be quite busy and probably under-staffed, under funded, etc. but something needs to be done to revive interest so we can finish this part of the project and start something NEW! I will help anyway I can, thanks.

Thanks for your consideration. jdrockster
Dust2Dust...Dust4Dust!

DanZ
Site Admin
Posts: 772
Joined: Fri Feb 27, 2009 3:44 pm
Location: Berkeley, CA
Contact:

Re: Variable speed auto-scan?

Post by DanZ » Tue Jan 25, 2011 6:04 pm

Hey there JD,

Thanks for your commitment to Stardust@home and all the hard work. I'm asking our tech Will about the variable speed. I would use it too!

I'm a little concerned about your interest in finishing "this part of the project and start something NEW." Do you just mean new movies, or are you thinking new project? If it's a new project, that's probably not going to happen very soon with Stardust. However, we are gearing up for a soon-to-be-launched Phase 4 with an all new scoring mechanism. All the details will be made public in the not too distant future. [If, on the other hand, you are interested in completely new science projects, check out: http://www.scienceforcitizens.net/finder/ and http://www.birds.cornell.edu/citscitool ... jects/find (with the latter listing much more than just bird projects!)]

New movies are also always in the works, but there are constraints that we cannot avoid and, beginning this month, a recent slow down. The last time we uploaded any was in December. For the full details, here's what I've been told:

S@H is searching only 4 new tiles right now -- they were scanned October & November and uploaded in December.

JSC CDL (Cosmic Dust Lab) are switched over to cosmic dust work at present, during this time they do not scan aerogel (6 months cosmic dust, 6 months aerogel). We are helping out by scanning some tiles here in Berkeley. Currently, we are having some much needed maintenance work done on our microscope, and then we will resume scanning.

CDL will switch back to Stardust aerogel in the spring, and at that time both Berkeley and JSC will scan tiles. One very important step has to happen before a tile is scanned: each tile has to be very carefully removed from the Stardust Collector Tray and installed in it's new permanent holder. Tile I008 has recently been extracted and will be scanned in Berkeley. That tile has not been previously searched during Phases 1 - 3.


Otherwise, I don't think we removed any old movies and, as you probably know, we purposely wanted dusters to re-examine Phase 1 & 2 movies because of the new search image we put forth in Phase 3 (see discussion here). I don't see any reason why we would remove the new Phase 3 movies either (although I'll inquire). But overall, we realize, for the veterans at least, that seeing old movies is not all that exciting, but for scientific reasons, it had to be done (re-examination of the old Phase 1 & 2 movies). I'm also pretty sure the VM launches a randomly selected movie at each start and randomly selects each following movie, but let me check.

Thanks again,

Dan

DanZ
Site Admin
Posts: 772
Joined: Fri Feb 27, 2009 3:44 pm
Location: Berkeley, CA
Contact:

Re: Variable speed auto-scan?

Post by DanZ » Wed Jan 26, 2011 4:41 pm

jdrockster wrote:After viewing many of the same old movies many times(not calibration movies), their distinctive features are quite obvious. It would be more interesting if a random generated number would start us at tile x,y,z quadrant a,b,c, etc. at each login. It was very exciting at the beginning of phase 3 to find a potential track, click it, and see that it was only viewed by a few people, and I was the first to see it. Many of those movies have still only been viewed by a few "DUSTERS". Did you remove or change those movies?
Well what do you know! Apparently the VM was set to look at just 4 new tiles for a while. That would explain why someone doing a considerable amount of dusting might start seeing some repeats. We will certainly re-enable the other older movies again (including those from Phase 3) as we optimize our searching strategy in Phase 4. Stay tuned!

Dan

nitram819
Posts: 90
Joined: Mon Mar 02, 2009 9:38 am
Location: Quebec, Canada

Re: Variable speed auto-scan?

Post by nitram819 » Thu Jan 27, 2011 11:09 am

I for one certainly would not mind going through the old movies. I had found unreported tracks in old (phase 1-2) movies during phase 3 , so the chances are still there.

And I hear that a new phase will be starting soon? I was curious to know when it will start approximately. I assume that it will only begin when JPL will again start scanning new tiles again, so, next spring maybe?

Martin

DanZ
Site Admin
Posts: 772
Joined: Fri Feb 27, 2009 3:44 pm
Location: Berkeley, CA
Contact:

Re: Variable speed auto-scan?

Post by DanZ » Tue Feb 22, 2011 1:19 pm

nitram819 wrote:And I hear that a new phase will be starting soon? I was curious to know when it will start approximately. I assume that it will only begin when JPL will again start scanning new tiles again, so, next spring maybe? - Martin
We really thought it was going to be January Martin, but other priorities, bugs, etc. I'm hoping for March! I will inquire with the team and see if that's a good guess.

Dan

jdrockster
Posts: 27
Joined: Wed Aug 16, 2006 11:14 am
Location: Northern California

Re: Variable speed auto-scan?

Post by jdrockster » Wed Mar 16, 2011 3:04 am

Dan,Thanks for following up on the tiles. I saw many new (old) movies from phase 1& 2 with this session. The other four tiles were repeating so frequently that I knew something must be wrong. I am sure many old tiles could use another look with the criterion established in phase 3. I will get a list of the early phase 3 movies that I marked with potential tracks. Many were badly focused and may have been re-scanned.

Do people who score in the top 100 still get to visit the lab? I live in Sparks Nevada but I visit family in S.F. frequently. I would love to see where this project happens. I retired as a materials/research engineer with the Dept. of Transportation and worked in many labs.
Dust2Dust...Dust4Dust!

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

Re: Variable speed auto-scan?

Post by greuti » Wed Mar 23, 2011 4:24 am

Hi,

Do you know, from time to time there're blog entries on the main page of Stardust@home?! Like this one from December 10, 2010: http://stardustathome.ssl.berkeley.edu/blog/?p=165

So we are viewing currently only two tiles, which are tile I003 and I008.

"I am sure many old tiles could use another look with the criterion established in phase 3" I am sure we have already clicked at anything that looks like a track.

If it's worth looking again and again at the same old movies, my suggestion would be to go online with two databases of focus movies. One that includs all movies from any phases or tiles of interest and one that only contains two or more new scanned tiles that frequently would be renewed! That would mean after the whole pool of dusters have viewed (all) the new scanned tiles, i.e. movies, in the first/early days/weeks of the release, they could then switch to the old movies when they're waiting for the next brand-new movies.

DanZ
Site Admin
Posts: 772
Joined: Fri Feb 27, 2009 3:44 pm
Location: Berkeley, CA
Contact:

Re: Variable speed auto-scan?

Post by DanZ » Wed Mar 30, 2011 6:31 pm

jdrockster wrote:Dan,Thanks for following up on the tiles. I saw many new (old) movies from phase 1& 2 with this session. The other four tiles were repeating so frequently that I knew something must be wrong. I am sure many old tiles could use another look with the criterion established in phase 3. I will get a list of the early phase 3 movies that I marked with potential tracks. Many were badly focused and may have been re-scanned.
Glad it's gotten better. There's been quite a bit of juggling of the movies that are currently playing lately, but it should remain stable for a bit - until we get to Phase 4 (which is looking good for April, perhaps May).
jdrockster wrote:Do people who score in the top 100 still get to visit the lab? I live in Sparks Nevada but I visit family in S.F. frequently. I would love to see where this project happens. I retired as a materials/research engineer with the Dept. of Transportation and worked in many labs.
Let me see if I can make it work for you JD. Send me a PM with possible dates.

Cheers!

Dan

philipgm
Posts: 1
Joined: Sun Aug 14, 2011 2:21 pm

Re: Variable speed auto-scan?

Post by philipgm » Sun Aug 14, 2011 2:38 pm

I use the iPad to view and the variable speed would be wonderful.
Cheers
Philip

DanZ
Site Admin
Posts: 772
Joined: Fri Feb 27, 2009 3:44 pm
Location: Berkeley, CA
Contact:

Re: Variable speed auto-scan?

Post by DanZ » Fri Sep 02, 2011 7:42 pm

jdrockster wrote:I like the concept of the auto-scan but rarely use it. I would use the feature more if I could control the speed of the device. I don't know how difficult it is to add such a feature; (variable speed), but it would really help me if I could match the focus speed with my eyes and brain on any given day.
Here's the answer I got, meaning maybe later!

Yes, we could do this. It hasn't been high on my priority list. Maybe after we get phase 4 nailed down we can revisit some of these nice-to-have requests...

I'll keep it on the list of things for us to think about.

Dan

Post Reply