It is currently Sat, 24-08-19, 18:12 GMT

All times are UTC




Post new topic Reply to topic  [ 55 posts ]  Go to page Previous  1, 2, 3, 4
Author Message
 Post subject:
PostPosted: Fri, 25-09-09, 0:26 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4634
Location: Hamburg, Germany
Chris wrote:
Maybe just always average the radii at 0 and 360 degrees? It'd be good to emit warning when the radii differ.

--Chris


That's what I did already, yet it needs a little more code modification.
Anyway, the problem is completely gone. Looks nice and clean now.

A "quickshot" approach means just inserting
into the initial reading loop:

$r[$l] = $radius;
if ($long == 360.0) {$radius = $r[$l-180];}

It also eliminates the problem with less code change. But taking the average is perhaps a little "better"?
I am really not sure about this...

Quote:
A little emacs keyboard macro did most of the work...

I just inserted into my Perl script loop the single line

print "$lat,$long,$radius\n" if ($long == 0.0 || $long == 360.0);

which reproduces your above output.

Fridger


Top
 Profile  
 
 Post subject:
PostPosted: Fri, 25-09-09, 13:55 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4634
Location: Hamburg, Germany
Hi all,

meanwhile I have finalized my Perl script (Version 1.1) for the ascii CMOD generation of the latest scientific shape model of Phobos by Peter Thomas/ Cornell U.. I rewrote quite a bit of the Perl code, which now gets rid perfectly of that nasty seam that we discussed earlier.

As ChrisL has noted, the reason was due to the input shape model data, since the radii differ a little in a range of latitudes for lon=0 and lon=360 degrees. So I averaged the 2 values throughout, but print out a warning where they originally differed. This now looks like so:
Code:
> ./phobos_cmod.pl < m1phobos.tab

* lat = 50 deg: radius(lon=0) != radius(lon=360) => ( 10.0356, 10.0373)
* lat = 52 deg: radius(lon=0) != radius(lon=360) => ( 10.0051,  9.9945)
* lat = 54 deg: radius(lon=0) != radius(lon=360) => (  9.9701,  9.9474)
* lat = 56 deg: radius(lon=0) != radius(lon=360) => (  9.8698,  9.8557)
* lat = 58 deg: radius(lon=0) != radius(lon=360) => (  9.7666,  9.7633)
* lat = 60 deg: radius(lon=0) != radius(lon=360) => (  9.7445,  9.7431)
* lat = 62 deg: radius(lon=0) != radius(lon=360) => (  9.7540,  9.7546)
* lat = 64 deg: radius(lon=0) != radius(lon=360) => (  9.7382,  9.7398)
* lat = 66 deg: radius(lon=0) != radius(lon=360) => (  9.6929,  9.6941)
* lat = 68 deg: radius(lon=0) != radius(lon=360) => (  9.6340,  9.6344)
* lat = 70 deg: radius(lon=0) != radius(lon=360) => (  9.5875,  9.5822)
* lat = 72 deg: radius(lon=0) != radius(lon=360) => (  9.5775,  9.5762)
* lat = 74 deg: radius(lon=0) != radius(lon=360) => (  9.5821,  9.5824)
* lat = 76 deg: radius(lon=0) != radius(lon=360) => (  9.5946,  9.5953)
* lat = 78 deg: radius(lon=0) != radius(lon=360) => (  9.6108,  9.6122)


Furthermore, I have added proper references to the script and the generated phobos_ascii.cmod file, polished a bit my comments and recommend to work on the ascii CMOD with cmodfix from Celestia. Apart from the elimination of the seam, there were no visible changes of Phobos from Version 1.1 compared with the original Version 1.0 of my script.

+++++++++++++++++++++++++++++++++++++
Here is the updated phobos_cmod.pl Perl script (Version 1.1) for download, including for your convenience, the scientific input data, the generated phobos_ascii.cmod and the final binary phobos.cmod, as obtained after applying the proper cmodfix command. You can try it in Celestia, but note it is matched to my new 4k Phobos texture and NOT to the present 1k airbrushed map in the Celestia distribution. Still it looks not too bad either.

Everything (but my new 4k phobos texture) is in a zip archive. Have a click

http://www.celestialmatters.org/users/t ... Phobos.zip

+++++++++++++++++++++++++++++++++++++

Finally, for a nice joint impression about my new matching pair: the shape model and the 4k texture, I made a high-quality mpeg-4 (Xvid) video (18.5 MB) in changing light and changing views. It works both with the Windows Media player and with vlc. While it looks pretty good, you can also judge the amount of remaining polar pinch (north and south) under changing light conditions.

+++++++++++++++++++++
http://www.celestialmatters.org/users/t ... phobos.avi
+++++++++++++++++++++
Depending on your line speed, you may have to first download the video before playing. After all, it's big 1024x768 pixel size...

Next, I try to improve on the polar pinch problem in my way...let's see what will come out of it ;-)

Enjoy,
Fridger


Top
 Profile  
 
 Post subject:
PostPosted: Fri, 25-09-09, 14:08 GMT 
Offline
Site Admin
User avatar

Joined: Thu, 30-08-07, 22:52 GMT
Posts: 2727
Location: France, South, not far from Montpellier
Just tested a fresh conversion, no more seam as expected. Nicely done!
Now, I'm a bit perplex because of the locations; what should we trust? The model, the texture or the default .locs file within Celestia distribution? Some labels seems to be off but as I asked above, what should we trust??


Top
 Profile  
 
 Post subject:
PostPosted: Fri, 25-09-09, 14:17 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4634
Location: Hamburg, Germany
ElChristou wrote:
Just tested a fresh conversion, no more seam as expected. Nicely done!
Now, I'm a bit perplex because of the locations; what should we trust? The model, the texture or the default .locs file within Celestia distribution? Some labels seems to be off but as I asked above, what should we trust??


Christophe,

you should first of all accept that the texture imaging, the shape model and the location labels are based on measurements and computer fitting and thus inevitably suffer from various sources of uncertainty. Notably, since for now, I had to combine 3 different image maps, each with its own specific uncertainty (since they were matched to different models!!)

In the course of this project, I have explicitly checked some time ago that the location coordinates in Celestia 1.6.0 correspond to the latest officially acknowledged values. There is, however, another more recent ESA based shape model that has been matched explicitly to the latest HIRISE/SRC imaging data which I also mostly have used. Unfortunately that newest model is NOT (yet) available in form of data. So we have to live with the best we can get and that's what I have (of course) provided! ;-).

But I have already addressed these issues in my phobos 4k texture thread, havn't you read that??

As a reminder from my discussion, here is the latest image based on that unpublished model and the SRC data

Click for big
Image

Fridger


Last edited by t00fri on Fri, 25-09-09, 23:20 GMT, edited 1 time in total.

Top
 Profile  
 
 Post subject:
PostPosted: Fri, 25-09-09, 16:18 GMT 
Offline
User avatar

Joined: Sat, 13-10-07, 18:18 GMT
Posts: 373
Just wanted to say "Thank you" for your efforts here Doctor Schrempp...

Much appreciated at this end. :wink:


Top
 Profile  
 
 Post subject:
PostPosted: Sat, 26-09-09, 13:15 GMT 
Offline
Site Admin
User avatar

Joined: Thu, 30-08-07, 22:52 GMT
Posts: 2727
Location: France, South, not far from Montpellier
t00fri wrote:
...I have explicitly checked some time ago that the location coordinates in Celestia 1.6.0 correspond to the latest officially acknowledged values....


Right, so for now I guess we can say that the locations are our best reference? (less uncertainties than the over all process maps/model?)


Top
 Profile  
 
 Post subject:
PostPosted: Sat, 26-09-09, 13:45 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4634
Location: Hamburg, Germany
ElChristou wrote:
t00fri wrote:
...I have explicitly checked some time ago that the location coordinates in Celestia 1.6.0 correspond to the latest officially acknowledged values....


Right, so for now I guess we can say that the locations are our best reference? (less uncertainties than the over all process maps/model?)


I tend to agree. That was at least the basis for my Phobos project.


But what is methodically much peferable, is to use scientific data as they were published and documented instead of attempting to scale by trial and error certain parts of published textures for an "apparently better agreement"! A scientific analysis incorporates many different considerations, notably also about errors and transparency of sources of errors. If there is agreement among the team that better data are available, these will be published in due time. For us that's "update time", but not before.

Scientific measurements only have (lasting) value if they contain a careful analysis of uncertainties/errors along with the central data values. TOGETHER with these errors, the results are perfectly well-defined, otherwise NOT. Often, it is more tedious to arrive at reliable error values than to obtain the central data. ;-)

A problem in Celestia is that we still have not managed to invent a visually attractive way of displaying uncertainties of scientific data that we use. Hence we always refer to the "center values" of measurements. Many laymen forget that within the published errors also other orbits, positions etc of celestial objects are EQUALLY probable! It is NOT correct that the center values of data that we always use for Celestia are the most probable values. This becomes even more subtle and relevant, if the frequent correlations between different data are taken into consideration.

Only in the presence of careful error estimates is it possible to make sensible statements about whether two different measurements of the same quantity are mutually compatible or not.

Fridger


Top
 Profile  
 
 Post subject:
PostPosted: Sat, 26-09-09, 14:49 GMT 
Offline
Site Admin
User avatar

Joined: Thu, 30-08-07, 22:52 GMT
Posts: 2727
Location: France, South, not far from Montpellier
t00fri wrote:
...A problem in Celestia is that we still have not managed to invent a visually attractive way of displaying uncertainties of scientific data that we use...


Yep; your .Sci package need to address this problem. Now I guess it cannot be a global display for everything on screen. I suppose the best would be the display of uncertainty per object selected IF the feature is ON. Now this is truly a not simple topic, so later we will need to discuss it unless you have already precise ideas on the topic...


Top
 Profile  
 
 Post subject:
PostPosted: Sat, 26-09-09, 15:06 GMT 
Offline
Site Admin
User avatar

Joined: Fri, 31-08-07, 7:01 GMT
Posts: 4634
Location: Hamburg, Germany
ElChristou wrote:
t00fri wrote:
...A problem in Celestia is that we still have not managed to invent a visually attractive way of displaying uncertainties of scientific data that we use...


Yep; your .Sci package need to address this problem. Now I guess it cannot be a global display for everything on screen. I suppose the best would be the display of uncertainty per object selected IF the feature is ON. Now this is truly a not simple topic, so later we will need to discuss it unless you have already precise ideas on the topic...


Certainly. I am thinking about an error display since a long while and I do know some things that I want to AVOID in this context ;-) . One foreground constraint is clearly that a display of errors should BOTH carry some quantitative information and should be pleasing to look at. Notably, it must not disrupt the beauty of the rendering of the Universe that one has set out to do in the first place...

Surely, an error display must remain optional and would not be a default setting ;-)
Fridger


Top
 Profile  
 
 Post subject:
PostPosted: Sat, 26-09-09, 15:52 GMT 
Offline
Site Admin
User avatar

Joined: Thu, 30-08-07, 22:52 GMT
Posts: 2727
Location: France, South, not far from Montpellier
Ok let's keep this for later, else Mona will turn crazy!


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 55 posts ]  Go to page Previous  1, 2, 3, 4

All times are UTC


Who is online

Users browsing this forum: No registered users and 3 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group