Some quick previews... [Large graphics inside]

General discussion about Celestia that doesn't fit into other forums.
Topic author
rthorvald
Posts: 1223
Joined: 20.10.2003
With us: 20 years 11 months
Location: Norway

Post #101by rthorvald » 02.10.2006, 00:03

fsgregs wrote:I now am collecting at least a month's worth of desktop wallpaper.


Thanks, i am glad you like it.

But, i would appreciate criticism in this thread, too. All these scenes are
works in progress, and could use fresh views. Such comments earlier
on has been to great help (ref. the city, for example).

While i am at it, here are a couple of pics from what i have worked on this
weekend: This is two different points in the path of the Grand Tour probe from Ran II,
now considerably tuned up: The first in orbit around the final stop, the
second during descent
of an onboard probe (after visiting all the outer worlds):

... The probe floats around in the atmosphere for 20 minutes before
disappearing (after 32 years of voyage!)

Image

Image

Writing a natural-looking ZYZ for this entirely by hand (depicting slingshot
/ gravity assists by 4 planets, several of which needed to be within centimeters to fit
with the animations) was not trivial!
Image
- rthorvald
Image

bh
Posts: 1547
Joined: 17.12.2002
With us: 21 years 9 months
Location: Oxford, England

Post #102by bh » 02.10.2006, 08:02

Err...cripes!
regards...bh.

Reiko
Posts: 1119
Joined: 05.10.2006
Age: 41
With us: 18 years
Location: Out there...

Post #103by Reiko » 05.10.2006, 12:00

I love that Hroenn pic. :)

Avatar
Chuft-Captain
Posts: 1779
Joined: 18.12.2005
With us: 18 years 9 months

Post #104by Chuft-Captain » 05.10.2006, 17:46

Hi Runar,

Sorry, no criticisms of the images themselves, just their sizes. :)
I'm wondering what you use to capture your images (or what compression you're using), because they are much larger than they need to be.
Perhaps you should process them with an image editor before posting.
eg. I opened each of your images above in my photo-editor, and then saved each one...

Code: Select all

             Before     After 
shore.jpg    410K     80K
thaeg.jpg    504K     90K
thcity.jpg    216K     38K
tour1.jpg    520K      75K
tour2.jpg    291K      44K


About an 80% saving!
Here's one for quality comparison (80K): [EDITED to display 80K file as an image instead of thumb]
Image

Your threads would be a lot easier to read if the images didn't take forever to download (and you'd probably save on your own diskspace as well).
Last edited by Chuft-Captain on 06.10.2006, 01:35, edited 1 time in total.
"Is a planetary surface the right place for an expanding technological civilization?"
-- Gerard K. O'Neill (1969)

CATALOG SYNTAX HIGHLIGHTING TOOLS LAGRANGE POINTS

Topic author
rthorvald
Posts: 1223
Joined: 20.10.2003
With us: 20 years 11 months
Location: Norway

Post #105by rthorvald » 05.10.2006, 19:28

Chuft-Captain wrote:Perhaps you should process them with an image editor before posting.
eg. I opened each of your images above in my photo-editor, and then saved
each one...


Hum.
The screendumps are originally PNGs, which i open in Photoshop for cropping
and then re-save with a modest JPEG compression.

Yes, the thread would be easier to read if i used heavier compression - or if i
used thumbnails. Which i would naturally do in any usual forum discussion.
But: this thread is entirely about the images. So, for this spesific
purpose, to me it makes no sense at all to reduce the quality further.

- rthorvald
Image

Avatar
Chuft-Captain
Posts: 1779
Joined: 18.12.2005
With us: 18 years 9 months

Post #106by Chuft-Captain » 06.10.2006, 01:29

So, for this spesific
purpose, to me it makes no sense at all to reduce the quality further.
That's fair enough, but did you click on the thumb posted above and compare the quality of my 80K fullsized image with your original 400K version?
(Note: I'm not suggesting changing the image dimensions or using thumbs. (I haven't changed the dimensions of your image, just it's physical filesize on disk)
I for one cannot see any significant reduction in quality in the 80K version ... (after all I didn't explicitly apply any additional compression, but simply opened it and saved it).

It's up to you whether you apply a stronger compression, but IMHO an 80% reduction in filesize without loss of quality or reduction in image dimensions should also be possible for you. Perhaps you should see what happens if you also open and save for a second time. (Maybe the conversion to JPG from PNG isn't happening properly the first time).

Smaller filesize (not dimensions) would also make your threads more popular and reduce the load on your webserver.
"Is a planetary surface the right place for an expanding technological civilization?"
-- Gerard K. O'Neill (1969)

CATALOG SYNTAX HIGHLIGHTING TOOLS LAGRANGE POINTS

Topic author
rthorvald
Posts: 1223
Joined: 20.10.2003
With us: 20 years 11 months
Location: Norway

Post #107by rthorvald » 06.10.2006, 02:35

Chuft-Captain wrote:I for one cannot see any significant reduction in
quality in the 80K version ... (after all I didn't explicitly apply any
additional compression, but simply opened it and saved it).
You are right in that in your example the difference isn??t too great.
My copy is a tad sharper than yours, but not by very much.

Please note, however, that you can??t expect this to be the case in
general; all images are different (amount of detail, variation between
neighboring pixels, color ranges etc)..

I do know a little bit about image manipulation after working in the field
professionally for more than twenty years (admittedly in a real darkroom
before the mac arrived) , and can assure you that if this was not so, we
would use high jpg compression for everything. But we don??t. Rather,
we tend to avoid it.

An example:
[1] Original screenshot, no JPG compression:
Image

[2] My usual compression rate applied, bringing filesize down from 284 to 52 kb
Image

[3] Bringing the copy [2] down to a comparable filesize (from your
example) by re-saving my jpg like you did - here fom 52 to 16 kb, i.e. less than 80%:
Image

As you can see, the difference is considerable between #2 and #3. Actually,
it is far greater than between #1 and #2 even though the variation in
pure filesize should indicate the opposite.


PS: I don??t have the original PNG doc from your example anymore, or i
would have performed the test on that one. But take a set of random
pictures through this, and more often than not, the end result will be bad.

Chuft-Captain wrote:Perhaps you should see what happens if you also
open and save for a second time
What happens is that you destroy more information. By re-saving a jpg
file you introduce more artifacts, in effect garbling the image.

Chuft-Captain wrote:Maybe the conversion to JPG from PNG isn't
happening properly the first time)

It happens properly indeed. My original tends to be around 1.3 to 1.4 MB,
while the copy ends up - usually - between 100 and 200 KB.

- rthorvald
Image

Avatar
Chuft-Captain
Posts: 1779
Joined: 18.12.2005
With us: 18 years 9 months

Post #108by Chuft-Captain » 06.10.2006, 03:25

Personally I would be just as impressed by #3 as #2 as it still conveys the nature of your work, even if with some artifacts, and I know it's going to look better in Celestia anyway!

Obviously presenting your work without loss of quality is important to you, and that's your decision, and in that light your comments are all valid, however, as I'm not simply trying to be negative, but constructive, here's a technique that you may wish to use to make your threads more accessible to narrowband users, while still allowing the higher quality JPG to be accessed with a click...

In both cases below I'm using the lower quality version as a large thumbnail linking to the higher quality version. (Of course youe could use smaller thumbs if you want)

Code: Select all

[url=http://runar.thorvaldsen.net/celestia/temp/testhigh.jpg][img]http://runar.thorvaldsen.net/celestia/temp/testlow.jpg[/img][/url]

Click on the image for a higher quality version
Image

Code: Select all

[url=http://runar.thorvaldsen.net/celestia/temp/shore.jpg][img]http://img157.imageshack.us/img157/5493/shorepm6.jpg[/img][/url]

Click on the image for a higher quality versionImage

Regards
CC
"Is a planetary surface the right place for an expanding technological civilization?"
-- Gerard K. O'Neill (1969)

CATALOG SYNTAX HIGHLIGHTING TOOLS LAGRANGE POINTS

Topic author
rthorvald
Posts: 1223
Joined: 20.10.2003
With us: 20 years 11 months
Location: Norway

Post #109by rthorvald » 06.10.2006, 04:03

Chuft-Captain wrote:I'm not simply trying to be negative, but constructive, here's a technique


Don??t worry, i understand and appreciate that.
Though the thumbnail trick is certainly useful in many circumstances, i
am not using it in this particular thread, since it isn??t purposeful here:

- It is obvious what the thread is about. It is also labelled with a warning
for "narrowband" users.

- Anyone that opens it, then, are going to click the full-size image anyway.
This means the thumbnail do not reduce any bandwidth cost at all:
it just adds to it, as it is an extra layer of data between you and the
image you open the thread to look at.

- Unless you are deleting your cache after each visit (which most people
don??t), the older images are already cached, so they do not eat up
any bandwidth either.

... So, the only instance where the thumbnail is useful here is for the rare
visitor that does not understand the "Large graphics inside" warning, or
alternately, is paranoid enough about his internet usage to flush his
browser cache daily. In light of this, i find it worthwile to present this material
in the way it is going to be viewed anyway. And, it is *my*
bandwidth, not shatters, that is taxed...

- rthorvald
Image

Avatar
Chuft-Captain
Posts: 1779
Joined: 18.12.2005
With us: 18 years 9 months

Post #110by Chuft-Captain » 06.10.2006, 04:25

I guess I'll just have to avoid your threads (even though I enjoy your images) in the future.
I tend to clear my cache at least weekly, so revisiting the thread can sometimes mean another 10-15 minute wait for it to load again.

BTW: A completely unrelated question. ..
In your Billow Maidens download (ML) [EDIT: DL'd over a year ago] you have a HTML page with celURL's in it. For me, none of these links don't seem to go to the correct locations. They seem to be off by considerable distances. Is it possible celestia's frames of reference have changed since you created that page? Have you had any feedback from others about this?
"Is a planetary surface the right place for an expanding technological civilization?"
-- Gerard K. O'Neill (1969)

CATALOG SYNTAX HIGHLIGHTING TOOLS LAGRANGE POINTS

Topic author
rthorvald
Posts: 1223
Joined: 20.10.2003
With us: 20 years 11 months
Location: Norway

Post #111by rthorvald » 06.10.2006, 04:38

Chuft-Captain wrote:I guess I'll just have to avoid your threads (even though I enjoy your images) in the future.
I tend to clear my cache at least weekly, so revisiting the thread can
sometimes mean another 10-15 minute wait for it to load again.
10 minutes? That seems a *lot*. There aren??t more than 3 to four pics on
a page, giving approx. 600 kb per page. Are you on dialup? Even so, it is slow:
many sites are bigger than that. How long does it take to open a typical newspaper website?

Chuft-Captain wrote:BTW: A completely unrelated question. ..
In your Billow Maidens download (ML) you have a HTML page with celURL's
in it. For me, none of these links don't seem to go to the correct locations.
They seem to be off by considerable distances. Is it possible celestia's
frames of reference have changed since you created that page? Have you
had any feedback from others about this?


Yes, i discovered that very early on. There is a thread about it here
somewhere, and a note on the Ran mini-site (the last page).

There is an error in certain versions of Celestia that makes huge
distances be calculated differently accross some versions. As i remember,
i - with the help of Selden - calibrated it to work correctly with the Windows
1.3.2 version, which was the current final release at the time. This was
incompatible with the similar OSX and Linux versions, as well as with all
older Windows versions.

This bug was fixed in later OSX versions, but i don??t know anything about
Linux ones.

- rthorvald
Image

Avatar
Chuft-Captain
Posts: 1779
Joined: 18.12.2005
With us: 18 years 9 months

Post #112by Chuft-Captain » 06.10.2006, 04:52

rthorvald wrote:10 minutes? That seems a *lot*. There aren??t more than 3 to four pics on
a page, giving approx. 600 kb per page.
That **subjective** estimate is based on page 7 of this thread (at least 2MB):

Code: Select all

             Before     
shore.jpg    410K     
thaeg.jpg    504K     
thcity.jpg    216K   
tour1.jpg    520K     
tour2.jpg    291K   



Yes, i discovered that very early on. There is a thread about it here
somewhere, and a note on the Ran mini-site (the last page).

There is an error in certain versions of Celestia that makes huge
distances be calculated differently accross some versions. As i remember,
i - with the help of Selden - calibrated it to work correctly with the Windows
1.3.2 version, which was the current final release at the time. This was
incompatible with the similar OSX and Linux versions, as well as with all
older Windows versions.

This bug was fixed in later OSX versions, but i don??t know anything about
Linux ones.

- rthorvald
I'm using win32 and Celestia 1.4.1. This may explain it, as at the time I downloaded RAN (prob version 1), the current version of Celestia would have been 1.3.2 I think.

I assume the latest RAN2 version will have celURL's compatible with 1.4.1 (unless there's still an incompatibility between OSX and XP?)

It looks as though repositioning of the Sun in 1.5.0 may also slightly affect ALL celURL's and bookmarks.
"Is a planetary surface the right place for an expanding technological civilization?"
-- Gerard K. O'Neill (1969)

CATALOG SYNTAX HIGHLIGHTING TOOLS LAGRANGE POINTS

Topic author
rthorvald
Posts: 1223
Joined: 20.10.2003
With us: 20 years 11 months
Location: Norway

Post #113by rthorvald » 07.10.2006, 01:58

Chuft-Captain wrote:That **subjective** estimate is based on page 7
of this thread
Anyways, 10 minutes suggests something is *wrong*. Perhaps a slow
connection between your and mine ISP.

But, you have given me something to think about.

A solution might be to convert older images to thumbs, and only keep the
current at full size. That would satisfy your (and others) concerns re
bandwidth, and still letting me work the way i prefer.

Chuft-Captain wrote:It looks as though repositioning of the Sun in 1.5.0
may also slightly affect ALL celURL's and bookmarks.

I *am* considering this, so, there will not be a Ran III release until well
after a finalized 1.5 is out.

- rthorvald
Image

Avatar
fsgregs
Posts: 1307
Joined: 07.10.2002
With us: 21 years 11 months
Location: Manassas, VA

Post #114by fsgregs » 08.10.2006, 04:02

It looks as though repositioning of the Sun in 1.5.0
may also slightly affect ALL celURL's and bookmarks.


WHOA!!! This is the first I've heard of this. I have HUNDREDS of CEL:URL's in use in my Educational journeys. Is there something about 1.5.0 that is going to make them no longer point to the spot I intend them to point to? :cry:

I sure hope not!! Tell me it ain't true!!!! :x

Frank

Avatar
Cham M
Posts: 4324
Joined: 14.01.2004
Age: 60
With us: 20 years 8 months
Location: Montreal

Post #115by Cham » 08.10.2006, 04:41

fsgregs,

sorry, but this is true. We must accept it, because now the sun will sit in the proper position, and there were some corrections related to the speed of light used by Celestia.
"Well! I've often seen a cat without a grin", thought Alice; "but a grin without a cat! It's the most curious thing I ever saw in all my life!"

Malenfant
Posts: 1412
Joined: 24.08.2005
With us: 19 years 1 month

Post #116by Malenfant » 08.10.2006, 07:02

Yep, the Sun's been in the wrong place (plus, it needs to be orbiting the solar system's barycentre now anyway).

I'd hope there'd be a conversion tool to convert old cel URLs to new ones, but it's a necessary change. The old location was set the way it was only because of an ugly fudge relating to parallax.
My Celestia page: Spica system, planetary magnitudes script, updated demo.cel, Quad system

Topic author
rthorvald
Posts: 1223
Joined: 20.10.2003
With us: 20 years 11 months
Location: Norway

Post #117by rthorvald » 08.10.2006, 14:27

Cham wrote:this is true. We must accept it, because now the sun will sit in the proper position, and there were some corrections related to the speed of light used by Celestia.


Hm, will this affect XYZ positioning, too? Will we have to modify *all* existing XYZ documents?

- rthorvald
Image

Avatar
selden
Developer
Posts: 10190
Joined: 04.09.2002
With us: 22 years 1 month
Location: NY, USA

Post #118by selden » 08.10.2006, 15:07

xyz trajectories and the other types of orbits continue to be relative to the bodies specified in the SSC catalogs that define them. e.g. Cassini's trajectory in Celestia currently is specified relative to the Sun, not the Solar System's barycenter. So is Saturn's orbit. Similarly, Titan's orbit is defined relative to Saturn, not Saturn's barycenter. As a result, Celestia's Cassini still flys by Titan following the same paths it always has.
Selden

Topic author
rthorvald
Posts: 1223
Joined: 20.10.2003
With us: 20 years 11 months
Location: Norway

Post #119by rthorvald » 08.10.2006, 15:13

selden wrote:Celestia's Cassini still flys by Titan following the same paths it always has.


Thanks: that??s the answer i was hoping for.

- rthorvald
Image

Avatar
selden
Developer
Posts: 10190
Joined: 04.09.2002
With us: 22 years 1 month
Location: NY, USA

Post #120by selden » 08.10.2006, 15:20

The problem with URLs is that they specify "universal" xyz coordinates, not relative ones, and the solar system is no longer located where it used to be :(

Chris is investigating ways to implement some form of backward compatibility, but it doesn't look easy.
Selden


Return to “Celestia Users”