Bug 63367 - Zoom level depends on window size
Summary: Zoom level depends on window size
Status: RESOLVED FIXED
Alias: None
Product: kstars
Classification: Applications
Component: general (show other bugs)
Version: unspecified
Platform: Compiled Sources Linux
: NOR normal
Target Milestone: ---
Assignee: kstars
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2003-08-28 13:28 UTC by Dik Takken
Modified: 2003-09-14 16:50 UTC (History)
0 users

See Also:
Latest Commit:
Version Fixed In:
Sentry Crash Report:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Dik Takken 2003-08-28 13:28:31 UTC
Version:           Recent CVS Snapshot (using KDE KDE 3.1.1)
Installed from:    Compiled From Sources
OS:          Linux

The zoom level should be independent on the size of the application window. As a result of this not being the case, I can't print an image of the whole night sky, because my display resolution is too low. Display resolution should not limit my viewing/printing in any way.
Comment 1 kstars 2003-08-28 16:17:12 UTC
Hello, 
 
Actually, the zoom level is not dependent on window size, it is fixed to the pixel 
size.  In other words, the Zoom Level is a floating-point number that describes the 
number of pixels subtended by an angle of one radian.   
 
So, to see the entire sky from a smaller window, you just need access to lower 
zoom levels.  As of now, the range of zoom levels is hard-coded, but we can 
certainly make it an adjustable option.  In the meantime, if you have a recent CVS 
version of KStars, you can easily modify the hard-coded limits in your local copy.  
In the file kstarsdata.h, find the "#define MINZOOM" line, and make that number 
smaller.  Then recompile and reinstall kstars. 
 
thanks for the report, 
Jason 
Comment 2 kstars 2003-09-13 02:19:38 UTC
Subject: kdeedu/kstars/kstars

CVS commit by harris: 

reduced MINZOOM to 200, which makes it possible to see the edges 
of the displayable area, even when the window is as small as 640x480.
Post-3.2, we will make MINZOOM and MAXZOOM adjustable parameters.

CCMAIL: 63367-done@bugs.kde.org, kstars-devel@kde.org


  M +1 -1      kstarsdata.h   1.65


--- kdeedu/kstars/kstars/kstarsdata.h  #1.64:1.65
@@ -65,5 +65,5 @@
 #define NCIRCLE 360   //number of points used to define equator, ecliptic and horizon
 
-#define MINZOOM 300.
+#define MINZOOM 200.
 #define MAXZOOM 1000000.
 #define DEFAULTZOOM 2000.


Comment 3 Dik Takken 2003-09-13 11:41:59 UTC
Thanks for the commit, 
 
Still, I think that the visible area should not depend on two parameters, but on only one. How does 
this relate to the Field of View feature? The idea of the FOV settings is to specify a fixed visible area of 
nxn degrees, to match the FOV of the telescope. This must be independent of window-size. If not, the 
user will have to find an answer to the question: Which window size matches that of my telescope?? 
Comment 4 kstars 2003-09-13 14:58:39 UTC
Subject: Re:  Zoom level depends on window size

On Saturday 13 September 2003 05:42 am, you wrote:
> Still, I think that the visible area should not depend on two
> parameters, but on only one. How does this relate to the Field of
> View feature? The idea of the FOV settings is to specify a fixed
> visible area of nxn degrees, to match the FOV of the telescope. This
> must be independent of window-size. If not, the user will have to
> find an answer to the question: Which window size matches that of my
> telescope??

No, the FOV subtends a fixed angular size.   So it will subtend a fixed 
number of pixels at a given Zoom level, regardless of the window size.

Making the Zoom level depend on window size instead of pixel size makes 
very little sense to me.  In that case, If I resize the window, I will 
effectively be zooming in and out.  As it is, making the window bigger 
has a much more intuitive effect: it shows you a larger fraction of the 
sky.  Isn't that what one wants when they make the window bigger?

Besides, which dimension would you use to determine the Zoom level, 
width or height?  It would look really strange and wrong to have a 
resize in one dimension dragging the sky along with it, while a resize 
in the other dimension just showed more (or less) sky (i.e., the 
current behavior).

Jason 
Comment 5 Dik Takken 2003-09-14 16:50:44 UTC
"No, the FOV subtends a fixed angular size. " 
 
Ok, so the FOV is defined to be a fixed area of the sky. We both agree on that. :) If I understand it 
correctly, the FOV can be either a subset or a superset of the sky that is displayed in the window. Is 
this correct? So, you could indicate the FOV by drawing a bounding circle around the centre of the 
window, for instance. In that case, resizing the window does not matter indeed. the FOV is a certain, 
defined part of the window. This is different from my definition of the FOV, that's why I didn't 
understand it. 
 
I was thinking that the FOV was a property of the window itself, restricting the window to display no 
more/less than the area defined by de FOV, no matter how big or small the window. In my view, the 
FOV *was* the window itself. Maybe you also understand my point of view now. In my point of view, 
resizing the window would not give you more sky to look at, only higher resolution. As a 
consequence of this, the Zoom level becomes meaningless. the FOV will govern the picture in stead of 
the Zoom level. 
 
"Besides, which dimension would you use to determine the Zoom level,  
width or height? It would look really strange and wrong to have a  
resize in one dimension dragging the sky along with it, while a resize  
in the other dimension just showed more (or less) sky (i.e., the  
current behavior)." 
 
You're right about this, it is a little less elegant, but you can use the diagonal of the window to cover 
a fixed angle. Most people will use more or less square windows anyway, so it will work fine.  
 
I think both your and my view are OK, maybe you can give the users choice between keeping the 
Zoom level fixed or keeping the FOV fixed. Both will work fine for most users, I think. 
 
Thanks, 
 
Dik 
Comment 6 kstars 2003-09-15 01:09:39 UTC
Subject: kdeedu/kstars/kstars

CVS commit by harris: 

Completed planned "arbitrary zoom" feature.  There is now a menu item to 
set the zoom level by specifying the angular size of the window.  This may 
also be of interest to the reporter of bug #63367, although it doesn't
exactly do what he was asking about...

(don't forget to "make install"; kstarsui.rc has changed)

CCMAIL: 63367@bugs.kde.org, kstars-devel@kde.org


  M +1 -0      kstars.h   1.61
  M +32 -0     kstarsactions.cpp   1.70
  M +2 -0      kstarsinit.cpp   1.71
  M +1 -0      kstarsui.rc   1.23
  M +15 -11    skymapdraw.cpp   1.35