"Tek" <tek### [at] evilsuperbraincom> wrote:

>
> The only squashing effect I'm aware of is when viewing through a telescope,
> which is the same flattening effect you get with a zoom lens, this image has
> a non-zoomed field of view so you wouldn't see squashing.
>
> The squashing is because the more you zoom in, the narrower your field of
> view, the closer to parallel the edges of your view become, so the less
> perspective you have in the image. With a wide field of view objects further
> away get a lot more perspective, so get smaller quicker, a zoomed lens keeps
> things much more similar sizes.

Hmm, that *would* seem to be the case, even in the situation I'm thinking of.
Maybe I'm wrong about the whole idea. But I'm imagining, say, the planet Mars:
If it were perhaps far larger in actual size (to take up an area in the sky two
or three times the visible diameter of the Moon, for example) while still being
so far away, it seems to me that the 'perspective' of the 'closer' and 'farther'
parts of the disc would be foreshortened. Based on the idea that, relative to me
as the observer at such a distance, the 'distance difference' between the closer
and farther parts would be quite small, relatively speaking.  I guess I just
need to do an experiment, to prove or disprove it to myself. (I have an image in
my mind of what I think it should look like--but maybe that's a false image!)

There's actually a 'down-to-earth' example that I've looked at, which *seems* to
show this effect:  It's a giant crane hoist (a *really* big one) at a Navy yard
near me. It has an elevated 'track' (that the crane runs on) which is probably a
good 300-feet long. When that track is rotated to be 'in line' with my line of
sight, it's near and far ends appear to be just about the same size
visually--almost no perspective distortion, in other words. (And I can only view
it from far away--not close enough to show the normal perspective distortion--so
an unfortunate side effect is that it 'looks' quite small.)

Time to experiment! :-)

Ken

Post a reply to this message

> "Tek"<tek### [at] evilsuperbraincom>  wrote:
>
>>
>> The only squashing effect I'm aware of is when viewing through a telescope,
>> which is the same flattening effect you get with a zoom lens, this image has
>> a non-zoomed field of view so you wouldn't see squashing.
>>
>> The squashing is because the more you zoom in, the narrower your field of
>> view, the closer to parallel the edges of your view become, so the less
>> perspective you have in the image. With a wide field of view objects further
>> away get a lot more perspective, so get smaller quicker, a zoomed lens keeps
>> things much more similar sizes.
>
> Hmm, that *would* seem to be the case, even in the situation I'm thinking of.
> Maybe I'm wrong about the whole idea. But I'm imagining, say, the planet Mars:
> If it were perhaps far larger in actual size (to take up an area in the sky two
> or three times the visible diameter of the Moon, for example) while still being
> so far away, it seems to me that the 'perspective' of the 'closer' and 'farther'
> parts of the disc would be foreshortened. Based on the idea that, relative to me
> as the observer at such a distance, the 'distance difference' between the closer
> and farther parts would be quite small, relatively speaking.  I guess I just
> need to do an experiment, to prove or disprove it to myself. (I have an image in
> my mind of what I think it should look like--but maybe that's a false image!)
If Mars was that big, you'd be able to see it's curvature, and Earth 
would orbit around Mars, not the Sun... In that case, mars would be 2 or 
3 times bigger than the Sun, and MUCH, MUCH more massive. 8 to 27 times 
the volume, and a larger density.
>
> There's actually a 'down-to-earth' example that I've looked at, which *seems* to
> show this effect:  It's a giant crane hoist (a *really* big one) at a Navy yard
> near me. It has an elevated 'track' (that the crane runs on) which is probably a
> good 300-feet long. When that track is rotated to be 'in line' with my line of
> sight, it's near and far ends appear to be just about the same size
> visually--almost no perspective distortion, in other words. (And I can only view
> it from far away--not close enough to show the normal perspective distortion--so
> an unfortunate side effect is that it 'looks' quite small.)
>
> Time to experiment! :-)
>
> Ken
>
>

Our perseption of perspective depends heavily on the distance that 
separate us from what we observe, and the relative depth compared to 
that distance.

In a perfectly transparent medium, any sphere whose apparent diameter is 
the same will look the same. Be it a 1 cm marble at 1 m or a 1 000 000 
km thing at 100 000 000 km.  In both cases, the diameter is 1% of the 
distance, and both will have the same apparent curvature.

In the example of the crane, you may be at about 5000 feet (a little 
under a mile) fom it.
The close part is at 5000 feet, and the far point is at 5300 feet. A 
difference of about 6%.
If you are near it, say at around 100 feet and on some scafolding, the 
close point is at 110 feet and the far end at 400 feet. A diference of 
72.5%.


Alain

Post a reply to this message

Alain <aze### [at] qwertyorg> wrote:


> > But I'm imagining, say, the planet Mars:
> > If it were perhaps far larger in actual size (to take up an area in the sky
> > two or three times the visible diameter of the Moon, for example) while
> > still being so far away, it seems to me that the 'perspective' of the
> > 'closer' and 'farther' parts of the disc would be foreshortened. Based on
> > the idea that, relative to me as the observer at such a distance, the
> > 'distance difference' between the closer and farther parts would be quite
> > small, relatively speaking.

> If Mars was that big, you'd be able to see it's curvature...

Yeah, that's probably true; I came to the same conclusion.

So I finally did a test(!), using a 'typical' camera angle of 67-deg. in
POV-Ray. (No telephoto, in other words.)  And the result is...I don't see much
if *any* perspective difference between small 'nearer' objects and large
'farther' objects. (If it's there, it's very subtle.)  Hmm, not at all the
result that I was expecting. I guess it's time to put this idea to rest. :-(

Or else 'cheat' the scene, to get what I want!  ;-P

Ken

Post a reply to this message

"Kenneth" <kdw### [at] earthlinknet> wrote in message 
news:web.4bdfacfc3d08c812ae92d9930@news.povray.org...
>
> So I finally did a test(!), using a 'typical' camera angle of 67-deg. in
> POV-Ray. (No telephoto, in other words.)  And the result is...I don't see 
> much
> if *any* perspective difference between small 'nearer' objects and large
> 'farther' objects. (If it's there, it's very subtle.)  Hmm, not at all the
> result that I was expecting. I guess it's time to put this idea to rest. 
> :-(

As Alain pointed out, there's no difference, and there can't be a 
difference. The camera doesn't know how big it is, so all sizes are 
relative, something 10m wide 100m away is identical to something 10km wide 
100km away.

> Or else 'cheat' the scene, to get what I want!  ;-P

This is very common in illustrations and movies: they get a telephoto image 
of the moon or a telescope image of a planet and stick it in the background 
of a non-telephoto scene, creating artificial flattening.

Personally I like the fact that ray tracers can do stuff more realistically. 
I did some test renders using real measurements of the size of planets 
viewed from orbiting moons, it's surprising how small the planet appears 
with normal camera angles. Even Jupiter viewed from some of its closer moons 
is pretty small. But then if you zoom the camera in you get the nice 
flattening effect and a very interesting image. I wanted something less 
realistic for my IRTC scene but it's nice to know what looks real. 
Unfortunately I don't have the test scene on this PC, so I can't show you, 
doh!

-- 
Tek
http://evilsuperbrain.com

Post a reply to this message

High!

Tek wrote:

> Personally my favourite is warpgate, closely followed by skyship and 
> dustland.

I like "Dustland" most... it's classical space art! Did you use an 
ordinary heightfield or did you model a whole isosurface moon?

See you in Khyberspace!

Yadgar

Now playing: The Kick Inside Of Me (Simple Minds)

Post a reply to this message

news:4be00b8f$1@news.povray.org...
>
> I like "Dustland" most... it's classical space art! Did you use an 
> ordinary heightfield or did you model a whole isosurface moon?

I started writing a macro to make a nice spherically-curved triangle mesh 
with varying polygon density, then I remembered that I couldn't spend too 
long on any of these scenes because I had to do so many for my irtc concept, 
so I abandonned that and went with a boring standard height_field! :)

-- 
Tek
http://evilsuperbrain.com

Post a reply to this message

"Tek" <tek### [at] evilsuperbraincom> wrote:

> > Or else 'cheat' the scene, to get what I want!  ;-P
>
> This is very common in illustrations and movies: they get a telephoto image
> of the moon or a telescope image of a planet and stick it in the background
> of a non-telephoto scene, creating artificial flattening.

Yes! In fact, this is probably where my wacky idea came from--old sci-fi
illustrations--and thinking that that was how the real world worked!! It's been
embedded in my mind for a *long* time.

Hmm, it makes me wonder about my 'big crane' example; I need to go and look at
it again...armed with my 'new knowledge'.  ;-)

Ken

Post a reply to this message