Light structures
Ioannis
Void sphere, with totally reflecting inner walls.
Second Structure:
Void torus, with totally reflecting inner walls.
(Point) Light source:
In sphere, off center and on center.
In torus, hidden from the observer and visible by observer.
Observer:
Placed inside void sphere and void torus, respectively.
Questions:
What would an observer see if one tries to look at the reflecting walls
while the light source is illuminating the inside of the structures?
Thanks much in advance,
--
I. N. Galidakis
http://users.forthnet.gr/ath/jgal/
Eventually, _everything_ is understandable
Sam Wormley
> First structure:
>
> Void sphere, with totally reflecting inner walls.
>
> Second Structure:
>
> Void torus, with totally reflecting inner walls.
>
> (Point) Light source:
>
> In sphere, off center and on center.
> In torus, hidden from the observer and visible by observer.
>
> Observer:
>
> Placed inside void sphere and void torus, respectively.
>
> Questions:
>
> What would an observer see if one tries to look at the reflecting walls
> while the light source is illuminating the inside of the structures?
>
> Thanks much in advance,
Depends on the distances... light has finite speed. If the distance is
much more than 100 light years... a human observer might not live long
enough to see a reflection... and it might be pretty weak.
Happy Hippy
> First structure:
>
> Void sphere, with totally reflecting inner walls.
>
> Second Structure:
>
> Void torus, with totally reflecting inner walls.
>
> (Point) Light source:
>
> In sphere, off center and on center.
> In torus, hidden from the observer and visible by observer.
>
> Observer:
>
> Placed inside void sphere and void torus, respectively.
>
> Questions:
>
> What would an observer see if one tries to look at the reflecting walls
> while the light source is illuminating the inside of the structures?
>
> Thanks much in advance,
looking? If so, a pattern of distorted
images of yourself in torus and
sphere, not?
More interesting, which I may build
sometime, a room totally of mirrors.
When you walk in, the door is so closely-
fitted as to be virtually invisible.
How long could you last in there?
John
Helpful person
macro...@internetcds.com
creation?
What is the structure of a very large light
wave while forming? Lets say a light year
long wave. If light is local then it should
take a year to form. I call this Light's
expansion. But what form does it take before
it grows to its full size? If its large it
will take significant time in order to form.
Light would take time to grow to its full
size. But before it reaches its full size
it must take a different form. It must
literally change shape to reach full size.
Androcles
"Ioannis" <morp...@olympus.mons> wrote in message news:1130786407.849507@athnrd02...
| First structure:
|
| Void sphere, with totally reflecting inner walls.
|
| Second Structure:
|
| Void torus, with totally reflecting inner walls.
|
| (Point) Light source:
|
| In sphere, off center and on center.
| In torus, hidden from the observer and visible by observer.
|
| Observer:
|
| Placed inside void sphere and void torus, respectively.
|
| Questions:
|
| What would an observer see if one tries to look at the reflecting walls
| while the light source is illuminating the inside of the structures?
Himself.
Androcles
donsto...@hotmail.com
John
***************
In a room with only mirrors - until you died of thirst.
Ioannis
Ο "Sam Wormley" <swor...@mchsi.com> έγραψε στο μήνυμα
news:fnu9f.497921$x96.495231@attbi_s72...
"light years"? Make those structures having radii of 10-20 meters and try
again. As for the light source, make that a 1000W incadescent bulb. What
would the observer see looking at different angles?
Surely as Androcles wrote, there exists at least ONE direction where the
observer will see his/her (distorted) reflection. I imagine this direction
would probably be against his feet if he/she is standing on the floor of the
sphere/torus, but is this the *only* direction where he/she will see
him/herself? Can he perhaps see a reflection of the light source somewhere?
Can anyone actually justify an answer?
Thanks,
Skywise
> First structure:
>
> Void sphere, with totally reflecting inner walls.
>
> Second Structure:
>
> Void torus, with totally reflecting inner walls.
>
> (Point) Light source:
>
> In sphere, off center and on center.
> In torus, hidden from the observer and visible by observer.
>
> Observer:
>
> Placed inside void sphere and void torus, respectively.
>
> Questions:
>
> What would an observer see if one tries to look at the reflecting walls
> while the light source is illuminating the inside of the structures?
>
> Thanks much in advance,
Sorry for the late reply, but this got me to tinkering around with
the idea in POVRAY (a graphical raytracer, not optical) as what you
describe is quite easy to simulate in that program.
Just playing around with the case of being inside a 100% reflecting
sphere shows a remarkable number of possible images just by changing
your location, the light source location, and where you look.
The light source moving around doesn't do much. If it's in the center
and you look at it, all you see is it. If it moves off to the side,
there appears to be two light sources except they get these long
multi-reflection tails going off into the distance not unlike what
you see when standing between two mirrors that face each other. What's
interesting is these tails are straight lines even though they are
reflections on a curved surface.
The biggest effect is the reflection of one's self. I am just using
a simple marble textured sphere as the 'observer' and placing the
camera on it. In the simple case of being about halfway from the
center of the sphere and looking towards the center, you see multiple
ringed reflections of yourself, concentric circles.
I can't wait to see what the inside of a torus looks like.
If you'll give me a few days, I'll render up a bunch of examples and
post them on my website. It'll look better if I do some animations but
that takes quite a while to render. Actually, it's not rendering,
it's raytracing, which is why it takes so long. Some of the little
320x240 size test renders I'm doing have tens of millions of rays
for just one frame.
Brian
--
http://www.skywise711.com - Lasers, Seismology, Astronomy, Skepticism
Seismic FAQ: http://www.skywise711.com/SeismicFAQ/SeismicFAQ.html
Quake "predictions": http://www.skywise711.com/quakes/EQDB/index.html
Sed quis custodiet ipsos Custodes?
Supernews sucks - blocking google, usenet.com & newsfeeds.com posts
platopes
Skywise wrote:
[snip]
>
> If you'll give me a few days, I'll render up a bunch of examples and
> post them on my website. It'll look better if I do some animations but
> that takes quite a while to render. Actually, it's not rendering,
> it's raytracing, which is why it takes so long. Some of the little
> 320x240 size test renders I'm doing have tens of millions of rays
> for just one frame.
>
> Brian
I'll just pop in here and say please do in fact post these things
of which you speak; they sound extremely cool.
You'll post in this thread when they're done?
p
Skywise
@z14g2000cwz.googlegroups.com:
>
> Skywise wrote:
> [snip]
>>
>> If you'll give me a few days, I'll render up a bunch of examples and
>> post them on my website. It'll look better if I do some animations but
>> that takes quite a while to render. Actually, it's not rendering,
>> it's raytracing, which is why it takes so long. Some of the little
>> 320x240 size test renders I'm doing have tens of millions of rays
>> for just one frame.
>>
>> Brian
>
> I'll just pop in here and say please do in fact post these things
> of which you speak; they sound extremely cool.
> You'll post in this thread when they're done?
I sure will. It may be a couple days though. I just got done with
the first set of renders for an animation of 60 frames and it took
over 2 1/2 hours. Since I have several different ones planned, well
you see how it adds up.
Brian
Jürgen Appel
> What would an observer see if one tries to look at the reflecting walls
> while the light source is illuminating the inside of the structures?
If ray optics approximation is valid:
Then you might want to look for information about circular (or spherical
respective torodial) billiards. The answer can be rather involved
depending on the observer's or the light source's position.
If the ray optics approximation does not hold, you have to solve the wave
equations with the boundary conditions given by a) the geometry of your
mirrors and b) the light source's position and shape c) the observer's
position and shape (and reflectivity).
After a long time (assuming at least some scattered light e.g. from the
observer) you will find the whole chamber at thermal equilibrium and the
observer will have the surface temperature of your light source and see
thermal blockbody radiation according to this temperature in every
direction.
Best regards,
Jürgen
--
GPG key:
http://pgp.mit.edu:11371/pks/lookup?search=J%FCrgen+Appel&op=get
platopes
Took a look at your website. Great pictures, especially lunar
eclipse.
CERN link is awesome.
A friend just told me about BOINC. Might be fun to set up a computer
just for that...
since computers can be found on the street!
p
Ioannis
news:11mj063...@corp.supernews.com...
[snip]
> If you'll give me a few days, I'll render up a bunch of examples and
> post them on my website. It'll look better if I do some animations but
> that takes quite a while to render. Actually, it's not rendering,
> it's raytracing, which is why it takes so long. Some of the little
> 320x240 size test renders I'm doing have tens of millions of rays
> for just one frame.
Thanks Brian. Yeah, when you are done, give the newsgroup a ring (or email
me). I will prepare a web-page with a description that links to your
animation/rendering shots.
Cheerio!
> Brian
> --
G=EMC^2 Glazier
it shows its structure and also how it is fanning out,and losing energy
the square of the distance. Seems strange that only three feet away from
a light bulb it is nine times dimmer. Bert
Scrim
"Skywise" <in...@oblivion.nothing.com> wrote in message
news:11mj063...@corp.supernews.com...
Sounds excellent, the animations in particular!
Other hollow mirrored shapes that come to mind: pyramids, cones,
ellipsoids.
Can someone set you up with distributed processing power to do the ray
tracing quickly and put together some extended 'fly by' type animations,
perhaps with a realistic human observer figure fitted with the camera in one
eye, so to speak? I'm sure people would be happy to contribute just to see
the results! Perhaps it would even be possible to fly around the shapes in
real time and find the best flight plan for visual effect!
Scrim
donsto...@hotmail.com
*****************************
But doesn't that depend on where the initial intensity was measured?????
G=EMC^2 Glazier
From that point measuring out an inch and I half is where I stopped
light. I"m proud to say my pictures are a million million times faster
than a strobe can do. Have been shown at EG&G,and at MIT strobe
department. Those that would like to see these pictures email me.
Photography is one of my first loves,and I have pictures that no one
else can take. Is my ego showing? I think so because I'm blushing.
Could just say I can tame the speed of light. Just think light can go
around the Earth 7 times in one second,and how do you divide less than
two inches into that. Its a very short tick Bert
AES
> > Just playing around with the case of being inside a 100% reflecting
> > sphere shows a remarkable number of possible images just by changing
> > your location, the light source location, and where you look.
> >
Just to throw in a somewhat relevant historical note:
One of the earliest inventions in the laser field was the elliptical
pump cavity: an elliptical cylinder with a high-reflectivity polished
and coated surface, pumping flashlamp along on focal line, and laser rod
along the other focal line, taking advantage of the fact that every ray
that leaves one focal point of an ellipse passes through the other focal
point after one bounce.
Light coming from points within a finite circle around the focal line
(or from a finite diameter light source along one focal line) doesn't,
however, in this case all pass through the same circle at the other
focal point or line. Some of it passes outside, i.e., some of the pump
light misses the rod, which is undesirable, since pump lamps unavoidably
have a finite diameter emitting region.
A guy in Germany -- Dieter Roess, I believe it was -- noted that if you
use an _ellipsoidal_ cavity (i.e., an ellipsoid of revolution), any
light ray departing from the cavity axis between one focal point and the
corresponding end of the cavity is guaranteed to pass through the cavity
at some point between the other focal point and the other end of the
cavity after a single bounce, although without any point-by-point
imaging. And although there is some slight aberration as you move off
axis, the same argument more or less holds true even for finite-diameter
light sources.
Net result: build an ellipsoidal pump cavity with the flashlamp running
from one focal point to the nearest cavity endpoint and the laser rod
similarly positioned at the opposite end, and you get noticeably
increased pumping efficiency -- though also a substantially larger and
more clumsy pump cavity structure. As a result, a few ruby lasers were
built this way, but the concept quickly died out (at least, AFAIK).
As an alternative, however, the same principal applied to an elliptical
cylinder -- including a circular cylinder -- says that if you shrink the
pump cavity down to where the pump lamp entirely fills the region from
one elliptical focal line to the cavity wall and the laser rod fills the
opposite space, you take advantage of the same principle, and get more
or less all of the lamp radiation striking the laser rod.
As a result "close-wrapped" pump cavities of this type became widely
used. The final design point, however, at least in some modern
lamp-pumped lasers, is to use inexpensive ceramic pump cavities of
pretty much arbitrary shape having diffusely reflecting walls, but made
with ceramics of such extremely high reflectivity that all the light
leaving the lamp eventually gets to the rod in any case.
(And of course the real bottom line is that pump lamps are obsolete at
this point; soon it will eventually all be DPSSs.)
hanson
"G=EMC^2 Glazier" <herbert...@webtv.net> wrote in message
news:22428-436...@storefull-3333.bay.webtv.net...
> Donstock The source that gave out the photons was
> surface electrons. From that point measuring out an inch
> and I half is where I stopped light.
>
Herbie, that point is where the edge of your film is, you incandescent
old fool! It shows up very clearly on your print. I have posted and
e-mailed that to you with the attachment of YOUR own pix.
Are you getting blind in you old age or do you feel the urge to con and
impress yourself after you take your Viagra? ahahaha... ahahaha...
>
[Hebie]
> Have been shown at EG&G,and at MIT strobe department.
>
ahahaha... AHAHAHA....
... and last time you posted your same tripe Sam Wormley told you:
http://groups.google.com/group/sci.physics/msg/aeb450ca359bef78
::::: I think, Herb, that you don't understand the physics (or lack of it)
::::: in your picture. It was nice of the quys at MIT to humor you.
>
[Hebie]
> I"m proud to say my pictures are a million million times faster
> than a strobe can do.
>
Herbie, cut off even more of the edge of your picture and repaste
in onto a black background and you can impress and convince
yourself and MIT that your pix is even faster now... AHAHAHA...
>
[Hebie]
> Those that would like to see these pictures email me.
>
By all means, do that Herbie. "A sucker is born every minute"...
ahahaha... See how many you'll get..... AHAHAHA.....
>
[Hebie]
> Photography is one of my first loves,and I have pictures that no one
> else can take.
>
.... Of course you do, Herbie.... ahahaha... That's because of your
own "Spin Theory" which gives your pixs that special Hebie-spin.
>
[Hebie]
> Is my ego showing? I think so because I'm blushing.
> Could just say I can tame the speed of light. Just think light can go
> around the Earth 7 times in one second,and how do you divide less than
> two inches into that. Its a very short tick Bert
>
[hanson]
ahahaha... "your ego"? Herbie.... No, Hebrie, don't be so humble.
It's your intelligence that comes thru so bright and shiny, even
with "Its very short tick".... Now, go and do what your Viagra is
supposed to be for, or there will be no more ticken and much less
ficken... ahahahaha..... AHAHAHA... Thanks for the laughs, Hebie.
ahahaha... ahahanson
... ahaha..
Ioannis
news:siegman-2490DA...@news.stanford.edu...
>
> > >
> > > Just playing around with the case of being inside a 100% reflecting
> > > sphere shows a remarkable number of possible images just by changing
> > > your location, the light source location, and where you look.
> > >
>
> Just to throw in a somewhat relevant historical note:
>
> One of the earliest inventions in the laser field was the elliptical
> pump cavity: an elliptical cylinder with a high-reflectivity polished
> and coated surface, pumping flashlamp along on focal line, and laser rod
> along the other focal line, taking advantage of the fact that every ray
> that leaves one focal point of an ellipse passes through the other focal
> point after one bounce.
Heheh! I was sure *somebody* would get the right idea about it, eventually
:-)
Updated web-page:
http://users.forthnet.gr/ath/jgal/math/Light.html
Skywise
news:To%af.31348$7m4....@fe2.news.blueyonder.co.uk:
> "Skywise" <in...@oblivion.nothing.com> wrote in message
> news:11mj063...@corp.supernews.com...
<Snipola>
>> If you'll give me a few days, I'll render up a bunch of examples and
>> post them on my website. It'll look better if I do some animations but
>> that takes quite a while to render. Actually, it's not rendering,
>> it's raytracing, which is why it takes so long. Some of the little
>> 320x240 size test renders I'm doing have tens of millions of rays
>> for just one frame.
<Snipola>
> Sounds excellent, the animations in particular!
> Other hollow mirrored shapes that come to mind: pyramids, cones,
> ellipsoids.
> Can someone set you up with distributed processing power to do the ray
> tracing quickly and put together some extended 'fly by' type animations,
> perhaps with a realistic human observer figure fitted with the camera in
> one eye, so to speak? I'm sure people would be happy to contribute just
> to see the results! Perhaps it would even be possible to fly around the
> shapes in real time and find the best flight plan for visual effect!
>
> Scrim
Unfortunately the software doesn't allow that. If I had a network
of computers then I could work on multiple animations at one time.
Or if I had somethign faster than this pIII/800mhz.
It's going to take longer than I anticipated. But I'm still working
on it.
For one thing, I origianlly was doing 60 frames per animation. I've
had to double that because the image changes so much at some points
that it just looks like a bunch of randomness strung together. When
viewing you couldn't readily see how the structure of the reflections
were changing as the viewer moved.
Some frames take a looooooong time to render. It's all the reflections.
Normally the software stops tracing a ray after only a few reflections.
But for this project I've set the max_trace_level to it's max of 256.
That slows the render by several times. Remember, this program is
actually tracing the light rays (from camera to scene) to compute the
image. No fancy pseudo reflections from a 3d render engine.
Ray tracing by it's nature is slow. Some of the runs are taking about
4 hours on my poor old machine. So I can't see quickly if I've made a
mistake, and I've made a couple.
In case anyone is interested, the home page for the program is here:
http://www.povray.org/
But I'm still plugging away at it. I have a render going on in the
background as I write this.
Skywise
and it took me a while to get all those megabytes uploaded.
Right now I only have the case of a sphere done. I'll be working
on the torus over the next week and add that to the page. I
expect that to look quite interesting.
The animations are a little weighty on the megabytes but nothing
is over 5 megs.
Hope everyone enjoys it!
http://www.skywise711.com/misc/reflections/reflections.html
AES
had a long list of answers to the related question:
"What do you see if you look at yourself
in a phase conjugate mirror?"
I can't find my original file, but some of the answers I remember
include:
* Yourself backwards.
* Yourself inside out.
* Yourself younger (since PC also involves time reversal)
and so on.
Skywise
@news.stanford.edu:
I remember this coming up once before. My asnwer was that you'd
see nothing. Just black.
Since a PC mirror reflects back along the incoming axis, let's
reverse the ray. As you look at the PC mirror, your sight line
goes from your pupil to the mirror and then straight back to
your pupil. Therefore what you'd see is a reflection of your
own eye, the darkness in the pupil. Unless there happened to
be some light striking the eye that reflected exactly along the
line of sight, I think all you'd see is black.
Ioannis
>
> OK, I finally got the webpage uploaded. I'm stuck with dial-up
> and it took me a while to get all those megabytes uploaded.
>
> Right now I only have the case of a sphere done. I'll be working
> on the torus over the next week and add that to the page. I
> expect that to look quite interesting.
Before you start with the Torus, it might be worth considering a few more
cases for the Sphereverse (x,y,z):
Observer at (0,0,0), light source at (0,10,0) or maybe at (0,10,10) and the
observer taking a 360 degree spin around the y-axis (himself), starting from
looking at himself against the closest wall point (0,0,-40). This would be
equivalent to looking around in a 360 degree fashion.
In the first case, the light source would be directly above him, so it would
be invisible throughout the entire spin. In the second case, the light
source will come to view when the observer's rotation is around 180 degrees
away from the initial view and then go off view again, as the observer
returns to his initial position, facing the closest wall point at (0,0,-40).
Another interesting case would be the same as the above, but with the
observer above the x-z plane some, as you have already done with the above
x-z variants on your site.
Maybe substitute any other suitable offset, instead of 10, if it works
better.
> The animations are a little weighty on the megabytes but nothing
> is over 5 megs.
>
> Hope everyone enjoys it!
>
> http://www.skywise711.com/misc/reflections/reflections.html
>
> Brian
--
I. N. Galidakis
Eventually, _everything_ is understandable