tap water falling
Peter Q
shoot the water falling from my tap using my G3, at a close distance with
the built-in flash. And here is the outcome...
http://61.144.232.237:8000/Images/pictures/Water_Falling.JPG
I think it's very lovely, in a way..
Peter
dirtycow
"Peter Q" <pmo...@cogeco.ca> wrote in message
news:l7wJa.36205$b67.4...@read2.cgocable.net...
Could do with resizing, but it looks great! Almost looks like crystal or
ice. Nice one!
Matt
Zol.
is quite dare I say it `Pretty` :) Zol.
"Peter Q" <pmo...@cogeco.ca> wrote in message
news:l7wJa.36205$b67.4...@read2.cgocable.net...
Marcia Beinhauer
David G. Stinner
nice shot.
Frank
wrote:
I stopped as soon as I saw it was over 1mb!
--
Regards
Frank.
Peter Q
ps: I didn't resize it coz i wanna show the max detail possible...but it IS
over 1M in size :(
Peter
"Peter Q" <pmo...@cogeco.ca> wrote in message
news:l7wJa.36205$b67.4...@read2.cgocable.net...
Mike
2300x1700 pic is beyond me... only people that can see it are those with webTV
and giant screen tv's...
Vince R
> Thanks for all of your comments :)
> ps: I didn't resize it coz i wanna show the max detail possible...but it IS
> over 1M in size :(
>
> Peter
>
>
> "Peter Q" <pmo...@cogeco.ca> wrote in message
> news:l7wJa.36205$b67.4...@read2.cgocable.net...
> > To capture waterfall, people often uses a longer shutter time. I tried to
> > shoot the water falling from my tap using my G3, at a close distance with
> > the built-in flash. And here is the outcome...
> >
You must really have a great deal of time on your hands...
> >
> > I think it's very lovely, in a way..
> >
> > Peter
> >
> >
>
>
>
--
Vince
Peter Q
them first..
Well I thought the title suggests that it's "tap water falling"
>
> You must really have a great deal of time on your hands...
>
I am just trying to capture some interesting moments out of people's daily
life,
and yeah, it only took 1/250 sec to capture this one :P
Peter
Bob Crownfield
like crystal glass!
>
> Peter
Mike
slightest idea of what this is if you didn't tell them" suggests you have
captured something that might keep people wondering, thinking about the image
long after they've seen it. That's what every photographer (except Vince,
apparently) strives to do. If he needs to be told what every image he sees is
supposed to be, I doubt he is the audience you, nor many of us, are looking
for.
As for having too much time on your hands, well, photographers rarely have
enough time to pursue the images they want, and the fact you took the time to
capture this one speaks of inquisitiveness and dedication, not some sort of
lackardly laziness, as again Vince seems to try to suggest.
My little "blast" earlier was only in the size - which made it hard to look at
- not the image.
Good job.
Mike
Todd Walker
pmo...@cogeco.ca says...
Interesting shot. Next time use f/4.0 or smaller to avoid the chromatic
abberations.
--
________________________________
Todd Walker
http://twalker.d2g.com
Olympus E20
Canon G2
My Digital Photography Weblog:
http://twalker.d2g.com/dpblog.htm
_________________________________
Peter Q
Explain please
Thanks
"Todd Walker" <twalk...@hotmail.com> wrote in message
news:MPG.19618db2e...@news-server.jam.rr.com...
Shaw Goh (NetPlus Micro Computers)
Stephen H. Westin
> purple fringes around bright spots.
Which, I think, are blooming artifacts rather than lateral chromatic
aberration. Notice that the artifact is just as pronounced in the
center of the image as at the edges.
--
-Stephen H. Westin
Any information or opinions in this message are mine: they do not
represent the position of Cornell University or any of its sponsors.
Roland Karlsson
news:uznk7p...@graphics.cornell.edu:
> Which, I think, are blooming artifacts rather than lateral chromatic
> aberration. Notice that the artifact is just as pronounced in the
> center of the image as at the edges.
>
Nope - it is chromatic aberration.
Look at the white spots in the upper part - they are red at
the top and blue/purple at the bottom. If you look in the
lower parts (where there are not so many examples) it is
the other way arround.
Surprising though that the aberration is so visible so near
the center of the image. This is probably due to the
fact that they are greatly overexposed.
Roland
Joseph Miller
> Nope - it is chromatic aberration.
>
> Look at the white spots in the upper part - they are red at
> the top and blue/purple at the bottom. If you look in the
> lower parts (where there are not so many examples) it is
> the other way arround.
>
> Surprising though that the aberration is so visible so near
> the center of the image. This is probably due to the
> fact that they are greatly overexposed.
>
>
> Roland
I think the explanation is very different and to me much more
interesting. It is not an aberration of the camera, but a result of
dispersion in the water. Water has a variable index of refraction with
wavelength, just like glass, and this water is acting like a very poor
lens. If fact, in some areas it almost acts like a prism. That is why
the "aberration" is so visible near the center of the image, and I
predict stopping down the lens will have no effect on it, as long as the
exposure remains very short.
Joe
Bart van der Wolf
"Roland Karlsson" <roland_do...@bonetmail.com> wrote in message
news:Xns93A4BE0AE571Dro...@130.133.1.4...
In my opinion that shows it isn't CA as we know it, but a combination of
blooming and angle of incidence on the sensor. It only shows up clearly on
(near) over-exposed areas bordering on dark areas, so the leakage is
visible. All digicams suffer from it to a degree, and it usually disappear
with apertures smaller than F 4.0 (which would make the CA conclusion more
plausible, but it is just restricting the angles of incidence). Also
remember that longer wavelengths (including IR) penetrate the silicon
substrate much more and are relatively out of focus. An additional factor is
the different reconstructed resolution of Bayer CFA red and blue versus
green.
Bart
Roland Karlsson
> I think the explanation is very different and to me much more
> interesting. It is not an aberration of the camera, but a result of
> dispersion in the water. Water has a variable index of refraction with
> wavelength, just like glass, and this water is acting like a very poor
> lens. If fact, in some areas it almost acts like a prism. That is why
> the "aberration" is so visible near the center of the image, and I
> predict stopping down the lens will have no effect on it, as long as the
> exposure remains very short.
Yepp - you are right!!!
Creeps back to my dark wet hole in shame :)
Roland
Roland Karlsson
$e4fe...@news.xs4all.nl:
> In my opinion that shows it isn't CA as we know it, but a combination of
> blooming and angle of incidence on the sensor. It only shows up clearly on
We are both wrong ;(
Joseph Miller has the correct explanation.
Let us just sneak out through the backdoor
and pretend we never were here :)
Roland
Roger Halstead
wrote:
>Thanks for all of your comments :)
>ps: I didn't resize it coz i wanna show the max detail possible...but it IS
>over 1M in size :(
In that case you still need to set the display size to a normal screen
even if it's as high as 1280 X 1024. Without the formatting the photo
gets displayed at the normal screen resolution of 70 to 90 some dots
per inch which means it's several screens wide and several high.
Saved as a jpg and then displayed in paint shop pro (or most any other
image program) puts the entire shot on the screen which looks far more
impressive than the unformatted one.
Using the following
<CENTER> <IMG SRC="Water_falling.jpg" HEIGHT=1280 WIDTH=1024></CENTER>
Will center the photo in the browser window and display it with the
given dimensions without losing the original resolution.
Maintaining the same aspect ratio you can set it for any screen size,
or just use width as 100% for a full screen display.
Always include a warning when posting links to *large* images. Those
of us with fast connections usually don't mind, but those with dial up
are going to have to wait a longgggg time and will most likely give up
in disgust, or think something broke.
As far as size, you could resize the image to 1024 X 768, or 1280 X
1024 and the viewer will se no difference between that and displaying
the original as a full screen image as they are limited by the
resolution of their displays.
BTW, when viewed as a full frame image in its entirety it is a very
impressive shot.
Roger Halstead (K8RI EN73 & ARRL Life Member)
www.rogerhalstead.com
N833R World's oldest Debonair? (S# CD-2)
Joseph Miller
Actually, in this newsgroup it's quite natural to blame it on the
camera. We are all so focused (cough) on camera performance that
anything that looks a bit strange is almost automatically attributed to
the camera.
Joe
Bart van der Wolf
> "Bart van der Wolf" <bvd...@nospam.nl> wrote in news:3ef89145$0$49114
> $e4fe...@news.xs4all.nl:
>
> > In my opinion that shows it isn't CA as we know it, but a combination of
> > blooming and angle of incidence on the sensor. It only shows up clearly
on
>
> We are both wrong ;(
>
> Joseph Miller has the correct explanation.
I'm afraid not. The same phenomenon can be seen on (dry) tree branches
against the sky.
Bart
Bart van der Wolf
No need for that....
The prediction that stopping down the lens will have no effect on it, will
not hold (my prediction ;-)). What's more, water (unless in the shape of a
small droplet) doesn't have a refractive or dispersion index high enough to
cause the (internal reflection/dispersion) effect shown in such a large body
of water.
Bart
Todd Walker
says...
> Using the following
> <CENTER> <IMG SRC="Water_falling.jpg" HEIGHT=1280 WIDTH=1024></CENTER>
> Will center the photo in the browser window and display it with the
> given dimensions without losing the original resolution.
>
That is the absolute WORST thing you can do. It forces the viewer to
download the whole image (all 1 or 2 megs worth) but doesn't allow them
to see it at full resolution because you are constraining the
proportions in the HTML code. You are wasting their time by doing this.
Resize the image to 800x600 if you want the whole image to be visible in
(just about) everyone's browser at once and compress it so that the file
size is around 200k. You will lose a minimum of detail and your viewers
will like you much better.
J...@no.komm
Todd Walker <twalk...@hotmail.com> wrote:
>In article <l7wJa.36205$b67.4...@read2.cgocable.net>,
>pmo...@cogeco.ca says...
>> To capture waterfall, people often uses a longer shutter time. I tried to
>> shoot the water falling from my tap using my G3, at a close distance with
>> the built-in flash. And here is the outcome...
>>
>> http://61.144.232.237:8000/Images/pictures/Water_Falling.JPG
>>
>> I think it's very lovely, in a way..
>>
>> Peter
>>
>>
>>
>
>Interesting shot. Next time use f/4.0 or smaller to avoid the chromatic
>abberations.
How do you change the f-stop of the water?
--
<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <J...@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
Roland Karlsson
$e4fe...@news.xs4all.nl:
> I'm afraid not. The same phenomenon can be seen on (dry) tree branches
> against the sky.
Ah ... but in those case the branches are mostly in corners
and in those case it is chromatic aberration in the camera lens.
In this case it is with a high probabality chromatic
aberration in the water itself.
Roland
George Kerby
Xns93A4D7A19E6Aro...@130.133.1.4, "Roland Karlsson"
<roland_do...@bonetmail.com> wrote:
Roland, you really must stop flagellating yourself!
______________________________________________________________________
Posted Via Uncensored-News.Com - Still Only $9.95 - http://www.uncensored-news.com
<><><><><><><> The Worlds Uncensored News Source <><><><><><><><>
Frank ess
"Roger Halstead" <rdha...@tm.net> wrote in message
news:0r7hfv8vcqgobml7f...@4ax.com...
| On Mon, 23 Jun 2003 15:17:42 -0400, "Peter Q" <pmo...@cogeco.ca>
| wrote:
|
| >Thanks for all of your comments :)
| >ps: I didn't resize it coz i wanna show the max detail
possible...but it IS
| >over 1M in size :(
|
| In that case you still need to set the display size to a normal
screen
| even if it's as high as 1280 X 1024. Without the formatting the
photo
| gets displayed at the normal screen resolution of 70 to 90 some dots
| per inch which means it's several screens wide and several high.
|
Which my Opera browser display entire, after one click. (30%. If I
recall correctly)
Frank ess
Bart van der Wolf
I see the same if I take an image of a ball of crumpled aluminium foil. On
optical axis with a large aperture, lots of color fringes. Smaller aperture,
almost none.
http://www.dpreview.com/learn/Glossary/Digital_Imaging/Blooming_01.htm
describes the blooming part, but doesn't specifically address the angle of
incidence effect, nor the resolution differences between green and blue/red.
Bart
Roland Karlsson
ghost_...@hotmail.com:
> Roland, you really must stop flagellating yourself!
He he, it is very temporary. Soon I will be
the usual arrogant myself :)
Roland
Roland Karlsson
> http://www.dpreview.com/learn/Glossary/Digital_Imaging/Blooming_01.htm
This is a copy of what Phil has to say about this picture:
" Anti-blooming gates, while mostly successful (and certainly
for more modern sensors) blooming can still be a problem in
very extreme exposures (very bright edge against a virtually
black edge) and is typically visible as either a vertical
streak or white halo extending for several pixels. The effects
of blooming often amplify the visibility of chromatic
aberrations. Blooming is really good at destroying the detail
of leaves shot against a bright sky. "
It is a case of blooming (the white overexposure) that amplifies
the chromatic aberration (the purple edges).
And ... if you shall be very precise, I don't really think
that the overexposed parts really should be called blooming.
It is probably a case of too much light that has spread to
nearby cells and overexposed them. Blooming is when charge is
spilling over to nearby cells - and this is usually something
that spreads in one direction only. making streaks. At least
- this is what I have learned from astro-photography.
Roland
Bart van der Wolf
SNIP
> And ... if you shall be very precise, I don't really think
> that the overexposed parts really should be called blooming.
> It is probably a case of too much light that has spread to
> nearby cells and overexposed them. Blooming is when charge is
> spilling over to nearby cells - and this is usually something
> that spreads in one direction only. making streaks. At least
> - this is what I have learned from astro-photography.
I don't believe that electrons have a preference to migrate in only one
direction. They spill over to all neigboring pixels, but are only visible in
the less exposed ones because they can add signal. This will result in color
fringes due to the nature of the Bayer CFA, which has little use for fully
saturated wells that could signify maximum signal and anything in excess of
it.
Bart
Roland Karlsson
> I don't believe that electrons have a preference to migrate in only one
> direction. They spill over to all neigboring pixels, but are only
> visible in the less exposed ones because they can add signal. This will
> result in color fringes due to the nature of the Bayer CFA, which has
> little use for fully saturated wells that could signify maximum signal
> and anything in excess of it.
>
Look here
http://www.stargazing.net/david/ccd/blooming.html
http://www.ccddirect.com/online-store/scstore/blooming.html
Don't find anything but streak blooming.
It is only photographers that talk about any other blooming,
and I have found no scientific/technical article about
this kind of blooming. It could just be light spillover instead
of charge spillover. Or - maybe astronomical sensors
differs to photographic sensors.
Roland
Bart van der Wolf
SNIP
> Don't find anything but streak blooming.
>
> It is only photographers that talk about any other blooming,
> and I have found no scientific/technical article about
> this kind of blooming. It could just be light spillover instead
> of charge spillover. Or - maybe astronomical sensors
> differs to photographic sensors.
AFAIK the streaks are in the readout direction, the bucket brigade
off-loading the charge through ajacent wells, and not being able to clear
all of the charge before receiving new charge from its neighbors.
The difference with astronomical CCDs is usually that those can be
monochrome (more light due to absent CFA), cooled to reduce noise, and
sometimes exposed from the (thinner) back side, thus avoiding the
attenuation of surface circuitry. The priciple stays the same.
Bart
Joseph Miller
I'm sorry Bart, but you are wrong. On astronomical CCDs the excess
charge does first bleed along coluums, or what we call in the parallel
direction There's just less of a barrier in that direction. I'm not
guessing or speculating. This is well-understood. It has nothing to do
with readout, and it has nothing to do with whether there are color
filters in front of the pixels. CCDs in digicams are different in how
they handle excess charge, as they have more complicated anti-blooming
structures. Nevertheless, CCD video cameras are in widespread use, and
you will often see bright lights on videos and on TVs (especially on
programs of nighttime sporting events) bleeding along coluums.
As for the colors in the water coluum, the image does not in any way
appear overexposed. Water has plenty of dispersing capability. When I
teach my classes about dispersion, I suggest they notice the flat
underwater lights the next time they are at a swimming pool at night.
They will see color fringes around the lights, and if the water is
choppy, the color pattens can be complex. This is the same physics as
that being shown in the water coluum. Why are you so resistant to this
idea? The next time you are in an art-glass gallery, look for a rod or
section of glass that looks like this water coluum. Look through it at
a uniform white background. The optical properties of water and glass
are very similar.
Joe
Bart van der Wolf
"Joseph Miller" <jmi...@surfnetusa.com> wrote in message
news:aM6dnYo4D_t...@surfnetusa.com...
SNIP
> I'm sorry Bart, but you are wrong. On astronomical CCDs the excess
> charge does first bleed along coluums, or what we call in the parallel
> direction There's just less of a barrier in that direction.
I did notice that particularly on some astronomical CCDs the colomnwise
propagation of excess charge is common. This seems to indicate anti-blooming
drains that only run vertically
(http://www.prinst.com/library_enc_saturation.shtml).
> I'm not guessing or speculating. This is well-understood. It has
nothing to do
> with readout, and it has nothing to do with whether there are color
> filters in front of the pixels.
Which raises the question, why in the colomn and not (as much) in the row
direction? My answer, it's in the design of the particular CCD. Normal
charge spill-over has no preference for direction.
> CCDs in digicams are different in how they handle excess charge, as they
> have more complicated anti-blooming structures.
That is apparently so. The manual of my Canon G3 states the possibility of a
vertical red bar showing on the LCD with extreme local overexposure, but it
also states that it will not be visible on the still images (it will on
movies). So at least this camera has a more symmetrical behavior to blooming
for stills.
> Nevertheless, CCD video cameras are in widespread use, and
> you will often see bright lights on videos and on TVs (especially on
> programs of nighttime sporting events) bleeding along coluums.
Exactly, movies and "real"-time LCDs mean rapid (perhaps interlaced)
read-out. Again, the spike/bar error occurs in the vertical direction, Why?
> As for the colors in the water coluum, the image does not in any way
> appear overexposed. Water has plenty of dispersing capability. When I
> teach my classes about dispersion, I suggest they notice the flat
> underwater lights the next time they are at a swimming pool at night.
> They will see color fringes around the lights, and if the water is
> choppy, the color pattens can be complex. This is the same physics as
> that being shown in the water coluum. Why are you so resistant to this
> idea?
I'm just not convinced by a water dispersion theory. It takes a very shallow
angle to the surface for internal reflection, and a very small body of water
(like a miniscule droplet in a rainbow) to reflect and disperse. A turbulent
thick stream of water seems to deviate too far from that ideal rainbow
condition.
I'm open for better suggestions, but the effect shown looks exactly the same
as color fringing around high contrast edges, such as branches, that is
common in digicams.
> The next time you are in an art-glass gallery, look for a rod or
> section of glass that looks like this water coluum. Look through it at
> a uniform white background. The optical properties of water and glass
> are very similar.
I guess we differ in the appreciation of what a refractive index of water
and a refractive index of glass means for dispersion. Crown glass can have a
refractive index of around 1.45 and dense flint glass approx. 1.9. The
dispersion of flint glass can be 4 times as strong compared to the lower
refractive index glass. Pure water has an even lower refractive index of
1.33 and thus a much lower dispersion than common optical glass.
Bart
Joseph Miller
>
>I'm just not convinced by a water dispersion theory. It takes a very shallow
>angle to the surface for internal reflection, and a very small body of water
>(like a miniscule droplet in a rainbow) to reflect and disperse. A turbulent
>thick stream of water seems to deviate too far from that ideal rainbow
>condition.
>
>
with the color phenomenon in the water column. It's simple dispersion.
A wedge prism disperses light without any reflection. In fact, I
worked with our optical designer once on a lens design, and we we were
able to cut down the chromatic aberration of the lens by filling one
part of it with water. Water had the right dispersive properties
compared to its refractive index so that it acted like a special glass
that was not in the catalog. The corrugated surface of the water column
presents steep angles of incidence to the incoming light, favoring
prismatic behavior. If I had the time, I would do the calculation and
show you how it is possibe. In fact you can even get dark areas in the
column from which no light appears to emerge. I can't remember if this
image had any. Well, I think we've beaten this to death. I'm done. I
suggest you do a simple experiment: look at underwater lights in a
swimming poor (standing outside the pool looking down through the
surface at an angle- no reflections are involved)and see if you still
believe that the dispersive properties of water are inconsequential.
Joe
>
>
Bart van der Wolf
SNIP
> Well, I think we've beaten this to death. I'm done.
Thanks for your (final?) attempt to convince me. I guess a side by side
comparison of the scene shot on film, and shot with a digicam, should at
least allow to eliminate the sensor influence.
Bart