«

On 2010-12-27 00:09:00 +0000, Troy Piggins said:

> * Paul Furman wrote :
>> <...>
>> Night Shots Dubai Boats Troy Piggins.jpg
>> Very nice, the colors are actually similar to Pete Attkins' harbor shot.
>
> Thanks mate
>
> I had hoped to do a little more for this submission, but time and
> weather haven't been kind to me.

Sorry for omitting to comment on both Paul's and your submissions.

I was very interested in this shot because it has a similar mix of
light sources and pollution that I have to contend with in my harbour
shots. I've spent enjoyable hours making curve and chroma adjustments
to your image in an attempt to learn two things: a more simple post
processing method than the one I've been using; why I'm obsessed with
reducing certain hues - green from mercury lamps, cyan from a cause
I've yet to fully understand, and orange from sodium lamps. All I've
learnt is that converting the recorded scene into "my style" is not
only very difficult, it's impossible to come up with a logical
explanation and corresponding procedure. Perhaps that's a good
conclusion - when software becomes able to emulate "my style" it will
be time for me to pursue a different hobby.

Thanks very much for your images, Paul and Troy.

--
Pete

On Wed, 29 Dec 2010 11:20:45 +0000, Pete
<> wrote:

>On 2010-12-27 00:09:00 +0000, Troy Piggins said:
>
>> * Paul Furman wrote :
>>> <...>
>>> Night Shots Dubai Boats Troy Piggins.jpg
>>> Very nice, the colors are actually similar to Pete Attkins' harbor shot.
>>
>> Thanks mate
>>
>> I had hoped to do a little more for this submission, but time and
>> weather haven't been kind to me.
>
>Sorry for omitting to comment on both Paul's and your submissions.
>
>I was very interested in this shot because it has a similar mix of
>light sources and pollution that I have to contend with in my harbour
>shots. I've spent enjoyable hours making curve and chroma adjustments
>to your image in an attempt to learn two things: a more simple post
>processing method than the one I've been using; why I'm obsessed with
>reducing certain hues - green from mercury lamps, cyan from a cause
>I've yet to fully understand, and orange from sodium lamps. All I've
>learnt is that converting the recorded scene into "my style" is not
>only very difficult, it's impossible to come up with a logical
>explanation and corresponding procedure. Perhaps that's a good
>conclusion - when software becomes able to emulate "my style" it will
>be time for me to pursue a different hobby.
>
>Thanks very much for your images, Paul and Troy.

For your own shots, it's not possible to do a white balance in the field?
Or is that beside the point?

On 2010-12-29 14:16:34 +0000, HocusPocus said:

> Pete wrote:
>
>> Troy Piggins said:
>>
>>> * Paul Furman wrote :
>>>> <...>
>>>> Night Shots Dubai Boats Troy Piggins.jpg
>>>> Very nice, the colors are actually similar to Pete Attkins' harbor shot.
>>>
>>> Thanks mate
>>>
>>> I had hoped to do a little more for this submission, but time and
>>> weather haven't been kind to me.
>>
>> Sorry for omitting to comment on both Paul's and your submissions.
>>
>> I was very interested in this shot because it has a similar mix of
>> light sources and pollution that I have to contend with in my harbour
>> shots. I've spent enjoyable hours making curve and chroma adjustments
>> to your image in an attempt to learn two things: a more simple post
>> processing method than the one I've been using; why I'm obsessed with
>> reducing certain hues - green from mercury lamps, cyan from a cause
>> I've yet to fully understand, and orange from sodium lamps. All I've
>> learnt is that converting the recorded scene into "my style" is not
>> only very difficult, it's impossible to come up with a logical
>> explanation and corresponding procedure. Perhaps that's a good
>> conclusion - when software becomes able to emulate "my style" it will
>> be time for me to pursue a different hobby.
>>
>> Thanks very much for your images, Paul and Troy.
>
> For your own shots, it's not possible to do a white balance in the field?
> Or is that beside the point?

You've made a very good point. I hope this answers your question...

The vessels I shoot are (un)loaded at the other side of an estuary.
There are at least five different light sources, two of which
continually change: the moving crane lights and the twilight. The range
of illuminant colour temperatures is roughly 2700 - 12000 K.

Here's a simple example (sorry myPicturetown shows them in the wrong order):

<http://www.tinyurl.com/2w2uvdy>

1_1118_orig.JPG is a crop from the camera JPEG. The sky looked totally
dark as I was taking the picture.

Settings on Nikon 35 mm FF camera:
WB: Auto, 0, 0
ISO: 200
Lens: 85 mm f/1.8
Aperture Priority: f/4 at 1 second
Exposure Comp: 0
Matrix Metering
Picture Control: Standard
Active D-Lighting: Auto

The other two JPEGs are different edits of the camera RAW. This is a
reject shot so I haven't performed my usual complex editing to convert
it into "my style", but I think it serves as a useful illustration of
the effects of mixed lighting.

--
Pete

On Thu, 30 Dec 2010 19:06:14 +0000, Pete
<> wrote:

>On 2010-12-29 14:16:34 +0000, HocusPocus said:
>
>> Pete wrote:
>>
>>> Troy Piggins said:
>>>
>>>> * Paul Furman wrote :
>>>>> <...>
>>>>> Night Shots Dubai Boats Troy Piggins.jpg
>>>>> Very nice, the colors are actually similar to Pete Attkins' harbor shot.
>>>>
>>>> Thanks mate
>>>>
>>>> I had hoped to do a little more for this submission, but time and
>>>> weather haven't been kind to me.
>>>
>>> Sorry for omitting to comment on both Paul's and your submissions.
>>>
>>> I was very interested in this shot because it has a similar mix of
>>> light sources and pollution that I have to contend with in my harbour
>>> shots. I've spent enjoyable hours making curve and chroma adjustments
>>> to your image in an attempt to learn two things: a more simple post
>>> processing method than the one I've been using; why I'm obsessed with
>>> reducing certain hues - green from mercury lamps, cyan from a cause
>>> I've yet to fully understand, and orange from sodium lamps. All I've
>>> learnt is that converting the recorded scene into "my style" is not
>>> only very difficult, it's impossible to come up with a logical
>>> explanation and corresponding procedure. Perhaps that's a good
>>> conclusion - when software becomes able to emulate "my style" it will
>>> be time for me to pursue a different hobby.
>>>
>>> Thanks very much for your images, Paul and Troy.
>>
>> For your own shots, it's not possible to do a white balance in the field?
>> Or is that beside the point?
>
>You've made a very good point. I hope this answers your question...
>
>The vessels I shoot are (un)loaded at the other side of an estuary.
>There are at least five different light sources, two of which
>continually change: the moving crane lights and the twilight. The range
>of illuminant colour temperatures is roughly 2700 - 12000 K.
>
>Here's a simple example (sorry myPicturetown shows them in the wrong order):
>
><http://www.tinyurl.com/2w2uvdy>
>
>1_1118_orig.JPG is a crop from the camera JPEG. The sky looked totally
>dark as I was taking the picture.
>
>Settings on Nikon 35 mm FF camera:
>WB: Auto, 0, 0
>ISO: 200
>Lens: 85 mm f/1.8
>Aperture Priority: f/4 at 1 second
>Exposure Comp: 0
>Matrix Metering
>Picture Control: Standard
>Active D-Lighting: Auto
>
>The other two JPEGs are different edits of the camera RAW. This is a
>reject shot so I haven't performed my usual complex editing to convert
>it into "my style", but I think it serves as a useful illustration of
>the effects of mixed lighting.

Right! I see the problem. I guess turning the camera around to have a white
card lit by the mixed lighting wouldn't work because the light levels are
so low. Praise be to RAW!

On 2010-12-31 03:41:22 +0000, Wasting Valuable Time Educating the Morons said:

> Wasting Valuable Time Educating the Morons wrote:
>
>> HocusPocus wrote:
>>
>>> Pete wrote:
>>>
>>>> On 2010-12-29 14:16:34 +0000, HocusPocus said:
>>>>
>>>>> <...>
>>>>> For your own shots, it's not possible to do a white balance in the field?
>>>>> Or is that beside the point?
>>>>
>>>> You've made a very good point. I hope this answers your question...
>>>>
>>>> The vessels I shoot are (un)loaded at the other side of an estuary.
>>>> There are at least five different light sources, two of which
>>>> continually change: the moving crane lights and the twilight. The range
>>>> of illuminant colour temperatures is roughly 2700 - 12000 K.
>>>>
>>>> Here's a simple example (sorry myPicturetown shows them in the wrong order):
>>>>
>>>> <http://www.tinyurl.com/2w2uvdy>
>>>>
>>>> <...>
>>>
>>> Right! I see the problem. I guess turning the camera around to have a white
>>> card lit by the mixed lighting wouldn't work because the light levels are
>>> so low. Praise be to RAW!
>>
>> It's not that the light levels are so low, but that the light levels from
>> each light-source hitting the card will be very very different from the
>> light-levels from each source that are hitting each subject.

Yep, that's the problem. The side of the vessels facing the camera are
illuminated mainly by twilight; the sterns have mixed sources; some of
the containers are lit mainly by the crane lights. The same thing
happens when shooting on a sunny day, but to a lesser extent: the
shaded and shadow areas of the scene are illuminated by higher
temperature sources.

>> RAW will not
>> matter in this regard either. You can adjust the color balance just as well
>> from any JPG file, since your resulting output format (monitor-display or
>> print) will always be well below any RAW or JPG color bit-depth anyway.
>
> There is also an advantage to using a JPG file output and setting the
> camera on auto white-balance in such situations that involve widely
> different light-source frequencies and temperatures. The camera will have
> already tried to find a happy medium between all the various sources so you
> have a good starting point. With RAW that is not so and will never be so.

Auto WB works far better than I expected. It's invaluable for twilight
shots, especially during the rapidly changing light at dawn and dusk.
My RAW editor uses all of the camera setting by default therefore its
starting point is effectively the camera JPEG.

I'm unable to transfer a WB setting from the editor back into the
camera as a preset. It's theoretically possible because the camera
provides two-axis adjustment of all presets. The lighting at the port
requires a significant shift to magenta on the green-magenta axis, but
setting it correctly in the field seems to be impossible.

> I
> know this from my own demanding macro-photography and photomicrography of
> live nighttime insects. Whereby I had to use UV (short, long, and near),
> and various incandescents (into near-IR), and fluorescent light
> combinations (fluorescents in cool, warm, and daylight versions) to attract
> and photograph certain species. Each species being more attracted by
> certain wavelengths and combinations of light. In order for them to remain
> and be photographed sharply at microscopic levels in natural poses the
> lights had to remain set at the same intensities which originally attracted
> them so they would stay still and content. Then too they would often shift
> their own positions between each different light-source until they found
> their most comfortable combination of wavelengths. All light sources had to
> be used in the resulting photographs. The camera's auto white-balance (then
> applied to the camera's JPG output) did a better job at finding a common
> denominator between these many different source wavelengths than could be
> done by using RAW.

For me, auto white-balance plus presets is by far the most useful
feature of digital cameras. AF, auto exposure, and instant playback are
useful, but I'm so thankful that I can now take pictures instead of
faffing around with sets of colour balancing filters

--
Pete

On Thu, 30 Dec 2010 19:06:14 +0000, Pete
<> wrote:

>Here's a simple example (sorry myPicturetown shows them in the wrong order):
>
><http://www.tinyurl.com/2w2uvdy>
>
>1_1118_orig.JPG is a crop from the camera JPEG. The sky looked totally
>dark as I was taking the picture.

I found that a very interesting picture. All of the lines of
light-reflection on the water are slightly curved. I don't think that
this can be due to any form of lens distortion as as the curvature
appears to be uniform across all cases. The explanation is, I think,
that the surface of the water is a slightly curved mirror. What we are
seeing is the effect of a (tidal?) current affecting the height of the
water surface. Bernoulli's theorm says the water level should be
slightly lower where the tide is flowing fastest and thanks to your
photograph we can see this on the surface of the harbour.



Eric Stevens

On 2011-01-01 22:01:37 +0000, Eric Stevens said:

> Pete wrote:
>
>> Here's a simple example (sorry myPicturetown shows them in the wrong order):
>>
>> <http://www.tinyurl.com/2w2uvdy>
>>
>> 1_1118_orig.JPG is a crop from the camera JPEG. The sky looked totally
>> dark as I was taking the picture.
>
> I found that a very interesting picture. All of the lines of
> light-reflection on the water are slightly curved. I don't think that
> this can be due to any form of lens distortion as as the curvature
> appears to be uniform across all cases. The explanation is, I think,
> that the surface of the water is a slightly curved mirror. What we are
> seeing is the effect of a (tidal?) current affecting the height of the
> water surface. Bernoulli's theorm says the water level should be
> slightly lower where the tide is flowing fastest and thanks to your
> photograph we can see this on the surface of the harbour.

Thanks, Eric. The lens is reckoned to have a totally invisible barrel
distortion of 0.04%.

Just had a look through shots taken on different nights, the curvature
varies from more pronounced to none at all. The curve is most
noticeable towards the shallow shore in the foreground.

This part of the river is a channel with dredged shipping lanes,
upstream to the left, estuary is off to the right.

I've never used tide tables before. I'm guessing that the tide height
is sinusoidal with time and that the water flow rate is zero at both
high and low tides. Using those assumptions I've calculated the
following:

The image in the link above was taken when the tidal water was flowing
from left to right (downstream); the tide was dropping at the rate of
0.44 m per hour.

The image I submitted to the SI shows no curvature. It was taken when
the tidal water was flowing upstream; the tide was rising at 0.16 m per
hour (giving one third of the flow rate).

Occasionally I've noticed motion blur on the vessels during a 30 second
exposure, which I thought was caused by the tide. Obviously it wasn't
because that changes by only 4 mm in 30 s. Perhaps it's caused by
pitching. E.g. a Panamax Container Ship has a pitch natural frequency
of 0.25 rad/s, which is a period of 25 seconds; this requires a shutter
opening time of 8 seconds or less to reduce the motion blur. That's my
theory, anyone like to blow it out of the water?


--
Pete

On 2011-01-02 03:51:57 +0000, Paul Furman said:

> Pete wrote:
>> Eric Stevens said:
>>> Pete wrote:
>>>
>>>> <http://www.tinyurl.com/2w2uvdy>
>>>
>>> I found that a very interesting picture. All of the lines of
>>> light-reflection on the water are slightly curved. I don't think that
>>> this can be due to any form of lens distortion as as the curvature
>>> appears to be uniform across all cases. The explanation is, I think,
>>> that the surface of the water is a slightly curved mirror. What we are
>>> seeing is the effect of a (tidal?) current affecting the height of the
>>> water surface. Bernoulli's theorm says the water level should be
>>> slightly lower where the tide is flowing fastest and thanks to your
>>> photograph we can see this on the surface of the harbour.
>>
>> The image in the link above was taken when the tidal water was flowing
>> from left to right (downstream); the tide was dropping at the rate of
>> 0.44 m per hour.
>
> Confirmed

After almost endless web searches I conclude that the curved
reflections are unlikely to be caused by the Bernoulli effect because
the velocity of the current reaches only 1 knot in this part of the
channel.

One possibility is that the wind caused the water level to change near
the shore: the wind at the time was NNW 2 to 6 knots (it was aligned
downstream) and the camera was pointing towards NE. From what I've read
it would need an onshore wind (coming from the NE) to make the water
level rise near the shore. The velocities seem too low to cause the
curved reflections. However, shots taken from the same point on other
nights show either no curvature or similar curvature.

Another possibility is wave energy propagation from the sea into an
estuary. Could wave diffraction have something to do with this
curvature? A sequence of shots taken with shutter times from 1 to 30
seconds show no difference in the curvature when present or
straightness when absent.

In future, I'll record both tidal and wind data in the EXIF comment
field. If I had the stamina I'd take time lapse sequences between high
and low tides on different occasions. Evidently, water is not as flat
as one might expect.

--
Pete

On Fri, 7 Jan 2011 19:05:30 +0000, Pete
<> wrote:

>On 2011-01-02 03:51:57 +0000, Paul Furman said:
>
>> Pete wrote:
>>> Eric Stevens said:
>>>> Pete wrote:
>>>>
>>>>> <http://www.tinyurl.com/2w2uvdy>
>>>>
>>>> I found that a very interesting picture. All of the lines of
>>>> light-reflection on the water are slightly curved. I don't think that
>>>> this can be due to any form of lens distortion as as the curvature
>>>> appears to be uniform across all cases. The explanation is, I think,
>>>> that the surface of the water is a slightly curved mirror. What we are
>>>> seeing is the effect of a (tidal?) current affecting the height of the
>>>> water surface. Bernoulli's theorm says the water level should be
>>>> slightly lower where the tide is flowing fastest and thanks to your
>>>> photograph we can see this on the surface of the harbour.
>>>
>>> The image in the link above was taken when the tidal water was flowing
>>> from left to right (downstream); the tide was dropping at the rate of
>>> 0.44 m per hour.
>>
>> Confirmed
>
>After almost endless web searches I conclude that the curved
>reflections are unlikely to be caused by the Bernoulli effect because
>the velocity of the current reaches only 1 knot in this part of the
>channel.
>
>One possibility is that the wind caused the water level to change near
>the shore: the wind at the time was NNW 2 to 6 knots (it was aligned
>downstream) and the camera was pointing towards NE. From what I've read
>it would need an onshore wind (coming from the NE) to make the water
>level rise near the shore. The velocities seem too low to cause the
>curved reflections. However, shots taken from the same point on other
>nights show either no curvature or similar curvature.
>
>Another possibility is wave energy propagation from the sea into an
>estuary. Could wave diffraction have something to do with this
>curvature? A sequence of shots taken with shutter times from 1 to 30
>seconds show no difference in the curvature when present or
>straightness when absent.
>
>In future, I'll record both tidal and wind data in the EXIF comment
>field. If I had the stamina I'd take time lapse sequences between high
>and low tides on different occasions. Evidently, water is not as flat
>as one might expect.

I hadn't meant to start you on this but seeing you are underway I will
join you.

I hadn't thought of the curvature as being due to the wind so much as
it is due to tidal effects. 1 knot of tide flow is likely to cause a
depression of about 1" in the local water level. 2 knots is worth
about 4" etc.

Apart from that, whenever there is flow, friction requires that there
be a slope in the local hydraulic grade line. So, if there are fast
and slow local tidal currents the surface of the sea will be neither
flat nor level. It is with the aid of this slightly wiggly mirror that
your photograph reflects the ships and the lights on the far side of
the harbour.

In your case, the surface of the sea is acting as a slightly curved
mirror. The important part of the curvature causes the reflected light
path to be concave to the left (?). I'm having to say this from memory
as the site is currently down. That says to me that relative to the
far side of the harbour, the near side of the water flow is slightly
tilted with the left side being high. This will be due to the very
slight gradient required to induce water flow from left to right.

What the current is doing on the far side of the harbour is a mystery
to me but I expect that local yachtsmen will be familiar with it.



Eric Stevens

On 2011-01-07 21:49:19 +0000, Eric Stevens said:

> On Fri, 7 Jan 2011 19:05:30 +0000, Pete
> <> wrote:
>
>> On 2011-01-02 03:51:57 +0000, Paul Furman said:
>>
>>> Pete wrote:
>>>> Eric Stevens said:
>>>>> Pete wrote:
>>>>>
>>>>>> <http://www.tinyurl.com/2w2uvdy>
>>>>>
>>>>> I found that a very interesting picture. All of the lines of
>>>>> light-reflection on the water are slightly curved. I don't think that
>>>>> this can be due to any form of lens distortion as as the curvature
>>>>> appears to be uniform across all cases. The explanation is, I think,
>>>>> that the surface of the water is a slightly curved mirror. What we are
>>>>> seeing is the effect of a (tidal?) current affecting the height of the
>>>>> water surface. Bernoulli's theorm says the water level should be
>>>>> slightly lower where the tide is flowing fastest and thanks to your
>>>>> photograph we can see this on the surface of the harbour.
>>>>
>>>> The image in the link above was taken when the tidal water was flowing
>>>> from left to right (downstream); the tide was dropping at the rate of
>>>> 0.44 m per hour.
>>>
>>> Confirmed
>>
>> After almost endless web searches I conclude that the curved
>> reflections are unlikely to be caused by the Bernoulli effect because
>> the velocity of the current reaches only 1 knot in this part of the
>> channel.
>>
>> One possibility is that the wind caused the water level to change near
>> the shore: the wind at the time was NNW 2 to 6 knots (it was aligned
>> downstream) and the camera was pointing towards NE. From what I've read
>> it would need an onshore wind (coming from the NE) to make the water
>> level rise near the shore. The velocities seem too low to cause the
>> curved reflections. However, shots taken from the same point on other
>> nights show either no curvature or similar curvature.
>>
>> Another possibility is wave energy propagation from the sea into an
>> estuary. Could wave diffraction have something to do with this
>> curvature? A sequence of shots taken with shutter times from 1 to 30
>> seconds show no difference in the curvature when present or
>> straightness when absent.
>>
>> In future, I'll record both tidal and wind data in the EXIF comment
>> field. If I had the stamina I'd take time lapse sequences between high
>> and low tides on different occasions. Evidently, water is not as flat
>> as one might expect.
>
> I hadn't meant to start you on this but seeing you are underway I will
> join you.
>
> I hadn't thought of the curvature as being due to the wind so much as
> it is due to tidal effects. 1 knot of tide flow is likely to cause a
> depression of about 1" in the local water level. 2 knots is worth
> about 4" etc.

Interesting and surprising.

> Apart from that, whenever there is flow, friction requires that there
> be a slope in the local hydraulic grade line. So, if there are fast
> and slow local tidal currents the surface of the sea will be neither
> flat nor level. It is with the aid of this slightly wiggly mirror that
> your photograph reflects the ships and the lights on the far side of
> the harbour.
>
> In your case, the surface of the sea is acting as a slightly curved
> mirror. The important part of the curvature causes the reflected light
> path to be concave to the left (?). I'm having to say this from memory
> as the site is currently down. That says to me that relative to the
> far side of the harbour, the near side of the water flow is slightly
> tilted with the left side being high. This will be due to the very
> slight gradient required to induce water flow from left to right.
>
> What the current is doing on the far side of the harbour is a mystery
> to me but I expect that local yachtsmen will be familiar with it.

Thanks very much for you time, Eric. So far, local knowledge has not
offered any explanation, but I will contact a couple of people who may
be interested to take it further.


--
Pete