cp5k fails in open forum

[Rick Stirling]

By way of clarification ... the below should read ....

'reciprocal of the sum of the reciprocals OF THE RESOLUTION of all the imaging elements'

i.e. ' Res = 1/(1/a+1/b+1/c+...)

If you plug some numbers into this equation you will quickly see
that the total resolution of any optical system can be no better
than the element with the lowest resolution.

It is feasable that two cameras with similar sensors would have
differing total resolution if some other imaging element were
significantly different. The lens for example.

cheers,
Rick Stirling

Is there any way to quantify this? Can this only be seen in full
size images or does it persist even when the images are shrunken?

Does focal length make any difference?

 

[kessler]

Hi Rick,

That certainly is a quanitfiable definition of resolution but it doesn't work in the desired direction (from the images).
-Mark

i.e. ' Res = 1/(1/a+1/b+1/c+...)

If you plug some numbers into this equation you will quickly see
that the total resolution of any optical system can be no better
than the element with the lowest resolution.

It is feasable that two cameras with similar sensors would have
differing total resolution if some other imaging element were
significantly different. The lens for example.

cheers,
Rick Stirling

Is there any way to quantify this? Can this only be seen in full
size images or does it persist even when the images are shrunken?

Does focal length make any difference?

--
[email protected]

 

[Brent E]

Now we all must call you mr philisophical!!!!
interesting quote...

Brent

(the grass is always greener ...),

in the begining maybe, at the top its just a different shade of
white balance.


--
beauty can only be comprehended in the moment before death, as
beauty bows to destruction - john keats

 

[Rick Stirling]

Sure it does, it has too. Physics and all that stuff ...

Lets take an random example ...

Lens resolution = 100 lpmm ( line pairs per mm )
Sensor resolution = 50 lpmm
printer resolution = 25 lpmm

Tot Res =

1/(1/100+1/50+1/25) = 1/(.01+.02+.04) = 1/.07 = 14.28 lpmm

cheers,
Rick Stirling

i.e. ' Res = 1/(1/a+1/b+1/c+...)

If you plug some numbers into this equation you will quickly see
that the total resolution of any optical system can be no better
than the element with the lowest resolution.

It is feasable that two cameras with similar sensors would have
differing total resolution if some other imaging element were
significantly different. The lens for example.

cheers,
Rick Stirling

Is there any way to quantify this? Can this only be seen in full
size images or does it persist even when the images are shrunken?

Does focal length make any difference?

--
[email protected]

 

[Phil Dunne]

I did the whole ummm and ahhh thing for a month spending hours each nite researching a camera. I was very tempted by the 707 but ultimately what it came down to is what I'd use the camera for. Being a landscape architect I needed good macro, accurate colour representation and wide angle. More pixels for large format printing would be good too. If you look at the image comparison tests the CP5K is a clear winner over the 707 and most other cameras in the 4/5mp range. Resolution is important but I find it a lot quicker to sharpen an image than it is to do colour correction. For those cameras that the CP didn't beat in the image comparison, it beat them on features and range of manual control.

It comes down to what you will use the camera for - work that out (it took me 2 weeks to realise this) and your choice will be much narrower. If you want to take happy snaps and action shots or lots of nite shots then I'd suggest the nikon is not the camera you require.
--
Cheers
Phil

cp5000

 

[kessler]

Hi Rick,
Thanks for taking time to explain, but I wasn't being very clear when I said:
"doesn't work in the desired direction (from the images)."

I understand that resolution can be calculated from the lens/sensor/printer... Can the ability to resolve fine detail be quantified from knowledge of the subject and an image (not a resolution chart). Even better, can it be quantified from just an image assuming a sufficiently detailed subject. This measure probably wouldn't be equal to resolution (defined below), but it would be a real world measure of the smallest detail the camera can capture.

When I go for a fancy dinner the waiter doesn't give me a tale of the powerful biceps of the archer who drew the finely craft bow and shot the razor sharp arrow in to the beautiful stag who was raise on a diet of pure oats and natural grasses given to him as he roamed free in the rolling hills of Ireland. He says there will be a great tasting piece of dead meat on my plate, and that is why I order it (actually I generally order vegetarian). Okay the analogy is not very good, but my point is taht what I care about is the image not theories of optics, especially since we don't don't know how the specifics of the interpolation scheme is factoring in.

Cheerios,
Mark

Lets take an random example ...

Lens resolution = 100 lpmm ( line pairs per mm )
Sensor resolution = 50 lpmm
printer resolution = 25 lpmm

Tot Res =

1/(1/100+1/50+1/25) = 1/(.01+.02+.04) = 1/.07 = 14.28 lpmm

cheers,
Rick Stirling

i.e. ' Res = 1/(1/a+1/b+1/c+...)

If you plug some numbers into this equation you will quickly see
that the total resolution of any optical system can be no better
than the element with the lowest resolution.

It is feasable that two cameras with similar sensors would have
differing total resolution if some other imaging element were
significantly different. The lens for example.

cheers,
Rick Stirling

Is there any way to quantify this? Can this only be seen in full
size images or does it persist even when the images are shrunken?

Does focal length make any difference?

--
[email protected]

--
[email protected]

 

[peter konzuk]

I like your formula and it may apply to physics but the variables involved in digital imaging on closer inspection would have to be far more complicated than this calculation. I suspect the +... are very significant such as metering, compression, lens mtf, system mtf, etc, etc, etc. In any case to confuse a dp review with science is unfair to the both reader and Phil. Just look at the resoultion chart data provided on this site. The variables not controled for include... focal length, 19.8 for the 707 and 13.6 for the cp5k not to mention a compression ratio of 7 vs 9 and exposure. And what about contrast? Even pop photography long ago realized the deficiencies of resolution chart testing. Contrast can be as important as resolution in a subjective impression of sharpness hence the significance of the modulation transfer factor measurements they use. Measure the contrast in those dp review resolution charts. Which camera imaged a truer black and white? Aspherical elements tend to improve contrast. Three of the cp5k's 9 elements are aspherical. I don't even want to get into the blown highlight issue but IMHO if the cp5k's lens was less contrasty wouldn't that solve the blown highlight issue. Perhaps? but unfortunately then you'd lose low contrast resolution, subtle tonal and colour gradations. Far better to have the option either by correct metering or in digital post processing.

There are lots of reasons not to like the cp5k but before you lend too much credence to pseudo resolution comparisons realize that these comparisons are not scientific. Realize that resolution is only one variable measuring image quality. Anyone can produce a sharp looking image without detail or vice versa. Kessler's question about the water and shutter speed in those image comparisons is very appropriate.

Sorry about the rant but when dp reviews start getting confused with science I get a little bent... I know you know this Rick I just wonder what other variables belong in the +... of your formula of which the lens is just one?

peter

'reciprocal of the sum of the reciprocals OF THE RESOLUTION of all
the imaging elements'

i.e. ' Res = 1/(1/a+1/b+1/c+...)

If you plug some numbers into this equation you will quickly see
that the total resolution of any optical system can be no better
than the element with the lowest resolution.

It is feasable that two cameras with similar sensors would have
differing total resolution if some other imaging element were
significantly different. The lens for example.

cheers,
Rick Stirling

Is there any way to quantify this? Can this only be seen in full
size images or does it persist even when the images are shrunken?

Does focal length make any difference?

 

[John Lai]

Some high definition F707 photos:

http://www.dpreview.com/forums/read.asp?forum=1009&message=2884268 Taken by a former CP 5000 owner.

Food13a.jpg from this page http://www.pbase.com/tigadee/people_of_malaysia

Best Regards
John

Does focal length make any difference?

Best Regards
John

There is an interesting thread in the open forum comparing 5mp
cameras.
Several people are claiming the cp5k has substantially less
resolving power then the 707 and possible other 5mp DCs. It's not
cp5k bashing, that is just one of the results. Do you think this
is the case?

http://www.dpreview.com/forums/read.asp?forum=1018&message=2912417

http://www.dpreview.com/forums/read.asp?forum=1018&message=2922156

I noticed in Phils review, where he showed the same image taken by
several cameras, the cp5k was missing detail in the water. I
thought that could be due to the lower shutter speed used. Any
thoughts?

--
[email protected]

--

--
[email protected]

 

[Rick Stirling]

You're absolutely correct Peter. That's why I added the '+...'

Identifying all the significant contributors to the total resolution of the system is a daunting task. I tend to understate and leave it to the reader to envision the 'greater picture'.

perhaps the '+...' was a bit too subtle. ;-)

And just for the record, I think that $ for $ Nikon has the best optics out there. That's why all of my cameras and lenses are Nikon and have been for 35 years. ;-)

Rick

There are lots of reasons not to like the cp5k but before you lend
too much credence to pseudo resolution comparisons realize that
these comparisons are not scientific. Realize that resolution is
only one variable measuring image quality. Anyone can produce a
sharp looking image without detail or vice versa. Kessler's
question about the water and shutter speed in those image
comparisons is very appropriate.

Sorry about the rant but when dp reviews start getting confused
with science I get a little bent... I know you know this Rick I
just wonder what other variables belong in the +... of your formula
of which the lens is just one?

peter

'reciprocal of the sum of the reciprocals OF THE RESOLUTION of all
the imaging elements'

i.e. ' Res = 1/(1/a+1/b+1/c+...)

If you plug some numbers into this equation you will quickly see
that the total resolution of any optical system can be no better
than the element with the lowest resolution.

It is feasable that two cameras with similar sensors would have
differing total resolution if some other imaging element were
significantly different. The lens for example.

cheers,
Rick Stirling

Is there any way to quantify this? Can this only be seen in full
size images or does it persist even when the images are shrunken?

Does focal length make any difference?

 

[Ken.]

Some high definition F707 photos:

I have seen that shot from dwight. Great shot! He really is loving that 707! The 707 is one of the best resolving carmeras.

To be fair I have to state that I have gotten shots just as clear as this with my 5000.

Ken

 

[Rick Stirling]

What about 001.jpg on http://www.pbase.com/tigadee/people_of_malaysia ? Why not use it to evaluate the resolution of the 707? It was taken by the same camera. FOOD13.JPG is not particularly remarkable either except the food looks good enough to eat. ;-)

What you have pointed out are photos ( at least your 1st example ) that are well exposed, properly focused with little or no camera shake - a set of conditions that also occurs occasionally with other brands of cameras depending on the technique of the photographer. ;-)

Rick Stirling

Some high definition F707 photos:

http://www.dpreview.com/forums/read.asp?forum=1009&message=2884268
Taken by a former CP 5000 owner.

Food13a.jpg from this page
http://www.pbase.com/tigadee/people_of_malaysia

Best Regards
John

 

[Rick Stirling]

See Peter's response and my reply above. I guess I was being a bit too subtle with the formula. ;-)

I should relate an experiance I had evaluating the resolution of 2 camera systems many many years ago. The numbers said that system 1 should be the hands down best performer as its lens was clearly better than system 2. But the images came out the wrong way. Camera 2 consistantly delivered much better images. It turned out that camera 2 had a much better system for keeping the film plane flat than camera 1 did. ;-)

cheers
Rick

Hi Rick,
Thanks for taking time to explain, but I wasn't being very clear
when I said:
"doesn't work in the desired direction (from the images)."
I understand that resolution can be calculated from the
lens/sensor/printer... Can the ability to resolve fine detail be
quantified from knowledge of the subject and an image (not a
resolution chart). Even better, can it be quantified from just an
image assuming a sufficiently detailed subject. This measure
probably wouldn't be equal to resolution (defined below), but it
would be a real world measure of the smallest detail the camera can
capture.

When I go for a fancy dinner the waiter doesn't give me a tale of
the powerful biceps of the archer who drew the finely craft bow and
shot the razor sharp arrow in to the beautiful stag who was raise
on a diet of pure oats and natural grasses given to him as he
roamed free in the rolling hills of Ireland. He says there will be
a great tasting piece of dead meat on my plate, and that is why I
order it (actually I generally order vegetarian). Okay the analogy
is not very good, but my point is taht what I care about is the
image not theories of optics, especially since we don't don't know
how the specifics of the interpolation scheme is factoring in.

Cheerios,
Mark

 

[kessler]

Wasit one of those fancy units that actually sucks the film to the film plane?

I should relate an experiance I had evaluating the resolution of 2
camera systems many many years ago. The numbers said that system 1
should be the hands down best performer as its lens was clearly
better than system 2. But the images came out the wrong way. Camera
2 consistantly delivered much better images. It turned out that
camera 2 had a much better system for keeping the film plane flat
than camera 1 did. ;-)

cheers
Rick

Hi Rick,
Thanks for taking time to explain, but I wasn't being very clear
when I said:
"doesn't work in the desired direction (from the images)."
I understand that resolution can be calculated from the
lens/sensor/printer... Can the ability to resolve fine detail be
quantified from knowledge of the subject and an image (not a
resolution chart). Even better, can it be quantified from just an
image assuming a sufficiently detailed subject. This measure
probably wouldn't be equal to resolution (defined below), but it
would be a real world measure of the smallest detail the camera can
capture.

When I go for a fancy dinner the waiter doesn't give me a tale of
the powerful biceps of the archer who drew the finely craft bow and
shot the razor sharp arrow in to the beautiful stag who was raise
on a diet of pure oats and natural grasses given to him as he
roamed free in the rolling hills of Ireland. He says there will be
a great tasting piece of dead meat on my plate, and that is why I
order it (actually I generally order vegetarian). Okay the analogy
is not very good, but my point is taht what I care about is the
image not theories of optics, especially since we don't don't know
how the specifics of the interpolation scheme is factoring in.

Cheerios,
Mark

--
[email protected]

 

[John Lai]

How dares he expose such an out of focus photo!

John

What about 001.jpg on
http://www.pbase.com/tigadee/people_of_malaysia ? Why not use it to
evaluate the resolution of the 707? It was taken by the same
camera. FOOD13.JPG is not particularly remarkable either except the
food looks good enough to eat. ;-)

 

[kessler]

Yes, I saw those two posts earlier; however, I previously understood that you were allowing for arbitrarly complex details that were not specified.

I think we are in a never ending loop (like the guy trying to get the water or the apple). Maybe there isn't a simple way to get at the resolving power from the images and maybe that is your point. The case is certainly well made that there is a complex way that can't be used in this scenerio.

What do you think, could there be a simple measure taken from the images?

-Mark

I should relate an experiance I had evaluating the resolution of 2
camera systems many many years ago. The numbers said that system 1
should be the hands down best performer as its lens was clearly
better than system 2. But the images came out the wrong way. Camera
2 consistantly delivered much better images. It turned out that
camera 2 had a much better system for keeping the film plane flat
than camera 1 did. ;-)

cheers
Rick

Hi Rick,
Thanks for taking time to explain, but I wasn't being very clear
when I said:
"doesn't work in the desired direction (from the images)."
I understand that resolution can be calculated from the
lens/sensor/printer... Can the ability to resolve fine detail be
quantified from knowledge of the subject and an image (not a
resolution chart). Even better, can it be quantified from just an
image assuming a sufficiently detailed subject. This measure
probably wouldn't be equal to resolution (defined below), but it
would be a real world measure of the smallest detail the camera can
capture.

When I go for a fancy dinner the waiter doesn't give me a tale of
the powerful biceps of the archer who drew the finely craft bow and
shot the razor sharp arrow in to the beautiful stag who was raise
on a diet of pure oats and natural grasses given to him as he
roamed free in the rolling hills of Ireland. He says there will be
a great tasting piece of dead meat on my plate, and that is why I
order it (actually I generally order vegetarian). Okay the analogy
is not very good, but my point is taht what I care about is the
image not theories of optics, especially since we don't don't know
how the specifics of the interpolation scheme is factoring in.

Cheerios,
Mark

--
[email protected]

 

[Rich Gibson]

Amen! Esoteric discussions about specifications aren't as relevant as figuring out how you want to use it and selecting the "better" one for that particular purpose.

Kess, tell you what. It's clear your agonizing over the choice. Why not set a date to decide...period! Then buy that camera and tell us what you got. Make it a week, not more. Otherwise Fall will be here and you'll miss a heck of a lot of shooting. Frankly, either the 707 or the 5000 are fine cameras. Now all you have to do is match their strengths with your needs.

Rich

It comes down to what you will use the camera for - work that out
(it took me 2 weeks to realise this) and your choice will be much
narrower. If you want to take happy snaps and action shots or lots
of nite shots then I'd suggest the nikon is not the camera you
require.
--
Cheers
Phil

cp5000

 

[kessler]

Kess, tell you what. It's clear your agonizing over the choice.

Yes, it is painfully obvious, and just painfull!!

Why not set a date to decide...period! Then buy that camera and
tell us what you got. Make it a week, not more.

Are you kidding me? a week? How about a month?

-Mark

Rich

It comes down to what you will use the camera for - work that out
(it took me 2 weeks to realise this) and your choice will be much
narrower. If you want to take happy snaps and action shots or lots
of nite shots then I'd suggest the nikon is not the camera you
require.
--
Cheers
Phil

cp5000

--
[email protected]

 

[Rick Stirling]

Yep! ;-)

Wasit one of those fancy units that actually sucks the film to the
film plane?

 

[Sean Keegan]

Well said Richard!

Two points.

1. When the 5000 was announced we had an unbelieveable number of
folks from Olympus & Canon (on this web site you can see their post
history. ) poor mouth the 5000 damning it with faint prase or
making outright macho bravado pronouncements. "Nikon really blew
it, yadda yadda, yadda." This doesn' t, as a rule go the other way.
Nikon devotees, for whatever reason, just don't go trolling much in
other forums.

2. Your question is tantamount to coming up tp someone and asking
"those guys over there said your kid is ugly, is that true?"
Better, best and worst, along with "blows away" aren't words that
will generally start enlightened conversations.

If you are sincerely interested in picking a camera read as many
different posts as you can which address strengths and weaknesses
and make your own assessments. Don't expect others to decide for
you or tell you which is better. All these D.C.s are great or they
wouldn't be so popular. They are different. I would not consider
making a purchase basing it on the worst features of the camera.
When I decided on my 5000 I knew the type of photography I tend to
do and looked for the camera which best met my needs. For example,
I vacation a lot and go into caves, ruins and castles and meseums.
Often I am in small rooms. I almost never do telescopic
photography. With the 19mm capability of the 5000 it met my needs.
It's not perfect though, 'cause I still have trouble stopping my
grandchildren's activities well enough.

Good luck with your search. I just don't think you'll get much
asking which one is better or which is worse.

The photo you posted is very nice. I don't want to, nor do I have
the skill to, critique it. My only comment is that I think I would
need a full size, raw image to see the issues they were discussing
in the other forum. Please post it if you can and maybe other
people can comment on it.

-Mark

Ken

I'm fairly certain he is not a troll. Actually, two different
people had done some comparisons. One of them was slamming on the
7i pretty hard. He may have been biased but I think he was trying
to report accurate conclusions.

Do you see any missing detail here? Please advize if I am missing
something. Taken with Nikon CP 5000.



My CP 5000 and CP 990 galleries

Ken Leonard

There is an interesting thread in the open forum comparing 5mp
cameras.
Several people are claiming the cp5k has substantially less
resolving power then the 707 and possible other 5mp DCs. It's not
cp5k bashing, that is just one of the results. Do you think this
is the case?

http://www.dpreview.com/forums/read.asp?forum=1018&message=2912417

http://www.dpreview.com/forums/read.asp?forum=1018&message=2922156

I noticed in Phils review, where he showed the same image taken by
several cameras, the cp5k was missing detail in the water. I
thought that could be due to the lower shutter speed used. Any
thoughts?

--
[email protected]

--

--
[email protected]

--

--
[email protected]

 

[Dave Oddie]

I read those other posts Kessler was referring to and I'm not sure
what they were talking about. They said that the 5000 with it's
5.24 megapixel ccd, only has about 2.47 megapixels of resolution.
I have no earthly clue what that means and it just plain doesn't
sound right. If I take a picture at the 2560 setting on my CP5K, I
expect an image that corresponds to the resolution I set. Am I
wrong?

In a word yes. You are confusing resolution with pixels. At the 2650 setting you will always get your 2560 pixels but if you stuck a milk bottle on the end of the lens you would not have a very sharp picture would you?

You would still have the same number of pixels but a poor image.

What they are trying to say is that the combination of the CP 5000 lens, the electonics that process the image and the CCD deliver an image that, depsite being 5 mega-pixels in size only shows as much resolution as a 2.5mp camera with a good lens and processing.

In other words they are slagging off one or more aspects of the CP 5000, I guess the lens. Put a poor lens in front of a 5mp CCD and it won't deliver very good resolution to the CCD for the CCD to record.

I find it hard to believe this is the case with he CP 5000.

Dave

Does that mean that if I have a 1.0 megapixel camera then I have .5
megapixel resolution? It seems silly. Some of the more technical
guys help me out here.

Julio

It is really not a big issue. A little bit more resolution here or
there?!?!?! All the 5mg cameras have enough resolution to spare. So
in your search for the camera to match you, you should look for
more important differences. The CP5000 might not be for you but not
because it shows a little bit less resolution in the water than the
Sony.... Good luck with whatever you choose.

I noticed in Phils review, where he showed the same image taken by
several cameras, the cp5k was missing detail in the water. I
thought that could be due to the lower shutter speed used. Any
thoughts?

--
[email protected]