K20 Dynamic Range comparison

What you really want, is a metering that is foolproof against blowing
highlights. Then you wouldn't need a new sensor.
No I want a sensor that responds similar to film.
That's all in the tone curve.
You cant just leave
your underexposed shots as is you know Erik.
No, that's why you apply a tone curve.
I sometimes wonder if
you ever have used film, the difference is very noticable.
The difference is that I know how to post-process. You seem to
want finished prints from the lab.
A metering system that force underexposes all the time is not a very
good photographic tool.
It's not underexposure by definition if it exposes to the right to the point
where it protects the highlights you want to protect. That is a good
exposure.
The highlights may or may not be important.
Digital has not much latitude in the HL area..needs more.
Negative exposure compensation.

Just my two oere
Erik from Sweden
 
I think we're talking about a sensor with a lot more highlight head room like a few more stops, sure you can use tonal curves to emulate better dynamic range inthe highlight but its not truely better, we're talking about being able to over-expose a shot by several stops and still be able to recover all the highlight detail, for this you need a much higher clipping limit
--
Mike from Canada

'I like to think so far outside the box that it would require a telephoto lens just to see the box!' ~ 'My Quote :)'



http://www.airliners.net/search/photo.search?sort_order=views%20DESC&first_this_page=0&page_limit=121&&emailsearch=mighty_mike88%40hotmail.com&thumbnails=
 
That's all in the tone curve.
Digital is linear, film is not. The tone curve wont make it respond in the same way
No, that's why you apply a tone curve.
Its not the same, and you should know this. Try it.
The difference is that I know how to post-process. You seem to
want finished prints from the lab.
I would refrain from comments like that. I do know how to post process, like it or not it is easier to get the look I want from film. That is my own view.
It's not underexposure by definition if it exposes to the right to
the point
where it protects the highlights you want to protect. That is a good
exposure.
Erik, I thought we went through all this a while back. I compared the film and digital shots..The film just nailed it..ala it had the DR to capture the contrasty scene, the digital blocked up the shadows..as I had to expose for the highlights.

There is more being captured on film..by some margin too. Sure pulling the digital around helps to bring it up, but its just dumping out highlights..making like harder all round.
Negative exposure compensation.
Erm...I think I know a fair bit about photography by now!

--



Clint is on holiday! Soon to return! ;-)
 
Hi Barry,

It is there, as simple as that really.
The 14bit NEF has a headroom I have never seen the like of.
It IS massive.

Using the right RAW converter, you can recover 1.7 stop of blown highlights consistantly and with good colour accuracy, sometimes more. Not film but we are getting there.

Now due to the noise performance of the sensor you can actually recover an unusually high amount of shadow detail too and with a surprisingly good colour accurcy.
Have you actually used one for a prolonged period of time?
My experience above mirrors that of a number of users of it.

But again at 5000USD it is supposed to deliver stellar performance.

That said, I fail to see why you would compare it to the K20D, D300, 40D, etc.
I'm impressed:

http://www.dpreview.com/reviews/nikond3/page20.asp

12 stops of DR in 14-bit RAW...
I take the numbers with a pinch of salt!

Yes it looks decent, still clipped though in one place, and its still
not up to negative film..I know, I ask a lot ;-)

--



Clint is on holiday! Soon to return! ;-)
--
Thomas

Nothing is foolproof to a sufficiently talented fool
http://main.duplophotography.com/
 
That's all in the tone curve.
Digital is linear, film is not. The tone curve wont make it respond
in the same way
What matters is the result you can get by applying a tone curve,
not the intermediate raw data.

"Highlight range" is not about the dynamic range of the medium,
it's the relation between a mid-grey which we are free to shift around
and the point where there is no more highlight detail.

It seems the reviews on this site create a lot of confusion by introducing
the concept "highlight range".
It's not underexposure by definition if it exposes to the right to
the point
where it protects the highlights you want to protect. That is a good
exposure.
Erik, I thought we went through all this a while back. I compared the
film and digital shots..
Yes, Barry, and you posted samples in a thread. Did you notice how many
people tried to explain to you that your test was flawed as a test of DR?
The film just nailed it..ala it had the DR to
capture the contrasty scene, the digital blocked up the shadows..as I
had to expose for the highlights.
People brightened your digital shot and made it similar to the film one and
there was no visible difference in shadow blocking in those samples.
There is more being captured on film..by some margin too.
That's very possible, but I've not seen it demonstrated yet.
Sure
pulling the digital around helps to bring it up, but its just dumping
out highlights..making like harder all round.
If you are happy about the shadow pushing ability, as you said, you
can buy more "highlight headroom" but using negative EC.

Just my two oere
Erik from Sweden
 
negative EC is detrimental to shadow perfomance, sure we can recover shadow better then highlights but i would prefer it if the shadowns could be brighter then black and then recovered down towards their actual range and so to with highlights... as in over-expose without clipping and then brring back and you'll end up with a much nicer, less noisy, more uniform and quite possibly more accurate colours result. i've done this with many cameras, i'd like to ability to do this with a few more stops of range in the highlight end before data becomes lost.
--
Mike from Canada

'I like to think so far outside the box that it would require a telephoto lens just to see the box!' ~ 'My Quote :)'



http://www.airliners.net/search/photo.search?sort_order=views%20DESC&first_this_page=0&page_limit=121&&emailsearch=mighty_mike88%40hotmail.com&thumbnails=
 
What matters is the result you can get by applying a tone curve,
not the intermediate raw data.
Makers apply a tonal curve to mimic that of film. However as I said, the two respond in different ways. And no degree of curve is going to change that. Fuji may have the right idea, I have no hands on with that one.
"Highlight range" is not about the dynamic range of the medium,
it's the relation between a mid-grey which we are free to shift
around

and the point where there is no more highlight detail.

It seems the reviews on this site create a lot of confusion by
introducing
the concept "highlight range".
As said, much more in the shadow end, but its no free ride as that brings noise with it with severe shadow pulliing. Nothing misleading about the highlight range, its not much on digital, its lots on negative film. On the reverse..its better shadow on digital than film.

In my own experience, the latitude in the highlight range is of more use, than in the shadow one. Film has it about where I want digital to be. I am well aware they dont respond the same, and as pointed out I have to meter differently. Out and about, film is just easier to use in this regard, because for contrasty scenes, it does a far far better job holding highlights.
Yes, Barry, and you posted samples in a thread. Did you notice how many
people tried to explain to you that your test was flawed as a test of
DR?
I dont think it was flawed, it showed it how it was. The only ones who disputed it were 100% digital shooters, because they cannot accept their beloved format is less than ideal in some situations. I on the other hand, are more than happy to point out the cons of film. Anyone who ever touched a roll of film, agreed with what I said.

As it happens, I took the film camera out to prove to myself that I didnt need film anymore. The unfortunate thing is, the results were a shock to me..having used digital for a while now, you forget how film does, and what it does well.
People brightened your digital shot and made it similar to the film
one and
there was no visible difference in shadow blocking in those samples.
As I said, yes you can shadow pull..but had I not adjusted my exposure for digital, I would have blown all my skies, gone forever. Sure you can on film too, but its much less a problem, and happens less often. Also, again..its easier and faster to just scan the film, as it gets it how I want it..the digital needs a lot of work to get there. Thats even before the other obvious resolution advantage of film comes into play too.

Dont get me wrong, I own both cameras, the last thing I wanted to say was my APS SLR didnt cut it too well.
That's very possible, but I've not seen it demonstrated yet.
When I finish off the roll of FP4 plus, I shall let you see the results.
If you are happy about the shadow pushing ability, as you said, you
can buy more "highlight headroom" but using negative EC.
Up to a point I can..but out and about its not so much fun, when I want to concentrate on the shots, not the metering.

--



Clint is on holiday! Soon to return! ;-)
 
Hi Barry,

It is there, as simple as that really.
The 14bit NEF has a headroom I have never seen the like of.
It IS massive.
Using the right RAW converter, you can recover 1.7 stop of blown
highlights consistantly and with good colour accuracy, sometimes
more. Not film but we are getting there.
Now due to the noise performance of the sensor you can actually
recover an unusually high amount of shadow detail too and with a
surprisingly good colour accurcy.
Have you actually used one for a prolonged period of time?
My experience above mirrors that of a number of users of it.
Only had a hands on with the D3 for a very short while, enough to appreciate the panzer tank build..but not enough to reach any IQ conclusions. I have seen printed D3 shots, and they are excellent, even high ISO ones at large sizes.

Yes the highlight recovery looks decent on the D3, a cut above many (bar fuji..which some would argue does not have enough res..6mp is fine for many subjects)

I know shadow wise you can do some big things, esp on FF sensors. The 6mp APS one I have is very good, excellent in that regard, I just wish it had at least a couple more stops in the HL range..that would be great.
But again at 5000USD it is supposed to deliver stellar performance.
Thats a lot of really nice film cameras, and lots of film! lol
That said, I fail to see why you would compare it to the K20D, D300,
40D, etc.
Point is, if the D3 is good for digital, but not up to film HL wise, that means the APS ones are not as good as the D3, and the small sensor jobs are just hard work and well short of it.

I would just like to see a bit more work on this, not just resolution. Fuji may have the right idea..but what will become of that??

--



Clint is on holiday! Soon to return! ;-)
 
MiightyMike:
negative EC is detrimental to shadow perfomance
I agree. But the premise was that Barry was happy with the shadows
so I think he should stop complaining about the highlight performance
until he's exhausted the latitude in the shadows.
Makers apply a tonal curve to mimic that of film.
I think they try to mimic how our eyes and brain remember the scene.
Anyone who ever touched a roll of film, agreed with what I said.
I think most people, except for the youngest, have used film. And
they don't all agree with you.
As it happens, I took the film camera out to prove to myself that I
didnt need film anymore. The unfortunate thing is, the results were a
shock to me..having used digital for a while now, you forget how film
does, and what it does well.
Personally, I find it much quicker to apply a curve on digital
then having the film developed and scanned. If there was an IQ
disadvantage of digital after the adjustment, it would be another
matter, maybe there is, but it wasn't demonstrated by your shootout.
If you are happy about the shadow pushing ability, as you said, you
can buy more "highlight headroom" but using negative EC.
Up to a point I can..but out and about its not so much fun, when I
want to concentrate on the shots, not the metering.
Right, so we are back where we started, a foolproof metering.

Just my two oere
Erik from Sweden
 
negative EC is detrimental to shadow perfomance
I agree. But the premise was that Barry was happy with the shadows
so I think he should stop complaining about the highlight performance
until he's exhausted the latitude in the shadows.
Ehrik, you are ignoring a few simple things! I have to correct the digital ones, I can just get my film shots printed..as is, no messing about.
I think most people, except for the youngest, have used film. And
they don't all agree with you.
Ok so point them out!
Personally, I find it much quicker to apply a curve on digital
then having the film developed and scanned. If there was an IQ
disadvantage of digital after the adjustment, it would be another
matter, maybe there is, but it wasn't demonstrated by your shootout.
Well leaving aside 6mp aint touching ISO 200 negative film for details (yes I know it flies in the face of what the luminous landscape says..but its just there to see), like I said, digital blows on highlights..simple as that..major downer IMO.
Right, so we are back where we started, a foolproof metering.
Go buy some film, see for yourself, talk is cheap..least I put something up to support my view.

--



Clint is on holiday! Soon to return! ;-)
 
As I said, yes you can shadow pull..but had I not adjusted my
exposure for digital, I would have blown all my skies, gone forever.
Sure you can on film too, but its much less a problem, and happens
less often. Also, again..its easier and faster to just scan the film,
as it gets it how I want it..the digital needs a lot of work to get
there. Thats even before the other obvious resolution advantage of
film comes into play too.
Resolution advantage? What kind of film are you talking about? Maybe the 50 ISO Velvia still slightly outresolves the K20D, but not by a large margin.

My old Kodachrome 25 shots have marvelous detail. But if anything ever was unforgiving to wrong exposure, it must have been that film... (not to mention that I only was really happy with a film that slow when I went trekking in harsh sun at 4000 meters altitude in Himalaya :-))
--
Espen
 
Resolution advantage? What kind of film are you talking about? Maybe
the 50 ISO Velvia still slightly outresolves the K20D, but not by a
large margin.
b&w film has the highest res, the techinical pan low ISO stuff, no photographic lens can out resolve it. Negative colour is next best..then slide, which of course has the advantage colour wise.

Even ignoring res, the highlight thing is a major issue, if we could merge the D3 with foveon, and slap some fuji DR in there..now we might be talking ;-)
My old Kodachrome 25 shots have marvelous detail. But if anything
ever was unforgiving to wrong exposure, it must have been that
film... (not to mention that I only was really happy with a film that
slow when I went trekking in harsh sun at 4000 meters altitude in
Himalaya :-))
Sure slide demands more exposure accuracy, but you have the choice of negative film also. With digital, you are just stuck with it. Dont get me wrong, digital whips film to bits at high ISO..by some margin, but the old stuff still has some pull in other areas.

--



Clint is on holiday! Soon to return! ;-)
 
Using the right RAW converter, you can recover 1.7 stop of blown
highlights consistantly and with good colour accuracy, sometimes
more.
Recovering blown highlights after one colour channel has blown has to do
with how much the colour filters mismatch the colour of the light source.
So it's variable rather than consistent. And if a colour channel has blown,
there is missing data so good colour accuracy can't be guaranteed.

Just my two oere
Erik from Sweden
 
Hi Barry,

It is there, as simple as that really.
The 14bit NEF has a headroom I have never seen the like of.
It IS massive.
Using the right RAW converter, you can recover 1.7 stop of blown
highlights consistantly and with good colour accuracy, sometimes
more. Not film but we are getting there.
Now due to the noise performance of the sensor you can actually
recover an unusually high amount of shadow detail too and with a
surprisingly good colour accurcy.
Have you actually used one for a prolonged period of time?
My experience above mirrors that of a number of users of it.
Only had a hands on with the D3 for a very short while, enough to
appreciate the panzer tank build..but not enough to reach any IQ
conclusions. I have seen printed D3 shots, and they are excellent,
even high ISO ones at large sizes.
Yep, has made my lie a lot easier in that respect.
Yes the highlight recovery looks decent on the D3, a cut above many
(bar fuji..which some would argue does not have enough res..6mp is
fine for many subjects)
Fuji still is the king of highlight recovery, but they are comming close.
I know shadow wise you can do some big things, esp on FF sensors. The
6mp APS one I have is very good, excellent in that regard, I just
wish it had at least a couple more stops in the HL range..that would
be great.
True, but I dare say that when you combine the recovering possible in both highlights and shadows for working with hihg contrast scenes then we have come a long way.
The 6MP sensor has some nice large pixels too.
But again at 5000USD it is supposed to deliver stellar performance.
Thats a lot of really nice film cameras, and lots of film! lol
aint that the truth:D
That said, I fail to see why you would compare it to the K20D, D300,
40D, etc.
Point is, if the D3 is good for digital, but not up to film HL wise,
that means the APS ones are not as good as the D3, and the small
sensor jobs are just hard work and well short of it.
True, not in terms of DR, but they are getting better and overall have gotten quite good.
I would just like to see a bit more work on this, not just
resolution. Fuji may have the right idea..but what will become of
that??
Fuji has one idea at least and it is going to be interesting to see what they come up with next.

--
Thomas

Nothing is foolproof to a sufficiently talented fool
http://main.duplophotography.com/
 
Using the right RAW converter, you can recover 1.7 stop of blown
highlights consistantly and with good colour accuracy, sometimes
more.
Recovering blown highlights after one colour channel has blown has to do
with how much the colour filters mismatch the colour of the light
source.
So it's variable rather than consistent. And if a colour channel has
blown,
there is missing data so good colour accuracy can't be guaranteed.
You may be right, but then again I have been working with the D3 for more than 3 months now and I have seen pretty consistant results... i cannot shed any light on the technical side, but I have spend 3 month putting it through its phases, that gives me a pretty fair idea of what it does in terms of real world usage.

--
Thomas

Nothing is foolproof to a sufficiently talented fool
http://main.duplophotography.com/
 
b&w film has the highest res, the techinical pan low ISO stuff, no
photographic lens can out resolve it. Negative colour is next
best..then slide, which of course has the advantage colour wise.
ah, ok. I used the really low iso b+w stuff only once, for reproduction work.
Even ignoring res, the highlight thing is a major issue, if we could
merge the D3 with foveon, and slap some fuji DR in there..now we
might be talking ;-)
:-)
Sure slide demands more exposure accuracy, but you have the choice of
negative film also. With digital, you are just stuck with it. Dont
get me wrong, digital whips film to bits at high ISO..by some margin,
but the old stuff still has some pull in other areas.
Sure. But for overall use, when you need at least ISO 200, one is far better off with digital. I'm just scanning some slides this evening, and even though I'm using a far too simple slide scanner, the grains of my old Fujichrome 200 shots can be really annoying. Or charming :-)
--
Espen
 
Example of exposing to the right by EV+2 at ISO800 with the K20Ds Extended dynamic range on so a not to blow the highlights, then pulled back in processing

http://forums.dpreview.com/forums/read.asp?forum=1036&message=27652798
--
Mike from Canada

'I like to think so far outside the box that it would require a telephoto lens just to see the box!' ~ 'My Quote :)'



http://www.airliners.net/search/photo.search?sort_order=views%20DESC&first_this_page=0&page_limit=121&&emailsearch=mighty_mike88%40hotmail.com&thumbnails=
 

Keyboard shortcuts

Back
Top