D100 Sharpness Revisited

[DavidMillier]

Jim

I have a D100 and still have my E10 (for 4 more hours!).

On A4 sized prints printed on my Epson 1270 on Epson Premium Glossy, images from the D100 are noticeably sharper, more detailed, smoother and have less noise than the E10 (ISO80).

This applies to jpeg and Raw images at ISO 200, 400 and probably 800 as well.

In essence at the ISO200 setting there is no noise at all on normally exposed images.

There is simply not an issue here.

Possibly if you routinely print very large prints there might be a noise issue - I haven't actually printed my own images at this size. I have printed D100/D60 comparison images at A3 from steve's digicams (the playground scene). The D60 and D100 versions are slightly different but they are both sharp, detailed and low noise. The D60 has a trace less noise but it isn;t significant.

Choose your camera to suit any existing lenses and on which camera suits you best. Image quality of all of them in practical situations is superb.

So can you name all the digital cameras that have less noise than
the D100?

If you can, you used all your fingers on one hand, didn't you
because there isn't more than 5 of them.

D30
D60
S1
S2
D1
D1X
D1H
DCS720
DCS760
1D
E10
E20

Of the 12 I can think of, I'd say 10 have less noise then the D100
at lower ISO's,
D30
D60
S1
S2
D1
D1X
D1H
DCS720
DCS760
1D

and 7 have less noise at every ISO then the D100.
S2
D1
D1X
D1H
DCS720
DCS760
1D

Is this suppose to boost the D100's integrety? Or were you hoping I
could only count up to five : )

--
Every Camera Has Short Comings,
some camera's fall short of coming!

 

[Ger Bee]

... is the area that sorts the men from the boys or the cameras from the toys.

Sport is the most difficult thing for any camera and photographer combination – probably 80% is down to sheer muscle in the equipment department and the remainder in the knowledge of the games – the sports photographers skills are different and he is even more heavily dependant on his gear than any other bracket – outside of specialities.

The lens is the key – we have two systems – manual – where the photographer anticipates action and pre-focuses etc. And AF – now AF needs some pretty heavy-duty support in both the lens and the camera.

If you are skilled enough in the manual lens arena then the D100 is fine – if you are relying on AF then the D100 may not be good enough. As you have used an F5 – and presumably with good results then the only real answer is a D1h – a D1x (but a little slower) or the Eos 1-D.

I have done very little with the D100 in the C-AF tracking mode but initially it did not beak any records and whilst it did work - in a test situation it missed a bit, caught up and missed again sort of thing – in a similar test my D1x was in focus on every capture with the same lens.

 

[j_hu]

Hi

[...]

I agree concerning sharpness but I disliked the pushed up noise
using USM... I was dissapointed by this even in print in moderate
sizes of about A4. And there is even noise in less sharpened images
using "normal".

[...]

Regards, A. Schiele

Hi Alfred.

Have you had a look at http://www.fredmiranda.com???

He provides tools to sharpen pictures without creating to much noise in otherwise homogeneous areas. If this really works (the sample pictures look rather convincing) it's $15 well spent, I think.

With kind regards
Jürgen

-------------------------------------------------------------------
In theory, there's no difference between theory and practice.
In practice, there is.

 

[Ger Bee]

A point that has arisen is the fact that the Fuji S2 JPEG is very good - almost as good as the Raw - also look at the Canon - JPEG v Raw is very similar.

Now if someone wants the best they shoot raw – but with the D100 someone HAS to shoot raw. The difference that I have seen myself is spectacular between the Nikon Nef and JPEG – there is no such spectacular difference between the other manufacturers Raw v JPEG modes (but Raw is always better).

I think the Nikon Nef can compete against all the rest with NC3 – but Phil also mentions that the D100 sharpening routines are different with NC3.

One may blame Nikon here – squarely – whether they deliberately did this to give a better image or try to force us to pay more is open to speculation – I know what I think.

This is not the dissapointing fact, but why not having a better
choice here? Untouched with RAW is ok, but for JPG output there
should be practically usable settings...

 

[Ger Bee]

... and NC3 with noise reduction?

I don't see too many people mentioning this. I thought it was cool, cleared up a lot of the coloured noise on my sample image without affecting the highlight which most other noise reducers – that I’ve tried - did soften the high lights as well.

 

[Anthony Medici]

Jim,

You are a funny man... But your statements aren't backed up with the facts. Based on the research you forced me to do since you didn't bother...

The D100 has more noise than the D60, 1D (only at ISO 100), D30 (only at ISO 100) and the S2.

The D100 has the same noise levels as the 1D at ISO 200.

The D100 has better noise levels than the D30 (ISO 200), D1, D1X, D1H, E10, E20, S1 and the DCS760

The documentation is below with references.

On page http://www.dpreview.com/reviews/nikond100/page16.asp
D100-1.25 @ ISO 200
D60 - 0.65 @ ISO 100, 0.86 @ ISO 200

On page http://www.dpreview.com/reviews/fujis2pro/page14.asp
S2 - 0.91 @ ISO 100, 1.11 @ ISO 200

On page http://www.dpreview.com/reviews/canoneos1d/page20.asp
1D - 0.98 @ ISO 100, 1.25 @ ISO 200
D30- 1.01 @ ISO 100, 1.32 @ ISO 200
D1X- 1.31 @ ISO 125, 1.47 @ ISO 200

On page http://www.dpreview.com/reviews/olympuse20/page13.asp
E20- 1.82 @ ISO 80

On page http://www.dpreview.com/reviews/olympuse10/page14.asp and http://www.dpreview.com/reviews/nikond1x/page22.asp

Using the old comparison method, the D1X is much better than the E10, D30, D1 or S1.

On page http://www.dpreview.com/reviews/nikond1h/page21.asp
D1H - 1.31 @ ISO 200
D30 - 1.08 @ ISO 200 (This could be ISO 100, it's hard to tell from the page.)

On page http://www.dpreview.com/reviews/kodakdcs760/page16.asp
The D1x's ISO 400 image is cleaner than the DCS 760's ISO 125.

Phil did not have a review (that I could find) for the DCS720 or the DCS720x.

I count four better at low ISOs than the D100, two which just came out (D60, S2) and two which are only better at less than ISO 200 (1D, D30).

None of the D1 series has lower noise than the D100 at ISO 200. SURPRISE!

--
Tony

http://homepage.mac.com/a5m http://www.pbase.com/a5m

So can you name all the digital cameras that have less noise than
the D100?

If you can, you used all your fingers on one hand, didn't you
because there isn't more than 5 of them.

Of the 12 I can think of, I'd say 10 have less noise then the D100
at lower ISO's,
D30, D60, S1, S2, D1, D1X, D1H, DCS720, DCS760, 1D

and 7 have less noise at every ISO then the D100.
S2, D1, D1X, D1H, DCS720, DCS760, 1D

Is this suppose to boost the D100's integrety? Or were you hoping I
could only count up to five : )

 

[alfred_schiele]

Hi

People keep harping that there is no sharpness issue, but there is.

If you shoot in JPEG.

So in harsh words, JPG is not really usable for "serious" work?

True you could shoot in raw (more noise) or use more sharpening,
also more noise.

Phil's comments (including the elaboration he's provided on this
forum) translate as follows.
If you shoot raw unsharpened, no "noise".
If you sharpen up converted raw, "the way most everyday users
will", there is "noise" -- more "noise" than "everyday JPEG".

Why this leads Phil to state conclusions that the "everyday Joe"
will interpret as criticisms of the hardware is beyond me. These
seem pretty clearly to be software issues and, in the case of NEF,
issues related to software that isn't even in the camera!

Noise is noise, regardless how experienced somebody is or not... A D-SLR is no P&S were I am not as critical.

"Everyday users" will sharpen up converted raw using "the camera's
defaults" and "NC3 default sharpening", including processing by a
relatively poor and stupid sharpening algorithm that makes no
distinction between areas of the picture that need sharpening and
those that don't. In other words, an aggressive unsharpen mask is
being applied to shadow and monotone areas.
I'd be interested to see the "noise" you'd get if you converted a
D60 raw image with sharpening off, then applied a dumb sharpening
algorithm that hits the shadow and monotone areas hard. "Noise",
lots of it.

I am sorry, but when I compare to my G2 the pure image output is much better without doing anything... just print. I know that it is not on pixel level, but what do I care if it looks better in print?

Obviously, Canon's in-camera algorithms and converter algorithms
are smarter than that. It is also obvious that Nikon's are not.

Here I agree.

That may make Nikon less suited to an everyday Joe, but that's not
me, at least when I'm shooting in NEF.

No, thats not the point but it takes much more effort and time to get the same results from D100 as "everyday Joe" with his simple stupid P&S... To be honest, I don't like spending hours for running intelligent batches or whatever to get good looking images and prints.

I am not talking art but average everyday quality...

[I'm looking for that one shot out of three "rolls" (i.e., 1 out
of 100 if I'm truly lucky) that's simply outstanding, worthy of
printing at 8x10 or 20x30 or more (!), and for that I'll want a
NEF. But I can't predict when I'll get so lucky, not in my
photography. Anything else I'll maybe want in a quality JPEG for
viewing on-screen, that's about it.]

So, what workflow gets the most out of the camera? Simply convert
with all sharpening OFF, then apply an intelligent sharpening
method like Fred Miranda's action or the methods
documented/explained step-by-step on luminous-landscape.com as
"high pass sharpening" or better yet "smart sharpening".

What an effort... why doing all this externally when other manufacturers do it in the box? All I would like is an option, either "pure" or "ready processed".

At that point, you have a finished product that has nothing to do
with the "standard all defaults" product evaluated by Phil.

With all this effort I could make a coolpix quality like nearly the same. So were is the real progress? (let's omit the handling please, it is for sure different)

Smooth, sharp, virtually without noise.

virtually? and what about the real thing... 8)=

Can't say whether Nikon will show up with something better w'r/t
JPEG for this camera.

Not only my eyes and feel but Phils numbers tell as well wich current D-SLR may exhibit more noise... And I am sorry, but it seems to be that I am not the only one who found this... I am not talking theory also, I shot the D100 for some time....

Regards, A. Schiele.

P.S. I really would like to find a way to manage this noise thing from D100 as I like the camera itself very much, but I am still searching and I doubt I will find solution that takes not more effort than I like... maybe QImage does better? I won't mind shooting NEF either.

 

[Anthony Medici]

P.S. I really would like to find a way to manage this noise thing
from D100 as I like the camera itself very much, but I am still
searching and I doubt I will find solution that takes not more
effort than I like... maybe QImage does better? I won't mind
shooting NEF either.

What noise thing? The D100 is one of the best. See http://www.dpreview.com/forums/read.asp?forum=1021&message=3158869

--
Tony

http://homepage.mac.com/a5m http://www.pbase.com/a5m

 

[Eric Hanauer]

DMillier the noise is THE number one issue which is keeping me
from getting the D100.

I too feel Nikon fell a little short with the D100.
Perhaps a firmware upgrade would change my mind, that is if one
came out to address this.

Jim K

Isn't it interesting how many experts there are out there who have never shot with a D100?

A happy D100 owner

 

[alfred_schiele]

Hi

P.S. I really would like to find a way to manage this noise thing
from D100 as I like the camera itself very much, but I am still
searching and I doubt I will find solution that takes not more
effort than I like... maybe QImage does better? I won't mind
shooting NEF either.

What noise thing? The D100 is one of the best. See
http://www.dpreview.com/forums/read.asp?forum=1021&message=3158869

To be accurate, the third best? (-;

It is very simple, beond all numbers: If my eyes are dissapointed then this is my measure that was not reached.

If there is no tweak that works for me to manage this then there is something left. Basic images form D100 (beond all sharpening discussion) do not look really "clean" in my eyes and they don't print perfectly "clear" either. Tweak is a must, without D100 images look not better than some P&S images in some regard... And as some others noted I found some slight tone separation in D100 images too. It is for sure all very slight, but sum it up and it explains the average basic image impression. And don't tell me we talk cheap cameras here... it is simply a lack of technology...

Should I tell you, that colour resolution of current 6MP cameras do not meet my imagination of 6MP pixel resolution I would like to see either?

Let's talk in some years again and you will possibly agree without any hesitation... but up to this we have to set comfortable with what is (hopefully) availiable or we have to leave it at all... and up to this I take the freedom to look at things very critical.

Regards, A. Schiele

 

[Ger Bee]

... with ICS+H.

So what do I think. Well the files do look better but do look sharpened. Now the problem I see is the way the sharpening favours the very bright spots – flash reflections from highly polished furniture, teeth and the catchlights in people’s eyes.

The problem then is I’m inclined to dismiss this as an over sharpened file – but there is more – on closer inspection one can see details in clothing, like stitching and logos and small writing become clearer etc.

It might work – I will persevere – as I said initially the very bright spots are definitely over the top but the rest of the image seems to benefit more than these damage it.

 

[Cribbage]

So in harsh words, JPG is not really usable for "serious" work?

I see obvious artifacting in D60, S2, D100, D1X, and 1D JPGs. I consider them unusable for print work even at 8x10, particularly if any manipulation is desired or required, period. At the same time, I understand many people don't notice the artifacting -- they are designed to be hard to notice by design. Again, I'm talking about artifacts, not "noise", not "sharpness".

I can take a person happy with their JPG photographs from any of these cameras and show them the artifacts -- they're always there -- and turn them cold on the whole approach.

I've shown many a friend or acquaintence the MPEG2 artifacts from their DVDs to the point that they find them hard to watch, at least within 10 feet of their TV screen. (And they blame me forever).

Call the JPGs from these cameras clean if you want, they're still compressed and "artifacty" to trained eyes.

Noise is noise, regardless how experienced somebody is or not

Nikon hasn't invested their time in RAW conversion tools that bundle intelligent sharpening. Canon has. Yet the cameras are contrasted, one as "noisier" than another. I just call that a sloppy confusion of hardware and software by the reviewer. Especially considering the fact that new software, that could be released at any time could turn the tables on the conclusions being represented as "gospel" about hardware.

I am sorry, but when I compare to my G2 the pure image output is
much better without doing anything... just print.

What is "pure image output"? The output of RAW conversion with intelligent sharpening?

Obviously, Canon's in-camera algorithms and converter algorithms
are smarter than that. It is also obvious that Nikon's are not.

Here I agree.

Nikon hasn't engineered or priced 3d party software out of the market. I like having Bibble and Qimage out there. If NC3 were free, they might not be.

Other companies would be happy if all you needed was their camera and their printer. I'm happy to have more options, even if I have a more complex workflow. I;d use the more complex workflow anyway, to have maximum flexibility and quality. I'm glad to have a less processed image to work with, in the first place.

Nikon has convinced me that they offer unprocessed RAW, if I turn off all in-camera processing. I couldn't say if the same is true for Canon, I wouldn't know.

To be honest, I don't like spending hours for running
intelligent batches or whatever to get good looking images and
prints.

To each his own. By my standards, you can't with any of the cameras, but then other people are happy to live with photos crawling with artifacts.

I am not talking art but average everyday quality...

OK, I am not.

With all this effort I could make a coolpix quality like nearly the
same. So were is the real progress?

Doubt it. You and I are clearly striving for different definitions of quality.

Not only my eyes and feel but Phils numbers tell as well wich
current D-SLR may exhibit more noise...

Phil's "numbers" compare software, not hardware. They presume a workflow I wouldn't recommend and certainly don't use.

 

[alfred_schiele]

Hi

... and NC3 with noise reduction?

I did. And I will experiment further. But I am not totally convinced by this method either. It removes most of the subtile details as well. I would really like to see this NR equalized to certain brightness levels, especially the darker ones. Maybe it would be acceptable then.

To add, Edge noise reduction is a fuzzy thing as well. It sometimes gives noise a even better visible structure, sometimes like neat image introduces under certain circumstances.

I don't see too many people mentioning this. I thought it was cool,
cleared up a lot of the coloured noise on my sample image without
affecting the highlight which most other noise reducers – that I’ve
tried - did soften the high lights as well.

It is not bad, but it does not keep the subtile details either, as Phil said "water-colour" like, and indeed I found this a very good description for what NC3 NR does.

I think only a "intelligent" smoothing-sharpening action can do better than what I currently tried. But I find the effort for all this a bit over the hill, it should be simpler with professional tools... I am no pro but isn't time=money...? I am really not obseesed by postprocessing for such purposes...

Regards, A. Schiele

 

[Phil Askey]

Call the JPGs from these cameras clean if you want, they're still
compressed and "artifacty" to trained eyes.

High quality JPEG's saved at Fine quality from the current crop of D-SLR's will be very hard to distinguish from TIFF's. But I'm assuming you have trained eyes, please examine the following JPEG and show us a crop of all the JPEG artifacts.

http://img.dpreview.com/reviews/CanonEOSD60/Samples/QualityRes/IMG_9872.JPG
(2,893 KB)

Phil's "numbers" compare software, not hardware. They presume a
workflow I wouldn't recommend and certainly don't use.

My numbers test the SYSTEM known as the digital camera. The digital camera hardware is useless without the "onboard software" = firmware. Unless you can find somewhere selling third party D100 firmware then you're pretty much stuck with that which comes with the camera (unless it's updated later).

--
Phil Askey
Editor / Owner, dpreview.com

 

[alfred_schiele]

Hi

Isn't it interesting how many experts there are out there who have
never shot with a D100?

In my case I can assure you I talk only about my real life experiences with the camera. Possibly too short but long enough to find out a lot I wanted to know...

A happy D100 owner

I had some defective pixels and maybe if this didn't happen I would be like you, somehow still "happy" too... but so I returned it.

There were simply some little things left that didn't fully satisfy me, mostly subtil aspects of image quality (not the basic sharpness discussion though). I expected much from such a D-SLR... possibly too much. And so I am hesitating to look for another one. Just to add: I personally liked the camera design most of the current alternatives. But what does it help if I feel that images are only average?

Regards, A. Schiele

 

[DavidMillier]

This is such an exaggeration.

I defy most people to tell the difference between a D100 raw and a D100 jpeg on an A4 print of a normal photographic subject (i.e. one not specially chosen to reveal the difference).

Yes raw is better, yes you can clearly see the difference at 200% on screen with an image of a res chart but in real photographs, viewed in a normal way, the differences are subtle and only noticeable when the picture is so boring there is nothing else to enjoy.

Now if someone wants the best they shoot raw – but with the D100
someone HAS to shoot raw. The difference that I have seen myself is
spectacular between the Nikon Nef and JPEG – there is no such
spectacular difference between the other manufacturers Raw v JPEG
modes (but Raw is always better).

I think the Nikon Nef can compete against all the rest with NC3 –
but Phil also mentions that the D100 sharpening routines are
different with NC3.

One may blame Nikon here – squarely – whether they deliberately did
this to give a better image or try to force us to pay more is open
to speculation – I know what I think.

This is not the dissapointing fact, but why not having a better
choice here? Untouched with RAW is ok, but for JPG output there
should be practically usable settings...

 

[DavidMillier]

Alfred

I cannot believe you can call the images average!

My opinion is that they rival medium format (if not in absolute fine detail, at least in overall look).

Images from the D100 are ten times better than I can get with film and a 2700dpi film scanner, being superior in sharpness, colour balance and grain.

What printer do you use?

ps

For a real look at what next generation cameras will be producing, try some stitched photographs. I did a 3 way stitch (14 Mpixels) as a test, and at A4 there wasn't really that much extra visible detail in the stitched shot over the single shot.

I suspect that unless printed bigger than A4, future superior cameras would not improve image quality much further that you can get from the current D100. 6MP and the large sensor is better than you need for A4...

Isn't it interesting how many experts there are out there who have
never shot with a D100?

In my case I can assure you I talk only about my real life
experiences with the camera. Possibly too short but long enough to
find out a lot I wanted to know...

A happy D100 owner

I had some defective pixels and maybe if this didn't happen I would
be like you, somehow still "happy" too... but so I returned it.

There were simply some little things left that didn't fully satisfy
me, mostly subtil aspects of image quality (not the basic sharpness
discussion though). I expected much from such a D-SLR... possibly
too much. And so I am hesitating to look for another one. Just to
add: I personally liked the camera design most of the current
alternatives. But what does it help if I feel that images are only
average?

Regards, A. Schiele

 

[Ger Bee]

• There's the rub,

I tend to agree that it is a lot of effort and whilst I might like the effect it is slow for me.

I've just tried the new Bibble but I'd have to buy a new PC to run it --- so I can't, in fairness even sample its worth.

But I find the effort for all

this a bit over the hill, it should be simpler with professional
tools... I am no pro but isn't time=money...? I am really not
obseesed by postprocessing for such purposes...

Regards, A. Schiele

 

[Ron Frank]

The auto focus on the G2 was simply the worst experience of my
life. Beautiful photos for general purpose and some decent macro so
the camera provides most of what I wanted BUT> > > > I still use film
cameras for important action stuff but the move to a Digital SLR is
being lookked at a lot sonner than I wanted.

The G2 was never designed to shoot sports action.

If the Nikon D100 can get me through 80% of my sports needs I
probably will go with it..The D60 is just not an option (even
current Canon enthusiasts admit it failings here) and the S2 sync
is making me hesitant even though I prefer its resolution.

Maybe the Nikon D1x? But doesnt the D100 have a slight edge in
resolution and speed is about the same? (cept for sync ).

I don't know..need some more time to figure this out.

Not really. The camera you want is right under your nose, and has been on the market for over a year. The D1x/h is what a sports photographer needs. If you can not afford one, then purchase a f5, or F100 and stick with film.

The D100 may prove to be OK at sports. But IMO the limit in speed is not going to make it a good sports camera. I'm talking about buffer speed, not just the focus which is also a factor, but one that can be worked around. Firing a burst of shots and then missing critical action while waiting for the camera to download images is unacceptable in sports photography.

The D1h is really what you need, the D1x if your a guy who can live with a bit less buffer and camera speed, or the Canon 1D.

Ron

Bad comparison.

The D100 is NOT an F5!

Have you worked with either an S2 or a D60?

Battery issues abound with the S2 and the D60 is so slow at
focusing, I have heard moans come out of people's mouths while
using it.

 

[alfred_schiele]

Hi

So in harsh words, JPG is not really usable for "serious" work?

I see obvious artifacting in D60, S2, D100, D1X, and 1D JPGs. I
consider them unusable for print work even at 8x10, particularly if
any manipulation is desired or required, period. At the same time,
I understand many people don't notice the artifacting -- they are
designed to be hard to notice by design. Again, I'm talking about
artifacts, not "noise", not "sharpness".

I agree. I can see some too, coloured sprinkling pixels, clumsy red/blue edges, and some more... and of course stronger JPG artifacts... I looked at the S2 shadows and...? fuzzy by JPG artifacts, not really smooth either.

Noise is noise, regardless how experienced somebody is or not

Nikon hasn't invested their time in RAW conversion tools that
bundle intelligent sharpening. Canon has.

Again I agree. This confirms what I think of the output files too...

Yet the cameras are
contrasted, one as "noisier" than another.

Regardless how it is created, it apperas like this. I don't care about the reason of noise if it visibly there...

I just call that a
sloppy confusion of hardware and software by the reviewer.

The user can not separate things to its fundamentals... unfortunately.

Especially considering the fact that new software, that could be
released at any time could turn the tables on the conclusions being
represented as "gospel" about hardware.

Yes, but will that happen? I hesitate to invest only on such theoretical possibilities... Yes, if I work hard enough, I can make my grandma fly too, but again, will this happen in reallity? I doubt... I would belevie more to see the next camera instead...

I am sorry, but when I compare to my G2 the pure image output is
much better without doing anything... just print.

What is "pure image output"? The output of RAW conversion with
intelligent sharpening?

Pure in this sense is what I get without additional explicit tweaking. As you said, there is always an connection between sensor, firmware or aquirering processes, but then you get a file in the end, don't you? This is the first impression you get and the result with least effort. Well RAW allows much more different "basic" output, but I am thinking of the "simple, staightforward" way...

Obviously, Canon's in-camera algorithms and converter algorithms
are smarter than that. It is also obvious that Nikon's are not.

Here I agree.

Nikon hasn't engineered or priced 3d party software out of the
market. I like having Bibble and Qimage out there. If NC3 were
free, they might not be.

Bibble for D100 is out now. Will have a look on it. But I would prefer a single converter wich simply works instaed of all the testing which one is better...

Other companies would be happy if all you needed was their camera
and their printer. I'm happy to have more options, even if I have
a more complex workflow. I;d use the more complex workflow anyway,
to have maximum flexibility and quality. I'm glad to have a less
processed image to work with, in the first place.

I said nothing against this strategy! But I would appreciate N* to add simpler methods to tweak out more of a good camera with less effort...

Nikon has convinced me that they offer unprocessed RAW, if I turn
off all in-camera processing. I couldn't say if the same is true
for Canon, I wouldn't know.

Well, I think I share your impression again here, also I didn't use Canon RAW myself and can only judge from different examples I found in the net.

To be honest, I don't like spending hours for running
intelligent batches or whatever to get good looking images and
prints.

To each his own. By my standards, you can't with any of the
cameras, but then other people are happy to live with photos
crawling with artifacts.

Maybe I am looking for something inbetween. And maybe I agree that current cameras need tweaking to get the better out of them. It depends on the purpose if it is worth to do this... But using a P&S on the other hand is no real alternative if you like a quick and responsive camera... you have to take the bitter pill of postprocessing? One of the aspects I dislike...

I am not talking art but average everyday quality...

OK, I am not.

I understood this. But shouldn't a good camera like D100, maybe let's include a better aquire software, do better?

With all this effort I could make a coolpix quality like nearly the
same. So were is the real progress?

Doubt it. You and I are clearly striving for different definitions
of quality.

(-; Was a bit callenging, I know.

I am sorry that I refer to my own imaginations, possibly like you do as well. If you like to you can tell me more about your work. Maybe I can learn a lot!

Not only my eyes and feel but Phils numbers tell as well wich
current D-SLR may exhibit more noise...

Phil's "numbers" compare software, not hardware. They presume a
workflow I wouldn't recommend and certainly don't use.

I think Phil tries to put it all together, a way he assumes average users would do as well. It is quite clear that there are some out who do different. Concerning myself I saw that I didn't get what I expected from the simple method but I am not sure how much effort I accept to get better or how far I'd like to go. A too complicated, too time consuming workflow can frustrate the same as a bad initial image... And remeber, I get not paid for what I do (It even does not pay back in any way...) and I have limited time for my hobby either...

Regards, A. Schiele.