D7 vs. CP990 contrast WARNING BIG!

What mode did you use for the D7? In my testshots, I saw big differences between RAW and JPEG shots (even when I converted the RAW picture to JPEG with the Minolta tool and compared it with a straight-from-the-camera JPEG).

The RAW pictures has less noise (most noticable with long exposures) and I didn't loose too much shadow detail with EV -0.7.
tc

These are just JPEG from the camera. Shooting RAW is a whole new experiment that I'm leaving for another day. Bryan

You will be impressed. My tests with exposures around 4sec showed that a ISO 200 RAW image has nearly the same amount of noise than a ISO 100 JPEG image.

If there's enough time for it, I mainly shoot in RAW (even I hate the size of the files).
tc

Bryan Biggers said:

These are just JPEG from the camera. Shooting RAW is a whole new> experiment that I'm leaving for another day. Bryan>

Click to expand...

crowning said:

You will be impressed. My tests with exposures around 4sec showed
that a ISO 200 RAW image has nearly the same amount of noise than a
ISO 100 JPEG image.
If there's enough time for it, I mainly shoot in RAW (even I hate
the size of the files).
tc

Click to expand...

Hi tc,

I have to admit, that the main thing that is keeping me away from using RAW much on the D7 is the Viewer utility. I don't like what it does to colors and the editing choices are minimal and it is super slow on my primitive computer. Some day programs like Bibble or Qimiage might support it, but I have not seen much sign of anything cooking from Mike Chaney, and we aren't supposed to ask

BTW, I think that (maybe) the main reason that RAW seems to have less noise is that the sharpening routines applied by the viewer are better than the in-camera routines, and the noise on the D7 seems to be closely tied in to the sharpening. Bryan

Bryan Biggers said:

compensate for that? Could I test this on different cameras by
taking a picture of something pure white and checking it in a photo
program? I wonder what 18% gray would look like in a photo program

Click to expand...

Maybe Phil has already done this test for us, he publishes White patch colors for all the cameras. The D7 gave 220,224,226 on White, and the 990 gave 231,231,231, the D7 actually metered the Whitel lower. Hmmm wouldn't the illumination level have to be the same though? Bryan

Now the second set of shots have turned into the small versions with the photopoint banner. It seems to take a few hours. Bryan

Bryan Biggers said:

Frank B said:

Are you sure your link is to the full size picture? PhotoPoint
may have added a downsized image which if you click at PhotoPoint
will change to the full size with a different link.

Frank B

Bryan Biggers said:

The really weird thing is...
I originally linked to the full size shots on Photopoint in my msg.
When I first posted the msg, the full sized shots appeared. Now,
about 6 hrs later, the full size shots have been reduced to those
reduced size shots with the "hosting provided by photopoint
banner". Photopoint must have done that, I don't know how or why.
What good is it if you can't link to some full size shots? Does
anyone understand what happened?

The originals are in this album -Bryan

http://albums.photopoint.com/j/AlbumIndex?u=785850&a=13585450&f=0

Click to expand...

Click to expand...

Click to expand...

Bryan Biggers said:

Bryan Biggers said:

2) Nevetheless this can show that what all the people call "dynamic
range" is 95% really only exposure / contrast / saturation
difference.

Click to expand...

Yes, this is what I'm trying to show. The D7 seems to be making
"different" exposure and contrast decisions than the 990 does, and
it makes the shots look, well, different, but it is not because the
camera does not have what most people think of as "dynamic range".
The D7 can be made to take shots that are pretty close to the 990.

Click to expand...

I hope that now we stop to speak about the sex of the angels...

Bryan Biggers said:

Bryan Biggers said:

3) In MY system (matrox+Philips brillance+ PSP calibration) I'm
undeiced what shot cleaily is better than other because:
a) the colours seem more natrural in D7 also if a little
oversatured for my taste

Click to expand...

Click to expand...

Bryan Biggers said:

That is my fault, I picked +2 saturation on the D7, I probably
should shoot with +1. I'm leaning more towards +1 now that I have
printed some of my +2 shots, they are too saturated on the printer.

Click to expand...

It can be. You have to try to find it. This seem to be the character of D7. Made for who like to experiment. Not a point & shot.

Bryan Biggers said:

Bryan Biggers said:

b) the colours have more appeal (also if not so close to the
reality) in the 990

Click to expand...

Yes. The 990 has a very real look. The D7 colors were actually very
closer to reality, including the overall warm tone, I have the
monitor set so I can see the scene out the window and the D7 colors
were just perfect, but if you didn't see the original scene the 990
shot actually "looks" better.

Click to expand...

Seen the same also in many other shots.

Bryan Biggers said:

Bryan Biggers said:

4) Shadows as I see in MY system:
a) the shadow gradient is cleaily visible in both two.
b) the shadows of D7 are darker than in 990 (exposition difference)

Click to expand...

Yes. The 990 seems to have a tone curve that brings the shadows up.
That can be good and it can be bad, it depends on what kind of look
you want.

Click to expand...

Yes, another question of personal taste.

Bryan Biggers said:

Bryan Biggers said:

c) the shadow gradient seem more natural in D7; The 990 seem to
have some "step" (see the left down corner)
d) the shadow gradients in these shots of the D7 seem more close to
the rendition of a Pro camera than the 990 (this probably due to
the lenses; please please don't flame me for this heretic
conclusion!)

Click to expand...

The other thing that I see on the D7 vs. 990 shots, (not so much on
this one) are a purple cast to some objects due to chromatic
aberration on the 990 that the D7 does not have. On some scenes you
can really see it. The Noise is different between the two cameras
too. The D7 noise is more fine grained, and at least on my printer
at 8x10 it disappears. The 990 noise is broader, more like an "oil
slick" and I can still see it on a print.

Click to expand...

Some days ago I seen a post from a French guy referring to a french site (made by him) where it was a test 990/D7. I don't want to discuss about the personal preferences, therefore it's not possible to say that the colour of 990 in a particular shot were it is coming with a +strong+ magenta cast is more accurate than the D7 (on the same subject) that don't have this cast. You can like or not, but you can't say that a subject that naturally is strong yellow and that from the shot come magenta is well rendered. You can like magenta more than yellow, but this is another question.

"We will never stop to say that are the lenses to make the photo more than the camera". Translation for the "how to do" of this site:

http://www.nadir.it

It's a site for Photo-art. Very high end target.
This site have been awared for a import premium for art & culture in Italy.

The site is in Italian, but the images speak well in all languages if have 5 minutes to visit it ;-)

This is the link of a menu that go the the shots of varius authors.

http://www.nadir.it/portf-immag_00.htm

Bryan Biggers said:

Bryan Biggers said:

6) at this level we have also to consider the diffrences due to the
display system capibility (video board+monitor+setup) : different
system can produce diffrent result more favorables to a camera than
to the other.

Click to expand...

Well, yeah, and I know that my monitor and printer are unique to
me, and what looks good to me might not look good on another
monitor.

Click to expand...

it's the print in a phototlab that can tell the last word. What kind of service do you use?

I heard that the Fuji photolab is able to bring out all the qualities of the D7 shots.

Bryan Biggers said:

Bryan Biggers said:

A conclusion test it can be some indoor shots with less sun light.

Click to expand...

Well, I picked the outdoor flower shot because of the critisism
that the previous indoor shot was "staged" to "bias" the test some
how. I'd rather shoot indoors though, I get a lot better control
and the wind isn't blowing so much. Where I liver there is a big
field of hay to the West, and the wind seems to really get going
over it so that it almost never stops blowing except around Sunset.
Bryan

Click to expand...

I know but if you want to value the shadow gradients, despite to what somenoe say, a shot in direct sunlight is not the best situation. It can be during the sunset but not in the middle of the day. (It's still some people here that go hunting the right light before to take a shot??? ;-))

The first shots you have been made (table shot) are more suitable to this scope IMHO.

Bryan Biggers said:

I'm still thinking about what that
means! If so, would setting the camera for -0.7 EV all the time
compensate for that?

Click to expand...

Hi Bryan,

Same for me, I have never been able to unambigeously interprete the real meaning of this issue.

For me, the most intuitive interpretation is that it has been empirically determined that the average scene equates to a brightness of 13%, separate from the fact that the eye (or brain if you will) interpretes 18% gray as in the middle beween black and white under the same illumination.

This would mean that the average scene is metered lower than what humans find average and this would mean that such a camera would underexpose which is not consitent with the eos3 story.

The other interpretation is that it has been determined that not 18% but 13% gray is experienced as halfway gray and that therefore such a camera would think that the eye is more sensitive than what was tought previously and hence, it would overexpose compared to an 18% camera. This is consitent with what was observed on the eos3

Stiil I think that the last interpretation is less convincing.

For nature shots ok, it could be that evolution has "calibrated" our visual system so that the dynamic range of our "hardware" (or should I say "wetware") is optimally used. The same thing happened to our color perception. But it would also mean that all previous experiments that determined the 18% result were completely wrong which is quite hard to believe. And the average scene we see today can harly be compared with the one that was present a few hundred thousand years ago.

On the other hand, if interpretation 1 is correct than that would mean the Canon engineers have been on the wrong track which is even harder to believe.

Very puzzling to say the least.

Jean Paul