G2 and F707 sample images (comparison)

The Sony reds certainly look overdone, but they aren't even close
to being clipped. The largest value I could find was 202.

Ron Parr
clipped in the sense that the red colors are pure red, sorry.
Clipped means clipped. If the Sony colors aren't clipped please
don't use that term.
yes, I agree to that, so I'll head off to my corner now.... =(
centre of sock F707: R188: G0: B0
centre of sock G2: R150: G15: B15
Good catch here! This is actually rather strange, because I don't
think any camera in Phil's tests has ever come up with zero in the
other channels on color tests for any primary color. Strange. Hmm.
yeah, I find it very strange that the green and blue colors are
negatively clipped ( haha ) ;^)
Could it be due to the color noise reduction? The blue and green components looked too noisy so it did away with them? Maybe a new and diffferent RGBG mask is being applied to the 5meg CCD?
However, isn't the Sony much closer to the red of the Red Sox
colors? The Sony may be overdone, but I think the G2 is much
further off than the Sony...

(It's been almost 20 years, but I did grow up in Boston, and the
Red Soz stuff was always seriously red.)

David J. Littleboy
Tokyo, Japan
 
I've had a play with a variety of posted samples, but Softtower has
now given us an excellent variety of images to compare, and since
my feeling is that the G2 comes pretty close to the truth most of
the time I've had a brief attempt at matching a couple of the 707
equivalents.

I started with the shop-front/streetscape, with the Smoothie King
sign and the red bicycle in it, with the aim of seeing how far I
had to go to approximate the G2 colour balance. The resulting
numbers look alarmingly large, but actually don't work too badly on
several of the other samples -- which bodes well since my eventual
aim is to come up with a workably small suite of actions. I stress
that these are quick and rough, but they give an idea of the
lengths one might need to go to. I've also long since decided (this
will be my first digicam) that auto WB is for the birds, a feeling
that Ulysses seems to endorse a bit later in this thread:
Most long-time F-series users will use the One-push by second nature. It's a habit we've developed. But as you've mentioned in another thread (or on your page), and as Phil has successfully argued, it would be far more an improvement to the Cyber-shots if they included more presets.
As for Auto w.b., it's always going to present a sort of crapshoot (no pun intended), as it will swing its color temperature dependent upon the colors in the field of view ...
Anyway, here are the Photoshop offsets I came up with. Both samples
had been shot with auto WB and I thought the G2 was a tad warm, so
I white balanced both images to start with using the grey mailbox
as a reference. The 707 pic was then treated thus:

Red: Sat -15
Yellow: Sat +12, Hue -6
Green: Sat -40, Hue -5
Cyan: Sat -45
Blue: Sat +10
Levels: Gamma 1.03

As I said, this is a very rough start. I plan to get a lot more
detailed using a colour chart when the 707 arrives, but the
exercise has left me feeling very optimistic about the 707's
underlying capabilities (while cursing it for not offering RAW :-(
).

Mike
I am going to try what you did and play around with that same picture and when you work with that picture do you always get the original or could you try it on the 800 X 600?
Randy
There's
another color, misinterprited indoors by F707. If you have that
"Crest" toothbrush at home, compare the color of "e" and "t" on the
pictures with original. (or look at G2).
Yes, I mentioned the "bathroom product" photo, and did notice that
the "C" looks more orange, and the others more cyan on the F707,
other colors and exposure being comparable.

Anyway, it's something picky people like me will have to take into
consideration .....
Count me in with that one! ;-)
 
Softtower,

It's a great comparison! Thank you!!!!! - Ochen xoroshee sravnenie dvux functsionalno raznix fotoapparatov.

These two very different cameras produce very similar images. The 707 images have a bit more contrast and saturation. But it depends on someone taste. Adjusting 707 images contrast and saturation I was able to receive to almost identically colored images. My computer setup includes two video PCI cards and two monitors. Win98 is capable to increase the desktop space. The monitors and the video card are identical. The monitors adjustment is also identical.

The Sony images retain color in high lights better (may be 14 bits per color helps) and the resolution is noticeably higher.

Several messages stated that these two cameras are very similar cameras and suggested happy photography with either one. It depends. These two cameras are very different and Sony represents with its 707 the next step in the development of the digital cameras. The 707 superiority will not last long also it will be another better one soon. The only negatives for 707 from my point of view are - Price, x5 only Zoom, size, no image stabilization. With these features improved the 707 will be the ideal camera for me (at least for some time) and I will accept the higher price.

I would select the Sony 707 over G2because: x5 zoom, 14 bits, contained CZ (must for a few microns pixel), low light capability and functionality (assume that $$$ is not a problem).
Leo
I've been looking at G2 and F707 for a long time. Could not decide
which one is for me. And as some people already know I bought both,
hoping to compare their images. I don't want to start a new G2 vs
F707 war, this message is addressed to people who are having hard
times trying to pick one of those two. I am not even saying which
camera I decided to keep. I know I've been known here for my
criticism, so I feel obligated to say this: Sony did a lot better
than I expected.

Here we go: http://kontsevoy.com/comparo

--
Eugueny

P.S. Files are still uploading from my home PC to WEB-server, not
all pictures are available right now, it will take about 20 more
minutes. ( 120MB)
 
Hey Eugueny

Thanks for the great job on these. Was an excellent follow up to our earlier discussions about "does image quality matter?" It's a big help for me (amateur) to see a real-life comparison like this. And appreciate your keeping your decision private to help facilitate a more balanced discussion.

Not to be a pig, but would love to see a couple of other shots

1) night time with auto white balance. The sony nightime preset was very yellow. But the auto WB is a quick solution?

2) indoor flash - especially with models. So many of us schmucks would also use the camera for family parties and such.

Ok, so i am a pig!

Also, was great to see that the portrait shots did not show nearly as much of the red cheeks as I had feared on the sony. So many previous shot's i'd seen exhibited a real flush look.

In any case, whether or not you're looking to add to the comparison gallery, this was a big help. If anything, it got me more comfortable with the idea of the f707. Now if I can be comfortable with it in my hands. (Any thoughts on tactility

Looking a gift horse directly in the mouth,
dave
I've been looking at G2 and F707 for a long time. Could not decide
which one is for me. And as some people already know I bought both,
hoping to compare their images. I don't want to start a new G2 vs
F707 war, this message is addressed to people who are having hard
times trying to pick one of those two. I am not even saying which
camera I decided to keep. I know I've been known here for my
criticism, so I feel obligated to say this: Sony did a lot better
than I expected.

Here we go: http://kontsevoy.com/comparo

--
Eugueny

P.S. Files are still uploading from my home PC to WEB-server, not
all pictures are available right now, it will take about 20 more
minutes. ( 120MB)
 
If we were talking about, say, extreme whites, then I'd fully
agree.
Except we're not. Quite the opposite in fact. One of the main issues that some people have with some of the G2 is that there is a lack of shadow detail. I have attempted to point out that this is largely down to the metering mode and had evaluative metering been used on the G2 shots I'm confident we'd have seen more shadow detail. This is not based on conjecture, but on ownership of a G2.
But we're not -- the lighter end of the stonework detail is
very comfortably inside the dynamic range of both cameras. We're
only dealing with upper mid-tones here. Exactly the same quality of
the 707's imaging jumped out in Phil's Tower Bridge comparison with
the D7. Ironically, some people (myself included) were initially
wondering whether the 707 was "smearing" detail in the interests of
noise reduction. Turns out to be not the case, most certainly not
in printed output.
Well sounds like you've made you mind up in advancel. I can only advise you as a G2 owner that the G2 has a superior metering mode which wasn't used in these shots. Since I have seen direct comparisons of what the G2 can do in center-weighted averaging mode compared to evaluative, I'm sure that the dynamic range of many of the G2 pictures in this comparison shoot could have been improved on to some degree (not that they aren't pretty good on the whole anyway).
I'm not carrying a torch for the F707, btw -- I personally think
the Sony colour is terrible. But I'm confident it's capturing all
the image data I will want to work with; and it has other
advantages, notably that big zoom range.
What is an advantage to person A is a disadvantage to person B. Personally I find the extra bulk of the F707 an unwelcome trade-off for the undeniably excellent zoom.

Martin
 
No, no, no!

You've got it all wrong. It's...

"Pocketa-pocketa-queep! Pocketa-queep! As Walter Mitty stared off into the mists and the rising sun..."

-Ed
"clippety clappety, clippety clappety"
... on whose bridge? ;-)))
No, no, no...
That's "trip-trap, trip-trap, trip-trap". That's not the same thing
at all.
Of course ... I was allowing that maybe you'd "got rhythm"
 
Not to be a pig, but would love to see a couple of other shots
Ahh... I just dropped off the UPS package with one of the cameras in it. ;-(
1) night time with auto white balance. The sony nightime preset
was very yellow. But the auto WB is a quick solution?
Manual white balance was pretty good though, and Sony's manual is very quick to set.
2) indoor flash - especially with models. So many of us schmucks
would also use the camera for family parties and such.
Bryan Siverly is going to receive his both cameras (G2 and F707) soon, I did not have time for the flash testing, but he promised to check how these cams comete in that department.

Eugueny
 
I don't follow the reasoning: The metering mode has nothing to do with the dynamic range. Please tell us precisely how you think evaluative metering would have helped. I see two possibilities:

1. More exposure: This would only exacerbate the blown white areas.

2. Less exposure: The dark areas are already about the same exposure as the 707 and about the same detail level, so while less exposure might help the presently overexposed white parts, how do propose that it would avoid underexposing the shadow areas and sacrificing detail there?

Rememer that even if the metering mode is "matrix" or "evaluative" the exposure applies to the entire frame.

Ron Parr
 
Bryan Siverly is going to receive his both cameras (G2 and F707)
soon, I did not have time for the flash testing, but he promised to
check how these cams comete in that department.
Just got the F707, and I was snapping pictures as it charges... so I'm anxious. :)

Anyway, I'll be happy to do these. The pictures I've taken so far with the F707 and the HVL-F1000 flash were pretty disappointing, so I want to make sure I'm not doing something wrong by reading the manuals. I'm noticing consistent over-exposure and a variable blue-tint (sometimes it appears, sometimes it doesn't on the exact same scene)... it just seems odd.

Anyway, when I'm sure what I'm doing, I'll compare both cameras and let the images speak for themselves.

Softtower, if you don't mind, I'd like to work out how I can send my samples to you and we can keep them on the same page. Would that work somehow? You want to send me an email on your preference?

Thanks!

Bryan
 
There is truth, and there is truth.

Truly, we've got to be careful about reading too much into the
comparisons as no two eyes see things the same, no two monitors are
calibrated the same, and no two cameras have the same strengths.
For instance, I find that the G2 consistently leans too blue in
nearly every circumstance. If I'd have to Photoslop the Dark Angel
images, I'd also have to do the same to just about every G2 image.
Six one way and half-dozen the other. We have to choose the color
cast of our poison here. Each camera still needs work.
I would like to point out the night shots of the swimming pool. The G2 gives us an otherworldly electric blue, while the F707 renders it an unappetizing green bog. The F707 is closer (at least to most film representations I have seen.

That brings up a point that I don't believe I have seen enough in these discussions. Film isn't all that terriffic at color truth either. One of the things that soured me on film was having umpety-ump brands and grades of film to choose from depending on what you were shooting—and you always had the wrong type in the camera.

Only professionals (who are using someone else's time & money) can afford to put up with that BS. With digital I just pop it, slop it, and drop it.

And anyone who says he is dissappointed that he would have to post process his gigi-pics and uses that as a condemnation of a camera can have my old Kodak Brownie and my best wishes.

-Ed
 
I don't follow the reasoning: The metering mode has nothing to do
with the dynamic range.
Of course it does. How do you figure that it doesn't?
Please tell us precisely how you think
evaluative metering would have helped.
The evaluative metering mode gives a more balanced exposure, with less blown highlights and less underexposed shadow. Ergo better dynamic range.
Rememer that even if the metering mode is "matrix" or "evaluative"
the exposure applies to the entire frame.
Yes of course, but evaluative / matrix metering makes a better assessment of the picture as a whole and in most cnormal ircumstances gives a more balanced exposure and hence more likelihood of more detail in the shadows and less over exposure in the highlights. Centre-weighted averaging means that unless the centre is well balanced, the exposure will lean towards over exposing some parts, or under exposing others.

Perhaps I cannot explain it that well, so here's a better explanation:

http://www.dpreview.com/learn/Glossary/Digital_Imaging/Dynamic_Range_01.htm

Looking at this page, in my experience evaluative metering on the G2 tends to produce pictures closer to the one on the left and centre-weighted averaging closer to pictures like the one on the right.

Martin
 
I don't follow the reasoning: The metering mode has nothing to do
with the dynamic range.
Of course it does. How do you figure that it doesn't?
We're talking about the dynamic range of the camera as a sensing device. This is constant. The dynamic range of a scene and the dynamic range that is captured by the camera at a particular setting as a function of the scene are different things.

The argument here is that this photo shows that the G2 does not have the inherent dynamic range as a sensing as evidenced by it's inability to capture as much range in the photo of this particular scene in comparison to the 707.
Please tell us precisely how you think
evaluative metering would have helped.
The evaluative metering mode gives a more balanced exposure, with
less blown highlights and less underexposed shadow. Ergo better
dynamic range.
Ahhh... I think I see how you're thinking about it. It's true that correct exposure will make you less likely to clip bright spots or lose dark spots due to incorrect exposure. However, it can't change the inherent low-high range of the camera. In the picture in question, the two cameras are assigning roughly the same brightness levels to the dark stuff, so the low end of the range is the same. However, the 707 gets more of the high end, so we conclude that it has more range.

Now, assuming that the contrast setings haven't interefered with the test by boosting the G2's high end to the poiint of clipping, this would show that the 707 has a wider range. There really isn't any way you can get around this. Evaluative metering isn't magic.

You might take the contrast argument and this might be a sound line of reasoning. I could counter that a good contrast enhancing function ought not to produce these kinds of results at the default level, i.e., the default level is either too high or it ought to back off if it's blowing out a good part of the picture.
Yes of course, but evaluative / matrix metering makes a better
assessment of the picture as a whole and in most cnormal
ircumstances gives a more balanced exposure and hence more
likelihood of more detail in the shadows and less over exposure in
the highlights. Centre-weighted averaging means that unless the
centre is well balanced, the exposure will lean towards over
exposing some parts, or under exposing others.

Perhaps I cannot explain it that well, so here's a better explanation:

http://www.dpreview.com/learn/Glossary/Digital_Imaging/Dynamic_Range_01.htm
I just don't see where you're going with this. You can't decrease the exposure on the G2 shot without losing detail in the shadow areas. The shadow detail is barely discernable as it is. How the shot is metered simply has nothing to do with this. It's that simple.

Ron Parr
 
Is there half press button to lock focus before shooting? This would help to capture moving squirrel. If it doesn't have a half press button, 707 is probably isn't for me. Tom
Thanks for the pics and the comparison. I have been enjoying
looking at the pics and they are both good cameras. Which camera
seemed to focus faster, was either camera faster or better on
something moving, and was the weight that noticable between the
cameras?
Hm... At full telephoto F707 is slower BECAUSE it's telephoto is
much more 'tele' ;-) Normally they perform the same (at least I
don't remember big difference, didn't feel it).

F707 is heavier. Actually when you attach a strap to it, F707
becomes very easy to 'wear' camera, just dangles behind your back,
"looking" down. You'll forget about it soon.

Both cams couldn't autofocus well on something moving (my wife
failed to take a picture of a jumping squirrel with both cameras -
they kept trying to focus on the tree, not on the squirrel) - very
common problem, I don't know a cam which can do that. (I've never
used an autofocus SLR, by the way).
 
I sense this discussion is going 'round and 'round with no resolution in sight. Perhaps I can add a different angle:

We would all like to see cameras compared uner the conditions we expect to use them. If we are a P&S shooter, then the relevant conditions are full auto everything all the time. For the photography enthusiast, we would like to see the camera used under the "optimum" conditions since we would like to all llike to shoot under optimal conditions. This takes considerable experience in understanding the performance of each camera to know what "optimum" looks like for each camera.

I think tha Phil, as does most professional reviewers, tries to strike a balance between full auto and optimum. After all, how many cameras does he review a year, maybe 30? He uses a mix of laboratory set shots and images from wandering about London. Softtower had less than one week with two new cameras, so he had to make some choices and compromises. I thought he did quite well for a first time attempt. Similar comparisons on this forum in the past, like S75 vs G1, were done under full auto conditions, where performance was clearly limited to the default parameters (like sharpeing). Again, this is quite relevant to those who use full auto all the time.

I beleive Martin's point is that evaluative metering would be much closer to optimum utilization of the G-2s dynamic range. My personal view is that anything less than RAW converted to TIFF is not optimum for the G-2, but that is a much more involved comparison. Afterall, if I had a G-2 that's how I would use it, so those conditions are very relevant to my view of "optimum". I think Ron's point is that absolute dynamic range is set by the hardware and invariant of camera adjustments (how well one exposes the shot). Of course what is seen in the comparison photos is dependent upon quite a number of variables, including exposure.

Just to broaden the example base, here is a recent thread about Pro90 settings where one user much preferred a lower contrast setting. He felt it provided more tonal range: it is a wonderful shot
http://www.dpreview.com/forums/read.asp?forum=1010&page=1&message=1625992

This is one of the reasons why I feel RAW is a useful feature, as one can adjust these camera parameters after the fact to get closer to the "optimum" usage.

The real point here is that these camera comparisons will never be useful to everyone, as there are so many shooting styles, so many versions of "optimum". There is no absolute best camera, its always relative to the photographer using it.
Regards, Mike K
 

Keyboard shortcuts

Back
Top