+400% dynamic range?

[adas]

I'm very tempted to buy the new Fujifilm S3PRO camera because of its 12Mp and increased dynamic range(the new Super CCD SR II ).Fujifilm claim a 400% wider dynamic range for this camera.Have anyone idea what that means?
To my calculations,if I assume for an ordinary camera a 5 stop dynamic range,would be 5*4=20 stops.Sound ridiculous.
Most of reviewers say it's only a moderate dynamic range boost, almost imperceptible.

[Tom W]

I'd say that its more like two stops. 200% is a doubling or 1 stop. Double it again and you have 2 stops. They're playing a little fast with the math.

I wonder what they're comparing it to.

I may be way off, but I believe that JPEG images are limited to something on the order of 5 stops anyway. You just get a bigger RAW image DR from which to create that 5-stop JPG. It's a positive step.

[slejhamer]

Reading Reichmann's review, it seems that they're playing a little fast with the megapixel numbers too.

The S3 is a six megapixel camera, but because of the shape of its pixels Fuji extrapolates the data into larger files.

From: http://www.luminous-landscape.com/reviews/cameras/s3-pro.shtml

[CyberDyneSystems]

Reichman's wrong on this one.
The S3 actually has 12 million photosensors.

6 Million of them are "standard" photosites,. and the other 6 million are the special "r" sensors that require more light to register.. thus offering the extended dynamic range as these 6 million smaller sensors will not "blow out" when the others do..

MR's own review illustrates the two differnet photosite sizes. I fail to see how this then can be interpreted as less than 12Million photosites?

He seems to be cofusing the issue with teh Fuji "SuperCCD" of older models.. where indeed a 6MP sensor in a diamond pattern was used to create an interpolated 12MP file size.

In this case there really are 12 million photosites. There just ahppens to be two differnt types.

Still I love the idea of Fujis advancements here.. It would be nice to have this sort of technology incorporated into a Canon camera.

[PacAce]

Reichman's wrong on this one.
The S3 actually has 12 million photosensors.

6 Million of them are "standard" photosites,. and the other 6 million are the special "r" sensors that require more light to register.. thus offering the extended dynamic range as these 6 million smaller sensors will not "blow out" when the others do..

MR's own review illustrates the two differnet photosite sizes. I fail to see how this then can be interpreted as less than 12Million photosites?

He seems to be cofusing the issue with teh Fuji "SuperCCD" of older models.. where indeed a 6MP sensor in a diamond pattern was used to create an interpolated 12MP file size.

In this case there really are 12 million photosites. There just ahppens to be two differnt types.

Still I love the idea of Fujis advancements here.. It would be nice to have this sort of technology incorporated into a Canon camera.

From what I understand, each pixel from the two sets of 6mp sensor data are merged to create one pixel of the image, albeit a dynamically extended one, but still one pixel so, in all practicality, the camera only has 6 mp of data in the final image, not 12 mp.

[CyberDyneSystems]

So Fuji is merging 12million pixels into 6 miilion pixels of data and then reinflating it back up to a 12MP file?

Weird...

I have to agree though that it is a shame that Fuji and Minolta and Pentax cant seem to price there offerings as competitively as they need to. This Camera is too much at $2,500.00

[PacAce]

So Fuji is merging 12million pixels into 6 miilion pixels of data and then reinflating it back up to a 12MP file?

Weird...

I have to agree though that it is a shame that Fuji and Minolta and Pentax cant seem to price there offerings as competitively as they need to. This Camera is too much at $2,500.00

The 12mp might just be the raw file. But the final image (in JPG) would only be 6 mp.

[CyberDyneSystems]

The uncompressed RAW file and largest jpeg file are both 4256 x 2848 (12.1 million)

So the files are 12 MP. (unless you select lower res jpegs.. as can be done in most cameras)

Your right Leo,.. I see it now.. I looked at DPReview .. he explains it more clearly,..
The S3 Pro utilizes Fujifilm's "extended dynamic range" SuperCCD SR sensor which features two photodiodes at each photosite (a single 'input pixel'). The 'S' pixel has normal sensitivity and captures the same range of light as a conventional CCD photosite, the 'R' pixel is smaller and has a lower sensitivity and is designed to capture detail above the saturation point of the 'S' pixel, the camera can then combine the information from the 'S' and 'R' pixels to produce an extended dynamic range and avoid the loss of detail due to over-exposure

So indeed.. the two S and R photo"Diodes" are part of a singke "photosite" the imgae files DO get blaown up to 12MP in the same way as previous "SuperCCD" cameras from Fuji have done.

[PacAce]

The uncompressed RAW file and largest jpeg file are both 4256 x 2848 (12.1 million)

So the files are 12 MP. (unless you select lower res jpegs.. as can be done in most cameras)

OK, I reread the S3 sensor specs and you are correct. Each S photosite and R photosite generates a pixel of image data via extrapolation. I initially thought that the the S and R data were merged together to get the extended data for each pixel. I stand corrected. :o

[CyberDyneSystems]

lol,.. Now we have done a complete reversal! :lol: :lol: :lol:

[Scottes]

I stand corrected.

Wow. A very rare event.


Now, if we ever caught Bob Gross saying this...

Nah. Nobody will ever catch him being wrong.

[PacAce]

I stand corrected. :o

Wow. A very rare event.
Now, if we ever caught Bob Gross saying this...

Nah. Nobody will ever catch him being wrong.

Does it matter that I lied when I typed that? I was actually sitting when I said that so it doesn't really count. :p

[robertwgross]

Wow. A very rare event.


Now, if we ever caught Bob Gross saying this...

Nah. Nobody will ever catch him being wrong.

You know, I thought I was wrong one time, but I was incorrect on that.

---Bob Gross---

[CyberDyneSystems]

:lol: :lol:

[Scottes]

You know, I thought I was wrong one time, but I was incorrect on that.
I know.
I remember that well.
I thought you were wrong, too, but I ended up being wrong twice!

Darn you Bob Gross you'll be the death of me.

[Bodryn]

For the benefit of folks new to photography I suppose that what they are trying to do is to make digital cameras as capable as film of capturing a wide dynamic range (brightness range), more like the human eye which has rods for dim light and cones for bright light and thus able to compensate much better for over or under-exposure. Should be a very nice idea if it works.

[scottbergerphoto]

....... more like the human eye which has rods for dim light and cones for bright light .
Is that why the not so bright are often called nim-rods? :p
Yeah, I know, I crack myself up. :rolleyes:
Scott

[Tom W]

Is that why the not so bright are often called nim-rods? :p
Yeah, I know, I crack myself up. :rolleyes:
Scott

Well, what about us Nim-cones? We're under-represented. :)

[PacAce]

Is that why the not so bright are often called nim-rods? :p
Yeah, I know, I crack myself up. :rolleyes:
Scott

Then when someone calls me a cone head, I should take that as a compliment? Meaning I'm bright? Oops. Just remembered. It wasn't conehead he called me. It was bonehead. :confused: :o

[Tom W]

You mean you're from this family:

http://www.tvacres.com/images/coneheads_small.jpg

:)

[CyberDyneSystems]

I think rods and cones are for detail (essentially black and white) and color info.. I could be forgetting my physiology lessons from 2 decades ago.

[Jesper]

Rods and cones.... S and R pixels.... hey, my eyes have two kinds of pixels too! :)

CDS, I also thought that the rods are for light intensity and the cones are for colour vision, but it's not that simple. See here: http://en.wikipedia.org/wiki/Retina
The cones respond to bright light and mediate high-resolution vison and colour vision. The rods respond to dim light and mediate lower-resolution, black-and-white, night vision.
The eye has about 120 million rod cells and 6 million cone cells. The rods are mainly for low-light vision and the cones for bright light and colour vision. There are also three different kinds of cones for different parts of the spectrum.

Note, because the Fuji S3 has pixels that somewhat resemble the way the human eye works, it doesn't automatically mean that it's better than cameras that work differently.

[Moppie]

it doesn't automatically mean that it's better than cameras that work differently.
As someone else said, its 200% better than what?

[adas]

Than the others,Moppie.
I see you people madly against the interpolation.Probably some of you don't know interpolation came along within cameras since the early days of digital because is cheaper.Not to mention that your favorite 6Mp camera has only 1.5Mp of true resolution in the worse case of monocromatic subjects like red flowers or green grass. Than is heavily sharpened by the camera or your software to get an acceptable sharp image.
Again, lot of you have read the DPR of the older Super CCD SR and talk about the new one(SR II).
One key feature of the new sensor is that the S and R pfotosensors no longer share the same cell,they have distinct cell each, arranged in the diagonal matrix.So they will capture distinct detail,therfore doubling the resolution.
Now, is this better or what?

[PacAce]

You mean you're from this family:

http://www.tvacres.com/images/coneheads_small.jpg

:)

Hey, where'd you get that picture of my Aunt Prymatt, Uncle Beldar and Cousin Connie? :confused:

:mrgreen:

[PacAce]

Than the others,Moppie.
I see you people madly against the interpolation.Probably some of you don't know interpolation came along within cameras since the early days of digital because is cheaper.Not to mention that your favorite 6Mp camera has only 1.5Mp of true resolution in the worse case of monocromatic subjects like red flowers or green grass. Than is heavily sharpened by the camera or your software to get an acceptable sharp image.
Again, lot of you have read the DPR of the older Super CCD SR and talk about the new one(SR II).
One key feature of the new sensor is that the S and R pfotosensors no longer share the same cell,they have distinct cell each, arranged in the diagonal matrix.So they will capture distinct detail,therfore doubling the resolution.
Now, is this better or what?

One might argue that one camera has a better sensor than another (take the Foveon X3, for example) but what it really boils down to is the total package and the sensor is just one part of it. Just my 2 cents.

[defordphoto]

What it really boils down to is what kind of photo can this thing produce? If it's a great photo then they have succeeded. If not, then they have failed.

Plain and simple.

But, it sounds like hype to me, knowing Fuji's camera history. I am skeptical, but I'll wait to see a photo before I toss this camera off the cliff.

[Bodryn]

There is an absolutely fascinating factual book by Dr. Oliver Sacks: The Island of the Colorblind. This is a factual book about a real island where people have a genetic defect where their eyes have rods but not cones, and thus are completely blind to color. This is not so simple a problem - as mentioned above, there are far fewer receptors in these people's eyes, and probably most of these people would be legally blind. They function best in evening twilight but poorly in broad daylight. This story was also featured on a TV documentary a few years ago. If you haven't read anything by Oliver Sacks, you must do so because he has an absolutely delightful writing style.

I've always been fascinated with colorblindness and color vision in general. It may interest people to know that it's been learned that women generally have slightly more sensitive color vision then men. I learned this in no uncertain terms at the dentist a few years ago when he called upon his female assistant to match colors for my small fixed upper bridge. I thought my color vision was perfect but with the subtle color differences of teeth, neither I nor my dentist nor even the dental lab could quite match the ability of his female assistant to determine precise color matches. This phenomenon is probably very important to all you folks trying to correct your photos for color.