12 million is 12 million is still 12 million

So Gabe, does my logic from all the white papers seem sound and correct? It feels right that the S3 in IDR mode is guessing a set of S and R pixels.

on the mixing of s and r pixels it will keep one of the two, but not both. lets say for easy thinking, they both have 8 stops of dr. the S is set to 0-7 and the R is set to 3-9. combined we have 0-9 the 10 stops. but it will use 0-1 from the S and 8-9 from the r. the mixing is combining the 2-7 from the s and r pix.
--
Kodachrome, they give us those nice bright colours
They give us the greens of summers
Makes you think all the world’s a sunny day, oh yeah
I got a nikon camera, I love to take a photograph
 
You are quite right, you also have a reasonable explanation for the
bump in resolution, the effect of dynamic range on perception.
Thanks for the even tempered response!
I'm always tempered with tempered people. I'm profoundly
un-tempered with unreasonable and dishonest people.
That's how you responded to a new poster here tonite, and his very first post.

What's the word for 'arrogant' in Italian?

--

This info has been brought to you by a 100% genuine S3 user/owner. Beware of substitutes!

Ghosts ghosts go away, please don't come back another day...
 
There are complet and seperate images being created.
right
If you convert an S3 raw with s7raw, you can actually cretae two seperate images.
right again
One form the S,
obe form the R.
more right again
Th eR image looks like cra_p, all except the highlights.
it must, see comment below to Gabe.

my point is you have a large amount of same data, from the 2 different sets of pixels. most of that data will get tossed. While the pic actually contains the 12.34mp of data, its only at 6.17 resolution. Now weather or not Fuji is actually using the R data to aid in IDR mode to guess the data for iterpolation I do not know and Fuji, so far as I have read, is not saying. Going off of this I just figured out one of my own questions.

In IDR mode its guessing a set of S and a R data. so its only guessing the res on another set, not 2, thats where my thinking of the perceived res got out of whack.
--
Kodachrome, they give us those nice bright colours
They give us the greens of summers
Makes you think all the world’s a sunny day, oh yeah
I got a nikon camera, I love to take a photograph
 
Go here and click on the "what is super ccd." It's an interesting little flash presentation.

http://www.fujifilm.com.au/digital/
Kodachrome, they give us those nice bright colours
They give us the greens of summers
Makes you think all the world’s a sunny day, oh yeah
I got a nikon camera, I love to take a photograph
Hey, stay away from blue blotter and don't stare at the sun man.

--

This info has been brought to you by a 100% genuine S3 user/owner. Beware of substitutes!

Ghosts ghosts go away, please don't come back another day...
 
That is according to the manufacturer - Fuji.

:-)

--
Kevin P Kitching
 
I could have
a one pixel CCD or CMOS and if I could move it around fast enough I
could output 22MP. The reviewers would call it a one pixel camera
but the output would be 22MP.
I'd still prefer a CMOS myself. So much cleeeener
when shooting over 'an hour' per image. I can't move it
around fast enuff myself, personally.

You Fugi folks are as anal-retentive as us stupid Canon folks
can be. Same in all forums on DP it seems.

If your viewfinders are as crappy small as our Canon v-finders are,
we share the same beef. If you don't see the ISO readily available
on the LCD or in the viewfinder, we share the same beef.

WayneB.
 
right says what the white paper said. but had prety colors and pics. they are alos saying they have a better guess. kind of like vulcans guessing vs humans guessing.

so it still comes down to in the S2 in I12mp the camera guesses 50% of the pixels. all used for res.

in the S3 in IDW24mp the camera still guesses 50% of the pixels, but now 25% ofthe pixels are guessed for resolution. becayse the IDW has 4 sets of data. The real Sand R, the guessed S and R. With the real and guessed S data for res.
--
Kodachrome, they give us those nice bright colours
They give us the greens of summers
Makes you think all the world’s a sunny day, oh yeah
I got a nikon camera, I love to take a photograph
 
I have used the S2 for a long time, and the S3 since January.

For the sake of simplicity, I think it is clearer to call the S3 a "very good 6MP camera" in terms of resolution. What is the problem with it? 6MP is fine for most purposes.

Yesterday when I read this thread I took a Raw file made with the very sharp Tamron 28-75, and converted it to a) a 12MP Tiff, and b) to a 6MP one. There was a lot of trees, foliage and other detail in the pic.

I then converted the 6MP file in photoshop (bicubic) to 12MP, and the result was rather clear: No additional detail whatsoever was visible in the 12MP pic.

So, as has been said several times already, the S3 has not 12MP of detail, but so-to-say, "12MP of DR and subtle tones" - a questionable formula, but more adequate than to say "7,5-8,3 MP or effective resolution", something that noone can really measure(bate).

My personal impression is that the S2 is slightly crisper than the S3, the S3 generally produces a smoother file, with a little bit less fine details, and that a 20d has also a little bit more detail than the S3. 2MP is certainly not "a world of difference", as I could read sometimes in postings, and everyone who says "the 20d will beat the S3 hands down" does not know what he is talking about.

8MP versus 6 will mean that any detail in the photo like - lets say a face of a person - is rendered by 8000 pixels instead of 6000 pixels. No big deal, and not enough to show something that is not visible on the 6MP photo. It will just give better sharpness and detail definition by a certain extent.

I personally prefer the general quality/ look of the S3 files. But that is my personal opinion.

In the end I don't understand why this has to be re- discussed a zillion times.

The S3 has certain qualities that no other dslr has, but detail resolution is not its unique feature. I would personally say that a balanced photo with natural looking tones is at least as important as fine detail, but this is a personal evaluation.

Btw, in real world conditions factors like lens resolution, camera shake, Dof, light conditions and the nature of the object photographed (nature, portraits, etc,) will probably have a bigger influence than a few megapixels more.

One last thing concerning the price of the camera: I dont feel the S3 is overpriced: It is not cheap thats true, but it delivers a DR that lets me shoot without a grad filter in almost 100% of all cases, and even in contrasted sun/ shadow sitiuations without blowing the highlights all the time. This saves a lot of post processing and lets me concentrate on the more important editing aspectsthan repairing the damage.

No other DSLR can deliver this. With the buffer upgrade and maybe a AWB update in the firmware I guess that the S3 will be THE wedding RAW file camera.

regards, Bernie
 
Yesterday when I read this thread I took a Raw file made with the very sharp Tamron 28-75, and converted it to a) a 12MP Tiff, and b) to a 6MP one. There was a lot of trees, foliage and other detail in the pic.
I then converted the 6MP file in photoshop (bicubic) to 12MP, and the result was rather clear: No additional detail whatsoever was visible in the 12MP pic.
Phil did the very same thing in his S3 review, an overlay. His results showed the native file to be slighlty better.

If I were testing, I would print the print the two files, as that is what really counts.

I did that with a 20D, and their about equal. Thom Hogan and Reichmann and Anders, and Dave Eckells have said the same.

And now you add another data point. :)

--

This info has been brought to you by a 100% genuine S3 user/owner. Accept no substitutes!

Ghosts ghosts go away, please don't come back another day...
 
Since 2 1/2 years that i have my S2 Pro, I never made any tests on that -and it never really came to my mind-, but, quoting Thom Hogan's complete guide to the S2 Pro:

"3024. Sets the image to the default 3024 x 2016 pixels. This is the maximum size produced with a one-to-one photosite to pixel correspondence."

Would a 3024x2016 original 6 Mp image print better/sharper on, say 10"x15", than an "interpolated" 12 Mp one, as certainly the interpolation process isn't lossless ?
Anybody ever tested this ?

Cheers,
Jean-Pierre
http://www.pbase.com/scherrer/s2
Equipment list in profile
 
Since 2 1/2 years that i have my S2 Pro, I never made any tests on
that -and it never really came to my mind-, but, quoting Thom
Hogan's complete guide to the S2 Pro:
"3024. Sets the image to the default 3024 x 2016 pixels. This is
the maximum size produced with a one-to-one photosite to pixel
correspondence."
Would a 3024x2016 original 6 Mp image print better/sharper on, say
10"x15", than an "interpolated" 12 Mp one, as certainly the
interpolation process isn't lossless ?
The interpolation happens first. This is happens during the demosaic process, and it is a necessary first step. Many do not realize this. Then it is downsized to 6MP. The 12MP file is the original. It will yield a slightly better print than the 6MP file, because some information is lost in the downsize from 12 to 6.
Anybody ever tested this ?

Cheers,
Jean-Pierre
http://www.pbase.com/scherrer/s2
Equipment list in profile
--

This info has been brought to you by a 100% genuine S3 user/owner. Accept no substitutes!

Ghosts ghosts go away, please don't come back another day...
 
I dont care how many pixels a camera offers or if they are real pixels or unreal or whatever.

I just care how much detail a final images shows, and IMO a S3pro image shows less detail than a d2x or 1dsII but more detail than most 6 MP cameras.
The thing that matters for me is what I see in the final result.
 
Oops. A very sensible post. It'll probably get deleted. :)
I dont care how many pixels a camera offers or if they are real
pixels or unreal or whatever.
I just care how much detail a final images shows, and IMO a S3pro
image shows less detail than a d2x or 1dsII but more detail than
most 6 MP cameras.
The thing that matters for me is what I see in the final result.
--

This info has been brought to you by a 100% genuine S3 user/owner. Accept no substitutes!

Ghosts ghosts go away, please don't come back another day...
 
I don't think so ! The S2 -at least- is a 6 Mp camera, though delivering an original image of 3024x2016 pixels, wich are internaly up-rezed to 12 Mp, but the s" -and most certainly S3- sensor only delivers a true 6 MP !!!
...but this is only my opinion, and i share it !

...and i still LOVE my S2 and the quality of image it delivers, be it 12, 6 or whatever number of pixels !?!

Cheers,
Jean-Pierre
http://www.pbase.com/scherrer/s2
Equipment list in profile
 
I don't think so ! The S2 -at least- is a 6 Mp camera, though
delivering an original image of 3024x2016 pixels, wich are
internaly up-rezed to 12 Mp,
Yes, but it needs to be uprezzed to change the output pixel orientation from diagonal to rectilinear horizontal/vertical. For example, if you go into the camera menu and select 6 MP output, the camera first interpolates up to 12MP, (as it must) and then down rezes to 6MP.

--

This info has been brought to you by a 100% genuine S3 user/owner. Accept no substitutes!

Ghosts ghosts go away, please don't come back another day...
 
Fuji refers to the S3 as having 12MP.
Read the small print (or even better, talk to their reps). It's a 6MP unit which interpolates to 12 effective MP - just as the S2 did.

The S3 is still 6MP - the two sets of photosites sample the same (6MP) area. The two are then combined (overlaid) which does not suddenly produce 12MP.

Try drawing a 6x6 grid on acetate, then draw another identical grid on another piece of acetate. Now place one directly over the other - how many 6x6 grids can you see?

--
Kevin P Kitching
 
I know exactly what it is. I am just telling you what the marketing material says.
Fuji refers to the S3 as having 12MP.
Read the small print (or even better, talk to their reps). It's a
6MP unit which interpolates to 12 effective MP - just as the S2 did.

The S3 is still 6MP - the two sets of photosites sample the same
(6MP) area. The two are then combined (overlaid) which does not
suddenly produce 12MP.

Try drawing a 6x6 grid on acetate, then draw another identical grid
on another piece of acetate. Now place one directly over the other
  • how many 6x6 grids can you see?
--
Kevin P Kitching
--

This info has been brought to you by a 100% genuine S3 user/owner. Accept no substitutes!

Ghosts ghosts go away, please don't come back another day...
 
“The Fujifilm FinePix S3 Pro posesses a very progressive sensor
design. Instead of any external interpolation method known to my
test laboratory, here the interpolated mode provides a real
increase of detail information. Using high-end, well-suited lenses,
the camera can compete with eight million pixel models of other
manufacturers.”
States "... can compete with eight million pixel models of other manufacturers.”

That in itself does not confirm that it is 8MP - it's still 6MP which produces a result which is as pleasing as 8.

Why not ask Fuji why they still maintain their S2/S3 is 6MP if everyone else thinks it is more?

Saves 'em developing a true 12MP sensor I suppose! :-)

--
Kevin P Kitching
 
Could be but the extra resolution would only be in the range of brightness where the two "images" overlap. It would be interesting to know how much overlap there is between the bright and dimmer images. The bright image is from smaller sensors so it likely has significantly less dynamic range from the dimmer image so this would perhaps limit the overlap in brightness space.
Some disagree with this. Some believe that the R pixels are also
being used for additional resolution data, since they are in a
slightly dfferent position and each has its own microlens.
--
Leon
http://homepage.mac.com/leonwittwer/landscapes.htm
 

Keyboard shortcuts

Back
Top