Go Back   Steve's Digicams Forums > Digital Cameras (Point and Shoot) > Fujifilm

Reply
 
Thread Tools Search this Thread
Old Dec 29, 2003, 6:41 PM   #21
Senior Member
 
Join Date: Mar 2003
Posts: 205
Default

Quote:
Sorry, that, was installed on S5000 - scam

SCAM:: a fraudulent or deceptive act or operation.

The SCCD format, itself, has been shown to produce more real resolution for a given number of photosites vs. a conventional sensor. But this is beside the point, and I already stated taht I find that the s5000 DOES NOT produce additional pracical resolution, though otehr FUji cameras do. But, I have only seen the s5000 advertised as '3MP effective', with 6MP referred to as 'file' size or 'interpolated'. Do you assert that you have seen fuji claim the s5000 is a 6MP EFFECTIVE camera? Please direct me to this....

Quote:
Super CCD IV HR have promised higher resolution
Let's address the Fuji s5000 by itself, disregarding other fuji SCCD cameras: it does resolve 'higher' resoluton on vertical/horiztonal ISO standard measurement charts then most other standard 3MP cameras. However, this is on the order of 3-5% total resolution, which in reality will have very little if any perceptible difference on a scene. Technically, it is 'higher'.... did the brochure state precisely how much 'higher'?

The real 'scam' is the fuji F700, actually. Inversely, the F700 resolves nearly as much data as some 5MP cameras wth it's 3MP spatial coordinates(the R and S sensors are paired into singular spatial coordinates, so in effect a 3MP sensor spatially), once again adding more evidence that the SCCD sesnro alignment can work. However, the F700 DOES NOT use the pairs correctly. Proven, beyond reasoanble doubt by programmers on dpreview's fuji forum, the F700 does not use(at least to any practical effect) the low sensitibity sensors on JPEG images out of the camera! The only way to make use of the sensor pairs is to use a custom written RAW conversion program, of which then the 4x dynamic range theat were advertised can be extracted.

Quote:
(Have you thought, why the minimum ISO is 200?). As a result we have a lot of digital noise.
The s5000 produces lower noise, proportionately, at higher ISO ratings than other cameras with the same size sensor chip. I will not disagree with the lens being not so great..... it's not. It is probably the reason for the inability for the SCCD to provide any more practical resolution on this camera.

-Chris

[/quote]
WmAx is offline   Reply With Quote
Old Dec 29, 2003, 9:33 PM   #22
Member
 
Join Date: Sep 2003
Posts: 63
Default

.
Quote:
Let's address the Fuji s5000 by itself, disregarding other fuji SCCD cameras: it does resolve 'higher' resoluton on vertical/horiztonal ISO standard measurement charts then most other standard 3MP cameras.
Let's go to basic - resolution of 3MP matrix is - (0.75R+0.75B+1.5G)/3 = 1MP. Yes, if we wold given ideal resizing algorithm we could resize image to 1 MP without loss of quality. 3 or 6 MP - a pure marketing tricks, people are buying "Megapixels" (like PC users buying MegaHerz). Given simplest algorithm I know - every 8 "Fujixels"
R x B x convert to 2 real pixels RG[iB] [iR]GB.
x G x G
From 6 MP we will left 1.5 MP keeping same resolution(detail level).
Let's check our calculations - every third pixel is "false" - 1.5 MP - 0.5MP(33%) = 1MP of real resolution. I can't see, how Fuji's interpolations trick can help with resolution. Do you really think, that they are using gradient interpolation or bicubic resizing in their camera? Tests of 3 and 6 MP (or 12 and 6) resolution shows that are significant difference in them, which shouldn't be happen.
Really, resolution now depends more from lens, that from matrix.
Comparing with other cameras just (probably) shows, that their matrix are superating their lens with resolution (that, I believe a scourge of this year matrix). We could talk more having list of cameras to compare
Klyak is offline   Reply With Quote
Old Dec 29, 2003, 9:45 PM   #23
Member
 
Join Date: Sep 2003
Posts: 63
Default

Quote:
The SCCD format, itself, has been shown to produce more real resolution for a given number of photosites vs. a conventional sensor. But this is beside the point, and I already stated taht I find that the s5000 DOES NOT produce additional pracical resolution, though otehr FUji cameras do. But, I have only seen the s5000 advertised as '3MP effective', with 6MP referred to as 'file' size or 'interpolated'. Do you assert that you have seen fuji claim the s5000 is a 6MP EFFECTIVE camera? Please direct me to this....
That was not so rough trick - it was announsed "SuperCCD IV" matrix. On the site we could see advantages of "SuperCCD IV SR" in BIG letters and drawings. Just digging .RAF from my S5000 I haven't found any presence of S sensors. Than I found in broshure mentioning of difference in HR and SR, and mentioning, that HR - means 6 MP instead of 3 of "SuperCCD III".
Just then I concluded - S5000's matrix not SR and, really not HR - keeping just disadvantages of beeing smaller size. In combination with cheap lens and software - giving a huge disappointment. My next camera willn't be Fuji for sure.
I came from the world of "normal" fotografy, and still can't understand, how they can sell you this quality for that money.
Klyak is offline   Reply With Quote
Old Dec 29, 2003, 10:10 PM   #24
Member
 
Join Date: Sep 2003
Posts: 63
Default

Sorry, I don't like big posts.
Quote:
Let's address the Fuji s5000 by itself, disregarding other fuji SCCD cameras: it does resolve 'higher' resoluton on vertical/horiztonal ISO standard measurement charts then most other standard 3MP cameras
Fuji's only argument for getting better resolution - rotating 45 and octagram form are putting pixels close, so we are tetting horizontal and vertical resolutions better (diagonal going off for a long walk). Greate, same effect we can achive just making pixels smaller in size. But then it came obvious disadvantages of it - need of more contrast lens and light (45 rotating just rotate your head away from this fact). Unusual form should recieve more light - other cameras are putting microlens to achive the same. There's no revolution at all.
45 rotation - a marketing trick from good old times, when they weren't obliged to put "effective" pixels count.
I even remember time, they've announced SR matrix like 6MP effective - 3S+3R, but in a month or two cleared this bullshit from their site.
Klyak is offline   Reply With Quote
Old Dec 29, 2003, 10:46 PM   #25
Senior Member
 
Join Date: Mar 2003
Posts: 205
Default

Quote:
Let's go to basic - resolution of 3MP matrix is - (0.75R+0.75B+1.5G)/3 = 1MP. Yes, if we wold given ideal resizing algorithm we could resize image to 1 MP without loss of quality
Actually, they are able to gather some useful spatical data from each point. Most importantly is to capitalize on the human response weighting; that is, humans responde to green luminance with greater sensitivity . So, it's a trade off.. they are able to achive more usable resolution for perception by focusing on luminance rather then on equal color/luminance you would have with an equally distributed sensor array. This is achieved at the expense of some low level problems(espcially color fringing due to color interpolation errors). The rasied efficiency is plainly obvious in measurment charts. A digital imaging sensor, at best including Krell effect, can resolve 0.6 or 0.7 factor approxijmation of it's total sampling frequency in a given plane. The measurement chart numbers agree with the number of photosites, that correlate that the method is useful for raising efficiency of the number of sensors you have to work with. They could, of course, put three indivicual sensors each under diffuse lens, and have a strict 1:1 ratio RGB sensor array. However, this would (1) possibly reduce efficinecy of the sensor due to the light loss of the diffuser (2) reduce overall efficiency of resolution with a given number of sensors since y ou are now distributing sensors evenly to all paramaters, ignoring human visual perceptive properties. Given the high cost of high density sensors as the digital imaging technology developed, this was the most efficient way to raise resolution while keeping costs low. Of course, as companines move into a directon such as Foveon has taken, this will not longer be aconcern since high densitiy sensors aer lowering in price rapidly. Their will soon be no compelling reason once cost is not a factor. But, it's still hard to break an old habit, though.

Quote:
There's no revolution at all.
45 rotation - a marketing trick from good old times, when they weren't obliged to put "effective" pixels count.
The evidence is contrary. The s2, in controlled tests aginst NIkon DSLR cameras of the same sensor frequency, using the same lenses, has been demonstrated to outresolve the standard orientation 6mp cameras. The same is true for Fuji consumer cameras. Just recently, PHil askey's controlled test on dpreview clearly showed the 3MP spatial resoluton points of the F700 to clearly outresole a 3mp Canon standard sensor camera. The F700 actually came quite close to rsolving the same detail as a standard 5MP Canon camera.

Il'll even share two precisely framed test samples, FOV within 2-3 percent.

Nikon D100 vs. s2 -- I upsized th D100 6mp file to teh same physical size as the s2 12mp output to match it's size:

http://www.linaeum.com/images/s2_d100.jpg

Fuji F700 vs. Canon A70 vs. Canon s50 -- I upsized all images to the 6mp size of the F700 for equal size/easy comparison.... i heavily sharpned all of the images to bring out all low contrast resolved data. The fuji will have more noise, since it was at ISO200 and the canon's at ISO50 for the test.... also, the jpeg compresison is higher in the inital JPEG output files form teh fuji:

http://www.linaeum.com/images/a70_s50_f700.jpg

This is not the only evidence to show the SCCD cameras are not comletey hype, but just two releveant examples. You can find the soruce images for both examples at dpreview's archives. Whatever the mechanism(due to the Fuji SCCD, just better bayer processing, whatever), Fuji cameras are outresolving standard bayer square layout sensors with the same number of spatial points. The differences, so far, have not been exactly small, at leat in realtive view of the claims made. The examples against standard alyout cameras of the same spatial resolution(MP) are quite signficant so far. SOme cameas, like the S5000 just plain don't deliver teh results..... but their are various reasons this is possible.

You can, of course, say anything you want. But I urge to research 'why' specific methods/properties are used before you attempt to simply state that they are invalid. Bayer mosiac systems have been demonstrated, and accepted, by the engineering community as a valid approach to maximize percieved spatial resolution efficiency with a given number of sensors avaialble.

-Chris
WmAx is offline   Reply With Quote
Old Dec 30, 2003, 4:10 AM   #26
Junior Member
 
Join Date: Dec 2003
Posts: 5
Default What a delight

I have just bought the Fuji S5000. It is just out ofthe box and I find it FANTASTIC Labored a few hours over the manual with the camera and am just delighted with the quality. A real pleasure is the zoom and the palcement of the buttons for ease of access. Thanks to all who recommended it.
tamirante is offline   Reply With Quote
Old Jan 7, 2004, 8:23 PM   #27
Junior Member
 
Join Date: Oct 2002
Posts: 6
Default got it?

I also got my s5000 and very happy with it. I chose it over the A80 and glad I did. Wanted the a80 only for portability, but I still have my old Sony DSC-p30 for those cases. 8)
vprtal is offline   Reply With Quote
 
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



All times are GMT -5. The time now is 6:02 PM.