Steve's Digicams Forums

Steve's Digicams Forums (https://forums.steves-digicams.com/)
-   Sigma dSLR (https://forums.steves-digicams.com/sigma-dslr-72/)
-   -   SD10 true resolution (https://forums.steves-digicams.com/sigma-dslr/32306-sd10-true-resolution.html)

code Aug 24, 2004 5:10 PM

SD10 true resolution
 
I just downloaded a sample pic (the 'institution') from the SD10 review, interpolated to 4500X3000 and compared it to the Kodak mentioned. As far as I can see, the SD10image contains more details, and with better colors too.?? If I didn't mess up the Foveon sensor seems to be one to watch :love:.

Lin Evans Aug 25, 2004 12:58 PM

Though I admire the enthusiasm of SD9/SD10 fanciers, the fact remains that these cameras have close to six megapixel bayer processed sensor resolution, but in no way do they remotely even approach 11 megapixel EOS-1DS Canon resolution which is considerably below the resolution of the 14 megapixel Kodak Pro line 14n/SLR/n range. They don't reach the eight megapixel resolution of the EOS-20D or EOS 1D Mark II either.

The SD9/SD10 are wonderful cameras which offer amazing resolution and some of the sharpest pixel level detail of any digital camera, but resolution is measured by an agreed upon standards calculated by lines pairs per mm and by lines per image height.Numerous professional reviewers have measured the resolution of the SD9/SD10 and found it approximately identical with the six megapixel Canon, Nikon and Kodak dSLR cameras and far below the highest resolution dSLR's.

As good as it is folks, let's not get carried away with exaggerations...

Best regards,

Lin

code Aug 26, 2004 5:43 AM

Hello Lin!

If a simple side by side comparison like this shows more real life detail in the Sigma, I guess
I don't care much about the experts conclusions at all.

They should consider how differences in sensor technology impacts their measuring results and
relevance to the task at hand, to capture real life detail:

Traditional sensors are divided into blocks of four pixels in RGGB (bayer) patterns, which means
a 14Mpix sensor contains 3.5M blocks of four pixels. Each pixel in these blocks are selectively
colorblind and sees either Red,Green,Green,Blue and nothing else, which requires both algorithms
and guessing to try to resolve finer detail, plus blurring to avoid aliasing effects.

In a Foveon sensor each Pixel outputs a complete RGB color signal of it's respective image detail.

So, you might be able to see more gray scale lines in these comparisons, but not necesarrily more
real life detail, which is what I care about in an image. :D


Lin Evans Aug 26, 2004 9:57 AM

Quote:

They should consider how differences in sensor technology impacts their measuring results and relevance to the task at hand, to capture real life detail:
Reviewers are very well aware of how differences in sensor technology impact measuring results and the facts are that the Foveon processor captures about the same amount of "real life detail" as a six megapixel bayer sensor not equal to or anywhere near equal to a 14 megapixel bayer sensor regardless of how much you "wish" that were the case. Do you really think that professionals who see all types of images from every digital camera manufactured can't render an objective, accurate, description and analysis while you build a "sand castle" of wishes based on your comparison of two images on screen and your subjective "impression" of what constitutes detail?

Perhaps in the history of digital acquisition there has not been so much misinformation spread as from those who argue that the Sigma SD9/SD10 produce more "real world" resolution than is possible. You mistake false detail beyond Nyquistforreal detail and confuse resolution and sharpness.

Continuing to argue that a 3.5 megapixel arraybuilt on slightly more than 10 million photosites produces "more real world resolution" thana 14 megapixel array only demonstrates a lack of understanding of the process. Arguing that "color resolution" is somehow "real" and that "black and white" resolution is somehow "different" simply shows a lack of understanding of the term "resolution." If the Foveon based Sigma can't reveal the same level of black and white detail as a bayer eight megapixel camea (see resolution charts of Canon EOS-1D Mark II, Canon EOS-20D versus Sigma SD9/SD10) then how can it render equal or greater "real world" resolution than a 14 megapixel Kodak? Do you think that somehow information which can't be resolved in black and white can with some "miracle" be revealed with color? How do you "think" black and white detail is captured with the Foveon sensor?

Resolution has a scientific meaning and a scientific way of being measured. In digital cameras it's been thoroughly measured and no amount of obfuscation with terms like "real world" is going to change that. You've apparently been listening to too many fanatics who bring way too much confusion to this arena. The Sigma SD9/SD10 has approxomately the "real world" resolution of a six megapixel bayer sensor, period. Not more, not less.

You are not going to make a case for the Foveon sensor having the equivalence of a 14 megapixel bayer filtered sensor to anyone with serious experience with both. If you want objective comparisons I'll be glad to give you samples of both. I use both the Foveon and Bayer cameras and have enlarged and printed samples from each. In fact I do this on a daily basis and have hundreds of comparitive images.

Do you own a Sigma SD9 or SD10? Do you also own a six, eight, eleven, fourteen or sixteen megapixel Bayer camera? I do and I can assure you that the Foveon based Sigma SD9 and SD10 perform about equivaent to my six megapixel Kodak, Nikon and Canon six megapixel cameras and not even remotely close to my eleven megapixel Canon or sixteen megapixel Kodak.

Please do the folks here who use and enjoy their Sigma dSLR's a big favor and don'tbegin your first posts on this forum with exaggerated claims which simply make us look like we don't have a clue to the rest of the world. The SD9/SD10 are great tools and stand on their own in the world of digital acquisition without unsubstantiated fanatical claims. If you have a Sigma, then post your images and let them speak for you, please!

Lin


code Aug 26, 2004 11:17 AM

Quote:

Please do the folks here who use and enjoy their Sigma dSLR's a big favor and don't begin your first posts on this forum with exaggerated claims which simply make us look like we don't have a clue to the rest of the world.



Maybe you should loosen up and read my initial post again??. I don't make 'us' look like anything.
It's a tip,and I expect people on the forum and elsewhere to be perfectly able to check it out and make up their own minds.

And..judging from the tone of your reply, I thinkI will have to say goodbye to you for now.:lol:







Lin Evans Aug 26, 2004 12:49 PM

I did read your initial message and replied to it and the "tone" is simply an admonition to not attempt on your very frst post here on this forum to introduce the same silly argumentsformerly made on other Sigma forums. You are absolutely correct, the people here can make their own judgments based on their experience. On the other hand, you still have not answered my questions about whether you actually have the SD9 or SD10 or are simply trolling for controversy?

Lin

code Aug 26, 2004 2:33 PM

Ok. I'll makeONE exception to the last sentence in my prior post. I downloaded the samples from this
sites review pages, so that they where accessiblefor everyone here to check out and test my comparisons
and findings for themselves.

No one here seems to have discovered any trolling for controversy, except you. And Whether I personally
own a SD10 or not is of no relevance to my post.

Lin Evans Aug 26, 2004 4:35 PM

Quote:

No one here seems to have discovered any trolling for controversy, except you. And Whether I personally own a SD10 or not is of no relevance to my post.
Since you and I are the only ones posting on this thread, 50 percent of those those "here" seem to have questioned whether or not you are "trolling for controversy."

The relevance is experience, or lack of experience with the cameras under discussion. When you make your very first post on a forum with what appear to be outlandish claims you will most likely be suspected of "trolling" and you have indeed made the assertion that you feel that the Foveon based processor produces "more details" and "better color" than the 14 megapixel Kodak. You couched this in the form of a question and I've responded with the evidence by reviewers but you persist in the unsubstantiated claim. On the issue of "better color" there is no substantive measue since the term "better" is the essence of subjective. So you are certainly welcome to express that opinion and though I may or may not agree, my opinion is no better or worse than your opinion.

Myresponse has beensimply to inform you that your observations concerning resoluton comparisonsare wrong by any qualified reviewer's experiences and by scientific measurements made by numerous, indeed ALL, reviewers who did resolution testing. Observations based on, as you admit, a single photo are at best anecdotal and have little or no relevance when there is overwhelming evidence to the contrary. Had you extensive personal experience with the two cameras under question and could produce scientific evidence substantiating you assertion then you, I, multiple reviewersand the forum would all benefit. If you don't and can't then please change your subject to something which can be agreed on or at least discussed from an objective platform.

Lin


code Aug 27, 2004 12:35 PM

As a side note, maybe this could help inspire Steve to stop posting photos for everyone to make up their
own minds ??

What I really need is a picture of a digicam, with an industry standard lined testcard up front and a figure
telling me how many lines/mm of resolution it manages to squeese out. :-)<--GRIN


code Aug 28, 2004 2:35 PM

I must admit the comparisons gave me something to think about, and since human nature is
to look for an explanation, here's my go at one:

Imagine the cameras looking at four squares with a color pattern and an alignment of:
GR
BG

Projected by the lens directly over a corresponding unit block of four pixels in a 14Mp
sensor. Each pixel in this sensor sees only one color, determined by this filter array:
RG
GB

Now, the camera red sensor looks at a green square, the two greens look at one red, and
a blue square, and the blue sensor looks at a green square, and we have a problem..

There's actually no way to resolve this accurately. The best you can do is use some
reasonably clever algorithm, but not much more, and this explains most of the the
differences seen in most digicams output.

In this case, the Foveon sensor has a one quarter of thepixels of the 14Mp sensor,
so the four squares align precisely over only one of it's pixels.

This pixel, however, is sensitive to all colors, and integrate the color output from the
four squares perfectly. Like if you printed the squares out, and backed off until only
the combined colors remained as one square.

The next I do is to interpolate the Foveon sample pixels from on to four pixels. The input
to the algorithm is perfectly coloured pixels, and the algorithm recreates each of the four
pixels based on it's neighbours colors,intensity, and pattern.

We are back to algorithms and guessing again, but the input this time is consistent, and
the information contained in the algorithm adds to that in the image itself.

Et Voila! a comparison manufacterer's of traditional sensors don't like at all!! :lol:


Note:

This explanation doesn't really try to explain why the Foveon has a higher resolution than
it's pixel number, but rather why bayer type sensors has a lower resolution than
most people think they have. How much lower the resolution is, is largely determined by the
quality of their algorithms ablity to resolve the above mentioned Problem.









All times are GMT -5. The time now is 8:58 PM.

Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2017, vBulletin Solutions, Inc.
SEO by vBSEO 3.6.0 RC 2