Q Looses the 3-D of the M?

Started Jul 23, 2014 | Discussions thread
Lin Evans
Lin Evans Forum Pro • Posts: 17,356
Re: what's 3-D?

Hi Richard,

As a scientist, I have difficulty at times with the process because time after time what we formerly held to be immutable fact turns out to be wrong. It's like trying to make a unified theory encompassing Newtonian Physics and Quantum Mechanics. It's just the nature of orderly investigation into observations. New discovery changes our understanding of reality on almost a daily basis.
I really have problems when scientists try to bend their own rules to fit a square peg into a round hole. So it is with subjectivism. It seems that some of us simply can't admit that there are things in this world which do not fall under the domain of science. In order to quantify we must be able to measure. Subjective things simply are not amenable to any form of incremental measurement which can accurately mimic reality. We use terms like good, better and best to describe levels of acceptance, but these terms have little meaning outside of some standard arbitrarily set. We can't measure the amount of good, better or best with microjoules or megahertz or any other incrementally known values thus they are beyond quantification in any relevant sense. Rather than just admit this and get on with what we can quantify, attempts are made which may create illusions of grandeur among scientists, but upon careful analysis are only yet more subjectivism veiled as "science."

So it is with subjective terms such as "like, enjoy, prefer," etc. One can say they "like" the "Foveon Look" compared with the "Bayer Look." and rather than just accept this, there is the inclination to argue that one must present some "data" to justify the subjectivism. There is no need to do so, nor is there any necessity of justifying the subjective opinion. To say that one is preferred over the other should be sufficient into itself. Just because another or multitudes of others might not share that subjective opinion doesn't invalidate it in any way. I like strawberry ice cream rather than chocolate ice cream. Perhaps more people in the world prefer chocolate ice cream but that in no way invalidates my opinion and preference.

It's a foolish waste of time in my opinion to argue that something doesn't exist because it can't be quantified or argue that it (a subjective item) could be quantified if only the sample size were larger or whatever.

I'll take myself out of this discussion because to me it's a waste of time and energy. Some will not be convinced and probably because it's "in vogue" to be "scientific." Science has its place in my world but it's not the be all end all and it's far from perfect.

Best regards,


richard stone wrote:


In a way, science is its own religion, inspiring faith where there should be no basis for faith, or belief. And I am a big fan of science, as it is generally far better than any alternative.

For me this problem is exemplified in the work of John Ioannidis, A Stanford University Med. School professor who has written extensively, and to some acclaim, on the problem of verifying, that is to say, attempting to verify, so-called scientific tests and results. He has focused on medical issues but the problem is universal. The basic problem is that one test, one experiment, announced with great fanfare, is generally the basis for some "scientific" conclusion. And those tests are extremely difficult to verify, and frequently, indeed almost generally, those results are NOT duplicated. Yet this unhappy result is typically ignored. This lack of ability to verify is going to be more the case as investigations and issues become more complicated, and obscure.

And yet the whole basis of science is that it is verifiable, in the sense that it is subject to being proved wrong, as compared with belief systems which propose to provide "answers" on another basis.

Worse still, funding and interest in proposing and attempting and understanding those confirming tests is a lot less available than for testing or investigating some interesting and exciting "new" concept. In short, the entire scientific programme starts to resemble a house of cards. Even so, it is still the best we can do. That does not mean that a "scientific" test will reveal some truth. It's just (sort of) the best we can do.

Proposing some simple test is easy enough, such as a double blind kind of test, but even conducting the initial "test" or experiment is just one part of a scientific undertaking. And based on the work of Ioannidis the first test is likely to be wrong, and probably misleading.

Still, for me, the idea or claim of a "3d" effect for Foveon has always bordered on the mystical. Not that I do not prefer the Foveon images, because I do. What we know is that X3 Foveon provides equal, or very close to equal resolution for all colors, and that the sharpness is at the pixel level. We know that Foveon has some interesting aliasing issues, which seems, if anything, to make the image "crunchier" or "crisper" rather than making it blurry, or subject to moire. And we know the image is not interpolated.


Lin Evans wrote:

Hi Roland,

It's a scientific fallacy to assume that the subjective is quantifiable. No matter what the population size it's impossible to quantify things such as emotions. Only by manipulating the data and arbitrarily assigning values is it possible to create the "appearance" of quantification. There is no way to accurately assess the intensity of an emotion. Making ballpark guesses is all that's possible and this doesn't result in real knowledge. You can argue this forever but you will not be able to quantify an emotional response because there is no scientific way to measure such. You can't measure it in ounces, grams, liters, volts, watts, frequency or any other known physical property. If you can't measure it you can't quantify it. Any attempts are simply guesses and in my opinion no better than an observer stating "he was really mad about that, or he was mildly upset, or she was furious." These are arbitrary divisions and have no scientific meaning. All these attempts do is to order the emotion in some ascending or descending manner. Whether the increments are linear or logarithmic or something else entirely are unknowable. Sorry, but we will never be able to agree on this if you hold that emotions are any more quantifiable because of the population size. There is no way to know that the population is even homogeneous. What may appear as a mild emotion in fact could be a strong emotion. The observer simply can't get inside the head of the observed.

Best regards,


Roland Karlsson wrote:

Lin Evans wrote:

The connection is that unsolicited comments made by non-Sigma users at multiple major photography and equipment shows on numerous separate occasions revealed an unusual number of comments about a "3D" appearance. These comments were not heard at other manufacturer's displays nor have they been reported as being commonplace as they were at the Sigma booths.

As far as I remember those claims were strongest for the old 3 MP sensor and maybe also for the 4.5 MP sensor. And this was for huge prints, even A0. Now A0 is one million square mm. So --- it is 3-4 pixels per square mm. Even at some disyance, you would have strongly visible aliasing.

I do not know - but maybe this has something to do with the claim?

What does this mean? I have no idea. But trying to quantify it is unjustified from a scientific perspective. Something appears to have a dimensional appearance or it looks flat. We could discuss possible reasons until the cows come home and still have no agreement because subjectivity is not quantifiable.

Yes, it is. It takes some population size and some amount of tests in order to show anything. But ... it is clearly quantifiable. For small groups it is harder though. But ... likewise can be said for all subjective opinions. They are very hard to evaluate. As we have seen any time someone here tries to make a comparison by having people look at images.

It's not about "prefering" a Sigma over some other camera at all. It's about perceiving a 3D appearance from a Sigma photo.

A fine line to walk. It would be possible that a flat look and not the irritating 3D look more often is to be preferred

-- hide signature --
 Lin Evans's gear list:Lin Evans's gear list
Olympus E-10
Post (hide subjects) Posted by
(unknown member)
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow