Q Looses the 3-D of the M?

Started 3 months ago | Discussions thread
Scottelly
Senior MemberPosts: 2,132Gear list
Like?
Re: what's 3-D?
In reply to Roland Karlsson, 2 months ago

Roland Karlsson wrote:

Truman Prevatt wrote:

Roland Karlsson wrote:

Some few images have a kind of well defined sharpness that helps giving the feel of 3D. OOF parts have just the right fuzziness and all parts in focus are very sharp.

However, that is one of the oldest tricks in the book to produce the illusion of depth.

Yes, the problem is that Bayer cannot really do it - not without downsampling at least.

I've actually seen as many examples of Bayer - 3D as I have with Foveon 3D. A good photographer will bring it out.

At 100% ?

It is possible with images that are not 100 % of course.

Down-sizing a Bayer image results in an image that still seems to have some blur. Try down-sizing 36 MP images from a D800 or D800E to 14.7 MP and comparing them to identical images from an SD1 or DP series camera with a Merrill sensor. You'll get a very similar look, but it will not quite match. I think this is why Sigma used to claim the full megapixels of their sensors as the true resolution. Today, with better processing algorithms and no AA filters, Bayer pattern CFA sensors are getting better, but they still seem to capture less than 1/2 the resolution of an equivalent Foveon sensor (native mode resolution) . . . if one existed. So far, unfortunately, we haven't seen a Foveon sensor at 24 MP (native image resolution). Hopefully I am not confusing people too much with what I'm saying here. What I mean is Foveon sensor native resolution has traditionally been 1/3 of the sensor's "pixels" (Merrill being a 44.1 MP (14.7 x 3) Foveon sensor, but only producing a 14.7 MP "native" image. The Quattro complicates things a lot, so I think we should use the "native" image resolution more now. The Quattro is truly a 19.6 MP native resolution, though not as contrasty as the Merrill sensor, which is a 14.7 MP native resolution. Sigma calls the Quattro a 29 MP sensor, which is how many points of data collection there are in the sensor, but they are also stating a 39 MP equivalent resolution. That is double the "native" resolution, and certainly valid, when considering the Nyquist analysis so often used to compare Foveon sensors against CFA sensors in the past.

In the final analysis though, there seems to be detail in the images captured by CFA sensors that lacks clarity. I think the lack of clarity is why the "regular" digital cameras don't seem to capture the same "3 D-ness" that the Foveon sensor cameras capture. As far as perceived clarity goes, the Merrill sensor appears to have a leg up on the Quattro per pixel, but since the Quattro captures more pixels natively, I believe the perceived clarity is as high or higher, when the total image captured is considered. I DON'T believe there is "Bayer-ness" in the Quattro images. I believe there is something about them that gives people the impression that they are more "Beyer-ish" though. Could we call it "X-Trans-ish?" Maybe. Maybe I should use the terms "CFA-ness" and CFA-ish." Or should I just say "blurred?"

I like the Foveon images because they are NOT blurred. I like the Quattro images because they appear more realistic in most instances and are higher resolution. I think I would like to see a full-frame 72 MP (24 x 3) Merrill sensor in a camera from Sigma though. I think that would be the ultimate Foveon sensor achievemenet. Later, after a year or two, Sigma could make a full-frame Quattro, producing a sensor with 39 MP "native" resolution (far more than the 72 MP Merrill sensor's 24 MP "native" resolution). THAT would also be quite an achievement.

 Scottelly's gear list:Scottelly's gear list
Sigma SD14 Sony DT 18-55mm F3.5-5.6 SAM Tamron SP AF 10-24mm F/3.5-4.5 Di II LD Aspherical (IF) +7 more
Reply   Reply with quote   Complain
Post (hide subjects)Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow