The only Foveon image you have posted in the Photo Gallery is Roofer 25ft+ up from 50ft with long-lens hand held camera 1/500th--ISO-100:



When you look at the Sigma SD9 specs you will see the effective resolution of 10.3 million pixels. But the actual size of the output image is 3.43MP. That's what everyone else sees on their screen when they download the image. We see an image that would be pixelated if you blow it up above 100%.

It's only the output image that counts. The number of pixels is simply the product of the height and width.

I recently posted an Extremely cropped image from a 45.7MP Z7. It was cropped down to 1563x1954, just over 3MP. Nobody would look at that image and claim that it contains 9MP or that the original resolution before cropping was 137MP.

The extra term (x 3) in Foveon's calculation is to calculate what they claim is "effective pixels", not MP. They are trying to compensate for the fact that their sensors were still small and didn't have much actual resolution. You fell for that but the rest of us know better.

You have posted only a couple of other Foveon images images anywhere on this site since you registered a year ago.

One of them was taken with an SD 14 - 2652x1708, 4.5MP.

If you think your Foveon sensors are so good, why not post some of your images. Put your money where your mouth is. If you can't, go away.

__________________________(reply)

I have never come upon such bullheaded idiologues as on this site. And you take the cake---One other one kept insisting that the roofer photo you dug up from a year ago had a colored fringe which he himself could only have inserted since my copies had no fringe. You can fiddle with vocabulary all you want the SD1-Merrill has three layers of pixels and totals 46 megapixels on the sensor. And the files are 45-50 megabytes of size data which makes an image as large as I will ever need. If I want to cover a wall, I will use my 5x7 in film or my 8x10 in a view camera and NOT screw around with your didly digitals of any branding. For now my SD-1 which was introduced at $10 grand and was used in the world trip by artist/photographer Mario Dirks and published in the book "OUR WORLD TOUR." You don't get better images under difficult circumstances which he surmounted with his Sigma SD-1---46megapixel camera, fitted with the best Sigma lenses?----So---will you be beating a dead horse just one more time in an attempt to get the last word????--------------ew

I realize that it's it will be difficult for you to have to admit that Foveon has been feeding you a false information but that's exactly what has been happening to you from the beginning. Here are the specs from the B&H site and they contain the same pseudo arithmetic:

You are still confusing megabytes with megapixels. You don't understand what Foveon means by "effective" pixels and you don't understand the difference between resolution and sharpness.

The reality is that when final image is 4704x3136 pixels, that's only 14.75MP.

The magic of Foveon is that it does away with the demosaicing step that undoes some of the sharpness. I have already seen that for myself when I removed the Bayer array from a 24MP Sony sensor and bypassed the demosaicing process to produce a monochrome image. Although I still ended up with 24MP, the images are as sharp using the same lens as what I can get from a 45.7MP Z7. That's not an increase in resolution, just an improvement in sharpness.

We were all hoping that you were not coming back but, since you are here again, how about posting some examples from one of your Foveon cameras? If you can't do that then we will know that you are not playing with a full deck.

Whether a pixel (picture element, essentially a set of three numbers) is created by processing three LAYERS of sensels or the data from some number of adjacent RGGB sensel "cells" (yes two greens for each red and blue), the result is about the same number of pixels. All Foveon is doing is stacking the sensels in a column instead of separating them into a grid and combining them with lenses, filters, and math. THEORETICALLY, yes, that has some great advantages. In practice, however, the technology still needs a LOT of work. As much as I wanted it to work when I first saw it introduced, it has disappointed me every time I've revisited it. The good news is, a Bayer sensor still does a great job of emulating how the human eye works. For now, it does a better job.

https://teachmephysiology.com/nervous-system/ocular-physiology/retina/

Foveon's approach has one clear advantage over Bayer but it is fundamentally misunderstood by its supporters and promoted by a lunatic fringe of logically impaired supporters.

Here is a more lucid explanation of the differences - Foveon X3 sensor

That article ends with an interesting statement:


The 12.7MP image from Bayer array is sharper than the "14.7MP" Foveon image. In other words, 14.7MP is not true.

It also begs the question, "What is superior color resolution?"

The color filter array (Bayer or X-Trans) approach is the universally accepted solution. All of the successful development has followed that path. Even the Sigma fp L uses a Bayer array to produce 61MP.