What is holding Sigma back with the new Merril(s)?

Started 5 months ago | Discussions thread
D Cox Forum Pro • Posts: 29,351
Re: What is holding Sigma back with the new Merril(s)?

Tom Schum wrote:

mike earussi wrote:

Tom Schum wrote:

mike earussi wrote:

xpatUSA wrote:

D Cox wrote:

A 20 Mpix Foveon is not enough for 24x36mm.

76 lp/mm at the sensor would be good enough for any lens except an Otus mebbe.

I've printed 24x36 images from my 14.7mp Merrill that look excellent when using Topaz Gigapixel AI (in Architectural mode) to enlarge the file. BTW, Foveon files uprez very well, much better than Bayer.

It would be interesting to find out if Topaz Gigapixel would do well with a fp L file that has been downsized to 1/4 original pixels. The quality of these pixels ought to equal that of the Merrill, pixel for pixel, should it not?

No, a Merrill sensor will still look different because of how it determines color. You'll have better DR with the Bayer but poorer color variations.

Sigma knows how to determine color from a Foveon sensor, so they might be able to get a close match for it with special processing of a Bayer image. Or am I guessing wrong here and it is impossible?

This is one of several reasons I am becoming interested in the fp L.

The colours I get from the fp are quite pleasing. I used the Color Checker to make a custom preset. But I think the Merrill colours are better.

My opinion is that using three rather narrow-band filters will give an uneven response across the spectrum, with hues shifted toward the dominant wavelengths of the filters. A Foveon gives wide-band filters with plenty of overlap, so there are two or three numbers for each wavelength.

Ted thinks this gives more noise, which may be true. I can live with a bit more noise if it means better colours.

Don Cox

 D Cox's gear list:D Cox's gear list
Sigma fp
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow