Gvie up on SD1? Pentax K-1/ar7ii? Need advice.

Hey guys...

After 3 weeks of back and forth with Sigma, they can't resolve the color problems that the cam is having with wide angle lenses--even their own wide angles. They're offering me a refund and I'm considering it.

I love the Foveon concept, but is there a, iso 100 IQ advantage for Foveon versus say a Pentax K-1? I can get a refurb k-1 for 1200. What in your mind are some things the SD1 does better?

I've been looking at the test images on imageresource.com and comparing zoo tigers on Flickr from the Sigma and the Pentax..... zoo tigers and the sonoran desert because I grew up there. The sigma pics indeed look different and somewhat filmic, but it's hard to argue with nearly double the dynamic range on the K-1 (7 vs 14). Detail looks comparable to my eye, more detail on the Pentax, arguably more color info on the sigma....then again if 1/4 of the image has gone to black or white because of DR then there goes that advantage.

Bottom line: I miss MF film. No cartoony flat digital greens, no problems with flesh tones/reds, no problems with highlights provided you're careful. And very little editing.

FF Foveon may never happen or may be beyond my budget for the next 3-4 years. I have a wonderful old EF 100-400 4.5-5.6 L Canon lens with no current use as a sold my 7d for the Sigma... Would sell the lens to get money for better sigma lens, but not confident in the system given the issues I'm having. Film gear I can't afford to use.

I need something for the next couple years that won't be a total money pit while i wait for FF foveon to maybe pan out.

PTemple wrote:

Hey guys...

After 3 weeks of back and forth with Sigma, they can't resolve the color problems that the cam is having with wide angle lenses--even their own wide angles. They're offering me a refund and I'm considering it.

Sounds like your particular unit may be the problem. They are plenty of users here who don't have colour issues with WA lenses.

What model camera are you referring to? I'm guessing an sdQ(H)?

I love the Foveon concept, but is there a, iso 100 IQ advantage for Foveon versus say a Pentax K-1? I can get a refurb k-1 for 1200. What in your mind are some things the SD1 does better?

You simply have to work differently with Sigma cameras. You have to slow down and meter carefully - learn how to get the most out of your gear and post processing. The results are can be very rewarding.

I've been looking at the test images on imageresource.com and comparing zoo tigers on Flickr from the Sigma and the Pentax..... zoo tigers and the sonoran desert because I grew up there. The sigma pics indeed look different and somewhat filmic, but it's hard to argue with nearly double the dynamic range on the K-1 (7 vs 14). Detail looks comparable to my eye, more detail on the Pentax, arguably more color info on the sigma....then again if 1/4 of the image has gone to black or white because of DR then there goes that advantage.

Bottom line: I miss MF film. No cartoony flat digital greens, no problems with flesh tones/reds, no problems with highlights provided you're careful. And very little editing.

FF Foveon may never happen or may be beyond my budget for the next 3-4 years. I have a wonderful old EF 100-400 4.5-5.6 L Canon lens with no current use as a sold my 7d for the Sigma... Would sell the lens to get money for better sigma lens, but not confident in the system given the issues I'm having. Film gear I can't afford to use.

I need something for the next couple years that won't be a total money pit while i wait for FF foveon to maybe pan out.

PTemple wrote:

Hey guys...

After 3 weeks of back and forth with Sigma, they can't resolve the color problems that the cam is having with wide angle lenses--even their own wide angles. They're offering me a refund and I'm considering it.

I love the Foveon concept, but is there a [100 ISO] IQ advantage for Foveon versus say a Pentax K-1? I can get a refurbished K-1 for 1200 [USD?]. What in your mind are some things the SD1 does better? <snip>

1) NO color aliasing:

http://kronometric.org/phot/sensor/fov/Color_Alias_White_Paper_FinalHiRes.pdf

2) The SD1 Merrill correctly processed in SPP is well-endowed with acutance, some say too much.

3) Just pop out the dust-cover, slap an IR filter on the lens - and IR photography is ready and waiting.

In your shoes, I would take the refund and get a K-1 ... or grab an SD15 - they're not noted for the kind of problems you've been describing.

SigmaChrome wrote:

PTemple wrote:

Hey guys...

After 3 weeks of back and forth with Sigma, they can't resolve the color problems that the cam is having with wide angle lenses--even their own wide angles. They're offering me a refund and I'm considering it.

Sounds like your particular unit may be the problem. They are plenty of users here who don't have colour issues with WA lenses.

What model camera are you referring to? I'm guessing an sd Quattro H? <>

Going by the thread title, I'm guessing an SD1?

Points to consider:

I considered the K1 when it first came out as my second camera but the lack of high quality fixed lenses put me off. I do pano stitching and near perfect lenses are a requirement for high quality pano stitches. But they have recently announced a 85 f1.4 which hopefully is the start of a high quality lens series.

I've seen direct comparisons of the K1 to the SD1M and even though the K1 resolution is higher in pixel shift mode it still has the same Bayer difficulty of picking up subtle color and tonal variations that the SD1M can see and Bayer can't, and using the SD1M with pano stitching I can have both.

DR is something the K1 is much better with than any Foveon chip. Every camera system has its trade offs, ultimately you have to decide which features are the most important to you.

As for your wide angle lens problems, I've never had any using Sigma's 10-20 f3.5 and the 8-16.

xpatUSA wrote:

SigmaChrome wrote:

PTemple wrote:

Hey guys...

After 3 weeks of back and forth with Sigma, they can't resolve the color problems that the cam is having with wide angle lenses--even their own wide angles. They're offering me a refund and I'm considering it.

Sounds like your particular unit may be the problem. They are plenty of users here who don't have colour issues with WA lenses.

What model camera are you referring to? I'm guessing an sd Quattro H? <>

Going by the thread title, I'm guessing an SD1?

Yeah. Some days you look and don't see.

You could get the Sony with the Sigma adapter to use your Canon lens, and get a DP0 to scratch your Foveon itch.  I've never used a DP0 but it has an excellent reputation.

Jan

Shots from the Pentax cameras that I have seen posted have had very nice colours, but of course look different from Foveon photos. Pentax make excellent lenses.

The great strength of the Sony A7rii is that it can be used with a very wide variety of adapted lenses. The more recent Sony models have very advanced (and complicated) focussing, so are ideal for wildlife photography.

mike earussi wrote:

Points to consider:

I considered the K1 when it first came out as my second camera but the lack of high quality fixed lenses put me off. I do pano stitching and near perfect lenses are a requirement for high quality pano stitches. But they have recently announced a 85 f1.4 which hopefully is the start of a high quality lens series.

I've seen direct comparisons of the K1 to the SD1M and even though the K1 resolution is higher in pixel shift mode it still has the same Bayer difficulty of picking up subtle color and tonal variations that the SD1M can see and Bayer can't, and using the SD1M with pano stitching I can have both.

It is the subtle hue variations that Bayer cameras don't record well. You can see an area of grass or a bush in a Bayer image which is the same hue of green all over.

DR is something the K1 is much better with than any Foveon chip. Every camera system has its trade offs, ultimately you have to decide which features are the most important to you.

As for your wide angle lens problems, I've never had any using Sigma's 10-20 f3.5 and the 8-16.

DMillier wrote:

I can understand in principle why this may be the case for Bayer sensors. But in pixel shift mode? What could be the explanation in pixel shift mode - this effectively bypasses the Bayer pattern...

Pixel shift enables each pixel to be unique, but the Bayer color is also limited by the fixed color filters on the chip. Foveon doesn't have this limitation.

mike earussi wrote:

DMillier wrote:

I can understand in principle why this may be the case for Bayer sensors. But in pixel shift mode? What could be the explanation in pixel shift mode - this effectively bypasses the Bayer pattern...

Pixel shift enables each pixel to be unique, but the Bayer color is also limited by the fixed color filters on the chip. Foveon doesn't have this limitation.

The Foveon colour filters (layers of silicon) are fixed too, but their more gentle slopes allow finer discrimination of hues than the narrow-band filters used in Bayer sensors.

Some say that the Foveon filters lead to increased noise.

DMillier wrote:

D Cox wrote:

mike earussi wrote:

DMillier wrote:

I can understand in principle why this may be the case for Bayer sensors. But in pixel shift mode? What could be the explanation in pixel shift mode - this effectively bypasses the Bayer pattern...

Pixel shift enables each pixel to be unique, but the Bayer color is also limited by the fixed color filters on the chip. Foveon doesn't have this limitation.

The Foveon colour filters (layers of silicon) are fixed too, but their more gentle slopes allow finer discrimination of hues than the narrow-band filters used in Bayer sensors.

Some say that the Foveon filters lead to increased noise.

Do we have sources for this?

Here's one from a paper by AVC/Foveon:

"3.3 Color matrix effects

The overlap in the spectral response of the three channels, while providing robust information on the wavelength of narrow-band sources, leads to relatively large off-diagonal terms in the color transformation matrix needed to produce tristimulus values. As an example, it can be easily seen that the blue channel in the tristimulus function is much narrower than the blue photo-diode response. To narrow this channel, a large component of the green must be subtracted. The green is reduced by the blue and the red to narrow the peak and the red channel is augmented with the green to shift the peak to a slightly lower wavelength. Figure 7 shows a typical matrix10 that incorporates those features.

While large off-diagonal terms have a negative effect on signal-to-noise ratio, this is generally not more significant than the effect of the absorption losses in color filter arrays. For color filter arrays using yellow-magenta-cyan or other overlapping bands to increase light throughput, large off-diagonal color matrix terms are also needed, negating some of the sensitivity improvement. Those arrangements still absorb at least one-third of the light."

http://kronometric.org/phot/sensor/fov/SD9%20sensor%20in%20depth.pdf

While not exactly false, the last paragraph seems to be a bit like marketing-speak.

In that same paper, they claim an SNR of 61dB for the green layer, which is just over 10 EV, FWTW ...

DMillier wrote:

...and the impact of all this on pixel shifted Bayer composite shots alleged failure to detect subtle hue changes?

Zero. I was responding to "Some say that the Foveon filters lead to increased noise".

Pardon me for posting something longer than a line or two ...

DMillier wrote:

I can understand in principle why this may be the case for Bayer sensors. But in pixel shift mode? What could be the explanation in pixel shift mode - this effectively bypasses the Bayer pattern...

I'm on a bit of a journey at the moment - audio hifi. I'm doing an electronics course with the aim of learning how to design and build stereo amplifiers and I'm learning about loudspeaker design. I'm also learning about DSP and measurement techniques.

As part of this journey I'm doing a certain amount of experimentation and online research. it is very evident (within minutes, really) that the hifi world is bereft of scepticism and full of lots of very passionate, sincere people just making sh*t up. it is extremely difficult to cut through the huge amount of made up beliefs to find out what is real.

This seems to be a common problem, especially amongst men (for some reason) in many technical fields that lend themselves to amateur hobbyist dabbling. I put photography in that set. i don't doubt the passion or the sincerity but filtering truth out of the noise is a struggle. Human beings (no matter how well educated) seem to gravitate much more easily to rumour, religion and fantasy than fact, and science. There is an almost total lackl of appreciation for the impact of bias and a lack of scepticism. We are much more "believing engines", than engineers. I'm rapidly gaining an appreciation of Ted's "pedantry" - he has standards...

This is very interesting. There are things that can't be measured or scientifically proved when passion and emotion come into the picture. I use a lot of "junk" that I love and wouldn't change for newer, more expensive and "better produkts".

Foveon is unique design, others just rely on traditional Bayer color filter and few offer monochrome without the filter, but from some reason no filter is more expensive.

K-1 is the cheapest one with pixel shift which can come closer to Foveon, but only works for subjects which remain static until four photos have been shot. More expensive choice would be higher megapixel number which is then downsampled to have more color resolution for smaller image. Basically you need quite a big difference to have obvious differences. I'm afraid you would have to get something like Fuji GFX 100 to have pixel shift look with non-static subjects when downsized to more typical resolutions like 8K.

The capacity to see minor difference in hue is due to the nature of the filters.  Pixel shifting will not alter the filter characteristics.

Assume the RGB filters to be bandpass with very abrupt cutoffs, no overlap in response by the R, G or B filter. Assume two monochromatic lights of different frequency but both falling within the G band. There will be no response from the R or B sites. There will be no way to tell the lights apart.

With the filter characteristics of the Foveon sensor there would be a response in all layers. The response from the two lights will be different. In theory one would be able to notice the difference.

In practice, in my 19 years of digital photography, with a Nikon coolpix990, Canon DSLRs and Panasonic m43s and all of the Sigma camera's, sd14 and later, I have had more problems getting the colours right (pleasing) from the Sigma's than from any of the others. But that may be due to my eyesight, monitor or printer...

Paula