Sigma sd Quattro H and the fp both with the 135mm Art lens



Here's an fp shot with a Russian Helios 44-2 (58mm f/2) lens which I've had for years but never used because it didn't focus beyond about 3m. I took the lens apart and adjusted the helicoid stop (filed a bit of metal off!!) and now it focuses just fine; close to infinity.

Not everyone's cup of tea! But a bit of fun

0df416fc9d2043b8b9c47c8d68f502d4.jpg
Excellent "swirl", Steve!

I have that lens and a 44M.

--
what you got is not what you saw ...
 
Thanks, Steve. The image from FP comes out darker, I brightened it a bit and vice versa. Other settings to the left of "0". For SDQ, I always use crispy, then denoise in Topaz DeNoise - the result is then great! Due to the different time of photography, I did not solve the colors. Sigma FP is close..

SDQ H
SDQ H

FP
FP

settings
settings
I don't understand, when comparing CAMERAS and particularly raw data, why people have to do adjustments of any kind at all !! Best use RawDigger:

9b6f6f7d92fc4c10bb312b59dca5c4e9.jpg.png

8f89f53d8ba94eb3bdcb8cf987f9ac8f.jpg.png

Ignore EV0 setting, it's on Auto for both.

fp
fp

sd
sd

:-D

--
what you got is not what you saw ...
 
Last edited:
Thanks, Steve. The image from FP comes out darker, I brightened it a bit and vice versa. Other settings to the left of "0". For SDQ, I always use crispy, then denoise in Topaz DeNoise - the result is then great! Due to the different time of photography, I did not solve the colors. Sigma FP is close..

SDQ H
SDQ H

FP
FP

settings
settings
I don't understand, when comparing CAMERAS and particularly raw data, why people have to do adjustments of any kind at all !! Best use RawDigger:

9b6f6f7d92fc4c10bb312b59dca5c4e9.jpg.png

8f89f53d8ba94eb3bdcb8cf987f9ac8f.jpg.png

Ignore EV0 setting, it's on Auto for both.

fp
fp

sd
sd

:-D
I'm not really sure what I'm seeing in Raw Digger. However, I think I shot the fp at EV -0.3 and the sd at EV 0

I hasten to add that was a **** up.

My original processing in SPP and LightRoom was the same with both camera images. Pretty much everything at zero. The fp image looked a little flatter. When I increased the detail button and took the Noise buttons to zero for the fp, it lost the flatness and there was little to chose between the shots other than colour.

Ted, I know you don't have a Quattro or an fp but what is your opinion in compararing the images presented here? I know my comparison photos are a bit slap dash but I do think they are useful.

I'd like to try a landscape comparison with more distant detail. I believe the Quattro will do better but for me, right now, I don't believe the image quality of the Quattro is, on the whole, sufficently better than the fp to worth continuing to use the Quattro as an every day camera. Mostly because of it's ISO 100 limitation. I'm pretty shocked I've come to this conclusion.

Thing is the SA mount Art lenses are stupendously good. And they fit on the fp.

--
Best, Steve
 
Last edited:
I don't understand, when comparing CAMERAS and particularly raw data, why people have to do adjustments of any kind at all !! Best use RawDigger:

9b6f6f7d92fc4c10bb312b59dca5c4e9.jpg.png

8f89f53d8ba94eb3bdcb8cf987f9ac8f.jpg.png

Ignore EV0 setting, it's on Auto for both.

fp
fp

sd
sd

:-D
I'm not really sure what I'm seeing in Raw Digger.
What I see is not a lot of difference betwixt the two.
However, I think I shot the fp at EV -0.3 and the sd at EV 0

I hasten to add that was a **** up.
Not a big one, though. Perhaps it accounts for the difference in the color of the bricks. After all, -0.3EV is only a difference of 1/8 e.g. 113 ISO vs. 100.
My original processing in SPP and LightRoom was the same with both camera images. Pretty much everything at zero. The fp image looked a little flatter. When I increased the detail button and took the Noise buttons to zero for the fp, it lost the flatness and there was little to chose between the shots other than colour.

Ted, I know you don't have a Quattro or an fp but what is your opinion in comparing the images presented here? I know my comparison photos are a bit slap dash but I do think they are useful.
I think it depends on the purpose of the comparison which I believe is to show that it is possible to get similar image quality from the two cameras one way or another, e.g. raw formats, converters, converter adjustments, output-referred adjustments.
I'd like to try a landscape comparison with more distant detail. I believe the Quattro will do better but, for me right now, I don't believe the image quality of the Quattro is, on the whole, sufficiently better than the fp to be worth continuing to use the Quattro as an every-day camera. Mostly because of it's ISO 100 limitation. I'm pretty shocked I've come to this conclusion.
Watch out for circling sharks which might speed in to tell you how good the Quattro is at over 100 ISO.
Thing is the SA mount Art lenses are stupendously good. And they fit on the fp.
--
what you got is not what you saw ...
 
Hi TN Args.

I didn't understand your post TN Args .. I love and respect Foveon and the work of the Sigma engineers as myself and would never knowingly offend them. The modesty and humility with which Mr. Kazuto Yamaki performs is always a pleasant surprise for me, and I really liked him.
The Sigma SDM/Q is the only camera I'm willing to use instead of my last Pentax 67II. I think Foveon is a gem and I'm very happy to use it. The fact that the Quattro sensor is assembled differently than the original Foveon 1:1:1 is a fact, not an insult. I myself bought an SD Quattro to go with my great SD1 Merrill. The SD Quattro is an excellent advanced camera and I appreciate having it as do others on the sigma forum.
Please keep calm and balance, use as much empathy as you can, so that you don't blame yourself for your careless actions in the future - just like me, I examine my intentions and correct my shortcomings as I go. None of us here are without flaws, maybe the older we are, the better, but it's not possible without learning..

Patience my friend, we are all friends on the sigma forum after all. Please don't misinterpret my post. Peter
I suspect Arg's problem is with you describing Foveon's other than Quattro's as "real" - which can be easily taken as disparaging of the Quattro. I further suspect that Args does not feel the Quattro is some kind of watered down Foveon, but is every bit as real and effective as the 1:1:1 design.

But I should really leave it to Args to answer, sorry for butting in!
You did well!
 
I don't understand, when comparing CAMERAS and particularly raw data, why people have to do adjustments of any kind at all !! Best use RawDigger:

9b6f6f7d92fc4c10bb312b59dca5c4e9.jpg.png

8f89f53d8ba94eb3bdcb8cf987f9ac8f.jpg.png

Ignore EV0 setting, it's on Auto for both.

fp
fp

sd
sd

:-D
I'm not really sure what I'm seeing in Raw Digger.
What I see is not a lot of difference betwixt the two.
Ah. That's interesting and kinda confirms my thinking, Thanks, Ted. And I'm using the 'lowly' fp rather than the 61 mp fpL, which according to Scott, should blow the doors of the fp and its 24 mp sensor. I dunno. I do know the fpL has an AA filter and the fp doesn't and many don't like that

Best, Steve
 
Last edited:
Steve, I processed the photos to be about the same brightness, since cameras are not the same (and photos are not exactly "metered"), brightness is by eye.

This is my best result from either camera. Cameras have different color rendering, photographed in different light.

SDQ H
SDQ H

FP
FP
 
Last edited:
I noticed the H image was raw Hi-Res. One thing about the Imatest "Spilled Coins" target is that the results are independent of magnification ...
Imatest says:

The Imatest Spilled Coins chart (released January 2013) has several advantages over existing Dead Leaves charts. Most importantly it is almost perfectly scale-invariant, and hence produces more reliable and consistent results. Scale-invariant means that the pattern statistics (especially frequency spectrum and contrast) are independent of magnification, i.e., do not vary with camera-to-target distance, cropping, or pixel dimensions. This greatly simplifies setup and analysis— advantages are described here. This requirement is met by random (or nearly random) patterns that have a 1/f frequency spectrum - a (1/f)^2 Power Spectral Density (PSD).
Did you do any analysis of the SD Quattro H image vs the fp L image, using the "spilled coins" chart in those sample images (full-size AND raw files available here on DPreview) Ted?
No Scott, I can't afford the Imatest app ...
... and, in any case, Imatest's chart is not incorporated into DPR's test image - meaning that Imatest's app can not analyze DPR's test image.

Imatest:

185427a405684cb8a0e750f23443713a.jpg

DPR:

7c18fb1fd71843b0b0a199442616bb89.jpg
You're looking at the wrong part of the DPreview image Ted. I was referring to the part to the right of middle, which looks like that Imatest image.

You can see it here betweeh the six fluffy balls and the photo of the girl.


--
Scott Barton Kennelly
https://www.bigprintphotos.com/
 
Last edited:
I noticed the H image was raw Hi-Res. One thing about the Imatest "Spilled Coins" target is that the results are independent of magnification ...
Imatest says:

The Imatest Spilled Coins chart (released January 2013) has several advantages over existing Dead Leaves charts. Most importantly it is almost perfectly scale-invariant, and hence produces more reliable and consistent results. Scale-invariant means that the pattern statistics (especially frequency spectrum and contrast) are independent of magnification, i.e., do not vary with camera-to-target distance, cropping, or pixel dimensions. This greatly simplifies setup and analysis— advantages are described here. This requirement is met by random (or nearly random) patterns that have a 1/f frequency spectrum - a (1/f)^2 Power Spectral Density (PSD).
Did you do any analysis of the SD Quattro H image vs the fp L image, using the "spilled coins" chart in those sample images (full-size AND raw files available here on DPreview) Ted?
No Scott, I can't afford the Imatest app ...
... and, in any case, Imatest's chart is not incorporated into DPR's test image - meaning that Imatest's app can not analyze DPR's test image.

Imatest:

185427a405684cb8a0e750f23443713a.jpg

DPR:

7c18fb1fd71843b0b0a199442616bb89.jpg
You're looking at the wrong part of the DPreview image Ted.
Oops!! I had never noticed that before, per se ...
I was referring to the part to the right of middle, which looks like that Imatest image.

You can see it here betweeh the six fluffy balls and the photo of the girl.

https://www.dpreview.com/forums/post/66922208
Yes, it could be a copy of the prior non-proprietary "Dead Leaves" target or DPR bought the Imatest improved version,

--
what you got is not what you saw ...
 
…of course I understand that Quattro 1:0.25:0.25 image results are not the same as Foveon 1:1:1. They have something from Bayer and something from real Foveon..
I prefer the terms Classic Foveon and Advanced Foveon, reason being those terms correctly describe the relative merits of the images. Plus, your choice is more than faintly insulting and unnecessary.
🧀

Just something to go with that wine.

😉
 
I don't understand, when comparing CAMERAS and particularly raw data, why people have to do adjustments of any kind at all !! Best use RawDigger:

9b6f6f7d92fc4c10bb312b59dca5c4e9.jpg.png

8f89f53d8ba94eb3bdcb8cf987f9ac8f.jpg.png

Ignore EV0 setting, it's on Auto for both.

fp
fp

sd
sd

:-D
I'm not really sure what I'm seeing in Raw Digger.
What I see is not a lot of difference betwixt the two.
Ah. That's interesting and kinda confirms my thinking, Thanks, Ted. And I'm using the 'lowly' fp rather than the 61 mp fpL, which according to Scott, should blow the doors of the fp and its 24 mp sensor.
Whoa Nelly! That's not quite what I said or meant. Here's what I actually said:

"I wonder if the fp L blows the doors off the fp and SD Quattro H. I guess it should, for the price."

After I get another SD Quattro H (and a Sony A7r IV), I think I'll get a Sigma fp L. I'll do a similar comparison test between the SD Quattro H and the fp L at that time (pretty far in the future, I guess, because first I plan to get a Fuji X-H2 with some lenses, and a better/faster new computer, with which I can edit the 8K video from that camera). Then at least I'll be able to tell if the fp L can blow the doors off the SD Quattro H. Yeah, I guess I'm a crazy person, with a severe case of G.A.S. (Of course I'll need some good lenses for that Fuji camera too. 😉)
I dunno. I do know the fpL has an AA filter and the fp doesn't and many don't like that

Best, Steve
--
Scott Barton Kennelly
https://www.bigprintphotos.com/
 
Last edited:
The crucial thing that was not mentioned here is that the sigma FP is full-frame and the SDQ H is not full-frame, therefore the same aperture used does not show the same depth of field for both images. In the pictures, it can be seen, for example, on the backs of the chairs in the middle, as well as in the background of the wall on the right. I would use a higher aperture on the Sigma FP to make up for this difference. Then the resulting sigma FP image would be sharper in the foreground and further in the background, and better comparable.

The image of the Sigma FP is close to the SDQ H after the most elementary adjustment.
This fact records another very important thing: Sigma's SPP software is simply the best for a BAYER type sensor for basic image processing!!! This is the right move by the Sigma engineers who have matched the image output from the sigma FP with the SPP very well. For those who want to photograph with a Bayer type sensor, the Sigma FP (+ great Sigma lenses) - but image processed in SPP - is the best solution.

175f1062d5fa48f39e7d28eabd737587.jpg

dc2417a5d12349cc84b827edc06c27d2.jpg

Peter

For me, however, the only in principle unbeatable solution remains image output in Foveon 1:1:1 quality (or SD Quattro).
 
Last edited:
I don't understand, when comparing CAMERAS and particularly raw data, why people have to do adjustments of any kind at all !! Best use RawDigger:

9b6f6f7d92fc4c10bb312b59dca5c4e9.jpg.png

8f89f53d8ba94eb3bdcb8cf987f9ac8f.jpg.png

Ignore EV0 setting, it's on Auto for both.

fp
fp

sd
sd

:-D
I'm not really sure what I'm seeing in Raw Digger.
What I see is not a lot of difference betwixt the two.
Ah. That's interesting and kinda confirms my thinking, Thanks, Ted. And I'm using the 'lowly' fp rather than the 61 mp fpL, which according to Scott, should blow the doors of the fp and its 24 mp sensor.
Whoa Nelly! That's not quite what I said or meant. Here's what I actually said:

"I wonder if the fp L blows the doors off the fp and SD Quattro H. I guess it should, for the price."

After I get another SD Quattro H (and a Sony A7r IV), I think I'll get a Sigma fp L. I'll do a similar comparison test between the SD Quattro H and the fp L at that time (pretty far in the future, I guess, because first I plan to get a Fuji X-H2 with some lenses, and a better/faster new computer, with which I can edit the 8K video from that camera). Then at least I'll be able to tell if the fp L can blow the doors off the SD Quattro H. Yeah, I guess I'm a crazy person, with a severe case of G.A.S. (Of course I'll need some good lenses for that Fuji camera too. 😉)
I dunno. I do know the fpL has an AA filter and the fp doesn't and many don't like that

Best, Steve
Whoa Scott! That’s a shed load of cameras.
Best, Steve
 
Last edited:
I don't understand, when comparing CAMERAS and particularly raw data, why people have to do adjustments of any kind at all !! Best use RawDigger:

9b6f6f7d92fc4c10bb312b59dca5c4e9.jpg.png

8f89f53d8ba94eb3bdcb8cf987f9ac8f.jpg.png

Ignore EV0 setting, it's on Auto for both.

fp
fp

sd
sd

:-D
I'm not really sure what I'm seeing in Raw Digger.
What I see is not a lot of difference betwixt the two.
Ah. That's interesting and kinda confirms my thinking, Thanks, Ted. And I'm using the 'lowly' fp rather than the 61 mp fpL, which according to Scott, should blow the doors of the fp and its 24 mp sensor.
Whoa Nelly! That's not quite what I said or meant. Here's what I actually said:

"I wonder if the fp L blows the doors off the fp and SD Quattro H. I guess it should, for the price."

After I get another SD Quattro H (and a Sony A7r IV), I think I'll get a Sigma fp L. I'll do a similar comparison test between the SD Quattro H and the fp L at that time (pretty far in the future, I guess, because first I plan to get a Fuji X-H2 with some lenses, and a better/faster new computer, with which I can edit the 8K video from that camera). Then at least I'll be able to tell if the fp L can blow the doors off the SD Quattro H. Yeah, I guess I'm a crazy person, with a severe case of G.A.S. (Of course I'll need some good lenses for that Fuji camera too. 😉)
I dunno. I do know the fpL has an AA filter and the fp doesn't and many don't like that

Best, Steve
Whoa Scott! That’s a shed load of cameras.
Best, Steve
Yeah, and I plan to get a few lenses for my Nikon first too, such as a 40mm f1.4 Art, a 24-35mm f2 Art, and a 105mm macro.

🥴

--
Scott Barton Kennelly
 
I don't understand, when comparing CAMERAS and particularly raw data, why people have to do adjustments of any kind at all !! Best use RawDigger:

9b6f6f7d92fc4c10bb312b59dca5c4e9.jpg.png

8f89f53d8ba94eb3bdcb8cf987f9ac8f.jpg.png

Ignore EV0 setting, it's on Auto for both.

fp
fp

sd
sd

:-D
I'm not really sure what I'm seeing in Raw Digger.
What I see is not a lot of difference betwixt the two.
Ah. That's interesting and kinda confirms my thinking, Thanks, Ted. And I'm using the 'lowly' fp rather than the 61 mp fpL, which according to Scott, should blow the doors of the fp and its 24 mp sensor.
Whoa Nelly! That's not quite what I said or meant. Here's what I actually said:

"I wonder if the fp L blows the doors off the fp and SD Quattro H. I guess it should, for the price."

After I get another SD Quattro H (and a Sony A7r IV), I think I'll get a Sigma fp L. I'll do a similar comparison test between the SD Quattro H and the fp L at that time (pretty far in the future, I guess, because first I plan to get a Fuji X-H2 with some lenses, and a better/faster new computer, with which I can edit the 8K video from that camera). Then at least I'll be able to tell if the fp L can blow the doors off the SD Quattro H. Yeah, I guess I'm a crazy person, with a severe case of G.A.S. (Of course I'll need some good lenses for that Fuji camera too. 😉)
I dunno. I do know the fpL has an AA filter and the fp doesn't and many don't like that

Best, Steve
Whoa Scott! That’s a shed load of cameras.
Best, Steve
Yeah, and I plan to get a few lenses for my Nikon first too, such as a 40mm f1.4 Art, a 24-35mm f2 Art, and a 105mm macro.

🥴

--
Scott Barton Kennelly
https://www.bigprintphotos.com/
Ha! A good pal of mine is a commercial freelance photographer and I’m amazed at the age and seemingly poor quality of kit he uses. He came over to us the other day to take a load of photos of spring flowers for a data base he’s building for his commercial landscaping clients. Our garden is pretty jammed packed for spring flowers, so, I could see him in action. Mostly he used a 10 year old Olympus OM5 micro four thirds and a macro lens he got off eBay for 50 quid. Plus some home made reflectors held in place by bamboo canes.

I offered him my DP3 Merrill thinking the improved image quality would help. Nope, no point as people only look at photos on a phone screen.

That’s real life. It’s eejit amateurs like me that buys new gear.

Scott, have a look at the getDpi website. The L mount forum has some good fpL pics. Very impressive. That maybe the only camera you will ever need 😉



--
Best, Steve
 
I don't understand, when comparing CAMERAS and particularly raw data, why people have to do adjustments of any kind at all !! Best use RawDigger:

9b6f6f7d92fc4c10bb312b59dca5c4e9.jpg.png

8f89f53d8ba94eb3bdcb8cf987f9ac8f.jpg.png

Ignore EV0 setting, it's on Auto for both.

fp
fp

sd
sd

:-D
I'm not really sure what I'm seeing in Raw Digger.
What I see is not a lot of difference betwixt the two.
Ah. That's interesting and kinda confirms my thinking, Thanks, Ted. And I'm using the 'lowly' fp rather than the 61 mp fpL, which according to Scott, should blow the doors of the fp and its 24 mp sensor.
Whoa Nelly! That's not quite what I said or meant. Here's what I actually said:

"I wonder if the fp L blows the doors off the fp and SD Quattro H. I guess it should, for the price."

After I get another SD Quattro H (and a Sony A7r IV), I think I'll get a Sigma fp L. I'll do a similar comparison test between the SD Quattro H and the fp L at that time (pretty far in the future, I guess, because first I plan to get a Fuji X-H2 with some lenses, and a better/faster new computer, with which I can edit the 8K video from that camera). Then at least I'll be able to tell if the fp L can blow the doors off the SD Quattro H. Yeah, I guess I'm a crazy person, with a severe case of G.A.S. (Of course I'll need some good lenses for that Fuji camera too. 😉)
I dunno. I do know the fpL has an AA filter and the fp doesn't and many don't like that

Best, Steve
Whoa Scott! That’s a shed load of cameras.
Best, Steve
Yeah, and I plan to get a few lenses for my Nikon first too, such as a 40mm f1.4 Art, a 24-35mm f2 Art, and a 105mm macro.

🥴
Ha! A good pal of mine is a commercial freelance photographer and I’m amazed at the age and seemingly poor quality of kit he uses. He came over to us the other day to take a load of photos of spring flowers for a data base he’s building for his commercial landscaping clients. Our garden is pretty jammed packed for spring flowers, so, I could see him in action. Mostly he used a 10 year old Olympus OM5 micro four thirds and a macro lens he got off eBay for 50 quid. Plus some home made reflectors held in place by bamboo canes.

I offered him my DP3 Merrill thinking the improved image quality would help. Nope, no point as people only look at photos on a phone screen.

That’s real life. It’s eejit amateurs like me that buys new gear.

Scott, have a look at the getDpi website. The L mount forum has some good fpL pics. Very impressive. That maybe the only camera you will ever need 😉

--
Best, Steve
Thanks Steve, but I have a plan. I figure on getting the Fuji to shoot 40 MP super-telephoto and video stuff first, followed by the Sony A7r IV for better results when shooting eclipses. Then I'll get an fp L, and finally have an L mount camera.

Some day I might even get a 100 MP Fuji, but I need to make a few giant prints first, so I can justify spending that kind of money on camera equipment. I'll borrow a friend's GFX100 S for a weekend trip to somewhere photogenic, and make some giant prints I can compare, before pulling that trigger though.

;)

--
Scott Barton Kennelly
 
Interesting that the photos supplied by the OP are leading people to conclude the Sigma fp has effectively the same detail or resolution capability as the sdQH.

It doesn’t. Well, probably not. (Because DPR never tested the fp, I substituted Sony A7 III for fp in the following test image, which DPR say has a weak AA filter compared to none for the fp. But a weak AA filter has *very* little effect on resolution compared to the blatant difference we see below.)

I took the DPR test image at ISO 100 for A7III and sdQH (X3F) and processed in the same software (Affinity Photo), roughly matched brightness and WB, and resized to same size (54 MP). Let’s look at a small area of the test image at 200%.

ec93683c7bf148ba94f52d05d7742ee8.jpg

7243af85fe4040bdac001112d04b43d8.jpg

The difference is stark. Is it all in the Sony’s weak AA filter? I very much doubt it.

cheers

[Edit: apparently the Panasonic S5 has a 24 MP full frame sensor and has no AA filter, just like the Sigma fp, so here is its equivalent image. Confirms my point.]

0edfe23bd2b3443c88a66f906dfb6395.jpg

--
"A picture is a secret about a secret: the more it tells you, the less you know." —Diane Arbus
 
Last edited:
Interesting that the photos supplied by the OP are leading people to conclude the Sigma fp has effectively the same detail or resolution capability as the sdQH.

It doesn’t.
Yep. Some of us even think that the only way to compare two sensors is to look at 8x10 prints produced therefrom ... ;-)
Well, probably not. (Because DPR never tested the fp, I substituted Sony A7 III for fp in the following test image, which DPR say has a weak AA filter compared to none for the fp. But a weak AA filter has *very* little effect on resolution compared to the blatant difference we see below.)

I took the DPR test image at ISO 100 for A7III and sdQH (X3F) and processed in the same software (Affinity Photo), roughly matched brightness and WB, and resized to same size (54 MP). Let’s look at a small area of the test image at 200%.
Normally I would whine about re-sizing but the Siemens Stars are I believe scale-invariant and, in any case, make your point well enough.
ec93683c7bf148ba94f52d05d7742ee8.jpg

7243af85fe4040bdac001112d04b43d8.jpg

The difference is stark. Is it all in the Sony’s weak AA filter? I very much doubt it.

cheers

[Edit: apparently the Panasonic S5 has a 24 MP full frame sensor and has no AA filter, just like the Sigma fp, so here is its equivalent image. Confirms my point.]

0edfe23bd2b3443c88a66f906dfb6395.jpg
--
what you got is not what you saw ...
 
Interesting that the photos supplied by the OP are leading people to conclude the Sigma fp has effectively the same detail or resolution capability as the sdQH.

It doesn’t. Well, probably not. (Because DPR never tested the fp, I substituted Sony A7 III for fp in the following test image, which DPR say has a weak AA filter compared to none for the fp. But a weak AA filter has *very* little effect on resolution compared to the blatant difference we see below.)

I took the DPR test image at ISO 100 for A7III and sdQH (X3F) and processed in the same software (Affinity Photo), roughly matched brightness and WB, and resized to same size (54 MP). Let’s look at a small area of the test image at 200%.

ec93683c7bf148ba94f52d05d7742ee8.jpg

7243af85fe4040bdac001112d04b43d8.jpg

The difference is stark. Is it all in the Sony’s weak AA filter? I very much doubt it.

cheers

[Edit: apparently the Panasonic S5 has a 24 MP full frame sensor and has no AA filter, just like the Sigma fp, so here is its equivalent image. Confirms my point.]

0edfe23bd2b3443c88a66f906dfb6395.jpg
I agree, in these test images, the Bayer images look rubbish compared with the sdQH.

But 'in the field' real life test, there didn't appear to be much difference.

Do you have the test image for the Sigma fpL ? Only for curiosity.

--
Best, Steve
 
What I find most interesting about this comparison is that the Foveon sensor is able to get more microcontrast/sharpness from a lens than a Bayer sensor can. In the center, where the lens is sharpest, the two images look fairly similar, but on the sides and in the corners the SD Quattro H is able to better resolve the slightly unsharp parts where the resolution is not at its finest. It's like the sensor is less confused by what the detail there should be, and it really gets the best out of the lens; the Bayer sensor in the fp smooths it out. This is true in all of the processed examples I looked at on this thread. What I dislike is the general grittiness of the Quattro H image when viewing it at 100%. The tonal transitions should be smoother.
 

Keyboard shortcuts

Back
Top