sdQH & fp with the 40mm Art f/1.4 - Kitchen Table Comparison

Ok, folks. Another amateur comparison to tear apart. This time with the 40mm Art /1.4 lens.

I did wonder if my previous comparison may have been a bit muddled by the long 135mm lens and possibly getting the spot focus points a little different. It's pouring with rain here, so, this is an indoors comparison using a tripod. Much more accurate focus point picking (front of vase). I've framed the two cameras as accurately as I can by moving the tripod. Now the focus point is much closer to the camera than with my previous comparison. I've also cropped the essential bit in the centre of the frame.

I shot at f/5.6 and f/2.0.

My observation is the sdQH is better as I can easily read the small print on the back of the Marmite jar. With the fp, it's a little more difficult to read. Still, pretty close between the two cameras.

First the f/5.6 shots

sdQH

fp

Now the f/2 shots

sdQH

fp

Thanks for this comparison Steve. I find it quite interesting, and I see the SD Quattro H definitely wins in both pairs of photos. Like you say though, the two cameras make photos that look pretty similar. I think the fp will probably make for a good companion to the FFF, when it finally comes on the market, primarily for its low light (high ISO) and video capabilities.

I would just like to add, my intention of this comparison (and the previous 135mm comparison) is to compare the finished photos which I have created rather than it being a camera vs camera comparison. This is why I included the f/2 comparison to see the quality of the separation and out of focus areas.

For the record, I'm way more interested in the visual interest a photo delivers rather than technical excellence when displayed at some enormous magnification. Sigma cameras do seem to deliver interesting outputs and I do wonder if it is a lot to do with first processing the raw images in SPP.

In my other 135mm comparison, I was skooled and found out that X3F files can be processed in Affinity and this is what DpReview do for their camera tests. I don't have Affinity but I would like to compare the outputs of processing X3F files in SPP and Affinity. Maybe there is no difference.

josepmg wrote:

It looks very well and the texture of the vase is very interesting. To give an opinion, I would like to play with the RAWs if you allow us. Thanks.

Sorry, I've deleted the raw files as they're nothing I would keep.

The vase is  by Wanda Patrucco, Isola dei Pescatori, Largo Maggiore, Italy.   It has a Raku firing which develops the crackle glaze and weird and wonderful colours caused by oxidising various materials like straw and wood during the firing.   The potter never really knows what will come out of the kiln with respect the these random colours marking the known base colour.  It's very skilled work.  Here's another pot with lid

furtle wrote:

I would just like to add, my intention of this comparison (and the previous 135mm comparison) is to compare the finished photos which I have created rather than it being a camera vs camera comparison. This is why I included the f/2 comparison to see the quality of the separation and out of focus areas.

How does that make it not a camera vs camera comparison? If you want to compare photos, just use one camera.

For the record, I'm way more interested in the visual interest a photo delivers rather than technical excellence when displayed at some enormous magnification. Sigma cameras do seem to deliver interesting outputs and I do wonder if it is a lot to do with first processing the raw images in SPP.

Sure but let’s be honest, what a photo delivers is up to you, not the camera, unless there is something technically challenging about the photo that makes some cameras less suited than others. With modern cameras they generally can all can deliver the photo you want, except for these edge use cases, when one camera will beat another at what you call “technical excellence when displayed at some enormous magnification”. It might be extra high resolution, or extra good detail retention in shadows, or extra high burst speed, or extra high AF accuracy and speed and tracking.

If you are “way less interested in that”, then honestly, pick any camera and get on with making your middle use case images of vases and back walls, or supreme photographic art, it’s all the same. They are all highly capable at such.

IMO people make a beginner’s mistake in overemphasising the differences in the straight-out-of-camera (SOOC) images, viewed unmagnified on normal sized screens, because then all you are really looking at are differences in colours and contrast and brightness, maybe sharpening, which, to be honest, are all fully adjustable in software…because it is nothing more than software settings anyway. The only justification I can see for putting strong emphasis on SOOC images is for people who are never going to tweak the look in software, but just leave untouched what is, after all and in effect, the camera’s JPEG file. For those people, fair enough, obsess over the look of the software settings between cameras. For the rest of us… it’s pointless, gear-obsessed fun.

Lenses, that’s different…although I could make the same argument in some areas of lens differentiation. People still overplay the ‘lens rendering’ card too, for middle use case photography not viewed at enormous magnifications. But at least there are inherent differentiators like focus transitions that are apparent at normal viewing size.

In my other 135mm comparison, I was skooled and found out that X3F files can be processed in Affinity

Yes.

and this is what DpReview do for their camera tests.

No. It’s what I did for my tests. Before Sigma released firmware that enabled Quattro cameras to output DNG files, DPR used X3F files processed in SPP, which caused many difficulties comparing cameras and knowing what differences are due to the camera and what is due to the different software. Since the sdQH review DPR have used the DNG files in Adobe Camera Raw, hence simplifying the comparison of cameras. Unfortunately the sdQH DNG files are not quite as good as its X3F files, but DPR argued that this has little effect in practice and most users will benefit hugely from using their own standard workflow, so they went with DNG. They also provided X3F files of the test chart photo for those interested, but it doesn’t come up in the DPR comparator tool.

I don't have Affinity but I would like to compare the outputs of processing X3F files in SPP and Affinity. Maybe there is no difference.

There is quite a difference. It has led me to conclude that SPP is just as much, or more, responsible for the ‘Foveon look’ as the sensor, in terms of colour and contrast tuning. In all likelihood it has always been the case.

TN Args wrote:

furtle wrote:

I would just like to add, my intention of this comparison (and the previous 135mm comparison) is to compare the finished photos which I have created rather than it being a camera vs camera comparison. This is why I included the f/2 comparison to see the quality of the separation and out of focus areas.

How does that make it not a camera vs camera comparison? If you want to compare photos, just use one camera.

What's the point on comparing two identical photos from the same camera, same lens and same processing?  Anyhow, it's a comparison of the finished images from two different cameras with the same lens.  The final image for each camera is a melange of the sensor, the camera software, the lens, the Raw program, maybe LightRoom and of course the screen to view the image or the print.

It's not just a test comparing the sensors, which I assume could be done with some scientific instruments.

What is apparent is the two final images I produced are remarkably similar with the sdQH edging the sharpness.

For the record, I'm way more interested in the visual interest a photo delivers rather than technical excellence when displayed at some enormous magnification. Sigma cameras do seem to deliver interesting outputs and I do wonder if it is a lot to do with first processing the raw images in SPP.

Sure but let’s be honest, what a photo delivers is up to you, not the camera, unless there is something technically challenging about the photo that makes some cameras less suited than others. With modern cameras they generally can all can deliver the photo you want, except for these edge use cases, when one camera will beat another at what you call “technical excellence when displayed at some enormous magnification”. It might be extra high resolution, or extra good detail retention in shadows, or extra high burst speed, or extra high AF accuracy and speed and tracking.

If you are “way less interested in that”, then honestly, pick any camera and get on with making your middle use case images of vases and back walls, or supreme photographic art, it’s all the same. They are all highly capable at such.

IMO people make a beginner’s mistake in overemphasising the differences in the straight-out-of-camera (SOOC) images, viewed unmagnified on normal sized screens, because then all you are really looking at are differences in colours and contrast and brightness, maybe sharpening, which, to be honest, are all fully adjustable in software…because it is nothing more than software settings anyway. The only justification I can see for putting strong emphasis on SOOC images is for people who are never going to tweak the look in software, but just leave untouched what is, after all and in effect, the camera’s JPEG file. For those people, fair enough, obsess over the look of the software settings between cameras. For the rest of us… it’s pointless, gear-obsessed fun.

Lenses, that’s different…although I could make the same argument in some areas of lens differentiation. People still overplay the ‘lens rendering’ card too, for middle use case photography not viewed at enormous magnifications. But at least there are inherent differentiators like focus transitions that are apparent at normal viewing size.

In my other 135mm comparison, I was skooled and found out that X3F files can be processed in Affinity

Yes.

and this is what DpReview do for their camera tests.

No. It’s what I did for my tests. Before Sigma released firmware that enabled Quattro cameras to output DNG files, DPR used X3F files processed in SPP, which caused many difficulties comparing cameras and knowing what differences are due to the camera and what is due to the different software. Since the sdQH review DPR have used the DNG files in Adobe Camera Raw, hence simplifying the comparison of cameras. Unfortunately the sdQH DNG files are not quite as good as its X3F files, but DPR argued that this has little effect in practice and most users will benefit hugely from using their own standard workflow, so they went with DNG. They also provided X3F files of the test chart photo for those interested, but it doesn’t come up in the DPR comparator tool.

You've confused me.  Earlier I asked if Affinity can develop X3F files and you said yes.   Now you say the DPR did initially use SPP to process the X3F files but that caused "difficulties".  If Affinity can process X3F files, why didn't the DPR testers use Affinity instead of SPP?

Anyway, it seems the DPR testers were given a lifeline with the DNG outputs from the Quattro and these are processed in Affinity and your example images of their tests show the sdHQ is visibly  better (at 200%) than  the images from the fp; and that's using the Quattro's DNG output which is "not as quite good" as the X3F output.

If DPR had processed the Quattro's X3F files in SPP, the difference between the Quattro shots and the fp shots should have been even more obvious in favour of the Quattro.   But in my two examples, that is simple not the case.  It is a close run thing between the two images.   It should be a home run for the Quattro but it's not.  I could do a third test with another Art lens but I'm not sure I'll end up with a different result.

I do wonder if Sigma have boxed themselves into a corner with the fp because the upcoming FFF's image quality  will need to be "night and day " better than the Quattro and by reasoning, also that from the fp / fpL.  Right now, in my real life comparisons, that's not the case.

I don't have Affinity but I would like to compare the outputs of processing X3F files in SPP and Affinity. Maybe there is no difference.

There is quite a difference. It has led me to conclude that SPP is just as much, or more, responsible for the ‘Foveon look’ as the sensor, in terms of colour and contrast tuning. In all likelihood it has always been the case.

The 'Foveon look'.  Umm.  Since my fp's final image outputs look very close to my Quattro final images, I'm inclined to think it's the 'SPP look'.  Sigma undoubtably have some secret sauce in their SPP programme and it appears to work it's magic on X3F files and the fp's DNG files.

furtle wrote:

TN Args wrote:

furtle wrote:

I would just like to add, my intention of this comparison (and the previous 135mm comparison) is to compare the finished photos which I have created rather than it being a camera vs camera comparison. This is why I included the f/2 comparison to see the quality of the separation and out of focus areas.

How does that make it not a camera vs camera comparison? If you want to compare photos, just use one camera.

What's the point on comparing two identical photos from the same camera, same lens and same processing? Anyhow, it's a comparison of the finished images from two different cameras with the same lens. The final image for each camera is a melange of the sensor, the camera software, the lens, the Raw program, maybe LightRoom and of course the screen to view the image or the print.

It's not just a test comparing the sensors, which I assume could be done with some scientific instruments.

What is apparent is the two final images I produced are remarkably similar with the sdQH edging the sharpness.

For the record, I'm way more interested in the visual interest a photo delivers rather than technical excellence when displayed at some enormous magnification. Sigma cameras do seem to deliver interesting outputs and I do wonder if it is a lot to do with first processing the raw images in SPP.

Sure but let’s be honest, what a photo delivers is up to you, not the camera, unless there is something technically challenging about the photo that makes some cameras less suited than others. With modern cameras they generally can all can deliver the photo you want, except for these edge use cases, when one camera will beat another at what you call “technical excellence when displayed at some enormous magnification”. It might be extra high resolution, or extra good detail retention in shadows, or extra high burst speed, or extra high AF accuracy and speed and tracking.

If you are “way less interested in that”, then honestly, pick any camera and get on with making your middle use case images of vases and back walls, or supreme photographic art, it’s all the same. They are all highly capable at such.

IMO people make a beginner’s mistake in overemphasising the differences in the straight-out-of-camera (SOOC) images, viewed unmagnified on normal sized screens, because then all you are really looking at are differences in colours and contrast and brightness, maybe sharpening, which, to be honest, are all fully adjustable in software…because it is nothing more than software settings anyway. The only justification I can see for putting strong emphasis on SOOC images is for people who are never going to tweak the look in software, but just leave untouched what is, after all and in effect, the camera’s JPEG file. For those people, fair enough, obsess over the look of the software settings between cameras. For the rest of us… it’s pointless, gear-obsessed fun.

Lenses, that’s different…although I could make the same argument in some areas of lens differentiation. People still overplay the ‘lens rendering’ card too, for middle use case photography not viewed at enormous magnifications. But at least there are inherent differentiators like focus transitions that are apparent at normal viewing size.

In my other 135mm comparison, I was skooled and found out that X3F files can be processed in Affinity

Yes.

and this is what DpReview do for their camera tests.

No. It’s what I did for my tests. Before Sigma released firmware that enabled Quattro cameras to output DNG files, DPR used X3F files processed in SPP, which caused many difficulties comparing cameras and knowing what differences are due to the camera and what is due to the different software. Since the sdQH review DPR have used the DNG files in Adobe Camera Raw, hence simplifying the comparison of cameras. Unfortunately the sdQH DNG files are not quite as good as its X3F files, but DPR argued that this has little effect in practice and most users will benefit hugely from using their own standard workflow, so they went with DNG. They also provided X3F files of the test chart photo for those interested, but it doesn’t come up in the DPR comparator tool.

You've confused me. Earlier I asked if Affinity can develop X3F files and you said yes. Now you say the DPR did initially use SPP to process the X3F files but that caused "difficulties". If Affinity can process X3F files, why didn't the DPR testers use Affinity instead of SPP?

Anyway, it seems the DPR testers were given a lifeline with the DNG outputs from the Quattro and these are processed in Affinity and your example images of their tests show the sdHQ is visibly better (at 200%) than the images from the fp; and that's using the Quattro's DNG output which is "not as quite good" as the X3F output.

If DPR had processed the Quattro's X3F files in SPP, the difference between the Quattro shots and the fp shots should have been even more obvious in favour of the Quattro. But in my two examples, that is simple not the case. It is a close run thing between the two images. It should be a home run for the Quattro but it's not. I could do a third test with another Art lens but I'm not sure I'll end up with a different result.

I do wonder if Sigma have boxed themselves into a corner with the fp because the upcoming FFF's image quality will need to be "night and day " better than the Quattro and by reasoning, also that from the fp / fpL. Right now, in my real life comparisons, that's not the case.

Well, considering the 20 MP per layer resolution, I don't think we'll see much difference between the SD Quattro H and the FFF. Still, I think the new sensor and camera will be more useful, because I think the camera will operate faster, offer better image quality from the same full-frame lenses, and offer a wider view with wide-angle lenses. I think it will have some of that Merrill detail in distant trees and be able to see through fog and mist better than the Quattros can too. I think it will probably offer a little more dynamic range than the Foveon sensors of the past too.

I guess we'll just have to wait and see though.

🤔

I don't have Affinity but I would like to compare the outputs of processing X3F files in SPP and Affinity. Maybe there is no difference.

There is quite a difference. It has led me to conclude that SPP is just as much, or more, responsible for the ‘Foveon look’ as the sensor, in terms of colour and contrast tuning. In all likelihood it has always been the case.

The 'Foveon look'. Umm. Since my fp's final image outputs look very close to my Quattro final images, I'm inclined to think it's the 'SPP look'. Sigma undoubtably have some secret sauce in their SPP programme and it appears to work it's magic on X3F files and the fp's DNG files.

-- hide signature --

Best, Steve

josepmg wrote:

furtle wrote:

If DPR had processed the Quattro's X3F files in SPP, the difference between the Quattro shots and the fp shots should have been even more obvious in favour of the Quattro. But in my two examples, that is simple not the case. It is a close run thing between the two images. It should be a home run for the Quattro but it's not. I could do a third test with another Art lens but I'm not sure I'll end up with a different result.ç

If you mask foveon's virtues with harsh processing, you'll barely notice the difference. To see them, a fine development must be carried out in SPP and the final processing in specialized software.

I like my OOC jpegs from my SD Quattro H, but it took me a while to figure out what settings to use in the camera.

josepmg wrote:

furtle wrote:

If DPR had processed the Quattro's X3F files in SPP, the difference between the Quattro shots and the fp shots should have been even more obvious in favour of the Quattro. But in my two examples, that is simple not the case. It is a close run thing between the two images. It should be a home run for the Quattro but it's not. I could do a third test with another Art lens but I'm not sure I'll end up with a different result.ç

If you mask foveon's virtues with harsh processing, you'll barely notice the difference. To see them, a fine development must be carried out in SPP and the final processing in specialized software.

I process the  X3F Quattro files in SPP.  Pretty much all the settings are neutral / middle except maybe the exposure may need tweeking.  I'm not sure this is considered harsh processing.  The X3F becomes a tiff file with SPP and that tiff file goes to Light Room to convert to a jpeg.  Again my LightRoom settings are all pretty much neutral.  Maybe exposure and blacks/whites will get a bit of adjustment as will cropping. if needed.   Maybe Light Room isn't specialised software.

Maybe I'm wrecking the Quattro images in some way?

furtle wrote:

josepmg wrote:

furtle wrote:

If DPR had processed the Quattro's X3F files in SPP, the difference between the Quattro shots and the fp shots should have been even more obvious in favour of the Quattro. But in my two examples, that is simple not the case. It is a close run thing between the two images. It should be a home run for the Quattro but it's not. I could do a third test with another Art lens but I'm not sure I'll end up with a different result.ç

If you mask foveon's virtues with harsh processing, you'll barely notice the difference. To see them, a fine development must be carried out in SPP and the final processing in specialized software.

I process the X3F Quattro files in SPP. Pretty much all the settings are neutral / middle except maybe the exposure may need tweeking. I'm not sure this is considered harsh processing. The X3F becomes a tiff file with SPP and that tiff file goes to Light Room to convert to a jpeg. Again my LightRoom settings are all pretty much neutral. Maybe exposure and blacks/whites will get a bit of adjustment as will cropping. if needed. Maybe Light Room isn't specialised software.

Maybe I'm wrecking the Quattro images in some way?

You deffinitely are "wreckibg" the detail from your SD Quattro H if you use the middle noise reduction setting Steve.

-- hide signature --

Best, Steve

Scottelly wrote:

furtle wrote:

josepmg wrote:

furtle wrote:

If DPR had processed the Quattro's X3F files in SPP, the difference between the Quattro shots and the fp shots should have been even more obvious in favour of the Quattro.

If you mask foveon's virtues with harsh processing, you'll barely notice the difference. To see them, a fine development must be carried out in SPP and the final processing in specialized software.

Maybe I'm wrecking the Quattro images in some way?

You deffinitely are "wreckibg" the detail from your SD Quattro H if you use the middle noise reduction setting Steve.

Interesting the apparent difference in "detail" between Quattro H/SPP6 and SD1M/SPP5. I left the noise reduction in the middle for all my Merrill shots with no deleterious effect like "wrecking".

Scottelly wrote:

furtle wrote:

josepmg wrote:

furtle wrote:

If DPR had processed the Quattro's X3F files in SPP, the difference between the Quattro shots and the fp shots should have been even more obvious in favour of the Quattro. But in my two examples, that is simple not the case. It is a close run thing between the two images. It should be a home run for the Quattro but it's not. I could do a third test with another Art lens but I'm not sure I'll end up with a different result.ç

If you mask foveon's virtues with harsh processing, you'll barely notice the difference. To see them, a fine development must be carried out in SPP and the final processing in specialized software.

I process the X3F Quattro files in SPP. Pretty much all the settings are neutral / middle except maybe the exposure may need tweeking. I'm not sure this is considered harsh processing. The X3F becomes a tiff file with SPP and that tiff file goes to Light Room to convert to a jpeg. Again my LightRoom settings are all pretty much neutral. Maybe exposure and blacks/whites will get a bit of adjustment as will cropping. if needed. Maybe Light Room isn't specialised software.

Maybe I'm wrecking the Quattro images in some way?

You deffinitely are "wreckibg" the detail from your SD Quattro H if you use the middle noise reduction setting Steve.

I put the Detail button in the middle and also the chroma and luminance buttons in the middle.  Sharpness is knocked back to - 1.0 but I re sharpen selectively in LR.

I tend to use standard rather than portrait mode.

What would you suggest I try?

Remember, I first bought a dp0Q when they produced horrible noisy and blotchy images with plenty of highlights clipping.  Subsequent updates to firmware and SPP improved the Quattro’s images immensely.

I’m happy to try other settings because it ain’t right my sdQH and fp images look quite similar.

Scottelly wrote:

I think it will have some of that Merrill detail in distant trees and be able to see through fog and mist better than the Quattros can too.

Looks like I've missed some prior discussion, Scott.

Is there a problem with the Quattros' rendition of foggy and misty images?
--
what you got is not what you saw ...

josepmg wrote:

furtle wrote:

josepmg wrote:

furtle wrote:

If DPR had processed the Quattro's X3F files in SPP, the difference between the Quattro shots and the fp shots should have been even more obvious in favour of the Quattro. But in my two examples, that is simple not the case. It is a close run thing between the two images. It should be a home run for the Quattro but it's not. I could do a third test with another Art lens but I'm not sure I'll end up with a different result.ç

If you mask foveon's virtues with harsh processing, you'll barely notice the difference. To see them, a fine development must be carried out in SPP and the final processing in specialized software.

I process the X3F Quattro files in SPP. Pretty much all the settings are neutral / middle except maybe the exposure may need tweeking. I'm not sure this is considered harsh processing. The X3F becomes a tiff file with SPP and that tiff file goes to Light Room to convert to a jpeg. Again my LightRoom settings are all pretty much neutral. Maybe exposure and blacks/whites will get a bit of adjustment as will cropping. if needed. Maybe Light Room isn't specialised software.

Maybe I'm wrecking the Quattro images in some way?

I do not criticize your processing as such. I only comment that the differences would be more evident with a more delicate development. This is my opinion, which obviously does not have to coincide with yours.
If you compare the processing that I have uploaded here

https://www.dpreview.com/forums/thread/4702495?page=7

with yours, I hope you understand what I mean by "harsh".

I see from your link to my other 135mm comparison, you would put the Detail button at the far right which is the most ‘crispy’ and then use it Topaz Denoise.  I’ll try this as I seem to have Topaz Denoise on my computer.  No doubt I bought it when my dp0Q was producing horrid, noisy images.

I would only comment that putting the detail button at the far right, most crispy point may be considered to be quite a harsh move.

furtle wrote:

josepmg wrote:

furtle wrote:

josepmg wrote:

furtle wrote:

If DPR had processed the Quattro's X3F files in SPP, the difference between the Quattro shots and the fp shots should have been even more obvious in favour of the Quattro. But in my two examples, that is simple not the case. It is a close run thing between the two images. It should be a home run for the Quattro but it's not. I could do a third test with another Art lens but I'm not sure I'll end up with a different result.ç

If you mask foveon's virtues with harsh processing, you'll barely notice the difference. To see them, a fine development must be carried out in SPP and the final processing in specialized software.

I process the X3F Quattro files in SPP. Pretty much all the settings are neutral / middle except maybe the exposure may need tweeking. I'm not sure this is considered harsh processing. The X3F becomes a tiff file with SPP and that tiff file goes to Light Room to convert to a jpeg. Again my LightRoom settings are all pretty much neutral. Maybe exposure and blacks/whites will get a bit of adjustment as will cropping. if needed. Maybe Light Room isn't specialised software.

Maybe I'm wrecking the Quattro images in some way?

I do not criticize your processing as such. I only comment that the differences would be more evident with a more delicate development. This is my opinion, which obviously does not have to coincide with yours.
If you compare the processing that I have uploaded here

https://www.dpreview.com/forums/thread/4702495?page=7

with yours, I hope you understand what I mean by "harsh".

I see from your link to my other 135mm comparison, you would put the Detail button at the far right which is the most ‘crispy’ and then use it Topaz Denoise. I’ll try this as I seem to have Topaz Denoise on my computer. No doubt I bought it when my dp0Q was producing horrid, noisy images.

These kind of actions tend to bother me, especially when the adjustments are global - affecting the whole image. Another one is "I always set sharpness to -2.0 and sharpen it later in X".

As if all scenes require exactly the same processing and as if all shot capture qualities (focus, etc.) are identical.

I would only comment that putting the detail button at the far right, most crispy point may be considered to be quite a harsh move.

Good one