Quick Comparison – Leica Monochrom, Sigma DP2 Merrill and Hasselblad 39CF by Michael Ma

January 14, 2013 Crazy Comparisons, follow ups, Guest Post, Leica, Medium Format, SIGMA 104

Quick Comparison  – Leica Monochrom, Sigma DP2 Merrill and Hasselblad 39CF

by Michael Ma

Hi Steve:

Huge fan of your site. I am lucky to own two pieces of equipment that you have reviewed, so I thought I’d contribute. My Leica Monochrome just arrive today and I decided to give it a spin in terms of image quality. Both the Leica and the Merrill DP2 are reviewed in detailed on your website and they are renowned for their image quality. Since I’m lucky enough to have a loaner Hasselblad with the CF39 digital back on hand, I’ve decided to do a quick and dirty IQ comparison using the Hassey as bench mark.

Conditions:

Dim room light

1.7 meters to subject

All images had gone through only contrast adjustment, no sharpening

On tripod

–

Hasselblad 39CF with 80mm CFE lens F2.8, F5.6 1/2, 1/4S

The Hasselblad yielded a pleasing overall image in terms of tone and rendering. With 39 mega pixels 49x36mm sensor, we don’t expect anything less. The 100% crop shows that even when wide open, the CFE lens is sharp and shows nice contrast. Note that the physical size of the Hasselblad image is almost twice as large than the Leica and Sigma.

Hassey picture: 

8371384287_c7dc8e528f_k-2

–

Hassey crop (click image for full size cdrop)

–

Next up is the Leica Monochrome with Summilux 1.4 ASPH latest version F2.8, F5.6

WOW, the Leica is sharp! I don’t have a M9 to test the comparison between the mono sensor and the regular CCD sensor, but the 100% crop looks sharper than the Hasselblad shot and you can see the details in the canvas texture. The image was over blown in exposure but the details are still nicely preserved. Very impressive!

Leica picture

Leica Crop – (Click image for full size crop)

–

Finally comes the Sigma DP2 Merrill.

I couldn’t believe my eyes when I saw the pictures. It is clearly the most rich and detailed of all three. The photo was shot with the lens wide open at F2.8. Astounding details and color. Now look at that 100% crop. The texture of each brush stroke is so vivid. Beats the Hasselblad hands down.

DP2 Merrill

–

DP2 Crop

Conclusion? Well this is a very clumsy test. But besides the poor testing conditions I think there’s a story to be told here. All three are great camera systems. The Hasselblad is older and the lens probably could have used with more stopping down. But this is also a 9000 dollar set up (used price). The Monochrome setup is 12K all in (when bought new). The Leica lens is incredibly sharp and the Monochrome retains so much details in the shadows. But the ultimate winner here is the Sigma DP2. At a tiny fraction of the price of either the Hassey or the Leica, it delivers the best results in color, details, and contrast.

Michael Ma

Related Articles

104 Comments

  1. Two late notes: The blue channel on the Foveon is very clean. You can get very good results in monochrome at 1600 ISO if you use blue only. Adding a Hoodman Finder is a vast improvement. The camera takes practice, but is worth it, I think.

  2. The Foveon is the better thing that exists. But the developing, in Sigma Photo Pro, turns out to be difficult because it is necessary to have big knowledge and vision of the color; and this it does not have the majority of the people.

  3. Here is the problem. Sigma should learn how to make sexy cameras like Sony is now doing with the A7R, and slot that Foveon sensor into a more useful camera body. They will sell shed loads of it that way. Also, develop a full frame Foveon for goodness sake. Just selling “little compact cameras” that can keep up with the big boys is NOT good enough!

    And Hasselblad and Phase One should realize that maybe, their cameras are just not as “good value” for the money as they want us all to think. Those prices should be brought back down to earth, with a thud!

  4. Late in joining the party but what the hell.

    I can’t help but notice that the CFV-39 result is a bit soft. I’m not sure if its a focusing issue but I want to know how the shot was taken. The CFV-39 is very demanding in terms of shot discipline – normally, one needs to shoot it in 1/3x the focal length if hand held. Even on a tripod, there are steps to be taken to get the maximum acuity out of the lens+sensor combination (mirror lock-up, use of cable release).

    Though I applaud the effort to do this test, the CFV-39 should have been sharp as from what I have seen and experienced. Just my 2 cents.

  5. I bought a Porsche 911 , it was useless, I tried to fit my wardrobe in it and it wouldnt fit! The designers at Porsche obviously don’t know what they are doing!

  6. Here is a good posting of kb2zuz showing how much CFA softens the image with a Hasselblad H4D-50MS.
    http://www.dpreview.com/forums/post/51617517
    Add the blur of the AA filter of an normal Bayer DSLR and you understand the difference in image detail compared to a multilayer sensor layout like Foveon.
    The Bayer matrix sensor patent is from 1975. Sony and others have Foveon like multilayer sensor patents. I hope there is enough pressure from the market to make them into products.

    These little dp2m srgb jpgs don’t show the image quality of a dp2m. I don’t feel restricted by a DP2M, but I used to use Fuji Velvia with Medium Format cameras (Mamiya 7II and Fuji GSW690) in the past. 🙂

  7. I made a small mistake in what I said. I meant to say that if the Hasselblad crop had been REDUCED, that would make it look smoother than it is. But the point remains: Either the Hasselblad crop or the Sigma crop was resized, so this isn’t a valid comparison.

  8. I’ve had the original dp2 for years and it’s my favorite camera. Just trying to accept the $1000 price tag for the Merrill and buy it. Listen, as you have shown, the Sigma is one of the finest imaging devices available. Also, I found out that the slow loading of it’s images can be sped up with very fast SD cards. Also, you can continue shooting as the images load for up to 6 images, that makes it more usable than my Dp2.

  9. No one seems to have noticed that the 100% crops from the Hasselblad and the Sigma are both the same size, yet the Hasselblad has a 39 MP sensor, and the Sigma has a 15 MP sensor. One of the crops has been resized. If the Hasselblad crop was enlarged, that would make it look more washed out.

    Judging from what I’m seeing, this can’t be a valid comparison. The lighting must have been different for the two color shots. Foveon and Bayer technology don’t produce images that are as different as we are seeing here.

  10. The Sigma results look awesome to me.

    The only real problem with Sigma as a camera maker is lack of a proper lens mount. They should probably team up (or just buy Pentax) and use the standard K mount ( rather then using their dumb proprietary variant of the K Mount ) then it would be fun or Just make cameras in every other lens mount.

    I remember in the 80’s every other smaller camera makers like Vivitar, Kalimar, Chinon, Ricoh and Cosina had the pentax K mount. Why cant we have that again with DSLRs?

    Sigma makes some great lenses now and has some good cameras but people need a system and should feel backed by a brand with a heritage. The resale value of a great pentax lens is still very high but the used SA mount lenses sell for next to nothing.

    I hope someone in Sigma is reading this. Pls pls pls…come up with a “Real” lens mount and we will buy!

  11. Before you can compare those Hasselblad and Sigma, did anyone here except me tried to match the 2 files the closest possible in all aspects (even in a simple program as i’ve done) and compare them after. The opinion you would have over the sigma winning hands down at first look would probably be different. The hassel clearly have a better and natural rendering. Also it looks a lot more 3D like, normal for such a sensor size (Sigma can cheat our eyes because of the big amount of details and huge contrast but at final it will never give the same aspect as a bigger sensor). The sigma gives a less clean and realistic result at final but for its price the Sigma clearly an interesting camera on specific uses.
    Sorry, i am not used to post pics online but i would agree to share them. If someone is able to post them in this page, please give me your mail tks.

  12. To sum it up from a technical point of view :

    OPTICAL
    ——-
    A. The foveon has excellent and ‘high fidelity’ optical qualities with true per-pixel color capture needing no interpolation and no low-pass filter which is inherently a ‘high fidelity’ concept.

    vs.

    B. Bayer has not so excellent optical qualities with only R or V or B measurement per pixel and then ‘guesswork’ image interpolation , plus low pass filtering, which is inherently a ‘low fidelity’ concept.

    ELECTRICAL
    ———-
    C. The foveon has difficult electric properties with problematic readout due to weak low light capture properties (hence the weak ISO performance). So although the optical properties (A.) are very good, its electrical implementation remains problematic.

    vs.

    D. Bayer has very good electrical capabilities, with very good readout (hence the high ISO capabilities of Bayer cams)

    USABILITY
    ———
    Here Sigma is just not a big enough player to keep up with Canon / Nikons R+D capabilities, their image processing is slow (I guess it’s expensive to develop the image processing chips and I guess that the big players can embrace much more performant technologies)

    …

    One of the points where Sigma messed it up is that they didn’t produce the SD1 as a full-frame camera (with same pixel count) resulting in much larger photo sites which would have helped to create a much better electric implementation of the technology, further reducing noise and improving signal readout, and thus providing rock solid dynamic range. Coupled with true 16 bit processing this would have created a slow but incredibly high-end camera.

    Also I don’t understand why they don’t port the technology to Medium Format which would be an ideal ‘natural habitat’ for the Foveon technology as the shooting speeds are slow, lighting conditions are good, size is of no importance and image quality is paramount. Again the fact to have even larger photosites on a even larger sensor would have helped to create a even better electrical implementation of the FOVEON topology, resulting in absolutely stellar performance.

    The main problem with FOVEON is Sigma who consider their cameras as a vehicle for selling more lenses instead of really making the best of the technology independently from Lens platform, this is further crippled by their weak R+D and production capabilities.

  13. It may be that Sigma products used to have quality issues. So did Honda, Toyota … lots of manufacturers now identified with high quality.

    If Sigma continues to make strong products like the Merrill, the tide of public opinion will change.

    I think it’s also important to remember that when the original DP1 came out, there was nothing else in the marketplace like it: the DP1 predates the introduction of m4/3, NEX, X100 – all of them. Was it rough around the edges? Sure. But if you wanted the highest image quality in a compact body, there were few choices at that time.

  14. I saw too many bad things about DP2, such as serious green and magenta cast, the horrible skin color, etc. It looks like this image also has some yellow cast, as well…

  15. I just shot a commercial visual with a Phase One 80MP camera. I’m working as art director so I ‘prototype’ the visual with my sigma SD1, both subject and setup where very similar from ‘prototype’ to final shot.

    While the SD1’s resolution is astounding … however it’s dynamic range / color depth is not the same as the Phase Ones … these commercial visuals get heavily retouched as well as loads of contrast / sharpening to make them ‘pop’ in a very competitive environment. The Sigma SD1 file just fries out faster, it can’t take heavy retouching as well as Phase One (or Hassy for that matter) files … I think that Sigma initially planned to market the thing as a Medium Format equivalent … as a professional user I think that they do not really hold up to that challenge though.

    • I can understand that. The thing that’s impressive about the Merrill is that it is a compact camera that costs $1000. The fact that it is even being compared to these other two cameras proves that it is a very special compact small format camera.

      • Totally agree, for the price their cameras deliver amazing quality, very high sharpness and optical purity. It’s a ‘no brainer’ if the cameras correspond to your shooting style.

        To David : Yes foveon sensor have certain issues with the green/magenta cast, seems to be one of the tradeoffs of the technology. Although Sigma always that Foveon is ‘the’ technology which makes sense, there are numerous tradeoffs due to weak penetration of light through the silicone filters which results in very low readout currents and thus high levels of noise in low light. Also the sensor samples all RGB values at 100% whereas the human has more of a 20%R 60%G 20%B color perception, so the sensors output needs to be adapted to these values which is also a somewhat tortured and not exactly ‘high-fidelity’ process, putting some strain on the files. I guess it’s partially for these reasons/tradeoffs that the mainstream industry didn’t go for the Foveon technology as you will never get rid of them, whereas the Bayer technology will provide higher and higher resolution figures every year which will eventually offset issues related to Low-Pass filters and so on …

        However the Sigma cameras are still an amazing tool providing unmatched visual sharpness and purity. If you’re mainly shooting in good lighting conditions, speed is not an issue and if you don’t need to overly tweak and push the files they provide fantastic out of camera results which need no sharpening or processing. Also for people shooting BW they are a fantastic option, due to the sharpness and accurate per pixel value. And the green/magenta is not that much of an issue (from my experience) except if shooting into direct sunlight I never had any problems with these.

        • To sum it up from a technical point of view :

          OPTICAL
          ——-
          A. The foveon has excellent and ‘high fidelity’ optical qualities with true per-pixel color capture needing no interpolation and no low-pass filter which is inherently a ‘high fidelity’ concept.

          vs.

          B. Bayer has not so excellent optical qualities with only R or V or B measurement per pixel and then ‘guesswork’ image interpolation , plus low pass filtering, which is inherently a ‘low fidelity’ concept.

          ELECTRICAL
          ———-
          C. The foveon has difficult electric properties with problematic readout due to weak low light capture properties (hence the weak ISO performance). So although the optical properties (A.) are very good, its electrical implementation remains problematic.

          vs.

          D. Bayer has very good electrical capabilities, with very good readout (hence the high ISO capabilities of Bayer cams)

          USABILITY
          ———
          Here Sigma is just not a big enough player to keep up with Canon / Nikons R+D capabilities, their image processing is slow (I guess it’s expensive to develop the image processing chips and I guess that the big players can embrace much more performant technologies)

          …

          One of the points where Sigma messed it up is that they didn’t produce the SD1 as a full-frame camera (with same pixel count) resulting in much larger photo sites which would have helped to create a much better electric implementation of the technology, further reducing noise and improving signal readout, and thus providing rock solid dynamic range. Coupled with true 16 bit processing this would have created a slow but incredibly high-end camera.

          Also I don’t understand why they don’t port the technology to Medium Format which would be an ideal ‘natural habitat’ for the Foveon technology as the shooting speeds are slow, lighting conditions are good, size is of no importance and image quality is paramount. Again the fact to have even larger photosites on a even larger sensor would have helped to create a even better electrical implementation of the FOVEON topology, resulting in absolutely stellar performance.

          The main problem with FOVEON is Sigma who consider their cameras as a vehicle for selling more lenses instead of really making the best of the technology independently from Lens platform, this is further crippled by their weak R+D and production capabilities.

  16. For all of those bashing DP2 colors: how can you judge color in these shots without seeing the original? Just by comparison? Sigma is way more detailed, and that’s obvious. Also, for some, that painting doesn’t seem a real old painting. It would be interesting to know how old it is to judge color without seeing the original. Hasselblad are for film, in m opinion. DP2, for the brave and patient.

    • Spot on, as I said earlier one really needs to see the original subject to make an informed judgement. But its clear to anyone (but the very blinkered) that the DP2 has produced the best image in terms of detail captured. The sensor in the Merrills give the best “film” experience since film :)))) – which I still use 80% of the time. You certainly have to have patience with the DP1/2/3 Merrill cameras but it is paid off and then some by the results. For point and shoot I would imagine the new Fuji X100S is going to be very hard to beat.

  17. Is it just me, or does the Sigma remind you of the old GAF 500 slide film?
    They could get Foveon much better… But first they have to figure out a new way to “dope” the silicon layers to make it better.

  20. My dream Leica M, with a Foveon sensor, and USB type connectors and bus between lens and body, and the hotshoe so we can have the EXIF/distance data and control the flash just like a Nikon.

    Maybe also do an M without the rangefinder/viewfinder so it’s like the Sony R1 ( but with interchangeable lenses of course ). That should make it a bit less expensive. Now wouldn’t that be just peachy ?

  21. Interesting comparison between 3 great cameras. It’s a shame that some people feel the need to take sides, it reminds me of the computer scene, there would be flame wars, insults etc. just because someone liked AMD processors, not Intel, or ATI graphics cards not Nvidia.

    Manufacturers are not on your side, why are you on their’s? Most cameras have their pros and cons, and practically all of them are better at taking photos than we are. Just enjoy what you have, and let others enjoy what they have.

      • Yup, but the wise companies know that getting people on their side helps build brand loyalty and hence sales, basic stuff, ask Apple 🙂

        • I hear what you’re saying but what do these companies really do to earn that brand loyalty?

          I’m still waiting for my Christmas card from Apple 😉

        • Certainly a company needs to make acceptable products to build brand loyalty, but they don’t need to be anything special, you’ve only got to look at 99% of products to know they are unremarkable.

          After a while I think brand loyalty becomes like supporting a football team, does not matter how many times they let you down, how many racist players they have, or the fact that not a single player even comes from your area. Once people pick a side, that’s it.

          Look at Apple and Microsoft, both have made their fair share of terrible products, and their fair share of very good products, but the haters will hate and the fans will cheer, I’m not sure if the actual merit of products has much to do with it.

          • Spot on, people make a choice spend their money and then often will defend their choice no matter if the product is good or not. Once someone has invested time and money into a companies mini world it is very hard for them to leave to another companies. This is how Nixon and Canon did business for years, once people invested in their lens system etc it was hard for people to change sides on a whim. Apple are now the Kings of duping people, but others are not far behind them!

    • Hard to say without seeing the original “painting” in the flesh, but the image certainly has far more detail than the washed out Hassey image, the Leica image looks good(ish) for detail but obviously hard to comment on the colours :))))

  23. Reading the comments it’s hilarious, some get upset this little upstart creams the more expensive aspirational brands,the DP2 for certain subjects in certain condition produces the best IQ currently available ,certainly in the sub £10k bracket! So many articles on the internet confirm this – fact, so for a camera costing £750 that is astounding. Yes it has it’s faults but then so does every camera, it’s like a primadonna genius footballer, treat it right and it will win games for you, wrong and you are in trouble:)
    The DP2 is certainly not a “jack of all trades” but it is master of some, and if those “some” are the ones that float your boat you should give one a try, if not move on……

  24. The Camera Store video, and some commenters here, sell the Sigma short in terms of its everyday real-world usefulness.

    I took the DP2 Merrill on vacation to Panama recently and you can see some of the results here:
    http://www.flickr.com/photos/alistairhenning/sets/72157632463296789/

    With recent firmware updates the autofocus is easily on par with most mirrorless cameras in the marketplace right now. Many of the shots in this Flickr set were taken from a moving taxi, or while riding a horse!

    A true testament to the sheer resolution the DP Merrills produce is just how much it’s possible to crop files and still end up with print-quality results.

    In the past 3-4 years I’ve owned m4/3, NEX, and Fuji X-Series, having given up on SLRs as too unwieldy for my style of photography. For my real-world use (professional studio, event, and travel photography) the Sigma Merrills right now feel like the best balance of image quality and simplicity/usability.

    • I agree. DSLR’s are dinosaurs to lug around. I have a DP1m and it is magnificent. One gets a medium format image in a small camera that one can easily carry all day. There is magic in the DPm cameras.

  26. Also, I will do another compairson tomorrow or maybe the day after during day light with the lens stopped down. The Hassey is not designed to be shot wide open under dim light, as I pointed out in the original post. We’ll see, but my guess is that the DP2 will produce very similar high quality results as will the Mono.

    • Tanks for all the works you have done to produce this comparison. I am a die hard M9 fan. There is certain crispness in the pictures that always draw me into them. Then came along the DP2M. It will be hard to believe how a lowly (relative to the M9) $999 beats out the giant. Sometimes I play with the pictures by blowing to 300%, not really a realistically useful magnification. But nonetheless it is a joy to see that the details still hold. In all practicality I still enjoy using the M9 due to the lens quality and the whole RF handling. But with this Sigma thing in my pocket I know I have a very solid tiny little machine that always back me up with quality.
      PS. I am still waiting for some Hassy digital back to drop in price.

  27. I’m glad this post generated so much heated debate!!! To answer some of the questions above:

    Hassey has better color fidelity? : Yes, under the ambient light of the time, the hassey picture looked most like the actual color of the paitning under such lighting. HOWEVER, the DP2 picture actually resembles the painting under normal daylight. I found the color to be a bit over saturated but hey, it was surely striking.

    Canvas texture is the least important in art reporduction, I agree. However the goal of the test was not to reproduce the painting. The texture of the canvas was vividly captured by the DP2.

    Hassey would have done a lot better If I stopped the CFE 80mm down further to F8. But under the lighting condition that was impossible. It’s not fair as the Hassey was not designed for low available light photography. The focus was spot on and the mirror was locked up. Just that the lens is a bit soft under comarison at F2.8

    The DP2 is the best pocket camera for me. I have gone ahead and ordered the DP1. I had planned to buy a used M9 or the new M for color (since I got the mono and no way i’m selling that), but now I will just buy the DP1. Yes it has slow auto focus at low light but I don’t use it for night street shooting. In day time it’s absolutely fine and the buffer is sufficient. It makes you think more before just machine gunning the shutter. Again it might be slower but is it slower than a Leica (where you need to manual focus and the buffer is equally terrible). It delivers better IQ than the M 9 for sure.

  28. So funny to read al the haters comments about the DP2.
    I’m sure they NEVER hold one or try to use before vomiting such bitterness.
    I bow down in front of these “image specialists” like Jonny or Mike Watkins that can judge a camera on ONE image. Really. Bravo guys… Capa and Guetty Images need first-rate photographers like you.
    James Nachtwey can now retire peacefully…

    I have a DP2M and my trusty M9P, two different user experiences, two different feelings, but when it comes to sharpness and overall IQ, the M9P blushes like a redlight…
    The Foveon has this “something more” that CCD or CMOS CAN’T have.

    Come on guys, aren’t you tired of discussing about something you obviously don’t know…
    Get a life ! (or a camera !)

    PS : and stop comparing 100% pixel per pixel images… Be serious for a while…

    • These cameras have such different scope. The Hassy kit is almost always plugged into C1 or something to post-process. Those flat images are a real boon to those who love to post process.

      Didn’t Merrill crave that “straight from the camera to screen” richness that the Sigma cameras produce? Wasn’t that the whole point of the Foveon system? It was meant to dazzle with colour and contrast – it never was a subtle platform.

      The M9 is in a class of its own again. The Kodak dye choices for the CCD colour array impart their own flavour.

      That’s really all there is to it…

      • Sorry just don’t see the M9 in a class of its own, well only if it’s the “price” class 🙂 The camera is very ordinary, there are much better camera’s for thousands less, as ever it”s the glass attached combined with sensor etc that makes the difference. In many cases the £750 Sigma will cream an M9 set-up, in other cases it won’t. I have used both and can’t say I have a loyalty to either brand, if anything Fuji is my flavor most of the time.

        • No Paul, you’re right… That’s exactly what I meant when I said the M9 is in a class of it’s own. It’s the only camera out there that commands a huge price, holds cachet, yet is still very carryiable and mobile. The opposite would be the ‘blad, which looks like a massive machine if you take one waking around a town to take “street photos”… 😉 I guess I was being a tad facetious and somewhat cynical while using my theory on the dye choices on the CCD as a way to round out the argument.

          If you look back over the years at some of the earliest Kodak CCD based digi-slrs you’ll see they too had “interesting” colour rendition out of the camera. Sometimes that look was fairly frustrating to the user…

          • Ok, well not sure who owns the rights to the Kodak CCD now?? Seems Apple and Google bought most of the IP from Kodak as they sold off the household silverware to pay off debts 🙁

  29. I like apples, you like oranges…..

    I own the Sigma DP2M, and I think because of it’s limitations it does make you consider a photograph a LOT more before taking it.

    Every time I look at those files though on my computer…I say “Wow!”…..everytime.

  30. I only briefly toyed with the idea of buying one of the Sigmas. But several factors quickly nixed those thoughts. AF not being as fast as some other options out there being the primary factor for me. The other one being the fixed focal length lens. As much I might like the results I see some folks getting from the Sigmas, I’m satisfied to accept a little less than perfect IQ if I can have fast AF to allow me to capture non-static moments rather than potentially missing them. And I like that if I get bored with or need another focal length that I have options.

  33. The Sigma result is very sharp and detailed, but looks too contrasty for an old painting, but must not be. For the Hasselblad – it looks like the focus is off, I can not imagine the IQ to be so low on a professional camera. These lenses and backs can usually deliver pixel sharp images.

  34. Thanks Michael. I now only own the DP2 Merrill, and the X-E1 with a few lenses. The DP2 M is used mostly on PT4Pano rig (atop a “MeFoto” tiny tripod with a tiny indexing head) and the resulting 3–7 portrait stitches rival any from MF that I have seen.

    Under the right circumstances, this camera is capable of making arresting images. The exposures cannot be pushed or pulled though without funky colours or serious noise—so best to get it as close to ‘right’ in the camera as you. Thanks for writing this.

  35. surely the Blad has the best colour rendition – the sigma looks so deeply saturated… but, which is a more faithful rendition of the original subject?

  36. great test. i own a dp2merrill+dp1merrill. i won’t swap them for anything else. but there are a few who don’t want to believe how good they are simply because they spend more for less IQ.

  37. Sigma is super detailed but I question the white balance and it has too much contrast. The highlights are blown and shadows too dark. My SD14 was always too yellow and needed white balance adjustment. You just can’t rely on jpgs. The Hasselblad has much better tonality but is soft and needs more contrast. It seems more colour accurate. I wonder if the M9 would have achieved the most pleasing result.

    • That is not what I see.

      The DP2M looses out against both other cameras by a significant margin, at least to my eye… limitations in tonal and dynamic range are quite obvious even to a casual observer.

      It would have been interesting to shoot the same topic with Velvia 50 or a similar film (on 35mm or medium format), scan the result on a high quality scanner and compare.

      The result might have been an eye opener…

      • For sure, film would be an eye opener. But, because digital images are the new normal some would not appreciate, especially the many who have never shot film. This was not a very good test. There are many good Sigma photos to be viewed on the web. I have one of the early pre Merrill cameras and for the way I shoot, at 50 ISO, it works very well.

  39. Steve, would also have liked to see a comparison with the Fuji X – either the XP1 or the XE1. I am sure they would have made an equal impression as the DP2

    • I was not the one who did the comparison but can say with 100% fact that the X-E1/Pro would not give the detail and sharpness of the Sigma. It doesn’t even beat my RX1 in tshi regard (doing these tests now) and the Sigma beats the RX1 for detail and sharpness.

  40. I guess there is a camera for everyone today. I like the DP2M because I think it give a unique look to the picture. It’s not just about sharpness. When choosing which camera to buy I just went on flickr and looked at the pics, and I ended up preferring the foveon look. In the end it is (or at least it should) be a matter of taste, simply. The other camera I would consider is a Leica M, but I don’t have enough funds at the moment…

  41. Two possible factors accounting for the less than sharper Hassy image:

    * Bayer interpolation as typical for color sensors – except for Foveon
    * Mirror lock up?

    How about putting some Ektar 100 in that Hassy instead?

    • People have gotten used to the smear from interpolation. I’ve fallen into that trap myself. I like using the Olympus Art filters because I don’t like looking at what are considered to be the “sharp” digital images. People have asked me in the past why when using the same lens, going from a film camera to digital, their pictures aren’t as sharp. Then, I explain about the AA filter. But photographers adjust to the slight smear from the filter and it has become the new normal. Mediocrity has become the new excellence.

  42. I agree the detail in the Sigma is stunning (but that 30mm f2.8 lens is one of the most underrated pieces of glass). However, I’d love to see more comparisons between the Hassy and the Sigma. The MM9 is whole different beast. They Hassy here seems to have better tonality (the Sigma looks harsh) and it seems soft but in a smooth not-quite-in-the-sweet-spot kind of way. No doubt the Sigma–if you can live with the faults–is a lot of bang for your buck!

  43. Have to say I am pretty impressed with the Leica and also the DP2. I also see some mention of post processing that could have helped the Hassy. While that is true and certainly the older cameras do need more great light to begin with.

    Andrew Lamb, what was it about the DP2 that you did not like in respect of portraits? I would be interested to hear about that.

    Here is a portrait taken with a Leica MP using a Kodak Portra 800 with zero post processing just as an alternative to the digital offers above.

    http://www.flickr.com/photos/37556068@N06/8368505130/in/photostream

    • Stephen

      I found the handling really awkward. I’m not a fan of holding a camera at arm’s length. I put the Sigma on a tripod to get round that problem. Unfortunately, it wouldn’t focus. It simply didn’t want to know.

      I managed to take a few frames, none of them in focus, and then the camera started to buffer.

      As I mentioned before, this could all be down to me not using the camera properly but every I time I used the camera when there was pressure to get the shot it never worked out.

      Interesting you are featuring shots taken with a Leica M. When I sold my DP2M I bought a Fuji 6×8 rangefinder (made for the Japanese domestic market). Love it!

  44. I think some of the perceived sharpness has to do with the higher pixel density on the smaller sensor cameras. Do all of these cameras have similar pixel densities?

  45. I envy you. You own expensive cameras. Surprisingly the DP2 looks better than the other two. But shouldn’t the Leica crop be better? I see a lot of noise!

  46. I have to agree with Mike here. The Hasselblad captures all the subtle tonalities of the painting that the Sigma could not.

    • I would even bet that a lot of the tonalities are there in the Leica as well, but it for being monochrome and using better post processing.

    • Well I agree with your agree!

      Look at the pictures more closely than a superficial glance.

      The colored objects on the left and collar for example. Look at the various colors in those and subtlety that you cannot find in the DP2 shot. The DP2 makes it look like digital pixelated GARBAGE.

      yes….*GARBAGE*

          • Great to hear all these “experts” talking from real experience with files from all these cameras.

          • Well that’s the point of posting them right? To talk about them? Agree or not, it’s kind of boring to not have discussion. Maybe you would prefer to continue to discuss not having a discussion?

            Dumbest comment ever award ^^^^^

          • One doesn’t have to have shot with a specific camera or sensor in order to judge it. Some of us have experience with medium format systems but regardless it doesn’t take much experience or expertise to evaluate the two colour images and deduce that the Sigma shot has been robbed of tonality and colour in this reproduction.

            The DP2M may be able of extraordinary things, but the shot and 100% crops presented here are certainly not evidence of that.

            Why do you appear to have a problem with this discussion retow? Those of us who are discussing the three images here have kept our comments to those three images and aren’t condemning the Sigma DP2M on one image alone.

          • To answer your question: I’m surprised about the emotions the little Sigma seems to provoke. I’m a M shooter for many years, including M9 and some of Leica’s best glass. For landscpae/cityscape and stills, the Sigma files beat the ones the M9 can produce. I have no MF experience. But for less than USD1k, the foveon sensor mated with an excellent lens punches way way over its weight class. And evidence can bee seen all over the web. Some of the more credible testers are even of the opinion that its files compare well to the ones of the Nikon 800. Se from reading the emotional posts I rather get the impression that others have a problem with this little David (DP2M).

          • Lots of folks introduce emotion when talking in support or defence of their favourite cameras. With your first response you seemed to be sucked into that wormhole yourself. Better to filter out the noise and ignore the emotional – it’s impossible and not even fun to attempt to change the minds of ultra-fans and uber-haters.

            A number of us looked at the three images critically but objectively and found that the article author’s contention wasn’t supported by the images he provided. That to me is a worthwhile discussion to have.

            I’ve seen some stunning images produced by the DP2M; the image referenced in the article certainly isn’t one of them, proving once again it is possible to take a bad picture with a good camera.

            Maybe the image was screwed up in post; or maybe the DPxM cameras well known sensitivity limitations got in the way of making a good capture of what appears to be a relatively dimly lit subject.

            Cheers
            Mike

      • I don’t know about color accuracy as I didn’t see the original painting and I’m not an expert, but if you apply some gaussian blur on the Sigma image (I actually applied few times a 0.5 pixel blur) I think you can start to see the subtleties that are “hidden” behind the micro contrast.

        The Leica Monochrome image is pixelated too because, just like the Sigma, it doesn’t have an AA filter.

  47. That’s really intersting. I’ve heard a lot of good thing about the Merrill, that it is a really great camera for landscape and studio (apparently, it’s high ISO performance is even worse than the M9’s). Still stunned by the detail it delivers.

    I admit I’m disappointed by the Hasselblad shot. I would have expected more detail on a medium format sensor, but then it is the lens that contributes a lot to the quality. Would have been interesting to see the result of that lens stepped down a bit more

  48. I can’t agree that the DP2M produced better colour rendition in this comparison.

    Sure the DP2M image is crisper but is it more truthful?

    This is impossible to tell without seeing the painting itself, but given the fine colour blending in the 39CF image, it doesn’t appear the DP2M is telling anywhere near the full story about this painting.

    Ignoring capture processing differences (in camera or afterward), if it is colour we are judging on I’d have to give the nod to the Hassy with an assumption that it is being more truthful… by quite some margin.

    Look at the crop of the image produced by the 39CF – there are subtle tones of lavender that are completely missing in the DP2M image and that’s just one example. Look at both side by side the – DP2M image looks like it has been posturized in comparison.

    The DP2M image doesn’t convey nearly as much tonality and colour as the Hassy image, but if you want to see the texture of the canvas, then yes I guess it’s a winner. Sharpening the Hassy image and rechecking focus would give you all that and more.

    • Agree completely. If contrast was brought up in the Hassy image to match the Sigma, detail would be greater – considering the file is twice as large.

      And I would take a flat, (truthful as you say) image over that Sigma image any day.

    • I agree. I shoot and print paintings for a living. While it is not perfect, I could work from the Blad file. The Sigma file is trash from a reproduction standpoint. The dynamic range and color is terrible.

      The weave of the canvas is the least important factor in art reproduction. In fact it gets in the way of a good print. What is really important is capturing the detail in both highlights and shadows with accurate color throughout.

      My tools of choice for this type work is a 4×5 with a BetterLight Super 6K back and a pair of North Light banks. I also use Apo Rodinar lenses. Everything gets cross polarized and is completely profiled for color management.

      • I agree (former 8″x10″ filma and 4″x5″ digital photographer), but comparing the price and considering that maybe the Sigma RAW was not developped to it’s full potential, we still have a Sigma winner here. The Hassy cost 10x more, but gives only a +/- 15% advantage over the quality. I use the DP3 in studio for high priced portraits, and the clients mouth stay open when they see the prints. Tried 5D M2 and D800 before, but did not like the results as much as from the DP3.

  49. Interesting. I owned a DP2 Merrill and found it to be an infuriating camera. It was great for landscapes (not my thing) but for portraits I thought it was terrible.

    I wanted to do a side by side comparison with my 5D II in a real-world test. Photographing someone, for a magazine, using my mono-blocks I took a few frames on the Canon and switched to the Sigma. I was fully expecting the Sigma to yield better results. Sadly, I never found out as the wretched thing wouldn’t focus properly and buffered almost immediately despite my using a fast SD card. I had to switch back to the EOS to get the job done. I tried again on another shoot and the same thing happened.

    I get the impression that the Sigma is like a highly talented pianist who is brilliant at home but who has a nervous breakdown the moment he or she steps on stage.

    It’s a camera that works well for very very slow, considered work and I’ve seen great examples from it online so, maybe, it was pilot error on my behalf but I sold it without any regrets.

    And I really hated the software….

    • I wouldn’t worry, contrary to these results, the Sigma compacts don’t outdo even FF, I live in Japan with easy access to all of these cameras (well not this exact MF cam, but I have several of my own) and testing the DP’s against the 5d3 and D800 showed an obvious concession, however I was expecting this until I start seeing all these crazed people begin to hold up the sigma’s as something they’re not, it seems like the fact they’re in truth ridiculous, in every way almost, makes them somehow desirable, like people think since they have to work for the image it’s obviously better… well, they have about the best colour rendition I suppose.

      Not sure with this result, and Leica’s are about the only thing here in Japan I can’t just waltz up to and stick my SD card into. But there’s documented proof arounf the interweb, I think TheCameraStore did the exact test i’m talking about, the sigma vs d800, you can see the results.

      http://www.youtube.com/watch?v=f3VjyHQiqdE

        • Hi Andrew,

          – You took the words right out of my mouth there;o) – stationary subjects at base ISO (maaaaaybe pushing it to 200?) – still I have somewhat of a soft spot for those Foveons… but to much of a one-trick pony for me at least, at present… Thank God theres more to images than mere sharpness – LOL.. I think the DP-crop comes of a little too rough perhaps? Like it’s ‘on’ some extra micro-hyper-contrasty kind of steroids?

          Anyways my own experience with the ‘little monster’ was’nt all about crispness, sharpness and what not but more a subtle depth to the images that i did’nt see in my other cameras. Admittedly it was’nt there all the times but when it was there, it was in the DP-files.It’s offcourse highly subjective and can be argued as being a Foveon-fanboys disproportionate clinging to His/Hers idea of superior magic tool… And maybe it is… In the end, I sold it… ;o)

          Best
          klehmann

    • Not sure about that…

      I had a DP1 which was also an IQ champion in its comfort zone (a relatively small zone). I understand the DP Merrill are much improved over the previous generation, but I would not give up the flexibility and ease of use of my Leica M8 to go back to a Sigma DP.

      • Applause. The first Real validation of it. All other pro Foveon is just fanware. The camera’s comfort zone is so special and niche, that any Sigma Foveon could only make up a good 3rd or higher camera. Its slow, as slow as a snail crouching up a building per picture. The AF is slow as any other disappointing very slow AF camera. It offers banding noise (heavily) in anything high iso, even mid-iso.

        It works for still black and white or colour targets, taking your time and very good light for iso50/100 shooting. iso200 already greyed the colours (good still for bw), iso400 already creates noisy pictures, iso800+++ create banding.

        I’ve got my Sigma DP2x replaced by Sigma Germany. The Sigma Engineer commented to me that the merill would be just better what she manages well and partly worse in what the sigma could not provide already before.

        Be selective what you want to use it for, than, yes, it’s a superb camera. Otherwise it’s a pile of shame. Merill being the expensive variant of. 1000usd vs. 600usd before.

        I needed a flexible camera, as I did not plan to take 2-3 cameras on a trip, just to cover what’s offered to me in light, dynamic motives, … Very happy with the X100 which is superfast in comparison, while many claim it to be slow. I confirm, it’s not fast, but so much more than the sigma, in speed and iso/light flexibility.

        Sigma DP only for the weared who know what they’ll use it for every second.
        X100s for people who want to capture anything they like at any time and light at excellent quality. x100 for the older people, like me. 😉

        Have fun.

        • Well, I have both the X100 and the DP2 Merrill, both with the latest firmware, and the sigma is faster than in the AF department. So, for good light pictures, the sigma has been my go to camera.

  51. The DP2M is a little IQ monster. My M9 is gathering dust since I have the Merrill. Nothing can touch the foveon sensor in its comfort zone.

  52. I’ve always been intrigued by Sigma’s compacts. It seems like they have really come of age. Every review seems to give them nothing but praise.

Comments are closed.