Sigma DP2 Merrill Experiences

[Hulyss]

Hello Quentin, (first I want to say I'm very happy to be on Luminous Landscape, it is a quality place populated with quality people, a real heaven coming from DPr or french forums ...).

I have my camera since the 11 of this month. And I recoup almost ALL your "thumbnail sketch of the good, bad and the ugly" despite some little variations and here is my own little return:



The good:

# What I like with the DP, being a former DP user since the DP1, is the ergonomic, yes. It is very user friendly, pure and simple.
# The lens is just awesome, coming from some Zeiss on my D700 and some use of a S2. It is why I protect it with a Helioplan UV SH-PMC 49 x 0,75.
# You can screw the lens hood over the filter, so it protect the filter too because the hood is sort of square windowed as you can see above.
# The sensor is what I qualify as a photo revolution, a real. Richard Merrill and his team really did something to the photography and this is not marketing; just facts.
# The colour output is really film like, using also a lot of film (and building a large stock of Kodak ...   ).
# The white balance might be not very consistent BUT she is very just. I almost never need to tweak it on a single view.
# The overall output at this focal is better than what can produce the SD1m (Amen to the dedicated marvellous lens).
# When correctly exposed, no really need of any PP. The image is just and pleasing (this is a big evolution in SIGMA digital world, trust me).

Some examples just for the eyes no pp applied and OFC no sharpening applied:

Picture

Crop

 
Picture

Crop


And, for the quality of the bokeh:

Crop for bokeh


Well... you get it.

The bad:

# Battery life is poor.
# The raw developer is poor (but ... efficient).
# Not good above ISO 100 (I'm harsh but that's my exigences).
# Banding when using Nik Software Silver Efex, in some ways.
# Shadow transition problems on human skin. At the transitions zones, the skin turn in a variation of green and red.
# Because of the above I think (for portraiture) this camera will be Black and White only (If you play with shadows, like me).

EDIT: The problem can be fixed via PP in PS by going in hue Red +5 and Yellow -15. This work and give perfect skin colours/transitions.



Conclusion

This camera is a very good camera if we like to be purist, to take our time to compose, seeking a soul in our shoots. There is some flaws at this young technology but it is promising it is why I continue to buy it. It contribute to the "effort" let say But I really want to see a 24x36 foveon one day. For me it will be a perfect colour street camera, landscape and B&W phototool; but, on humans, my primary camera will stay my D700 and my Zeiss (for now ^^ I wait a firmware).

EDIT: The camera can now be used as a portrait camera by using the PS routine above.

[michael]

I've had the DP2M for less than a day and only had time to do a few quick test shots. But I have to say that I am knocked-out by the image quality. Some of the issues that I saw with the SD1 gone, some of them are still there. The two reviews that I've read in this thread seem to provide a very appropriate perspective.

Lot's more to say in future but, this is not a camera for pussys.

Michael

[Hulyss]

I've had the DP2M for less than a day and only had time to do a few quick test shots. But I have to say that I am knocked-out by the image quality. Some of the issues that I saw with the SD1 remain, some of them are still there. The two reviews that I've read in this thread seem to provide a very appropriate perspective.

Lot's more to say in future but, this is not a camera for pussys.

Michael

  for sure

EDIT: I modified my review, look at my EDIT in my previous post. Skin color problem can be fixed via PP.

Example:



After R+5/Y-15

[Quentin]

Huyliss

Nice work on the skin tones.  Tends to reinforce the view that Sigma are not great in the raw software department.  No surprise there, then!  I don't have the lens hood but I am awaiting the close up accessory lens.  Great review, similar conclusions to my own (which I find reassuring).

Michael,

Glad we are apparently on the same page regarding image quality.    Definitely a camera that rewards the careful and thoughtful photographer.

[kgelner]

Thanks for the review, I enjoyed it... one thing of note though on using manual focus in combination with AF lock.   When in AF Lock if you switch to manual focus, it automatically sets the manual focus distance to where the AF was locked at.  So you can use this to switch to manual mode to verify exactly where the AF landed, and make slight adjustments from there if you wish.

Note that there seems to be some bug in switching between manual and AF modes that at times it reverts the AEL setting back to being AEL, instead of AFL...

[JayWPage]

I wonder if some of the DP2 Merrill users could comment on the dynamic range. What other camera is it comparable to?

[Quentin]

I wonder if some of the DP2 Merrill users could comment on the dynamic range. What other camera is it comparable to?

It is excellent.  The Cathedral shots I posted earlier in this thread show good detail in the stained glass windows without the need for bracketed exposures.  Expose for the shadows, and (within reason) you can recover the highlights.

[Dave Millier]

It is excellent.  The Cathedral shots I posted earlier in this thread show good detail in the stained glass windows without the need for bracketed exposures.  Expose for the shadows, and (within reason) you can recover the highlights.

Hi Quentin

As a long time Foveon user (SD9, SD14, DP1), my reservations about the technology have always been: 

1) Luminance aliasing. This has always been there, given the absence of an AA filter. The new sensor seems no different: there are always visible jaggies on things like angled grass stems. Lot's of people seem oblivious to this but to me it's one of the biggest trademarks of digital imaging. I've been looking at a lot of images from the new sensor and it still seems very obvious to me.

2) Funky colour. Every version of SPP and every new Sigma camera yields a different colour response.  I find Foveon has a tendency to overly yellow/orange foliage in warm light and tends to blow reds very easily giving a flourescent radioactive look and also hue shifting to magenta or orange

3) Colour casts. My SD9 is too yellow, my SD14 too green/yellow. Fortunately, DNG profiles help a lot. The DP1 was quite neutral except for an occasional tendency to go magenta

4) Green edges. The DP series has been prone to a green cast at the edges.

5) Shadow noise, especially large blotches that conventional NR software can't deal with (usually considered the result of the huge matrix operations needed to extract RGB colour from the weird colour space of Foveon)

Can you comment on these aspects with the new camera?

    (my 6 year old insisted on the smiley)

 

[KirbyKrieger]

Hi.  I am interested in some perspective on using the DP2M while zone focussing.  Sean Reid extols the virtues of the DP cameras for this method of working, in part (probably large part) because they have a thumb wheel dedicated to setting the focus distance (and the control has clearly marked distances on it).  As far as I can tell, the DP2M has removed this wheel and relocated focus distance control to a ring around the lens.  This ring has no fixed relationship to focus distance, and of course no scale.  As such, while the DP2M is leaps ahead of the DP2 in IQ, it seems it is not a tool that can be used for zone focus work.  Is this correct?  Is there any easy (preferably off-eye) way to set the focus distance?

Related: is the optional viewfinder usable?  Reviews of VF for other DP cameras have mentioned that the VF is a bit small and distorts, but not too small or too distorted to get in the way of use.  Comments from anyone who has tested one?

Thanks.

[michael]

The DP2M does not lend itself to zone focusing. I know my friend Sean is a big fan of that style of working, but I'm not, especially with a longish focal length lens. Look at a DOF table that uses a small circle of confusion and see what your DOF would be. It's very small and so if you using zone focusing most of your picture will be slightly out of focus.

In the days of film when you could use a large COF this made some sense with very wide lenses, but not these days.

I didn't mention it in the review, but the DP2S has a focus confirmation light on the top panel where one can see it if using an optical VF. I didn't bother with the one from Sigma, I have a 40mm Voigtlander that works well.

Michael

[KirbyKrieger]

The DP2M does not lend itself to zone focusing. ... [Much goodness snipped.] ... I have a 40mm Voigtlander that works well.
Michael

Many thanks for covering what I asked, what I needed to know but didn't ask, and more  .  Here's to good health.  --Kirby.

[Jason Denning]

Hi

Can anyone comment on the dynamic range, I like to be able to raise the shadows sometimes significantly and coming from digital medium format I can so wondered how this fairs? I would live to get access to a raw file or tiff file if anyone is willing to upload one.

Thanks

Jason

[Quentin]

Hi Dave

See my comments below

Hi Quentin

As a long time Foveon user (SD9, SD14, DP1), my reservations about the technology have always been:  

1) Luminance aliasing. This has always been there, given the absence of an AA filter. The new sensor seems no different: there are always visible jaggies on things like angled grass stems. Lot's of people seem oblivious to this but to me it's one of the biggest trademarks of digital imaging. I've been looking at a lot of images from the new sensor and it still seems very obvious to me.

I have not noticed any.  Sometimes the appearance of aliasing is actually the edge of pixels which would disappear if the image  is rezzed up- something I saw with scan backs, which of course are also RGB for each pixel (and therefore sharp)

2) Funky colour. Every version of SPP and every new Sigma camera yields a different colour response.  I find Foveon has a tendency to overly yellow/orange foliage in warm light and tends to blow reds very easily giving a flourescent radioactive look and also hue shifting to magenta or orange

There is some occasional purple and yellow blotching in underexposed shadow areas.  Reds do blow a bit more easily than they should, so you need to be a little careful.

3) Colour casts. My SD9 is too yellow, my SD14 too green/yellow. Fortunately, DNG profiles help a lot. The DP1 was quite neutral except for an occasional tendency to go magenta

Michael mentions cyan skies in his review.  The overall cast on my camera using auto  is a little green.  Correctable either in Sigma Photo Pro and not significantly worse than other cameras I have used.

4) Green edges. The DP series has been prone to a green cast at the edges.

Still there in shadow or underexposed areas.  Probably the wost problem (for me ).  Something Sigma need to work on to provide some kind of correction in their software.  Of course it is possible to address the issue in Photoshop, but it takes time and is a particular issue with something the DP2M is otherwise very well suited for - stitching

5) Shadow noise, especially large blotches that conventional NR software can't deal with (usually considered the result of the huge matrix operations needed to extract RGB colour from the weird colour space of Foveon)

The biggest noise issue with the DP2M is hatching or banding in areas of even colour, like skies.  Again, Michael has mentioned this, and it does seem to come and go depending on the image.  With a well exposed image processed with "0" in Luminance noise slider (really, the mid-position), it is not visible in normal operation.  Turn all NR off, and/or manipulate the sky and it becomes increasingly evident.  I kill it altogether by running Neat Image selectively applied to the sky  (or other even areas) or turning up NR in Sigma Photo Pro

But lets not get carried away here.  There are these issues, but there are equally other areas where the DP2M does not have issues that other cameras do have, like colour artifacts and moire, which are just as much trouble to deal with as any of the issues with the DP2M. And then of course there are the extraordinary benefits in resolution and acutance / sharpness not seen outside of megabucks MF backs (if at all), meaning a more significant ability to upscale images to match much higher resolution Bayer sensor cameras.

The DP2M is not a perfect camera; it is a camera that rewards work in the hands of a skilled photographer.  It would probably be a disaster in the hands of the metaphorical soccer mom or casual user, but they are not the Luminous Landscape constituency  - are they?

Can you comment on these aspects with the new camera?

    (my 6 year old insisted on the smiley)

I can match that
 

[photodan]

for sure

EDIT: I modified my review, look at my EDIT in my previous post. Skin color problem can be fixed via PP.
... After R+5/Y-15

Hulyss,

The skin color correction looks excellent in the profile image of the girl's arm and torso profile. However I noticed that the color of a portion of the girl's garment changed from a definite yellow to a definite orange type color.  Do you happen to remember what was the true color - was it the yellow (as shown in the pre-correction image) or orange (post correction image) or something in between?

I am very interested in this camera however it's a concern to me if in order to get rid of obvious color transition artifacts that I'll need to do extensive color correction on local areas.

Thank you.
Dan

[Rob C]

Remember film?

;-)

Rob C

[Hulyss]

Hulyss,

The skin color correction looks excellent in the profile image of the girl's arm and torso profile. However I noticed that the color of a portion of the girl's garment changed from a definite yellow to a definite orange type color.  Do you happen to remember what was the true color - was it the yellow (as shown in the pre-correction image) or orange (post correction image) or something in between?

I am very interested in this camera however it's a concern to me if in order to get rid of obvious color transition artifacts that I'll need to do extensive color correction on local areas.

Thank you.
Dan

Hello photodan,

Yes the yellow top changed colour. The real colour is more orange than yellow but, even the D700 is not very right on this. I would have done a mask, only on the zone; better than processing the whole image.

If there is no yellow in your shoot, you will have no problems at all. No luck for me 

Hi

Can anyone comment on the dynamic range, I like to be able to raise the shadows sometimes significantly and coming from digital medium format I can so wondered how this fairs? I would live to get access to a raw file or tiff file if anyone is willing to upload one.

Thanks

Jason

Hello Jason,

Here is a link to the DR of the SD1m. The DP2m seems to have more DR than the SD1m, even with the same sensor. This is strange.

http://www.techradar.com/reviews/cameras-and-camcorders/cameras/digital-slrs-hybrids/sigma-sd1-merrill-1088902/review/page:5#articleContent

[Hulyss]

A little test I did today. Ok they are not the same tools and the test isn't scientific, at all. But I like it (Be care, big files. let them charging. Exifs inside).

First the DP2m
http://www.hulyssbowman.com/tempo/DP2m.jpg

Second The S2 + 70.
http://www.hulyssbowman.com/tempo/S2.jpg

[KirbyKrieger]

A little test I did today. Ok they are not the same tools and the test isn't scientific, at all. But I like it (Be care, big files. let them charging. Exifs inside).

First the DP2m
http://www.hulyssbowman.com/tempo/DP2m.jpg

Second The S2 + 70.
http://www.hulyssbowman.com/tempo/S2.jpg

Thanks for posting.  I'd rather have the DP2M file -- but I can't conclude anything from that.  

My DP2M arrives Monday.

[Ligament]

Hi All,

Attached two files I took tonight with the DP2 Merrill and the vertical banding problem.

I have seen the vertical banding quite often in the 100 or so images I've taken since I purchased it.

How are you all avoiding this, and secondly how are you processing it out?

I converted from RAW in the Sigma Photo Pro software, 16 bit, ProPhoto space. Only adjust white balance and allow lens profile chromatic aberration to work. I try not to use anything else in SPP as it is terrible.

Open in Lightroom and make my adjustments there. Lightroom cannot get rid of the vertical banding so I do all my noise reduction and vertical banding adjustments in Nik Dfine 2.0.

Attached before and after Dfine 2.0.

I should comment that the microcontrast is impressive and I find myself dialing back the clarity slider in LR more often than not. The images are quite sharp and although some could benefit from sharpening, I'm finding it is not worth the noise enhancement.

The colors seem odd to me with a greenish cast to them that I am still trying to work out. I live in Seattle, WA with rather blue overcast light at all times and a very high Kelvin color temperature on average. Perhaps this has something to do with it, although on my bayer sensor cameras I have no such green issues.

You can download the RAW file, my TIFF file from SPP, and my post DFINE file here:
https://www.dropbox.com/sh/irihi9zhucl69jg/4Ap-ghH3Td

[Quentin]

To remove banding I either (1) turn up NR to full in Sigma Photo Pro, and selectively blend-in the NR reduced sky on a new layer, and/or (2) run Neat Image in Photoshop, again selectively applied to the sky (layers, or history brush), or whatever area needs it.