The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Sigma DP1 Quattro Shots

Quentin_Bargate

Well-known member
Some quick further testing has thrown up an issue.

This is a house image, tweaked in Photoshop, showing good across the frame sharpness.



However, look at this crop of the tree line above the house



And compare it with a shot taken shortly afterwards with the DP2M (bear in mind the different focal lengths so the DP2M trees appear larger)



Something is going on in the branches that reminds me of my old Kodak 14nx

This concerns me because it might affect landscape shots with trees taken against a bright background. Other than this, the image quality from the DP1Q is very good.
 

Quentin_Bargate

Well-known member
I am starting to think that while the new Quattro sensor arrangement works 90% of the time, Bayer type colour errors are creeping back in because only the top layer is full resolution. I think the above effect is liked to this and I see subtle signs of colour errors in some DP2Q shots. If it is, then we all better hang on to our Merrills...
 

Ohlhaus

New member
Thank's a lot for the first impression - Hope Jpeg is now usable and ISO 800 may be an option- I will wait for a later statement from you - If available :)

Enjoy the playtime with your dp1q
 

mezzoduomo

New member
The original Merrills are looking better all the time. I've become accustomed to their limitations, and it seems like efforts to overcome those limitations might be creating unintended consequences.
 

vjbelle

Well-known member
I don't own one but have toyed with it. Your posts above have made me really rethink this. I was really hoping that this would be an improvement over the original version but with this artifact I am now very alarmed. It appears to be a masking effect.... I would think something that could be rectified but is a killer for me so far.

Thanks for your postings regarding this camera.....

Victor
 

capital

New member
Hi Quentin,

In my comparisons of the DP2M and DP2Q I talked with you about on luminous landscape, I saw the same issue in my tests of trees set against sky that you are seeing here with the DP1Q. And is indeed a "bayer" type of problem.

It is quite unfortunate that this new layout introduces the potential for this type of artifacting.

While Sigma DPP greatly improved red color resolution with SPP6.0.6 as well as shadow detail rendering, Sigma's software is still a work in progress.

I have also noticed that Quattro files in SPP are prone to having some other interpolation type of artifacts, especially in some very fine repeating details, such as the chimney grill work example I posted on luminous landscape thread.

One further issue I found bad about SPP renderings thus far are in night photographs where a very rough stair step pattern occurs over highlight areas that is simply not present in Merrill files. Merrill files actually have a very pleasing roll off of highlights. I have even tried all the setting related highlight retention and recovery in SPP for the Quattro files and they fail to match excellent default Merrill rendering.

The net result of all this is that short of Sigma addressing these issues it mitigates any feeling towards upgrading to this new imaging layout. In fact, I have actually gone back and purchased a refurbished DP1s because it retains the good aspects of the x3 layout while also having more color in the shadows.
 

Quentin_Bargate

Well-known member
Unfortunately, I have to agree.

Under the right conditions, the new Quattros can produce image quality that exceeds that of the Merril's by a small margin. But I am increasingly of the view that this is at the expense of some Bayer type aritfacts. Not as many as a with a pure Bayer sensor, but some nonetheless, and that defeats the purpose of using a Foveon based camera.

I am concerend. Someone needs to wake Sigma up to the issue. current Bayer based cameras have reached a level of sophistication that is actually better that the example from the DP1Q I posted above. Processing power and cheap memory render redundant any argument about the need to keep raw file sizes down, which is the only justification for the new sensor design.


Hi Quentin,

In my comparisons of the DP2M and DP2Q I talked with you about on luminous landscape, I saw the same issue in my tests of trees set against sky that you are seeing here with the DP1Q. And is indeed a "bayer" type of problem.

It is quite unfortunate that this new layout introduces the potential for this type of artifacting.

While Sigma DPP greatly improved red color resolution with SPP6.0.6 as well as shadow detail rendering, Sigma's software is still a work in progress.

I have also noticed that Quattro files in SPP are prone to having some other interpolation type of artifacts, especially in some very fine repeating details, such as the chimney grill work example I posted on luminous landscape thread.

One further issue I found bad about SPP renderings thus far are in night photographs where a very rough stair step pattern occurs over highlight areas that is simply not present in Merrill files. Merrill files actually have a very pleasing roll off of highlights. I have even tried all the setting related highlight retention and recovery in SPP for the Quattro files and they fail to match excellent default Merrill rendering.

The net result of all this is that short of Sigma addressing these issues it mitigates any feeling towards upgrading to this new imaging layout. In fact, I have actually gone back and purchased a refurbished DP1s because it retains the good aspects of the x3 layout while also having more color in the shadows.
 

capital

New member
I emailed Sigma Japan about the chimney grill work example a month or so ago but did not ever get any response. Some of the issues are quite clearly a problem with SPP, like this chimney example, as can be seen in the accompanying comparison to raw luminance data that artifact does not exist.

The issue of blue bleed in branches against blue sky is also a bit perplexing, especially since the panchromatic top layer is blue biased! Here it might seem to suggest that Sigma SPP is being a bit too heavy handed in reconciling photo responses of the three layers. As a matter of comparison, SPP6 dot releases below 6.0.6 adopted a rather dumb algorithm for introducing noise reduction (even at lowest setting!) in grey values of about 90 or lower which artificially reduced resolution and lead people to think that foliage and other details had a watercolor look.

Sigma unfortunately has adopted a paternalistic attitude towards SPP processing, something they need to seriously readdress.
 

Rand47

Active member
I've been following the Quattro discussion for a while now. In anticipation of the new cameras, I bought a lens hood for the DP2Q, and the remote release. If any US based photographer would like to have them, PM me and I'll ship them to you for free, still in the boxes.

Just this last week I made my decision to stick w/ the Merrills for now, and ordered the DP1M to complete the set.

Thanks, everyone (and especially Quentin) for your objective testing and practical examples of the DPQs. For me, even though I can see the increased resolution in some of the sample images, they still lack a certain something "crisp" that gives the appearance of greater detail rendering with the Merrills.

Best regards,
Rand
 

rjp85

Member
I now have all three Merrills. I will also pass on the Quattro. Maybe next year we'll have another 1:1:1 sensor.

I was expecting the DP1M to have lots of CA and soft corners, but mine seems to have excellent corner performance, is incredibly sharp, and CA is minimal. I'm impressed, considering the things people have said about the DP1M. It also came with the latest firmware. Woohoo!
 

foveon

Member
beside the sensor thing, the DpP1 has a 19mm lense and in this lenght no lense is perfect corner to corner, especially open.
 

Quentin_Bargate

Well-known member
Driftwood, Grand Cayman



7 Mile Beach



The DP1Q performs quite well in relatively even light, but the shadows are noisy if you try to enhance the shadows (worse than a Merrill). Still, working quite well during my mainly business) trip to Cayman (arrived yesterday so still a bit Jet lagged...)

I have set the DP1Q up to take three bracketed shots 0.7 stops apart in an attempt to avoid blown highlights.
 
Top