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!

First real review of the DPQ2

Hulyss Bowman

Active member
I just post here some food for the brain :

The thing is that the Q seems to be underdevelopment even on the market. Might be SPP, might be the firmware, might be the sensor ... who know ? It is not the lens; that's sure. So we might find (more often than we want) a sort of noise under some conditions. A pattern noise not as aleatory as the Merrill. A clear noise. Then I remembered that Foveon have patented and idea. This idea was to create "light holes" into the top layer of the sensor to bring light to the bottom layer.

Here : FOVEON PATENTS you can find a load of interesting infos.

This one, particularly, attracted my attention : US2011057238 (A1)

I think they use it for the first time on the Quattro resulting in a more geometrical global image, at the pixel level. I mean, all the stuff is done on the top layer no ? So, for SIGMA it is cheaper to print and assemble the two bottom layers, 4.9 Mp each, high yield. They focus on the top layer then. I'm sure a lot of work have been done to reduce the chromatic noise "on sensor". The patent above allow it : It simple allow more light to reach the bottom, roughly.

So the Quattro is simply an old technology coupled with a sophisticated grid assimilated at the "blue layer", a grid of 19.6 Millions of pixel with a big part of those pixels being depleted... and this is visible. It have an impact on the vivid and organic rendering we used to see and mucho liked with the Merrill. Organic is the term. On the Merrill, you see the pixels. But the overall balance, clarity and contrasts make it just more real. With the Q you have more resolution (not that much) but a more structured sensor. And this is visible. You see the pixels, you see the pattern and you have a lot less contrast (certainly due to the depleted photo collectors).

Since the top layer is not complete and coherent in the blue wavelength you lose luminance. You loose contrast and clarity. So overall, the DP2 Quattro looks good when you reduce it to the Merrill ... It look "foveon" resized at 15 Mp. When you resize a Q full picture to 15 Mp you have a stunning result. But you do not have any more the power of 19Mp resolution. A 19,6 Mp full X3 normal foveon might have flaws but would explode the Q, propelling it to Jupiter and beyond. On detailed landscapes, the Q just turn "pointillism" and structured. Not organic like old foveons. Some may like. But as soon as I shot with it I didn't recognized a foveon. No way. Remember me when I wanted to sale my DP2x to take the first sony nex. The nex was not a fovoen at all and returned the camera and bought a DP2x again.

With the Q this is not that extreme, ok. You have a stunning resolution, it is more quick, colors are ok (same for the DP3m) but it seems a part of the soul left the shell.

So, how to make the quattro shining, really shining ? My idea is making the top layer denser; 24 Mp, why not, maybe a lot more. Then the camera record the datas and down-sample it to 16,5Mp via the processor. Then !!! You will have a pure image: Effective Pixels 16.5 mp / Sensor photo-detectors 24mp. So they must go Higher in MP on the top in order to increase final output in size. It is maybe actually the case on the Quattro, who know ?

A good firmware upgrade for SIGMA would be to introduce a special raw mode in the Q. The foveon mode. Make a raw not medium nor small. Make a Merrill raw of 15 Mp. Then, ppl who like foveon rendering will use it and the other who need only resolution and pointillism will use the full mode.

Then we will have something superior than the Merrills : An odd shaped Merrill without the flaws of the Merrill due to "improved" top layer.

What do you think about it, guys ?
 

ggibson

Well-known member
Unless you are talking an A7r it is not even going to be close.

;)
Actually, after comparing the Quattro to my A7 in a number of real-world shots, I came away with the conclusion that the A7 is quite close in detail to the Quattro. And actually, the A7 seems to have better color reproduction in some small details.

After my test shoot, I like the output of the Quattro quite a bit, but I'll leave the Q vs. M comparison to those who can shoot both side by side. The Quattro definitely had that "Foveon look" compared to my A7 though.
 

Kyndel

Member
I just post here some food for the brain :

The thing is that the Q seems to be underdevelopment even on the market. Might be SPP, might be the firmware, might be the sensor ... who know ? It is not the lens; that's sure. So we might find (more often than we want) a sort of noise under some conditions. A pattern noise not as aleatory as the Merrill. A clear noise. Then I remembered that Foveon have patented and idea. This idea was to create "light holes" into the top layer of the sensor to bring light to the bottom layer.

Here : FOVEON PATENTS you can find a load of interesting infos.

This one, particularly, attracted my attention : US2011057238 (A1)

I think they use it for the first time on the Quattro resulting in a more geometrical global image, at the pixel level. I mean, all the stuff is done on the top layer no ? So, for SIGMA it is cheaper to print and assemble the two bottom layers, 4.9 Mp each, high yield. They focus on the top layer then. I'm sure a lot of work have been done to reduce the chromatic noise "on sensor". The patent above allow it : It simple allow more light to reach the bottom, roughly.

So the Quattro is simply an old technology coupled with a sophisticated grid assimilated at the "blue layer", a grid of 19.6 Millions of pixel with a big part of those pixels being depleted... and this is visible. It have an impact on the vivid and organic rendering we used to see and mucho liked with the Merrill. Organic is the term. On the Merrill, you see the pixels. But the overall balance, clarity and contrasts make it just more real. With the Q you have more resolution (not that much) but a more structured sensor. And this is visible. You see the pixels, you see the pattern and you have a lot less contrast (certainly due to the depleted photo collectors).

Since the top layer is not complete and coherent in the blue wavelength you lose luminance. You loose contrast and clarity. So overall, the DP2 Quattro looks good when you reduce it to the Merrill ... It look "foveon" resized at 15 Mp. When you resize a Q full picture to 15 Mp you have a stunning result. But you do not have any more the power of 19Mp resolution. A 19,6 Mp full X3 normal foveon might have flaws but would explode the Q, propelling it to Jupiter and beyond. On detailed landscapes, the Q just turn "pointillism" and structured. Not organic like old foveons. Some may like. But as soon as I shot with it I didn't recognized a foveon. No way. Remember me when I wanted to sale my DP2x to take the first sony nex. The nex was not a fovoen at all and returned the camera and bought a DP2x again.

With the Q this is not that extreme, ok. You have a stunning resolution, it is more quick, colors are ok (same for the DP3m) but it seems a part of the soul left the shell.

So, how to make the quattro shining, really shining ? My idea is making the top layer denser; 24 Mp, why not, maybe a lot more. Then the camera record the datas and down-sample it to 16,5Mp via the processor. Then !!! You will have a pure image: Effective Pixels 16.5 mp / Sensor photo-detectors 24mp. So they must go Higher in MP on the top in order to increase final output in size. It is maybe actually the case on the Quattro, who know ?

A good firmware upgrade for SIGMA would be to introduce a special raw mode in the Q. The foveon mode. Make a raw not medium nor small. Make a Merrill raw of 15 Mp. Then, ppl who like foveon rendering will use it and the other who need only resolution and pointillism will use the full mode.

Then we will have something superior than the Merrills : An odd shaped Merrill without the flaws of the Merrill due to "improved" top layer.

What do you think about it, guys ?
Interesting thoughts Hulyss, but I do not have the knowledge to go into a debate about this, but I agree, that this Quattro does not have that "Foveon-look" or what we shall call it, and that was what got me to buy all 3 Merrills, and I like them all, even they are different, and with some shortcomings, I had hoped would be adressed without sacrificing anything, so I am not sure I will buy the new Quattro.
 

Kyndel

Member
Thanks, I tryed - DP1M vs. Quattro (take care not to take the SD1M instead of DP1M, I did the first time) ............at iso 100, still life.

Look at the red fabric (with other kinds of red fabrics showed at the same place), ...............there is NO details from Quattro !!!!!!!

But beautifull details from DP1M, and we know the DP2M is even better !!!!

The ruler: A lot of cameras show this not pure white, DPMerrills does, and I love it !!!, but not the Quattro.

I am sorry.
 

G43

New member
I compared it to D800E
The red fabric goes very wrong with the DPQ. So does the knitting bundles where orange and red almost looks the same.
The sharpness however looks nice with the DPQ
 

foveon

Member
I compared this doll inside ISO 200 from DP1m and DP2q but the images arent made very carefully. Beside the difference in focal lenght there are differences in POV and even in focus. so for me. its hard to tell something
 

PaulO

New member
I continue to be frustrated by the lack of color management in SPP and consistency with other color managed environments. When I output from SPP (ProPhoto RGB) and open in PS the colors are rendered differently. This reduces setting the color controls in SPP to guess work. Fortunately, I can usually correct the colors in PS. But I retain doubts as to whether I have set the color controls in SPP optimally. Just my 2 cents.
 

darr

Well-known member
I am a subscriber to REID REVIEWS and to those that are not, you may be interested in more reviews of the DP2Q:

Hello,

Today I published a full field review of the new Sigma DP2 Quattro. Coming up next will be a report on side by side studio tests of the DP2 Quattro, Sigma DP2 Merrill and Leica T. I'll also be looking at the Quattro's potential as a BW camera.

Cheers,
Sean Reid
Publisher
Welcome to ReidReviews
 

Kyndel

Member
Well, there is still hope (the red fabric I and others have written about, there is a clear difference when RAW)

UPDATED - Sigma dp2 Quattro First Shots posted - adding a raw conversion to the mix


and here is the comparometer = You can put Quattro vs. Nikon D810 now = Not bad exept for the red fabric, but then there is the explanation in the link ABOVE, and also exept for the not so white ruler, but then again = The Nikon D810 does not have that either

Imaging Resource "Comparometer" ™ Digital Camera Image Comparison Page
 

Kyndel

Member
Is it real serious to compare a DP1m to a DP2q, why they dont use a DP2m?
I agree it would be much better to use the DP2M as a comparison (I asked them long time ago to do a test about the DP2M), but the problem is: They have never done a thorough test about it, like they have about the DP1M,therefore they do not have the pictures.
 

Hulyss Bowman

Active member
Interesting thoughts Hulyss, but I do not have the knowledge to go into a debate about this, but I agree, that this Quattro does not have that "Foveon-look" or what we shall call it, and that was what got me to buy all 3 Merrills, and I like them all, even they are different, and with some shortcomings, I had hoped would be adressed without sacrificing anything, so I am not sure I will buy the new Quattro.
Yes, enjoy the Merrills. I will certainly get the DP3Q and make a proper review because in one year from now, maybe, the technology would have improved, like in the Merrill.
 

Malina DZ

Member
I said that a while ago (what you see in spp is not what you get after conversion)
And there's a big difference in contrast when you export B&W images from SPP to TIFF 16-bit and open them in LR (on 10-bit monitor). SPP has a lot to improve.
 
Top