Sigma DP1 Merrill review - Conclusion and pixel comparison.

After yesterdays rave review it may come as a surprise that I'm probably going to be selling the DP1 Merrill and the DP2 Merrill. Why? Well these are pretty much luxury cameras. The extraordinary image quality won't sell me any more pictures because people buy my images from thumbnails. They never get to see the quality until they have paid for it. Plus I'm going to get down to two cameras. This is non-negotiable and the DPM's are just too restrictive for what I do. There are all sorts of situations where I need telephoto lenses and lenses wider than 19mm, I need faster operation and much more versatility.

It is actually a privilege to take pictures with these cameras, but because of the reasons I outlined in the post yesterday, like everybody else Bayer sensor cameras offer me so much more. Maybe if Sigma offered or were going to offer some kind of small mirrorless interchangeable lens system with the lenses on the DP1M, DP2M and DP3M, then I might reconsider, but being honest, probably not even then.

By the way, if you want more details on the DP1M here they are. 

The 19mm lens has a little more CA than the DP2 Merrill and at wide apertures the corners can be slightly softer than the centre, but not by much. However that centre is probably slightly sharper than the DP2 Merrill. Thats it. Not much else to say really. Its probably the sharpest results from any camera ever made at ISO 100 this side of a Hassleblad MF back, and it may even produce results that are sharper than that. Its that good. But thats the end of the story.

There is absolutely no point in recounting just how bad it is at high ISO's. And boy is it bad. Its like nothing happened with digital camera sensors for the last 10 years. If you know what this camera is then you know all this anyway. You'll know I have to take 5 batteries out with me for a days shooting, you'll know there is no built-in IS so slow shutter speeds don't work well at all. You'll know there's no viewfinder so I have to bolt on a Sigma optical one, which like the rest of these things is nothing more than an approximation of what you'll get. OK so there's the SD1. Why not get one of those? Well because its heavy and there are few lenses that do the sensor justice. The 35mm f/1.4 is a great lens but its seriously big and heavy and the last thing I want is to carry something around that will torture my back even more.

So some kind of conclusion maybe? Well again its mostly repeating myself. These DP Merrill cameras have one great pro and lots and lots of cons. The pro however can't be found anywhere else. There just isn't any other camera in the DSLR / Mirrorless / Rangefinder marketplace that offers the level of image quality at ISO 100 that these cameras do. The lenses are optimised to the sensor, so you are going to get better results than with an SD1 and the majority of Sigma lenses, so thats only partly a competitor. As I've written before if you upsize one of the files from these cameras they are pretty close to what you can get from a 36MP Nikon D800E and possibly better than what you can get from the basic D800. If you make prints you can make huge prints from these cameras and they will reproduce up to A2 with no problem at all.

However if you want to see the magic, because that is sometimes what it strikes me as being, leave the files at their native size and scroll around at 100% just to see what you were close to when you took the photograph. These cameras and this sensor produce forensically clear images. Just finally to give you an idea of whats going on I've put together two blowups to show the difference between Foveon and Bayer sensors. 

One taken on the DP1 the other on a NEX-6 with a similar lens the 19mm Sigma prime on the Sony.

I've blown up the area to the right of the red button on the left remote control to 3200% which shows the image at the pixel level. The Sony image is sharpened, the Sigma image isn't.

There is a discrepancy in scale here (14MP v 16MP) and a different colour rendition
but I think you can clearly see the difference.
Both images at ISO 100 f/8. You can clearly see how the three foveon layers combined produce an entirely different image, which is sharper and with much more differentiation between different colour areas, and indeed pixels. I can't say for sure, because I'm no engineer, but I'm assuming that this is because the Sigma actually records whats there with its three sensors and the Sony with its Bayer array is filling in what it thinks should be there after recording it through an AA filter. However whatever the reasons, it isn't hard to see why the Foveon sensor(s) is capable of such dramatically superior results, within its workable limits. It simply captures more of what is actually in front of it.

As I said at the start of my piece yesterday, why don't all digital images do this? Why can't we have this level of definition in all cameras? Well we probably can't because of commercial reasons. Every camera manufacturer isn't going to ditch the Bayer sensor and start again. It would be madness if they did. Apart from that who else but nerds like me look at files at 3200% !!!  

And thats the real reason. People don't really care that much. The Bayer CMOS sensors do their job, give us acceptable images that print and reproduce well and of course allow us to take pictures in all sorts of situations that the Foveon sensor doesn't. And ultimately its what the image consists of rather than how it looks and the technical quality, thats important. 

But then assume you have extraordinary warm evening light illuminating the glorious landscape in front of you and you have a Bayer sensor camera and a Sigma Foveon camera with you, and you have your tripod. Which one do you choose then??

Join the Soundimageplus Blog Readers Group at Google+

For comment and discussion - join me over at Google+