"M9 Worst Image Quality" says DxOMark

It only took DxOMark three years to finally getting around to testing the full M9 platform (M9, M9-P and now M-E) and they summarize it with "...these cameras offer the worst image quality DxOMark have tested on a full frame sensor, with the exception of the 10-year-old Canon EOS 1Ds." Really?

Yes, we get it already. High ISO on the M9 platform isn't the greatest. It's also not the be-all, end-all of digital photography. Comparing "CMOS sensors used in the latest Nikon and Sony DSLRs" to a three year old product isn't just apples and pears, it's laughable.

Leica M Series: CCD Sensors just aren’t as good as the latest CMOS offerings

The 24x36mm full frame CCD sensor utilized in the Leica M Series rangefinder cameras produce significantly poorer raw image quality, compared to results from DSLRs featuring a CMOS alternative. In fact, with a DxOMark Overall Score of 68, or 69 for the Leica M9, M9-P and ME Type 220, these cameras offer the worst image quality DxOMark have tested on a full frame sensor, with the exception of the 10-year-old Canon EOS 1Ds.

No doubt Leica enthusiasts will assert we’re comparing apples and pears, and the advantages the Leica M System offers in terms of simple control, portability and discretion, as well as first class engineering, are more important.

It’s fair to say too these results purely examine the data from the sensor, excluding the impact top quality Leica M Mount glass will have on the real world results. At base ISO sensitivities Dynamic Range and Color Depth scores aren’t quite as far away from competition, like the Canon EOS 1Dx and 5D MKIII, as the DxOMark Overall Scores might suggest.

Compared to the evolution of CMOS sensors used in the latest Nikon and Sony DSLRs however, the DxOMark Scores indicate a gulf in image quality, not only at base ISO but importantly at mid range sensitivities up to ISO 1600. While image quality drops off very quickly on digital Leica M cameras as you increase the ISO, the top-end DSLR competition keep delivering the goods at staggeringly high sensitivities.

We will now have to wait and see if the new and much anticipated Leica M Typ 240, sporting a new 24Mp CMOS sensor, addresses this problem and puts Leica back in the digital camera game.

This article was mentioned by (tech) Onliner in Belarus!



I am seriously perplexed by this. If the sensor in the M9P is so poor, then why do the images look so great? Or am I simply deluding myself? Why would the M9P score lower than the M9 and M-E? Is it sample variation or is the sapphire glass somehow interfering with the sensor?

DxOMark is trying to assign a single number to the camera - as if it would indicate how good/bad it is, in comparison to other (and completely different cameras). It doesn't matter how good a camera scores if you don't like shooting with it. They're different tools. And it's really just part of the picture. It's also no secret that the M9 platform isn't a stellar high-ISO performer. At the native ISO of 160 however... Furthermore, they're comparing a 3+ year old CCD sensor to the latest CMOS and without lenses. As to why the numbers vary between models, that's a pretty good question. Sample variation? Margin of error? In the end, all that matters is what your own eyes see and prefer.

Fully one third of their scoring is Sports! What idiot would use a rangefinder M9 for sports and why would anyone enter a sports test for a review?

To be fair, their bread and butter is the dSLR, which are more typically used in sports. Their tests just don't really mesh well with the rangefinder.