Jump to content

Leica and DxOMark {MERGED}


Agent M10

Recommended Posts

  • Replies 142
  • Created
  • Last Reply
Guest tanks
Not bad. About where I expected from recent reviews. The M9 rated dynamic range is 11.7. Add 1.5 stops to that and it comes in at 13.2. Pretty close to DxO's rating of 13.3 Evsfor the M.

 

Overall the individual numbers are very close to the top rated cameras. Wouldn't you say that sensor performance is now not a reason to buy all of these top cameras? It now has more to do with lenses, AF, MF, RF and other characteristics that would be important to an individual. Maybe, now we can concentrate on photography.:)

 

Exactly. M was rated -1 stop below the top Nikon and Canon in low light ISO. That is where Leica lens shines, and makes up for it in spades.

 

Looks like moving to CMOS was the right move for Leica for sure.

Link to post
Share on other sites

Not bad. About where I expected from recent reviews. The M9 rated dynamic range is 11.7. Add 1.5 stops to that and it comes in at 13.2. Pretty close to DxO's rating of 13.3 Evsfor the M.

 

Overall the individual numbers are very close to the top rated cameras.

 

Yes, it is a significant improvment over the M9, on the other hand, in terms of noise the sensor is more than half a stop behind the top ranked Nikon and Sony FF cameras. Furhermore, the most recent 24 Mpixel APS cameras (Nikon D5200) has almost the same sensor performance. Whether all that matters is off course another question.

Link to post
Share on other sites

Advertisement (gone after registration)

Anyone remember the very detailed comparison by Erwin Puts of the Nikon D3X and the M9 ?

 

From memory it was neck and neck up to 640 ISO regarding dynamic range then the Nikon pulled away and was about 2 stops ahead at higher ISO.

 

If DxO was really useful I wouldn't see the Nikon D7000, a camera I know well and have made many A-B comparisons with, showing 80 against the M9's 69......

 

It's like Top Trumps, which some people like :cool:

 

Anyone want to measure THD of a high quality valve amp in comparison to a low quality transistor, then listen ?

Link to post
Share on other sites

There is - its you. In our quest for ever better results we forget that we are the ultimate decider. I can compare two image by looking at them - and whilst difficult, its not impossible to obtain identical files if you are really that determined. Numbers are all very well, but just how many of us can translate them into visual image 'quality' I wonder?

 

You are of course right, but my -or your- eyes do not sell internet clicks. Pretty graphs and meaningless ratings do...:rolleyes:

Link to post
Share on other sites

The DXO results seem generally consistent with CMOSIS claim of a 76 dB signal to noise ratio. The high ratio suggests that CMOIS design with the analog to digital converter on the sensor chip puts them ahead of Canon and as a practical matter equal to what Sony makes for the Nikon flagships. The improvement in M ISO that DXO reports over the M9 seems consistent with the 1.3-1.5 EV improvement over the M9 reported by those examining files from pre-production M cameras. My take on these data is that the new M is in fact what many of us have asked for: a rangefinder camera with the form factor close to that of the M3, an ability to use the best lenses available for hand held cameras, and a sensor that is the state of the art or very close to it. :)

Link to post
Share on other sites

Anyone remember the very detailed comparison by Erwin Puts of the Nikon D3X and the M9 ?

 

From memory it was neck and neck up to 640 ISO regarding dynamic range then the Nikon pulled away and was about 2 stops ahead at higher ISO.

 

If DxO was really useful I wouldn't see the Nikon D7000, a camera I know well and have made many A-B comparisons with, showing 80 against the M9's 69......

 

It's like Top Trumps, which some people like :cool:

 

Anyone want to measure THD of a high quality valve amp in comparison to a low quality transistor, then listen ?

 

I recall Puts concluding that the d3x had better image quality than the M9, but I cannot find said comparison on his new site. DxOMark found the M9 to have a DR of 11.7 vs 13.7 for the d3x, DxOMark measure camera DR's at base iso. The d3x did indeed perform better at higher iso's in Puts and DxOMarks' comparison. The DxOMark overall score for the d3x was 88, whilst the M240 scores 84, and the M9 reached 69, so the three year old d3x scored slightly higher that leicas newly released M240.

 

In the past I was concerned about such findings, but have come to realise these are just numbers and dont add to my enjoyment of photography. I sold my d3x and now own an MP with a 35 lux, and an R9 with a 100/2,8, both of which I enjoy more than my old d3x

Link to post
Share on other sites

I am waiting for somebody explaining us, that the "big step forward" of the M 240 compared to the M9 is just within the margin of errors, so that the test really shows that both models are more or less the same.

 

I am also sure that the following statement from DxO will be quoted in other circumstances:

 

"It’s interesting to note too that no RAW smoothing is applied to files from the new model, as was previously the case from ISO 400 on earlier versions."

 

How did they find out this?

Link to post
Share on other sites

On a sidenote, in the DxO lens tests, a Samyang 35/1.4 is rated the same as a Distagon 35/1.4, with the Samyang 2.0 being "better" than the Zeiss...:eek:

But then, the Canon 4.0/300 L is rated in the lowest 10% with 4 out of 20 for "sharpness";)

Link to post
Share on other sites

Well, in some of Sean Reid's comparisons of lenses you might see results which seem to imply that a lens which costs about 20% of another one is better in many respects. I just cannot remember the name of the producer of the very expensive lenses which don't always look best.

 

Unfortunately Reid has no rating system - you have to do the boring job of looking at photos.

Link to post
Share on other sites

"It’s interesting to note too that no RAW smoothing is applied to files from the new model, as was previously the case from ISO 400 on earlier versions."

 

How did they find out this?

They are calculating the correlation between neighbouring sensor pixels. If there is no correlation, there is no smoothing. See http://www.dxomark.com/index.php/Publications/DxOMark-Insights/Half-cooked-RAW/Noise-reduction. According to DxOMark, the new M does not apply in-camera noise reduction to raw data while the M9 does (at most ISO settings). Which should come as a surprise to some people.

Link to post
Share on other sites

All this talk of comparing todays high end equipment in laboratories makes me wonder how legendary phototographers from the past managed to shoot with such archaic equipment?! :eek:;)

An old trick - they judged it all by eye:D!

Link to post
Share on other sites

All this talk of comparing todays high end equipment in laboratories makes me wonder how legendary phototographers from the past managed to shoot with such archaic equipment?! :eek:;)

 

I don't for one moment think that the legendary photographers would not use the best equipment available for their needs:

 

Ansel a D800E ?

Capa a RX1 ?

Henri a Leica MM ?

 

While I often find the pointless technical disputes on some forums depressing, I think people who pay this much for equipment do have a right to see the purely technique scores.

But I prefer to think about "zone" quality

 

i.e. we can talk endlessly about the minute differences between a D800E, D600 and RX1, but IMHO they are all in the same zone of quality in terms of DR vs ISO vs Detail. You buy one of the other based on handling, taste and other characteristics apart from IQ.

 

There are oddballs, the DP Merrills are stunning, but its all ISO 100 work and Lightroom not allowed :(

 

So its an exciting market today, lots of choice, mostly excellent quality.

 

No excuses anymore, its all down to the photographer!

Link to post
Share on other sites

They are calculating the correlation between neighbouring sensor pixels. If there is no correlation, there is no smoothing. See DxOMark - Noise reduction. According to DxOMark, the new M does not apply in-camera noise reduction to raw data while the M9 does (at most ISO settings). Which should come as a surprise to some people.

 

Yes, since other testers found a decrease in resolution at high ISO settings, which they attributed to "smoothing" by noise reduction.

 

I have to admit, that I could not see this decrease in the examples given - but I believed in the statements that there is something to be seen.

 

Now either the decrease does not exist, or there is something else at work besides noise reduction.

Link to post
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...