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!

DXO Em-1

tashley

Subscriber Member
Well - it's here.

Pretty much exactly as I would have expected: the ISO performance takes a small hit by the addition of PD pixels, the DR and colour depth are a touch better. It doesn't score fantastically overall (though it is the best MFT so far) and fails to catch up with the better NEX cameras, as one might expect. And it is, predictably, miles off the M240.

HOWEVER... of course it has IBIS and, in my experience so far, that means that it is often possible to keep shooting at lower ISO as the light drops than with much of the competition. Given that the colour depth and DR are only slightly (and in real world terms largely irrelevantly) worse than some of the cameras it competes with, one might reasonably argue that if the ISO performance is 'real world adjusted' as a result of the IBIS, it is worth a 78 rather than a 73. I also award it an extra five points on the basis that the forthcoming 12-40 F2.8 sounds like the only zoom on pretty much any system that is 'sharp enough' across the frame at all focal lengths to meet my needs. There's no point having amazing colour, DR and ISO performance if the lens is soft at the edges..,

Consequently, love DXO though I do, I hereby award the EM-1 a total of 83 points. But then I deduct two points because Olympus have no idea how to configure an Auto ISO system so as to allow shutter speeds to drop appropriately to focal length in order to keep ISO low - so I have to jump through some weird hoops to make it work as it should.

So that's an 81. Pretty nice!

:D
 

Godfrey

Well-known member
I ignore DXO entirely. I've never found their numerical ratings indicated anything that made any difference to what I want a camera to do. Just don't see the point of all their measurement and analysis ... I can see what the camera does with my own eyes far more clearly.

G
 
V

Vivek

Guest
I ignore DXO entirely. I've never found their numerical ratings indicated anything that made any difference to what I want a camera to do. Just don't see the point of all their measurement and analysis ... I can see what the camera does with my own eyes far more clearly.

G
I do as well. :)

It is a pity, the highly regarded (in some circles) focusnumerique also "mess up" things.

For example: Test Sony A7R - Focus Numrique

They were trying show how the A7R ranks in terms of noise. The whole thing turned out to be a great advertisement for the Leica M. :ROTFL:

Tim still remains an odd man out (I say it in the best possible way), in my opinion.

This post, Tim, reminds me of a certain cuisine where they make use of Tabasco sauce and other bottled/canned goods for cooking (we were watching DVDs of "Chef", recently. :D )
 

retow

Member
The result is neither impressive nor is it disappointing, but simply the expected. The sensor does not seem to support the hype around the camera. Maybe the next generation will.
There is obviously a not too subtle difference between mft and aps-c sensors, still. I find the "it does not matter for most users and uses" or "in the real world" argument to explain away key IQ differences amusing.
 

tashley

Subscriber Member
Ooo Vivek, you're SO RUDE :D

Obviously when I want to cook for guests, I use the finest, freshest, most local and seasonal produce available - but my solitary vice of bottled ingredients is not a guilty one: I say it loud and proud - "I use MFT in public sometimes and I don't think it 'tastes' too bad... as long as you don't compare it to haute cuisine at the same sitting...'

I do as well. :)

It is a pity, the highly regarded (in some circles) focusnumerique also "mess up" things.

For example: Test Sony A7R - Focus Numrique

They were trying show how the A7R ranks in terms of noise. The whole thing turned out to be a great advertisement for the Leica M. :ROTFL:

Tim still remains an odd man out (I say it in the best possible way), in my opinion.

This post, Tim, reminds me of a certain cuisine where they make use of Tabasco sauce and other bottled/canned goods for cooking (we were watching DVDs of "Chef", recently. :D )
 

Godfrey

Well-known member
The result is neither impressive nor is it disappointing, but simply the expected. The sensor does not seem to support the hype around the camera. Maybe the next generation will.
There is obviously a not too subtle difference between mft and aps-c sensors, still. I find the "it does not matter for most users and uses" or "in the real world" argument to explain away key IQ differences amusing.
The sensor (vis-a-vis some absurdly high expectation of vastly increased sensitivity, acutance, or whatever the IQ noodniks seem to be desperately looking for) is not a huge part of the "hype around the camera" for me.

Ergonomics, responsiveness, improved capability with my existing FourThirds lenses, durability, consistency: These are what I am enthused about. Image quality was never even a moment's thought ... I knew that would be fine without even having to think about it.

G
 

Jorgen Udvang

Subscriber Member
I find DXO difficult to pronounce. I'll start using ADOX instead. Easy to pronounce. Goes better with my OM-2 as well. DXO never tested the OM-2, did they? ;)
 

ptomsu

Workshop Member
I never gave too much attention to DXO and I find their methods just absolutely nothing saying for real value of a photographic tool.

For me the strength of the m43 system with EM1 lies in the overall availability of different lenses, the usual high quality of these lenses (at least the non budget ones) and if the EM1 is just as good as the EM5 that is more than enough. BUT - it is even better! So this is currently the dream system with the EM1!
 

tashley

Subscriber Member
I have to say that if I were to be asked to rank my stupidly large collection of cameras in terms of their 'native' IQ (i.e. a lens-independent assessment of their overall file quality) I think that I would come up with a relative order pretty much exactly the same as they way they appear at DXO. So, as I have written elsewhere, if you really understand what DXO does and doesn't aim to tell you, it can be an extremely useful way of telling, before you buy a camera, how good its IQ will be...
 

etrigan63

Active member
I'm with you on this one Tim. DxO may not be a perfect reflection of real-life performance, but it does give us a relatively quantifiable method for trying to rank cameras by their abilities (most of which are subjective anyway). I am more interested because that now means they have E-M1 profiles for their excellent DxO Optics Pro application.

The name "DxOmark" was inspired by "PCMark" and "3Dmark" benchmarking tools for computers. Both of those have often been accused of not reflecting real-world performance, but they are certainly beloved by PC review sites. They are also very capable of beating the crap out of your PC, so you can at least get an idea of how your system will behave under unusually heavy loads.
 

raist3d

Well-known member
DXo: it is what it is. Can be used well or abused. If you have to spend cash and you are undecided between two options because they both look good, a DXo look for a sensor performance can solve the deal breaker.

Or helps to understand where your tool falls in the universe of sensor IQ and if you need more of something, where to look for.

The problem is - at least for the EM5/EM1/m43rds and the like is the constant lie propagated in many places that it is just as best APS-C in terms of image quality. Many may say that's not what they are looking at but that sure is one of the first things that is looked at by reviews, etc.

The truth is pretty simple- everything else equal the EM1 will not perform as good as a Sony APS-C sensor. I mean, Sony makes both sensors, you don't think they are just going to give a m4/3rds sensor something new or anything.

And the part that should be seen is that, hey! That's fine! The current m4/3rds sensor is pretty good and more than good enough for a wide range of work. People should just pick according to need and preferences, no need to try to be soothing you are not or put down something you have a bit less off etc.

There was a point where both Olympus and Panasonic where clearly left behind in sensor tech but they finally are at the place relative to the same technologies more or less for their sensor size.

It's all tradeoffs. Pick smaller and less IQ (but still good) or higher IQ and a bit bigger (APS-C) or ultimate high IQ and bigger (FF, but less than medium format).

I would just look at DXo mostly as a guidance of what to expect more or less when using the equipment and even different models in the same format. For example- it's good to know that the sensor in the EM5/EM1 is that much better than the old E-5 sensor, for someone using 4/3rds and looking for a real upgrade.

And of course, if you pick something up, you like it and you are doing and getting what you want, then it really doesn't matter.

- Ricardo
 

jonoslack

Active member
Hi Ricardo
The problem is - at least for the EM5/EM1/m43rds and the like is the constant lie propagated in many places that it is just as best APS-C in terms of image quality. Many may say that's not what they are looking at but that sure is one of the first things that is looked at by reviews, etc.
Actually - it seems to me that the constant lie is that there IS much difference between APS-C and µ43 - at least, if you're satisfied with the 4:3 aspect ratio.

They are usually compared in terms of sensor area, but IMHO the best reality check is to compare the pixel pitch:

Nikon D7100 (best APS-C): 3.9µ
Olympus E-M1 (best µ43): 3.7µ

Nikon D610: 5.9µ



That's 5% difference between APS-c and µ43 - honestly - that doesn't mean that the image quality cannot be as good. The difference between APS-C and full frame is where it really matters.

The reduction from 24mp to 16mp is mostly taken up by the width of the sensor.
 

raist3d

Well-known member
Hi Ricardo


Actually - it seems to me that the constant lie is that there IS much difference between APS-C and µ43 - at least, if you're satisfied with the 4:3 aspect ratio.
Sorry I can't quite agree. Since the EM5 I see lots of reviews saying it's just like the same. Same with previous (look at Robin and a few others).

They are usually compared in terms of sensor area, but IMHO the best reality check is to compare the pixel pitch:

Nikon D7100 (best APS-C): 3.9µ
Olympus E-M1 (best µ43): 3.7µ

Nikon D610: 5.9µ
The surface area is about 38% more for APS-C vs m4/3rds. That is a significant difference. That's like one stop and the tests do corroborate this. Open up a K-5 file - 16 MP and compare vs the EM1. The K-5 sensor has deeper shadows, tonality, DR. True used 14-bit RAW. You hardly even get a clear noiseless sky at ISO 200 to begin with.

By the way your quote of the pixel pitch vs the D7100 is completely misleading- the D7100 has 24 MP, not 16MP. What this means is that you can downsample the D7100 for an effective cleaner image to match the 16MP. A more valid comparison would be 16MP D7000:


4.78µm

That's a (4.78-3.7)/3.7 * 100 = 29% difference, not 5% difference. Sorry but that's to me significant.


That's 5% difference between APS-c and µ43 - honestly - that doesn't mean that the image quality cannot be as good.
Yes, it necessarily means it won't be everything else equal, because it' snot 5%, it's about more about 1/3rd.

The difference between APS-C and full frame is where it really matters.
I can't agree- I say it depends. To me a full stop DR/ISO is a notable difference.

The reduction from 24mp to 16mp is mostly taken up by the width of the sensor.
Look at the pixel pitch I mentioned above. When you have 24 MP the downsampling quality is proportional to that increase. And remember it's not just ISO, but tonality (14 bit) , DR, and color sensitivity.

Given the K-5/D7000 are 16MP, I think the comparison vs the EM1 is more than valid and there is indeed a notable difference between m4/3rds and K-5/D7000. The later clearly have deeper shadows, higher ISO performance, etc.

Now, if this makes no difference to your photography and what you want to get out of the camera that's entirely another matter.
As you know I am using a Q7 for a lot of my photography and for a lot of the subjects I do the sensor is not making a major difference in the quality images from a photographic point of view that I am getting for this kind of work. That doesn't mean I am going to think it's just the same as say a Sony 1'' sensor.

- Ricardo
 

jonoslack

Active member
HI Ricardo
you can argue this around and around - I'm not getting into a Pi$$ing match with a porcupine (not to suggest you're a porcupine :)).

There are lots of different ways of cutting this particular cake, I could start talking about vertical linear comparisons being more relevant than area, but honestly. it won't get either of us anywhere.

What bugs me is that the classic sensor size comparisons (showing the top right) very much mitigate against 4:3 sensors (which is of course relevant if you're going to cut your images down to 3:2, but not if you aren't).

I'm not arguing against the DxO sensor scores (82 for K5 II vs 73 for the E-M1) I'm certainly not denying the extra stop of dynamic range either . . . although, as Tim points out, the excellent IS on the Olympus will often give you that stop back. . . but I'm not very convinced that it's very significant - I have to say that the number of times I run out of dynamic range with the E-M1 are few and far between. . . . . . . on the other hand I never found one decent zoom to go with the Pentax (and that IS significant). . . . and if I'm going to shoot primes then I'll take the full biscuit and go full frame.

I was just disputing your statement that it is a "constant lie propagated in many places".

Nobody is denying that the sensor on µ43 is smaller - just that it doesn't have a significant effect on image quality (I think dPreview just said that). Significant being the operative word here I think.

There are lots of factors that go to make up good image quality. As sensors get better and better, sensor size becomes less and less of a key factor (IMHO of course)

All the best
 

Georg Baumann

Subscriber Member
There are lots of factors that go to make up good image quality. As sensors get better and better, sensor size becomes less and less of a key factor (IMHO of course)

All the best
Hi Jono, Just a thought, meant as a sidenote.

I am often struck by the sorta academic discussions, Nyquist, Raleigh, pixel pitch, you name it, that can be found on many fora concerning technological aspects. - German fora in particular are a flippin pest in that regard! You would think that most of them would be able to engineer and market a much better IQ260 / A7r / D800 / Em-1 etc., if only they could do as they wish! :ROTFL: -

IMHO, Image quality got better and better in spades over the past decade, down to the point that some DSLR can stand up to MF quality from not so long ago costing an arm and a leg.

In a way, the Magnum CEO nailed it for me in an interview with the Britiish Journal of Photography, explaining their move towards a subscription model, when he stated, "The photography world has changed, everyone is a photographer today!"

To my simplistic understanding, in my world, the question is simple. Does current technology give me the quality I want in a print, and at what cost. To be honest, thinking back not so long ago, I was amazed what quality I was able to pull from a Olympus E-1 file in print.

Ah well, sorry for ranting, back to my chinati :grin:
 

raist3d

Well-known member
HI Ricardo
you can argue this around and around - I'm not getting into a Pi$$ing match with a porcupine (not to suggest you're a porcupine :)).

There are lots of different ways of cutting this particular cake, I could start talking about vertical linear comparisons being more relevant than area, but honestly. it won't get either of us anywhere.
Oh come on now :) You are the one who brought up the photo site size ;-) What I posted is not an opinion but a fact. There's nothing wrong to have the discussion ;-) What I said is pretty straight forward. Color sensitivity/richness/DR/ISO are all a difference here. Whether that difference makes a difference to you as a photographer is a different thing. Hey, like I said, I am shooting a Q7!

What bugs me is that the classic sensor size comparisons (showing the top right) very much mitigate against 4:3 sensors (which is of course relevant if you're going to cut your images down to 3:2, but not if you aren't).
Well again, it all depends, as I said. Keep in mind 3:2 is full frame and there's clearly a vast photographic history supporting that kind of ratio/format. I actually prefer a 4:3 ration more overall, unless I am going for a more cinematic feel.

I'm not arguing against the DxO sensor scores (82 for K5 II vs 73 for the E-M1) I'm certainly not denying the extra stop of dynamic range either . . . although, as Tim points out, the excellent IS on the Olympus will often give you that stop back. . . but I'm not very convinced that it's very significant -
First the DXo overall score is kind of meaningless, what is interesting is to see the individual areas and match those accordingly to the photography you do.

There's also IBIS or IS in other systems too. While the Olympus IS is great, it doesn't make the other IS useless. The K-5 under comparison has actually as you know body IBIS and was actually the first to even compensate for tilt.

I have to say that the number of times I run out of dynamic range with the E-M1 are few and far between. . . . . . . on the other hand I never found one decent zoom to go with the Pentax (and that IS significant). . . . and if I'm going to shoot primes then I'll take the full biscuit and go full frame.
Well again, as I said, it's up to each personal needs. For you that works and that's great.

I was just disputing your statement that it is a "constant lie propagated in many places".
See below :)

Nobody is denying that the sensor on µ43 is smaller - just that it doesn't have a significant effect on image quality (I think dPreview just said that). Significant being the operative word here I think.
Thanks for proving my point above :) Yes, dpreview said that. Robin Wong says it's just as good as best APS-C. You see Kurt saying that dpreview said exactly that and he misquoted dpreview. Dpreview also says Xtrans Fuji is clearly above the EM1 performance, but nobody mentions that.

People could see this for themselves. Of course, significant here is going to be something that varies per photographer, it's up to each photographer.

There are lots of factors that go to make up good image quality. As sensors get better and better, sensor size becomes less and less of a key factor (IMHO of course)
All the best
Certainly but I see a convenient line drawn very often. Why stop at m4/3rds? Why not say 1'' Sony/Nikon 1 is more than good enough? As I mentioned I totally agree m4/3rds can be more than good enough for a wide range of photography. That's different from say that the difference is so minimal as to be just like APS-C. Which is what is being said often. Yes, 1 stop is quite significant- to me. But to someone else it may not be.

- Ricardo
 
Last edited:

raist3d

Well-known member
To my simplistic understanding, in my world, the question is simple. Does current technology give me the quality I want in a print, and at what cost.
And this is pretty much one of the main points I have been saying all along. The problem I point out is that I see a lot of the 4/3rds, m/43rds owner going on pretending it's just as good as best APS-C or so close there's no difference, in what seems to me an attempt to convince themselves it's ok to shoot that format or something.

What I say is another way of what you said, accepting the format, the quality it gives you and cost and moving on. They all have tradeoffs.

- Ricardo

PS: I understand what you say about technical discussion but not everyone doing technical discussions are on equal terms. Some of us actually do get out and create photographs instead of trying to tell companies how to design better sensors :)

I invite you to take a look at some of my more recent work in the "Fun with Pentax Q7" thread, over in the Pentax forum.
 

jonoslack

Active member
Oh come on now :) You are the one who brought up the photo site size ;-) What I posted is not an opinion but a fact.
No no, Ricardo - you started this with the 'lie' word - these are all opinions (both yours and mine), and different constructions on matters of scale - enough said I think.

. . . . and 'better' is such a complicated concept :salute:
 

raist3d

Well-known member
No no, Ricardo - you started this with the 'lie' word - these are all opinions (both yours and mine), and different constructions on matters of scale - enough said I think.

. . . . and 'better' is such a complicated concept :salute:
This was my claim:

...the like is the constant lie propagated in many places that it is just as best APS-C in terms of image quality.
I stand by it. I can make the list if you want.

Notice I didn't say all places. I elaborated that this tendency seems to have gone up a bit and some blogs that should know better are saying it. I also completely agree that this does not mean you can't get excellent image quality in that format, and that it's up to one preferences. I even provided further elaboration of that by the reminder that I shoot with a Q7 a lot.

There's nothing complicated in Better DR/ISO/Gradation/color sensitivity - it is pretty factual, not an opinion. Whether the camera system, or the ergonomics appeal more to you (said in general) is another matter (I bring up the Q7 example again).

The whole photo site thing you brought up was was clearly off :) And hey, cool but you did bring it up. Nothing wrong with owning to it ;-)

:angel:

Cheers,

- Ricardo
 
Top