Discussion in 'Other Systems' started by RT_Panther, Dec 15, 2012.
Battle of the Mirrorless - Part 1 (Low Light Performance)
Interesting, doesn't look that good for the OM-D so far, but as he says this is only part 1.
Look forward to seeing the rest of the results.
Thanks for sharing.
I think the results for the OM- D are about where I would expect them to be against APS-C sensor based cameras such as the NEX series etc
As we all know, and as the writer pointed out the overall performance of any camera is related to several factors more of which are yet to be discussed rather than just absolute low light performance
The images produced on the threads in this forum show clearly that the OM-D produces superb bright colourful sharp photos regardless of any individual performance measurements
The ability to do so is the most important 'performance measurement' and people here should perhaps not get hung up on individual performance comparisons to other cameras
The OM-D is like all Micro Four Thirds cameras, a compromise made to obtain a more manageable lighter more compact photography package.
The sacrifice that has to be made for the sake of size and weight is inevitably a little loss in performance in some areas
Olympus has (very much to their credit) reduced the margin in performance to a level where they have gained a huge number of converts to Micro Four Thirds from APS-C and FF sensor based equipment
This is not coincidental but based upon the images that are seen to be produced by the camera which are apparently close enough for these 'converts' to make them want to ditch their larger heavier alternative equipment
Any comparison work against other cameras should include that of the size and weight advantage which is made in return for the slight performance losses
I dont think many would argue that these slight losses are a small price to pay for the gains made in other ways
The bigger the sensor, the better it seems.
I know one thing for sure. Color and OOC-output of the E-M5 is one of the best in it class.
It is unclear from the text, but I suspect that the tester "matched" the field of view of the Olympus to the APS cameras by equalizing the long side of the image, like most reviewers do. This means an effective short side crop of the 4:3 sensor. If he would have matched the diagonal or, even better, the short side, the EM5 would have had more pixels in the same area.
Personally I am not interested in the 2:3 aspect ratio, I prefer more square formats like 4:3 and 5:4. I would therefore match field of view based on the short side of the image, since I would crop the long side of the 2:3 images in print anyways.
There is often an inherent flaw as well in these sorts of tests. As an example I noticed when I got my X100 that in Auto ISO it was bumping up the ISO way higher than I recall needing with my EP-2. For a test I set both cameras at something like 1600 ISO and took a load of pictures of a wall in our house and found I had to dial in -2/3 EV compensation to get both cameras to the same shutter and aperture. One of the contributors on here all also posts on dpreview said the same thing in comparing his X-pro1 and OMD so it seems to be a general trend between the two companies. In practise this means that I personally believe the claimed incredible high ISO performance of the Fuji's is only about half as good as claimed, there probably no more than half a stop to 2/3 stop better than m4/3 in reality. For these reasons I now tend to read very carefully how testers set up these sort of tests.
What a crock. Hardly surprising that a 24MP camer would hold more detail than a 16MP one. That doesn't say anything about the noise itself where the E-M5 is clearly better than the NEX7 (I have both, love both, shoot both and have many real world comparisons of my own). Or the ability of a camera to find focus in low light.
Low light performance is a mix of different factors. Detail retention, noise and the ability and speed of a camera to actually lock on focus. Not just how one camera can hold a bit more detail information because it has more pixels.
Then he makes a definitative statement about colour being better. Colour preference is subjective. It's not like Sony, Canon, Nikon or Olympus have glaring holes in the colour spectrum. They're just different.
It seems to me like another site presenting a personal opinion as fact.
In my opinion, these sort of tests should be done on FULL AUTO and also FULL MANUAL with the same settings on all the cameras with no exposure compensation etc.
That way a potential customer can see how each manufacturers handles it's auto mode and also direct comparison of each system at a standardized setting. On full manual, it may lead to some cameras being underexposed or over.. but that's the point.
Also, most of the comments should be written as "as expected, the APS cameras handle noise better then the m43s and much better then 1" technology. So no surprises found"
At the end of the day, a camera is just the mechanism to take photos. If you look at almost any photograph taken more than 10 or 15 years ago and all the way back to the earliest photograph you realise that the ISO we now have is an incredible luxury.
high ISO in itself is not a panacea to all photographic ills
that said, I am enjoying using 3200, 6400 and even 8000 iso on my OMD... do I need even higher?... probably not, though when I buy my next camera it will probably give me a stop or two more... but in the meantime I am more than happy
OMD 75mm ISO 3200
PB220030 by kevinparis, on Flickr
6400 ISO..... this was in a bar where it was too dark to read a magazine... not a perfect exposure due to user error - there is a bit of banding...but I did get the shot I wanted
PC070125 by kevinparis, on Flickr
75mm ISO 8000 - a dimly lit dinner - even in colour this shot is more than acceptable for my needs
PB230119 by kevinparis, on Flickr
the camera, the lenses, the IS and most importantly the skill and understanding of the photographer are what makes a photo... high ISO performance is just a small part of the equation
Fuji systems also weren't included...
Just remember it's PART ONE & not the whole test...:smile:
I looked through the article. I thought the E-M5 did just fine. This is one of the tests that would most naturally give the edge to the bigger sensor, and from what I saw it wasn't much of an edge in results. THe message I took away was if low light performance is everything to you get a D800
I did my own comparison between the E-M5 and a NEX-5N (with the same 16MP sensor as found in the F3, 5R & 6 tested here) and I reckon that the Sony is at least 1/2 stop cleaner than the E-M5 at higher ISOs after making allowances for the differences in stated versus actual ISO sensitivities. Theoretically that means that the 16MP NEX cameras are better in low light, but of course that is before considering whether the NEX system has lenses available with equal maximum apertures, which as we know is not always the case.
Where you do have lenses of equal maximum aperture you then have the oft-quoted "depth of field control advantage" of the larger sensor, but do I want to shoot everything wide open with a paper thin depth of field? Only if I have a single subject or am shooting at a flat plane. To equalise depth of field you'd be shooting the APS-C sensor with an aperture closed down by 1 stop and the ISO raised by 1 stop to compensate.
It's not for me which is why I still have my Sigma DP2 :smile:
It al comes down to why do we crave higher iso... is it to shoot in darker places? or is it to keep shutter speeds up?
to take a photo you need light... light enough to see the subject... then you need an aperture and a shutter speed that will capture the image. you do have to have some realistic expectations regarding this... you are never going to freeze the movement of your running kid in a dimly lit living room with the kit lens...even if your iso went to two stops beyond what the top cameras today offer.
3200 ISO.. which is way beyond what anybody had 10 years ago... is pretty damn amazing... 6400 is very acceptable... 8000 is workable
I am sure in 5 years we will have totally noise free 128000 iso.... not so sure that good photos from good photographers will be very different...
Which proves what an utter waste of time the test was then to not include the mirrorless system with the claimed best high ISO performance.
FUJI will be included in part 2
The EM-5 was not as much different from the APS sensored cameras as the APS were from the FX sensored Nikon! Although m4/3 has gotten better in ISO performance so have the others and I figure it will remain so, at least a stop or so difference. BUT how many of you are actually shooting at those extremly high ISO settings??? On a rare occassion I might bump it up to 1600 or even 3200 but very rarely. I think being able to even have ISO settings that high is amazing.
I agree, probably 3200 is the most I ever use and even then rarely.
For me at least it's academic how cameras perform at 6400 and above.
Hahhahaah .. interesting but useless ... I dont think many of online reviewers or bloggers actually use these cameras at all. Their interest seems to be how camera A performs against B , B against C etc etc .. In this modern era of technology none of these cameras are inferior and they do the job if someone knows how to handle them . What I have learnt so far is that such comparasions are useless and pointless and dont make anyone a wise photographer.IMHO , We should start analysing ourselves now rather than trying to find faults with cameras.
Separate names with a comma.