By Request: Full size from RAW Olympus OM-D E-M1 Images
Hello to all! I just arrived back home from Ireland after an 18 hour travel day, delayed flights and missed connections but hey, it was worth it just to go to the gorgeous castle and put the E-M1 to test in a beautiful environment (see my initial report HERE). The weather was not so hot on the day at the Castle which required me to up the ISO on the camera but I still feel the images are superb. I have had a few email requests for full size images, from RAW from the camera. So here you go!
Enjoy and have a great rest of the weekend! I will be back Monday morning!
You must click on each image to get the full size. Text is embedded on each image with the lens used as well as ISO.
It’s very clear that a small Micro Four Thirds or APSC sensor simply never beats a full frame sensor. There will always be about two stops difference in high ISO performance. Also the dynamic range differs from a full frame sensor. So, if you need less noise and generally better image quality, full frame is the best option. If you need a portable lightweight camera, full frame also is the best option (thank you Sony). However cute these little MFT and APSC cameras are, I think they are a dying breed.
Hi Steve
Wonderful Images I need to know when I shoot in RAW I see a white frame around the image in viewfinder and in Olympus Viewer not sure how to get rid of it but when I export to JPEG the White frame is not there
hello steve. I am very interested to know if I can make prints of a meter or so with files of this wonderful machine
Amazing that the whole camera world seems to believe that a larger ff sensor “captures more light” and produces less “low light noise”. It does not. FF lenses has bigger apertures than m43 lenses and they therefore put more light on the sensor which then produces less light. By “aperture” I do not mean f/ratio, but linear aperture as measured in millimeters.
Big hole in the lens = more light put on the sensor = less noise
So it all comes down to how large lenses you are willing to carry. Large lenses will always collect more light and produce less noise when light is scarce. Ask astronomers who build lenses with 5 meter apertures.
Forget sensor size. As long as the sensor is made to fit the lens, the only thing that matters for noise is the size of the aperture that brings in the light.
Arne, is it not the case that the size of the individual photo sensors have an impact on signal/noise performance, and that for a given pixel count a larger sensor can have larger sensors – leading to improved low light performance over a smaller sensor with same pixel count? This is a genuine question by the way – I am interested in the answer (academic though it is – I am an OMD using m4/3 convert).
sorry, spotted a typo should have said “larger sensor can have larger PIXEL SITES….”
No, the size of the pixels are not important, but the number of pixels are important.
Given a lens with a certain size aperture in mm you know the total amount of light that falls on the sensor. If you divide that by the number of pixels, you then get the total amount of light per pixel. The total amount of light per pixel gives you the basic noise level.
It is impossible to make a good low light camera with a small lens, but it is quite possible to make a good low light camera with a small sensor. Look at the Hubble telescope. The lens aperture is over 2 meters(!), but it doesn’t matter that the sensor is still small.
Of course, you can make a fairly good low light camera with fairly small lenses. I love my E-M5. 🙂 But up-to-date cameras with much bigger lenses will always have lower low light noise.
Many smaller cameras, like the E-M5, are also approaching the point where their low light noise is just good enough for almost all common use cases. It will matter less and less that a bigger lens would have given you even less noise. Ten years ago it was very important to have a fast PC, today most people don’t care because almost all PCs are fast enough for them. Instead people look for portability.
Thanks Arne. Of course small as the OMD lenses are physically – there are some pretty decent max apertures available, which as you point out is key to the low light capabilities of cameras like the OMD.
Thinking over this, it would seem that for a given pixel count (within reason of what can be squeezed on a smaller sensor), there is no advantage in a larger sensor which derives from its size alone – other than the possibility of achieving a shallower depth of field for a given aperture size (which is an advantage in some applications, though not all). Or am I missing something?
DOF also depends on aperture size, not sensor size.
(However, if all you have is f-ratio then one to get the aperture size is to multiply by sensor size.)
Good comments Arne. Some “total light” people have some strange theories and ideas. If you extrapolate their theories then it is impossible for a phone to take a photograph and Nikon wasted money on the 1 series. Their pet saying is ‘total light on the sensor’ this of course means nothing. Its a lot like having a measure of tea in the cup, or a length of string, How long was it? or being ginormously huge.
If one takes a picture with a 35mm sensor and lens. Then magically removes the sensor and replaces it with a m43 sensor of the same pixel density and efficiency, one gets exactly the same results as a quarter size cut out in the middle of the 35mm shot. AOV, DOF, noise the whole lot.
So if one wants to print on the side of a house get a 35mm. If A3 is the largest you will ever print occasionally then Nikon 1 sensor size up will suffice.
Anyway noise software is very efficient. I use Topaz its great. One does not clean dishes in a dishwasher with only water so why rely on fixed camera settings when the light is variable..
I’m not sure what is spectacular with this output; looks okay, but it’s just that. Many good APS-C cameras using Sony sensors had this level of quality 3 years ago.
Sony makes the sensor
A camera is not only about the sensor size for output quality.the highlight of M43 system is easy to use and good balance.Smaller than APS-C system but the same or better image quality(better lens).and the EM1 has high speed AF and bigger raw butter(51,better than 7D,nearly same D4),good weather proof ability,even freeze proof.more silent and stabilization（no mirror action).All the feature are good for geting better output quality easily.Big sensor size camera doe not mean you always can get bigger quality image, sometimes you need tripod for them and sometimes you can not take it with u so you get nothing.Lots of people always say bigger sensor get more light,but I say bigger sensor need more light as well.I am food photographer and I shot food mostly by nature light.For example,I use Olympus EM5+M.Zukio60mm F2.8@ F4 ISO200,1/30 shutter speed by nature light.if shot by full frame for the same DOF,the aperture is F8.if I want keep the same shutter speed,I need more light,maybe using flashgun or increase the ISO till 800.So ISO200 vs ISO800,the high ISO advantage of full frame is not in there,full frame downsize till M43 size maybe can get the same output quality.you want higher output quality,you need strobe,you need more light to light on the subject.So it is not just only a camera,it is a system..
For my options,I think Steve turn the” auto level” on when shot these photos.
Auto Levels? Ummm. No. 1st of all, there is no Auto Level when shooting in camera. These were from RAW using ACR.
sorry,steve,it is not level,it is gradation. there are 4 options,auto,low key,normal,high key
If that be set auto,seem like the level move toward right,so some noise will come out from the shadow area.
Another thing is the ACR is not good for Olympus files,the color,tone are very different..
These were all from RAW and even if the camera was set with the Gradations to on or auto, it would not matter in the RAW file.
Any chance you could post an actual RAW file, unconverted? There are some RAW tools out there that are starting to come available so I would like to see how they do with a RAW file.
I really like this camera and want to buy one as an ‘everyday’ camera so as not to have to lug my 5dmk3 everywhere… but I am worried about the sensor.
I went to a launch in London and tried it out. It felt fantastic, etc., so I went and bought an SD card and took some shots to look at later at home – I could see large fine JPEGs only as I didn’t have any RAW converter. I was surprised at the reticulation (very apt term, Chris), even at low ISOs. I bought a GF1 when they came out about 4 years ago or so and looking at those files now, they do not exhibit the same sort of… well, ugly traits, at least not this prominently anyway.
I’m not talking about characteristics relating to sensor size, dynamic range, bokeh, etc. I’m fine with that side of the coin and all sensors will have evidence of noise, digital structure, etc. in the end if you keep on peeping in, even at low ISOs. But anyway, there are m43 sensors that at this point in time are clearly better in this area regarding reticulation, even 4-5 year old ones. I took the same SD card to a camera shop and took some shots with an EP-5 with the same lens (45/1.8) and it seemed a lot better. Maybe there is some pre-production firmware funnyness?
Please tell me I’m wrong because Olympus have got so much right here!
On another point, in all I can glean off the internet, all tests regarding continuous focus tracking seem to show horses, bikes, etc., going basically across the screen… not very hard test of tracking ability. Or they’re shot form far away where the focus system does not have to work as hard as, say, when a bicycle coming quickly straight at you from, say, 20m to 2m. Why is this?
thanks for any insights!
Andy Elliott
” On another point, in all I can glean off the internet, all tests regarding continuous focus tracking seem to show horses, bikes, etc., going basically across the screen… not very hard test of tracking ability.”
Yes, it would be nice to see examples on how it does handle fast subjects, expecially with 4/3 lenses – some reviews say that it is better than with E-M5 and acceptable, others that is very good, almost on par with 4/3 bodies.
However, after a week’s concerted resistance, a smidgen of hope and a dollop of anxiety, I have placed my order and slapped down the deposit!!
Does the first girl portrait show severe banding in the shadow areas or am I mistaken?
Looking to that on a 42″ full hd screen I don’t see any banding, perhaps it depends on color calibration? I’ve done it recently, check yours…
Yes, it does need refreshing, but that wouldn’t affect banding I would think. We’ll see.
Thanks for these samples. Really shows better skin tones compared to the (VIVID!) jpegs, and also how much better the delivery is w/o the automatic sharpening of the jpegs.
I see very little difference in IQ to my used $500 OM-D E-M5 but if I had the money I would upgrade for the extra buttons and peak focusing. Is the extra HDR button easy to use?
I can’t figure out how to get a HDR shot with my OM-D. Since it was a cheap camera it came without box and manual.
Download the manual here: http://www.olympusamerica.com/cpg_section/cpg_support_manuals.asp?id=1583
Great images, the fuji has it beat though in image “quality”. In usability it probably blows xpro out of the water; seems it’s a tradeoff.
Steve, get yourself over here more often, uk and eire have amazing coastlines, and the joke about the weather is, “if you dont like it a different weather will be over in half an hour”
I was seriously considering moving to 35mm (D600/6D) until I started looking at lenses. Virtually all the available lenses bar the extremely costly models suffer from soft corners, huge vingettiing, CA etc. By the time you stop down to f5.6/8 to improve you are not really gaining much at all. People keep comparing lenses as if they will behave the same across formats. This is not the case.
Yes, FF does have its challenges as well which is why you need the big buck lenses to make them really sing. The good Micro 4/3 lenses will not have this issue 🙂
If these images are from the camera JPG particularly all settings of camera should be evaluated and considered not just the exposure settings. For example if I relied on the EM5 in camera JPG I would consider reducing sharpness to below zero and turning on noise reduction pending the ISO level.
The great thing about the EM5 and I am sure the EM1 is that the camera can be set to your liking to produce great results from out of camera JPG. You just need to know how to do it and experiment with what works for you as even skin appearance can be modified.
JPG’s are not my main interest as I like post processing from RAW. In this regards the EM5 does more than any enthusiasts needs and is hard to beat in this segment.
The EM1’s added fancies will surely entice those ruled by GAS and those that finally realise they don’t need the extra weight to capture the moments and memories.
The EM1 does not replace a higher end FF camera rather is an alternative that needs to be well considered.
Great images Steve! Looks like your trip was a blast and well rewarded. I really need to get myself a 45mm.
Looking at these full size; this is not really what skin should look like, is it? The sensor is just to tiny to deliver anything close to FF. But: How big must the print be to see any difference between this and FF?
Raw files implies no white balance tuning, no noise reduction, etc. Keep that in mind. I sold all my FF canon gear (5d + L lenses) and switched to m43, never had any regrets as I never liked the files of the 5dII. Of course, my M240 delivers gorgeous pictures, but for work, EM-5 is very reliable and consistent.
To be honest I’m quite disappointed by the noise.
The background of the first picture taken at 640 ISO is really noisy. I don’t get that with my EM5.
Even the second picture is noisy while taken at 200 ISO.
Is that coming from the transformation into JPEG?
Well, there are a couple of reasons for the noise. 1st and foremost, Micro 4/3 will NEVER give you a smooth silky clean image when viewed at 100%. I also turn off and never ever use any noise reduction. (most leave it on as default, I turn it off as I prefer sharp and noise to mush). So when you have zero NR and viewing a 100% file at ISo 400 and up you will see noise when compared to a full frame sensor. Just physics, and has always been like this – same with the E-M5, GX7, etc.
Also, when shooting in that weather and using – Exposure Comp as I did in some of the images it will also exaggerate noise a little. At the end of the day, and as I said in the video, there is no way it will ever match the output of a full frame camera. If it did it would be game over for every full frame made because the E-M1 feel, performs and is much smaller than any of them. It would then cost $4000 instead of $1300.
Micro 4/3 is not meant to compete with full frame bu it does compete with APS-C. This camera slaughters, kills and destroys ANY Fuji for usability and speed and even video as well as amazing Lens availability.
A camera is much more than the IQ alone. To enjoy it, it has to be high in the USE department as well. The E-M5 was and the E-m1 is even more so.
I just received a GX7 and have been messing with it for a few hours. Seems like a nice camera as well, and has the same Micro 4/3 characteristics.
As for the noise at 100%, there will always be some, even at ISO 200 but in print or resized web shots you will not see it. Only for those stressing over 100% on screen views 🙂
Dear Steve,
beautiful shots and interesting comments.
Concerning the ‘superiority’ of FF…sorry, but I am not so sure at all.
You claim physics, as everybody does, but which physics?
Recently, I have seen stellar images taken with a Pentax Q 10, crop factor 5,53.
So, why is FF better? Okay, the larger sensor may capture more light, but I am afraid that there is not one single physical reason which might underline an advantage of FF compared to any other formats.
Thank you for seriously considering my ‘query’ – I would be most interested in your kind reply.
Again, congrats for those wonderful images from Ireland!
I’m not sure either. I’ve just acquired an as new Leica Digilux 2 from 2004 (2/3 sensor, 5MP). The first OOC jpeg results are very pleasing and to my eye demonstrate that the relationship between sensor size, pixel size and lens quality is perhaps the most decisive factor rather than overall high numbers in one department or another. And of course I get manageable file sizes……
But I’ll still lust after an EM-1 ! And thanks for the pictures, Steve. I particularly liked all three portraits.
I used to have a D700 with which I keep a really good quality when I print a color picture in 100x150cm.
Do you think that it’s also possible with the E-M1?
not EM1,the E5 with best 4/3 lens will beat D700 at ISO100,200,300,400.
over ISO400,D700 win.
And why is that?
yes, use a fractal calculator, like Alien Skin Blow Up in Photoshop and magnify your photo by reallocation of the pixels. You will gat a full IQ shot without major loss. Fractal Design Print Pro is a similar soft, and, you can fraction the photo in single print sheets to assembel when doing billboard size.
Steve,
Watching pictures at 100% is not the way to view a photograph…..(you know, I know but most people seem to forget). 100% is for editing and not for viewing. In the days of old I viewed my pictures through a Focusscope…..to check the focus in the darkroom. I could focus on the pixel/grain. But I would not advice anyone to view my pictures that way :). If you print you print at 300 dpi (or 200 dpi). A decent monitor has a 92 pixels per inch pitch, so 100% means viewing at 92 dpi…..so to view at 300 dpi again you should view your shots at about 33% or at about 50% for 200 dpi. At 200 dpi a 4608 shot is 23 inch wide…..or more or less 60 cm……now when I used the Focusscope I used a Nikon F3 and FP4 125 (exposed at 200) or Kodachrome 64 (exposed at 80). I could blow up my shots to 60 x 40 cm and 50 x 40 cm but I liked the grain only at 13 x 8 cm (and sometimes even smaller).
So please people, stop fretting about the pixel peeping qualities of a camera. Nine times out of ten the IQ of the camera will outshine the IQ of the printer (mine is an R2400 and that produces/produced great prints at 30 x 40 cm from both my OM-D and my former GF1).
As for size…..yes you can blow up a digital picture almost limitless (compared to analog). Shamael is right about the IQ but of course you will loose fine detail (just as with the D700 that only had 12 Mpixels to start with). But that is okay…..I’ve seen Ellen von Unwerth shots made with a 5D blown up to that size……a little loss of detail at close range (20 cm which was followed by a reprimande from the museum warden (and he of course was right)) but at normal viewing distance of about 3 meters…..they where outstanding.
Greets, Ed.
I have Olympus E-p3 + several lenses and have been awaiting reviews of the OMD-EM1 with great anticipation. I am quite disappointed in the images, mainly because of the noise characteristics. Noise levels are not too bad though not great, but the noise doesn’t look at all like film grain, more like reticulation which was an unwanted effect that could happen during film development, causing the emulsion to ‘ripple’. I wouldn’t be spending near £1500 for this camera based on these images.
Fuji time
I agree to a large extent, Chris. The noise (reticulation is a good description) even at even 400 iso is not so “clean” or film-like and the files are not that malleable – maybe its the moggy exterior light that Steve encountered on the days he shot with it, though some of the interior portrait shots also display an unpleasant texture in the shadow areas. The price is fair for what it delivers, but still not a cheap camera if one has to build up a new system from scratch.
Perhaps we are expecting a magic bullet of a camera in every new model – compact and FF-like quality with a truck-load of features. Maybe manufacturers are building up expectations and hype too much with every new model release.
I too was waiting for this Olympus with great anticipation. I wanted a compact camera with high IQ and fast AF. This initial review has made me again realize that there is just no real replacement for a FF sensor and its IQ advantages – not yet.
For a small sensor this camera seems amazing and, no doubt, it will sell like hotcakes. It does what it does damn well from these wonderful images that Steve has posted. The new features and bells and whistles will appeal to many, but for those who just want a simple compact box that delivers FF quality … well, there is nothing really to replace FF. Perhaps we are expecting far too much. I think I will wait to see and hear more from Steve before making any decision. A Big Thank You, Steve, for the hard work, passion and dedication you always share with us.
I think you may be doing the rounds with this comment. trying to cause disruption as this well known troll commented on 43 rumors
http://www.43rumors.com/great-picture-collection-from-magnum-photographer-moises-saman-and-guess-what-camera-he-uses-for-his-work/comment-page-1/?comment_view=flat
Shenkie
2 days ago |
I am quite disappointed in the images of the em5/1 mainly because of the noise characteristics. Noise levels are not too bad though not great, but the noise doesn’t look at all like film grain, more like reticulation which was an unwanted effect that could happen during film development, causing the emulsion to ‘ripple’. 100% views off raw looks like shit even at low iso. This is the problem with m43. Pro’s never be spending near 1500 euro for this camera based on these images.
All sounds very similar?
I think you may be doing the rounds with this comment
Change that to someone may be doing the rounds.
Chris or should we say Viezevure as you are known on another website?:
Nice talk about the lens. Looks like a good one. The pictures i am not sure tho. To be honest I’m quite disappointed by the noise. The background of the pictures taken at 640 ISO is really noisy. Even at low iso it does not look real good. Noise levels are not too bad though not great, but the noise doesn’t look at all like film grain, more like reticulation which was an unwanted effect that could happen during film development, causing the emulsion to ‘ripple’. I wouldn’t be spending near 1500 euro for this camera. But its still good for a point and shoot. The wife and kids can not wait to get their hands on one.
http://www.43rumors.com/new-e-m1-reviews-and-tests2/comment-page-1/#comment-353854
Or should I have said Shenkie?:
I am quite disappointed in the images of the em5/1 mainly because of the noise characteristics. Noise levels are not too bad though not great, but the noise doesn’t look at all like film grain, more like reticulation which was an unwanted effect that could happen during film development, causing the emulsion to ‘ripple’. 100% views off raw looks like shit even at low iso. This is the problem with m43. Pro’s never be spending near 1500 euro for this camera based on these images.
http://www.43rumors.com/great-picture-collection-from-magnum-photographer-moises-saman-and-guess-what-camera-he-uses-for-his-work/#comments
And here where someone called Shenkie or should I say Chris said they had bought an E-M1?:
I forgit to mention I bought GX7 for my vidoe needs and E-M1 for family picture’s although I’m concerned about small sensor and big nose!
http://www.43rumors.com/new-e-m1-reviews-and-tests2/#comments
Looks like Shenkie, Chris or Viezevure you meant to say noise and not nose?
Great performance , congrats !
Impressive, the top photo with the horse and woman going through the water is my favorite. The gentleman is also one of my favorites! Good Job!
Yes , very good quality indeed. I especially like the portrait of the gentleman. What was the Castles’ name ?
Thank you. Very impressive.