E-M5 vs. E-P3: RAW Image Quality

This is a "rewrite" of my previous blog concerning JPEG image quality of these two cameras. It can be found here.

Below I have combined test charts shot with E-M5 and E-P3 at various ISO settings. Settings while shooting:

• Lens Zuiko D. 50mm f/2 Macro @ F4
• Equal aperture and shutter speed settings in both bodies at corresponding ISOs. (These cameras have equal sensitivities)

RAW files were opened in Lightroom 4.1 (Release Candidate 2), normalized and cropped equally. E-M5 crops are 100%. E-P3 crops were enlarged to the same size during Lightroom Export. All images are slightly sharpened and noise reduction applied in Lightroom. I kept the values for these the same for every file.

I chose to enlarge E-P3 files because I am interested in seeing quality in same size prints. This is closest to that situation on screen. Just remember: To see any real difference in sharpness at ISO 200 to 400 you need to print larger than A3. 

In JPEG comparison E-M5 was markedly better in every regard. At higher ISOs E-M5 gained about two stop improvement. I wrote then: This result comes from new sensor with more resolution and better pixel per pixel quality and thinner low pass filter. Now I know better, a lot of it comes also from improved JPEG algorithms. 

With RAW images differencies are not as big in a test target like this. This test target shows now mostly differencies in resolution and noise. E-M5 has a 14% advantage in linear resolution and thinner low pass filter. This is best shown in rhombus structure. Starting from ISO 800 up, E-M5 shows gradually more and more its better sensor. Further tweaking of the relation between sharpening and noise reduction would have benefitted E-M5 more than E-P3. With more demanding subjects E-M5 gives even more headroom for adjustments. I would say that for me the total gain is almost 2 stops in the way I tweak my images in Lightroom. With E-P3 I never went over ISO 1600. Now the limit will be slightly under ISO 6400.

Image below E-M5, M.Zuiko 12mm f/2 (Same image is shown shot as JPEG in previous blog).

• ISO 200, @F6.3
• Adjusted in Lightroom 4.1 (RC2)
• Crops 100% from brightest and darkest area without any modifications

Image below E-M5, Zuiko D. 12-60mm f/2.8-4

• ISO 200, @F5.6, underexposed on purpose.
• First image: opened into Lightroom 4.1 (RC2) at default settings, no adjustments
• Second image: adjusted by +5 stops in Lightroom, full Highlight correction (-100); two simple adjustments just to show the pushability
• Third image is a 100% crop. This radical opening of shadows brings out luminosity noise, as expected. But that´s all. No smearing to mention (grass is always difficult, stone had this mottled surface), relatively sharp details.

-p-

Olympus E-M5: Dynamic Range - RAW

I posted in February a blog on E-M5 dynamic range. At that time it was about jpeg images. Now we have two popular and well known converters supporting E-M5, which enabled me to take another look with raw files. These converters are Lightroom 4.1 (Release Candidate 2) and CaptureOne 6.4.

This graph shows the same E-M5 and E-P3 dynamic ranges (orange and red) for jpeg images as shown in my previous blog. They are actually measured from raw files converted into jpegs in Olympus Viewer 2. This software emulates in camera jpeg process, which means I can tweak the jpegs to show optimal jpeg quality.

Blue and green lines  show what can be achived from E-M5 and E-P3 raw files. I opened raw files into Lightroom and CaptureOne and measured dynamic ranges at various ISOs. Both software showed almost the same numbers for E-M5. With E-P3 there was slight variation at some ISOs but there was no trend in favor of either software. What you can see here is the average performance of these software. The biggest and maybe most important difference was for E-M5 at ISO 200. With Lightroom I was able to read a 12.6 EV dynamic range and with CaptureOne it was 12.0EV.  At ISO 400 they both gave the same 12EV.

When you look at this graph, please do not take exact numbers too seriously. Look at trends E-M5 versus E-P3 and raw versus jpeg. While saying that I think it is no mistake to have E-P3 dynamic range at 10EV, which is the same as DxO Mark result for the same camera. Also two well respected software giving practically same numbers (sans ISO 200) for E-M5 should be no mistake either.

I have now shot a few thousand images with E-M5. Those images show the same trend in real life situations compared to E-P3 as this graph.

-p- 

E-M5 vs. E-P3: JPEG Image Quality

Below I have combined test charts shot with E-M5 and E-P3 at various ISO settings. Settings while shooting:

• Lens Zuiko D. 50mm f/2 Macro @ F4
• JPG Large Fine (I saved also RAW files to be compared later) (note: there is no visual difference between files saved at Large Fine or Large Super Fine jpeg compressions at 100%, unless you start to do heavy post processing)
• Picture Mode: 4 Muted @ Contrast -2, Saturation -2, Sharpness -2, Gradation Normal (note: normal means Olympus standard jpeg gamma curve)
• Noise reduction & noise filter: OFF
• Equal aperture and shutter speed settings in both bodies at corresponding ISOs

JPG files were opened in Photoshop, normalized and cropped equally. E-M5 crops are 100%. E-P3 crops were enlarged to same size through Bicubic Smoother interpolation. All images are slightly sharpened. No noise reduction applied. (All test shots have gone through automatic normalization which sets certain amount of lowest values of lowest channel to zero. In E-P3 shots you can see more black outline between white and black than in E-M5 shots. Anything stronger than what I did would lead into losing detail in shadows. I left corrections as they are, and that´s why some E-P3 scales look lighter than the same ones with E-M5. You can copy test charts and tweak them if you want to have another "look".)

I chose to enlarge E-P3 files because I am interested in seeing quality in same size prints. This is closest to that situation on screen. Just remember: To see any real difference in sharpness at ISO 200 to 400 you need to print larger than A3. 

E-M5 is markedly better in every regard: sharpness, signal to noise ratio, dynamic range... . At higher ISOs E-M5 gains about two stop improvement. This result comes from new sensor with more resolution and better pixel per pixel quality, thinner low pass filter and impoved JPG processing.

I will post later more images and compare RAW files.

-p- 

Image below E-M5, M.Zuiko 12mm f/2 

• ISO 200, @F6.3
• JPG Large Fine
• Picture Mode: 4 Muted @ Contrast -2, Saturation -2, Sharpness -2, Gradation Normal
• Noise reduction & noise filter: OFF
• Crops 100% from brightest and darkest area without any modifications

Image below E-M5, M.Zuiko 45mm f/1.8

• ISO 1600, @F2.5
• JPG Large Fine
• Picture Mode: 4 Muted @ Contrast -2, Saturation -2, Sharpness -2, Gradation Normal
• Noise reduction & noise filter: OFF
• Normalization and slight sharpening in Lightroom 3
• Crop 100%, no noise reduction applied

Olympus M.Zuiko 12-50mm f/3.5-6.3 in comparison

During this time of year we have not much light even during daytime and the weather in Southern Finland has been quite cloudy and rainy. Because of that I decided to do a traditional test target shoot to compare the new zoom with some of my other lenses. 

Above is my test setup. I used two targets, one in the center and the other in the right upper corner. I shot the same "subject" (marked here as grey area) with different focal lengths. Of course distance varied accordingly to keep (subject) area the same. I used studio flashes and used their power settings to keep exposure on sensor constant in every picture. Accuracy was 1/10 stop. This way the differencies in lens diaphragm did not affect the results. I also focused center and corner targets separately to compensate for possible curvatures in lenses´focal planes. All test pictures were shot at ISO 200. Camera was Olympus E-P3. RAW images were opened in Lightroom 3.6 and I did a basic normalization of tones, correction of chromatic aberration and added sharpening. Those are the same basic tweaks I would do to real pictures. Possible distortions were not corrected as can be seen in corner shots.

Focal Length 12mm

The first comparison shows the center of 12-50mm zoom at full aperture F3.5 and closed at F5.6. Focal length is widest, 12mm. These are 100% crops, like all other target images here. The image quality gets better with smaller aperture as would be expected here.

 

Olympus´4/3 series zoom lens Zuiko D. 12-60mm f/2.8-4.0 is the perfect comparison lens for any lens having the same focal length range. Here we can see center targets. This result is in line with Olympus´MTF curves below: 12-60mm zoom is a great lens, and especially its resolution (orange curves) is higher than with the new zoom. You can find more of these MTF curves at Olympus web sites. They are good tools when comparing lenses.

 

In upper line of this picture we have the corners of 12-50mm zoom at apertures F3.5 and F5.6. Contrast and resolution gets better with smaller aperture, but radial (sagittal) and tangential (meridional) lines have a very different resolution. It can be also seen in MTF curves, where continuous line is sagittal and dashed line is meridional transfer function. For comparison we have corner shots with 12-60mm zoom at F3.5 and and M.Zuiko 12mm f/2 prime at F4.0. 12mm lens is not too good in corners, but it has this reasonable level all ready from F2.

Focal Length 14mm

 

As a further comparison lens we have M.Zuiko 14-42mm f/3.5-5.6 II kit zoom. Above its results at 14mm focal lenght and apertures F3.5 and F5.6. Left side images are from center and right side images are from corner. Now 12mm and 14mm focal lengths are so different that they should not be compared directly. That´s why I have below a common focal length of about 19mm.

Focal Length about 19mm

 

Here I set the shooting distance with Lumix G 20mm f/1.7 prime. After shooting with it, I shot with those three zooms from the same spot while setting zoom or focal length so that the cropping was always equal. Zoom rings did not show 20mm because Lumix actually has not a 20mm focal length. It is maybe close to 18,5mm. These crops are from center. Lumix 20mm and Olympus 12-60mm zoom show almost identical performance. Wide open 12-50mm zoom has a good resolution but its contrast is not up to 14-42mm zoom.

 

Looking at corners, 12-50mm zoom has again nice resolution but contrast performance is very modest. Thus it is practically a draw with kit zoom. Maybe it comes as a surprise for some people but Lumix 20mm is not much better in the corners. Number one from these lenses is quite obvious.

Closing down to F5.6 makes Lumix 20mm markedly better in the corners than these zooms.

Focal Length 42mm

 

This comparison has on the left centers at full aperture and on the right corners, again at full aperture. 12-50mm zoom gets better all the time as focal length grows. 14-42mm zoom on the other hand sees especially contrast dropping at its longest focal length.

Focal Length 50mm

 

Uppermost, on the left, we have 12-50mm zoom´s center at full aperture. All the other crops are from the corner. 12-50mm zoom shows just slightly better contrast in corners when closed to F8. Still contrast performance is lower than resolution. As new lenses I included M.Zuiko 45mm f/1.8 at the same aperture F4 as earlier were Lumix 20mm and M.Zuiko 12mm. 50mm is 4/3 series Zuiko D. 50mm f/2 Macro. 

Macro setting

 

As close up test I did a watch image. At macro setting 12-50mm has a 43mm focal length. It can not be changed as zoom ring does not move when macro setting is engaged. Largest aperture is F6.0. Here we have the closest focusing distance of 200mm.

I tried apertures F8 and F11. The smaller aperture shows already effects of diffractions but on he other hand the better depth of field can be a bigger benefit. How ever, the 12-50mm lens is capable of very nice images with close up subjects. This is a 100% crop.

For comparison once more on the left the new 12-50mm zoom and on the right Zuiko D. 50mm f/2 Macro. Aperture is F8. Focusing differs between these images, please look for the sharpest details in both images. 

Conclusion

Zuiko D. 12-60mm f/2.8-4 is one heck of a lens. I guess that´s old news. The images should tell quite clearly the differencies. However, we must remember that this kind of very precisely made test target comparison is a lot harsher than any normal shooting. The differencies in normal shooting would not be in the same league, there are too many disturbing factors. Even slight differencies in exposure, focus or camera shake could obscure a lot. But, yes this is what you get when everything is optimized.

M. Zuiko 12-50mm f/3.5-6.3 is a very reasonable lens for photography with its longer focal length and especially for close up shooting. Otherwise it fits better for video because of its silent and fast focusing and silent power-zoom. For video it has plenty of quality. You can check this by scaling test shots into 50%. Then they correspond to Full HD quality. 

-p-

Olympus M.Zuiko 12-50mm 1:3.5-6.3

I got a production sample of the new Olympus zoom lens, M. Zuiko 12-50mm 1:3.5-6.3, the other day. It is the first weather sealed lens in Olympus PEN M.Zuiko series. A weather sealed lens among ordinary bodies, well, that means there will be soon a weather sealed body. I would guess that the waiting time is probably not even long. The new lens has attracted some net controversy since it´s release, so it is interesting to see what it is good for.

Above the new zoom in size comparison with 4/3 series 12-60mm 1:2.8-4 and m4/3 series 14-42 kit zoom.

Build

The new lens is slightly larger than the 14-42mm kit zoom in shooting position. The lens has a solid feel, while it´s outer structure is plastic. Both zooming and focusing are internal, so the lens has no external moving parts other than the focusing and zoom rings. The narrow focusing ring is on the front of the lens. The zoom ring is fully electronic. When set for manual, it works like a normal zoom ring. In E-position, it works as power zoom switch. Zoom speed depends on the strength of the twist. Manual and e-mode are selected by moving the zoom ring back and forth. The selected position is shown on the right side of the lens frame. On the left side there are L-Fn and Macro buttons. When the latter is pressed, zoom ring can moved to a third position, which naturally is macro. The closest focusing distance in macro mode is 0.2 m and the minimum image area is 36x48mm. L-Fn in turn, is specially designed as AF lock for video shooting, in situations where you want to prevent the continuous focusing from locking on a wrong subject.

Half of lens elements are of special glass. DSA means Double Super Aspherical, a lens that can substitute a group of lenses in traditional design.

L-Fn and Macro buttons on the left side of lens.

Window on the right side of lens shows now that zoom ring behaves like a traditional zoom ring (while it actually is electronic). Moving zoom ring (on the right) to the right activates power zoom. Pressing Macro button and moving zoom ring another click sets the lens for macro shooting.

On paper

From Olympus provided data I would expect that this lens is in optical quality (contrast and resolution) between the kit zoom 14-42mm and the 4/3 series zoom 12-60mm 1:2.8-4. MTF curves can be found on Olympus' websites. They, too, show a relatively uniform quality throughout the image area. The new lens will not stand up, particularly in resolution, up to the 12-60-millimeter zoom in wide-angle position. However, the longest focal lengths should not differ at all in practice. That in turn means that the new lens should give in practice very visibly better image quality compared to the kit zoom at longer focal lengths.

The first impressions

Finland is not really a wonderland to test such a lens during this darkest time of the year. I shot yesterday the first batch of images  to get an idea of handling and usage in practice. All images were now shot handheld, and then at full aperture out of necessity. ISO sensitivities when shooting outdoors with this kind of lens during the darkest time of the year are too high for verification of previous contrast/resolution estimates. That´s for the next blog. The first impression, however, is in favor of them.

It is noteworthy that the image when opened in Lightroom shows practically no vignetting at any focal length at full aperture. Distortion is notable at 12mm position, and disappears almost entirely by 17mm. Chromatic aberration is also strongest at the widest setting and decreases / disappears when the focal length increases. Distortion, and chromatic aberration are easy to remove with the Lightroom tools, and you can make a specific total correction preset per focal length.

Below are a few pictures, I will return to a more accurate comparison between those previously mentioned zooms after Christmas.

-p-

 

Focal length 20mm, F4.6, 1/60s, ISO 800

Focal length 50mm, F6.3, 1/100s, ISO 1600.

Focal Length 50mm, F6.3, 1/80s, ISO 1600.

Focal Length 12mm, F3.5, 1/60s, ISO 640.

Update on December 25, 2011 at 11:49 by Pekka Potka

I know the zooming is electronic but when it is set to manual, does it feel like a manual lens?

There are so many versions of the feel of manual zoom among traditional lenses... The feel of movement in 12-50 is smoother and it has a better balanced resistance than in my copy of 12-60. There is a typical "electric scratching" noise while turning the ring on manual mode. End points of zoom range are well defined also with audio. Focal lengths are shown in viewfinder (and in viewfinder only).

Also, can you give the breakdown on aperture/focal length? Where does it hit f/6.3?

This graph shows largest F-numbers at various focal lengths based on viewfinder information:

-p-