Focal length and focus distance

You'll notice that the focal length of most lenses is specified at infinity focus. It turns out that some lens designs imply a change of focal length, as the focus distance changes.

Here's a comparison of two quite different lenses. The Nikkor 200mm f/4 AIS is a traditional fixed focal tele lens. Compact, light, and reasonably fast, it is a classic lens. It is also rather simple from a mechanical point of view. If features "only" five lens elements.

The other lens is the Panasonic Lumix G 45-200mm f/4-5.6. Since it is a zoom lens, and it also has special lens groups for the OIS, it features a whopping 16 lens elements.

They are both shown here, the Nikkor 200mm lens with the Nikon-M43 adapter mounted. The 45-200mm lens is shown zoomed to maximum tele, to be comparable with the Nikon lens:

They are also quite different when it comes to focusing. The Nikkor has a traditional focus mechanism, which simply moves the entire lens assembly forward when going from infinity to close distance focus. In the picture below, it is shown focusing at infinity (left) and just below 2 meters (right).

The Panasonic Lumix G 45-200mm f/4-5.6, on the other hand, features internal focusing. This is very practical, since it means that the length of the lens stays the same regardless of the focus.

Also, the internal lenses moving around are much smaller, and faster to move about. This makes the focus faster, and requires less juice from the batteries. And the 45-200mm does focus very fast indeed!

The negative side of internal focusing, is that it can affect the focal length when changing focus. Let's compare the two lenses when focused far and near. The upper and lower images are taken at exactly the same place, but with the focus placed at the parked car (upper), and the foreground flower (lower).

What we see here, is that when focused far away, the field of view of the lenses are mostly the same. The Lumix has a slightly wider field of view, but there is hardly any significant difference. When focused at 2 meter distance, however, the field of view is significantly wider with the Lumix lens.

Mostly, this is not any problem at all. When recording video, however, it can be annoying if the field of view changes significantly during focus. This feature is usually referred to as "focus breathing", as the objects recorded will pulse in size as the focus moves back and forth.

With contrast detection autofocus (CDAF), this can in fact be a big problem during video recording, as the camera must jog the lens focus back and forth to verify that the focus is correct. You can see these focus movements in a video recorded using the Lumix 20mm f/1.7 pancake lens. The Lumix 20mm f/1.7 lens features a traditional moving lens assembly, and does not suffer from focus breathing. This is good, since the continuous autofocus operation is not very visible in the video, but on the other hand, the lens is probably not as solid and weather resistant with this construction.

I first noticed this change of focal length when examining the bokeh of the 45-200mm at 200mm, compared with the Nikkor 200mm lens.

Lumix G 14-42, new kit zoom

For the introduction of the Panasonic Lumix G2 and G10 cameras, a new kit zoom was launched. Much to the dismay of Micro Four Thirds users, since on first sight it looks like a dumbed down version of the old kit zoom.

First of all, it has slightly worse technical specifications: The long end of the zoom is 42mm, while the old had 45mm. Next, the new kit lens has got a plastic mount, while the older has a metal mount. The new zoom also lost the OIS switch: Switching OIS on or off is now done through the menus.

In terms of ergonomy, the new lens also lost the rubber zoom ring. It now features a plastic zoom ring, which gives somewhat less friction when operating it with your fingers.  Some users of the old lens experienced that the rubber zoom ring came loose.  This will not be a problem with the new lens, since there is no rubber ring.

Now, the change of the long end focal length doesn't bother me. 42mm and 45mm is basically the same field of view, there is no significant difference here. Also, the plastic mount, if done properly with good quality materials, is probably solid enough. After all, this is a very light weight lens, and in normal use, it doesn't need as strong support as larger lenses.

What about other aspects? Some reports indicate that the sharpness of the new lens is not as good as the original Lumix G 14-45mm lens. I cannot comment this, since I haven't used both.

Here's an analysis of the sharpness and bokeh of the lens.

The GH1 kit lens, the Lumix G HD 14-140mm, is specified with an aperture range from f/4 to f/5.8. However, while zooming from wide to tele, it closes down very quickly. So it is fair to say that this is essentially an "around f/5.6" lens, with a bonus brightness in the short end.

What about the other kit lenses? This diagram shows the relationship between the focal length and the maximum aperture for the three kit lenses:

For the 14-42mm and 14-140mm kit lenses, these values were sampled by using the actual lens. For the 14-45mm lens, I took the values from various reviews off the Internet.

It looks like the new kit lens has slightly better speed at f=25mm: f/4.6, compared with f/4.9 for the old kit lens.

I also added the aperture data for the premium Olympus standard zoom, the Olympus M.Zuiko Digital ED 12-50mm F3.5-6.3 EZ power zoom. This lens is comparable with the other kit lenses in the short end, but the aperture closes down very quickly as the focal length increases. I think this is consistent with the Olympus M 4/3 design philosophy, which generally puts compactness ahead of maximum aperture.

The main purpose of the kit lens, is to be cheap and good enough for most beginners. I'm guessing that the size and number of the glass lens elements is an important contributor to the price.

The diameter of the front element of the 14-42mm lens is 13% smaller than that of the 14-45mm lens. And that means the area is 25% smaller:

It is quite remarkable that Panasonic has essentially retained the specifications, while shrinking the front element so much.  Of course, reducing the front lens diameter is not necessarily good for the image quality.  It could lead to more vignetting at max aperture, for example.

All in all, I think this will be a pretty successful lens. Some early reports indicate slightly worse sharpness, however, for the target audience that may not be a problem. The cheaper construction means that Panasonic can sell them in kits at a lower price point, which they will need now that the competition has gotten their systems launched.

The autofocus speed of the newer 14-42mm lens is very good.

Lumix 20mm compared with Sigma 30mm

A lot of people have complained that the Panasonic Lumix 20mm f/1.7 pancake lens is overpriced. To have a look at this statement, let's compare it with a lens in the same price range, the Sigma 30mm f/1.4.

The lenses have a lot in common. They share the same price tag in my market, and they do essentially the same job. The Lumix 20mm lens has a slightly wider field of view, and the Sigma 30mm has half a stop larger aperture. But these differences are not very significant. When I compare their fields of view, I refer to the Sigma 30mm being used on an APS-C camera, for which is was designed.

Here they are both:

The Sigma 30mm lens is shown with the supplied lens hood, which is very nicely designed. Sadly, the Lumix 20mm does not come with a hood, but I have put a step down ring on it, which acts as a compact hood.

As is apparent from the image, these lenses are very different in size. The Sigma (left) is 77x59mm, 430g. To the right, the Lumix is 25.5x63mm, 100g. Adding the supplied hood to the Sigma lens will make the difference even larger, of course.

When mounting the lenses to cameras, they look like this:

To the left is the Panasonic Lumix GH1 with the Lumix 20mm lens, and to the right is the Pentax K10D with the Sigma 30mm lens.

Image quality

What about the image quality? I tried to take the same picture with both setups, to see how they compare.

GH1 + Lumix 20mm @ f/1.7, 1/60 second, ISO 100 (click for larger image)

K10D + Sigma 30mm @ f/1.7, 1/45 second, ISO 100 (click for larger image)

Note that I used the same aperture on both lenses. The Sigma lens was stopped down from f/1.4 to f/1.7 to be comparable with the Lumix 20mm, which was used at the maximum aperture.

What we see straight away, is that the Lumix provides a wider field of view. I also think that the Pentax colours are more pleasing straight from the camera. Of course, using the RAW files you are free to adjust the colours of either images as you want.

To make the images easier to compare, I also added 100% crops from two sections of the images (click for larger image):

Here it is quite apparent that the Lumix lens is the sharpest. Also, there are less purple fringing artifacts in the Lumix image.

Now, we know that Chromatic Apperation (CA) artifacts are corrected for in software in the Panasonic Lumix G-series cameras. So the Lumix lens has an advantage here, in that these artifacts are automatically removed. Still, as a user of the systems, I care about the end result, not how it was achieved. And the end result is most certainly a lot better using the Lumix lens.

The Pentax setup uses phase detection autofocus (PDAF). With a large aperture lens like the Sigma 30mm f/1.4, this means that you can worry about the precision of the autofocus. Some camera body/lens combinations suffer from front-focus or back-focus. The setup might need expensive calibration to avoid these problems and achieve the best focus.

With the GH1 and Lumix 20mm lens, though, you get contrast detection autofocus (CDAF). With this system, you are ensured the best focus every time, as long as you set the focus region to suit your needs.

Size

The Lumix lens is a lot smaller and lighter. For me personally, that is a huge advantage. It could be a disadvantage for some users, though. Some customers might not take you seriously if you show up at a photography job with a small lens like this. When they pay for a photographer's services, some expect to get a person with a big camera and lens. For most users, though, this is not an issue.

Conclusion

From my point of view, these lenses, which do more or less the same job and are priced similarly, do not have an equal value. I much prefer the Lumix 20mm lens, which gives me better images, and is easier to lug around. I like the hood supplied with the Sigma 30mm lens. And having the option of using half a stop larger aperture is nice. However, you're not very likely to use the Sigma 30mm lens at f/1.4, since it is not very sharp wide open.

One could argue that my example image is not the most relevant for this type of lenses. These lenses are made for low light images of people, in which the corner sharpness doesn't matter too much. Also, while I haven't studied it carefully, I have a feeling that the bokeh from the Sigma 30mm lens is better. The Lumix 20mm bokeh is certainly very adequate, though.

I have also tested the autofocus of these two camera/lens combinations. The Pentax/Sigma combination is pretty fast in terms of autofocus, but not as fast as the Lumix. Also, the Pentax/Sigma makes much more noise when focusing.

Firmware v1.3 for Lumix 14-140 and AF speed

In the beginning of October 2010, a new firmware version 1.3 became available for the Panasonic Lumix 14-140mm f/4-5.8 superzoom lens. The new version is said to give faster AF speed, as well as faster startup time.

I have previously tested the AF speed of the Lumix 14-140mm. My conclusion was that the AF speed was very fast, albeit with some hunting in the longer end of the zoom, at close to the minimum focus distance. So I wanted to see if the new version improved the focus speed.

Again I used the same setup, with a LEGO figure as object to be photographed. The LEGO figure was placed at a distance of 0.5 meters, which is the minimum focus distance for the lens. After turning on the camera, I pressed the shutter release button, and measured the time until focus was achieved. The lightning was dull, about 6 EV.

I used the Panasonic Lumix GH1. Here's the experiment using firmware version 1.2:



And using the firmware version 1.3:



The time until focus was 0.76 seconds with firmware version 1.2, and 0.68 seconds with the latest firmware version. This is hardly any significant difference, but there is still some improvement.

I was not able to reproduce the focus hunting at 140mm that I experienced in my first comparison. The first test was done using firmware 1.1, so I guess this was fixed already in the version 1.2.

Bokeh comparison @ 20mm

I have two lenses that can do 20mm, so I thought it would be interesting to compare their bokeh. Both the Panasonic Lumix 20mm f/1.7, and the Panasonic Lumix 14-140mm f/4-5.8 zoom comprise the 20mm focal length. Apart from this, they are of course very different. At f=20mm, the superzoom has a maximum aperture of f/4.4, which is quite a bit smaller than the f/1.7 that the fixed 20mm prime can do.

To compare the lenses, I have taken the same picture with various apertures. The centre region is in focus, and the focus distance is around 60cm, just above the minimum focus distance for the 14-140mm superzoom lens.

First the picture with the Lumix 20mm f/1.7 at maximum aperture. The picture is rescaled and sharpened a bit.

Second, the same image taken with the Lumix 14-140mm at 20mm f/4.4:

Click for larger images.

The make the comparison more sensible, I have looked at 100% crops from the image at various apertures with both lenses (click for larger image):

And here are 100% crops from another region of the image. It was a bit more windy in the second series, so the leaves were not always in the same spot.

Conclusion

It is not surprising that the bokeh for the 20mm prime lens is the best. For out of focus highlights, the bokeh is pretty even, but has hard edges. On the other hand, it is not entirely round, especially when stopping down. All in all, the bokeh is pleasant, and has no significant ringing. The superzoom bokeh is a bit "dirty".

You'll also note some differences in sharpness. Note, however, that a higher ISO value was used for the f/8 and f/11 images for the 20mm prime lens. So some lack of sharpness could be due to using a higher ISO. It can still be noted that the superzoom lens appears very sharp.

The 14-140mm also has more flare, which is due to it's more complicated construction. The 14-140mm lens has 17 lens elements in 13 groups, while the 20mm pancake only has 7 elements in 5 groups. Generally speaking, the less number of lens elements, the less problems with flare you're likely to see. Of course, many other factors also affect flare.

Lumix 12.5mm 3D lens

Ok, so this lens has just been announced, and obviously I have not tried it. So this is based on what I have read about the lens so far.

Since writing this, I have made a review based on using the actual product.

Contrary to what some people say, I actually find this lens pretty interesting. Some say that Panasonic should rather spend the time developing "real" lenses. However, I think that this lens might have some real creative potential, and it also serves to bring more interest into the Micro Four Thirds system. So what's not to like?

Focal length

The lens is specified to have a focal length of 12.5mm. However, don't let that fool you into believing that this is a wide angle lens. Far from it. The two separate lenses project image circles that are smaller than the full sensor. Hence, there is an additional crop factor to take into consideration.

The field of view is comparable to a 65mm lens on traditional 35mm cameras, corresponding to 32.5mm on Four Thirds cameras. Hence, this is a long normal lens. Observing that 32.5 / 12.5 = 2.6, we can conclude that this lens has an additional crop factor of 2.6, beyond the crop factor of Four Thirds lenses, which is 2.

On the Panasonic Lumix GH2, the lens produces stereo images of 2048x1536 pixel resolution. With the GH2 having 4602x3456 pixels, this means that each of the two stereo images comprise 20% of the image sensor. Here is an illustration:



Aperture

With a fixed aperture of f/12, the lens gives a deep depth of focus. Which is good for a lot of applications. On the other hand, at f/12, you will need to push the ISO up unless being outside on a sunny day.

3D effect

Obviously, the 3D effect is achieved by having the two separate lenses seeing slightly different perspectives. The lenses appear to be very simple constructions, with four lens elements in three groups.

The stereo base, the offset between the two lenses, is 10mm, corresponding to about half of the sensor width. This looks like a very short distance, and I am unsure if any significant 3D effect can be achieved with this short lens offset. Perhaps the 3D effect can be boosted in post processing?

Usage on non-compatible cameras

When used with a 3D compatible Micro Four Thirds camera, you can probably use the lens in a seamless fashion. The viewfinder will probably not show a confusing picture like the illustration above, but rather one of the subsections, to make framing easier. Compatible cameras at the moment include Panasonic Lumix G2 and GH2.

With an older, non-compatible camera, like G1, GF1, etc, I'm guessing the lens can be used like all others, with the exception that the two lenses will not illuminate the whole sensor. You may need to set the autofocus point to somewhere off centre, for that reason. Perhaps you will also need to change the exposure mode, since parts of the sensor are not exposed.

Of course, if the lens is used with non-compatible camera, i.e., anything older than the G2 or GH2, then you must convert the image into 3D in post processing.

Using the Lumix 20mm as portrait lens

A portrait lens has two main properties. First, the focal length is fairly long, so that you can take a picture of a face at sufficient distance to avoid distortion. Second, the aperture is large, so that you can focus on a face and have the background out of focus.

Typical high end portrait lenses for film based cameras are 85mm f/1.4. The focal length corresponds to around 42-45mm for Micro Four Thirds. Many zoom lenses available for Micro Four Thirds will give you this focal length option. However, the largest aperture of these zoom lenses is generally much smaller than f/1.4.

One prime lens which comes close is the Lumix Leica 45mm f/2.8 macro. However, while this lens has a larger aperture than the zoom lenses, it is still far from f/1.4, and I am sure some will find it limiting for portrait usage.

Another lens comes close in the aperture department. It is the Panasonic Lumix 20mm f/1.7 pancake. However, this lens is quite a bit wider than most portrait lenses. So can you use it for portraits?



Let's take a look. When talking about portraits, let's say that we mean a picture in which the face fills the majority of the frame. When using the Lumix 20mm pancake lens, this is what we get, when going close enough to fill the face into the frame:



As you can see, the face is distorted. The nose and chin looks too large. This is due to getting very close to the face. The distortion is not a property of the wide lens. The distortion is related to the distance to the face. The closer the distance, the more distortion. With a longer focal length, you can stand at a longer distance when photographing, to avoid distortion.

So let's try a longer focal length. Here is the same face filled into the frame using the Lumix 45-200mm f/4-5.6 at 45mm:



The effect is very clear: 45mm is sufficient to get a proper perspective. The nose and chin now looks natural, and not artificially large, like in the picture taken at 20mm.

This picture was taken at 1m (3.3 feet) distance, which is the closest you can get with the 45-200mm lens. This is probably no coincidence. I'm guessing Panasonic designed the lens so that it could be used to take a headshot at the shortest focal length.

It is also interesting to see that the out of focus discs (bokeh) are about similar sizes in the two images. This means that f/4 at 45mm gives about the same background blur as f/1.7 at 20mm, when filling a face into the entire frame. Of course, the 20mm image was taken at a closer focus distance.

In a previous paragraph, I said that the distortion is related to the distance to the subject. So the solution is very simple: When using the Lumix 20mm pancake lens, take a step backwards. With the same distance as in the 45mm image above, the perspective should also be the same. Here is the outcome when using the 20mm lens at 1m (3.3 feet) distance:



In this example, there is no apparent distortion. However, the downside is of course that the face only fills a part of the frame. Also, the background is less blurred, since the focus length was larger.

The Lumix 20mm lens is very sharp. Especially in the centre region. So what you can do, is to crop the centre out of the image. Here is what I get using that method:



Of course, you might want to include a bit of background, and not crop as tightly as I have done here.

Conclusion

So can you use the Lumix 20mm lens for portraits? Yes, but don't stick the camera in the face of the subject. At a distance of about 0.7m (2 feet) or more, the face distortion should not be noticeable. If you need to, you can crop off a bit of the borders, to get a closer portrait.

Of course, you could take a so called "environmental portrait", in which you don't just shoot the face, but include some more of the person, and some elements around him or her. In that case, it makes good sense to use the 20mm lens. In fact, I'd say the Panasonic Lumix 20mm f/1.7 pancake is a perfect lens for environmental portraits.

Here is an example environmental portrait taken with the GH2 and the 20mm f/1.7 lens.  The image parameters are: f/2, 1/60 second, ISO 1600.

You could even use the Lumix 14mm f/2.5 wide angle prime lens for environmental portraits. As long as you keep a distance to the face of around 2 feet or more, you should not get any excessive distortion. On the other hand, if you photograph a sitting person head on, his knees will look abnormally large this way, due to the perspective.

Lumix 20mm distortion correction

One of the key features that separates quality optics from the cheaper variants, especially for wide angle lenses, is the the handling of distortion. In this respect, the Panasonic Lumix 20mm f/1.7 pancake lens is special, since in fact it features quite a bit of barrel distortion. However, this distortion is corrected by software inside the camera, so the user will never notice the fact in the first place.

However, the RAW files can be opened with third party software, to reveal the true image as captured by the sensor. In this case, I used UFRaw to convert the image to JPEG. Below is the original JPEG out of camera (left) and the RAW image with distortion uncorrected (right). The images were rescaled and sharpened. Click for larger images.

Out of camera JPEG	Uncorrected, from the RAW file

For these images, I used the 3:2 aspect ratio feature of the Lumix GH1 camera.

As you can see, there is some distortion in the right image. To correct it in The Gimp image processing software, around -13.5 Lens Distortion adjustment is needed.

Due to the distortion correction, all of the sensor surface is not used in the final image. Hence, the image area from the border is lost, as the image is adjusted and rescaled back to 12 megapixels.

In the image below, the area outside the white box corresponds roughly to the pixels lost during the distortion correction:



This means that when using the Lumix 20mm lens, about 7% of the sensor area is lost. The 12 megapixel sensor becomes 11 megapixels using the automatic distortion correction. The end user may never notice, however, as the output image is converted back to 12 megapixels anyway.

The upside of this is that when photographing non-straight shapes, you can probably get away with using the uncorrected image. This means that you gain some details in the borders, and a wider effective angle. For example, if you are photographing nature, flowers, people, animals, etc.

Most Micro Four Thirds lenses employ software distortion correction. Kit lenses like the Panasonic Lumix G 45-200mm f/4-5.6 and the Panasonic Lumix G HD 14-140mm f/4-5.8 do. One notable exception is the Panasonic Leica Lumix DG Macro-Elmarit 45mm f/2.8 1:1 Macro, which has all the distortion correction done optically.

Dual focus

In macro photography, the depth of field is commonly a constraining factor. At high magnification, even using f/16 might not give enough depth of field, if the object is not narrow. And closing down the aperture beyond around f/11 is not adviceable due to diffraction.

One way to get around this, is to take more than one photo, and change the focus distance between them. The photos can then be merged in photo editing software, to yield one single photo where more than normal is in focus. This technique is some times referred to as focus stacking.

When having a lot of photos to merge, it is best to use some software to do the merging automatically, since it involves a lot of tedious, manual work. However, when only merging two photos, it can be done quickly in any photo editing software that feature layers, e.g., The Gimp.

Here is an example. I took this picture of two LEGO minifigures. The figures were placed in a simple macro soft box to get even lightning. I used the Panasonic Lumix Leica 45mm f/2.8 1:1 macro lens for the image, and placed the autofocus area on the figure faces for focus. I used f/5.6, which is far from enough from getting both figures in focus at the same time.

Photo 1: Left figure in focus:



Photo 2: Right figure in focus:



Merging the two images is a matter of loading them both into the same image in two different layers, and then removing the out of focus areas. Note that some lenses will give a different field of view depending on the focus distance. Hence, you may need to rescale one (or more) of the photos when merging.

This merged image would have been impossible to take with only one exposure: