Sunday, 19 January 2014

Comparison @ 100mm

As the Micro Four Thirds lens lineup grows, there are more and more lenses covering the 100mm focal length. I have compared the image quality of four of them here:



From left to right: Lumix G 100-300mm f/4-5.6, Lumix X PZ 45-175mm f/4-5.6, Lumix X 35-100mm f/2.8, Lumix G 14-140mm f/3.5-5.6. The lenses are shown with the original supplied hoods.



The lenses


These lenses are all quite different, so let's start by looking quickly into them. Here is a diagram showing the focal length and aperture range of the four zoom lenses:



The Lumix G 100-300mm f/4-5.6 is a very long tele zoom, usable for, e.g., spectator sports, birds and wildlife. It is not very well suited for general use, as it starts at 100mm, which is already a very long focal length. The lens is quite good optically, and is reasonably priced. Like all Panasonic zoom lenses, it focuses very fast, and virtually inaudibly.

The Lumix X PZ 45-175mm f/4-5.6 is a compact tele zoom lens with power zoom. The zooming can only be operated electronically, which is good when you want to zoom while recording videos. It also works well for stills, as you can use the zoom ring much like a normal mechanical zoom ring. The lens is very good optically, and is a good choice for those who would like a compact tele zoom lens, at a some what steeper price. There are also more moderately priced alternatives.

Launched with the GH3 in 2012, the Lumix X 35-100mm f/2.8 is Panasonic's interpretation of the classic pro tele zoom lens. Sporting a constant f/2.8 max aperture across the focal length, it is generally regarded as a very good lens. It is also one of the most expensive Micro Four Thirds lenses available.

Finally, the Lumix G 14-140mm f/3.5-5.6 was launched in 2013, replacing the first version. Compared with the first version of the superzoom lens, this new version is smaller, lighter, and sharper. I think it is finally the successful "travel zoom lens" that we have been waiting for. I would strongly recommend this lens to those who use a Panasonic camera, and would like one single lens to cover most daily uses.

The image quality comparison


To test the lenses, I set the Panasonic GH3 camera on a tripod, used ISO 200, and set the 2s shutter delay to avoid camera shake. I photographed a box placed around 1.5m from the camera, and made sure the box surface was perpendicular to the camera. The lightning was artificial indoor lights, which can be quite challenging for the camera in auto white balance mode. Here are the images:

Lumix G 100-300mm f/4-5.6Lumix X PZ 45-175mm f/4-5.6
Lumix X 35-100mm f/2.8Lumix G 14-140mm f/3.5-5.6

The first thing we can note, is that the auto white balance did a somewhat different job with the different lenses. The white balance came out the best with the Lumix 45-175mm and Lumix 35-100mm lenses.

Second, even if I set all the lenses to 100mm, the field of view is different. There is a simple explanation to this: The focal length for a lens is always specified at infinity focus. In this example, I was focusing at about 1.5m distance. Many lenses, especially those with internal focus, change the focal length as you focus closer. This is called "focus breathing", and you can see an example here.

The biggest culprit in this respect is the Lumix 100-300mm. Again, the reason is simple: This lens has the longest minimum focus distance, 1.5m. And as we are very close to the lens's minimum focus distance, the focus breathing effect is the biggest. All other lenses in the test focus closer. The Lumix 35-100mm appears to be the least susceptible to focus breathing.

Getting into the subject of image quality, let's start by looking at 100% crops from the image centre:



Looking at these centre crops, we can see that all the lenses are impressively sharp in the centre when wide open (using the maximum aperture), with the exception of the most expensive lens, the Lumix X 35-100mm f/2.8. It could even have been a bit better when stopped down to f/4, and does not render perfectly sharp until stopping down to f/5.6.

And here are 100% crops from the top right corner:



Rendering the image sharply in the corner is the most challenging test. And we see clearly that even when stopped down, the lenses are not as sharp as in the centre.

The most sharp, in terms of being able to resolve the offset raster printing pattern, is the Lumix 14-140mm, which is surprising, as one would normally expect the worst performance from a lens with a long zoom range. Again, we see that the Lumix X 35-100mm f/2.8 is the worst in terms of sharpness.

Conclusion


First of all, the good news: All of the recent, normal aperture Panasonic zoom lenses are performing very well. I came to the same conclusion in a previous comparison at 140mm too.

The bad news is that the expensive Lumix X 35-100mm f/2.8 is not performing very well at 100mm f/2.8. I guess this is not so unexpected after all: At 100mm, I am using the lens at the longest focal length, which tends to be where the performance is the worst. In my experience, though, stopping the lens down just a little bit to f/3.2 does help quite a lot.

Also, the lens has a much brighter aperture than the rest at 100mm, with the difference being close to two stops for the Lumix 14-140mm and the Lumix 45-175mm. A larger aperture makes it more difficult to design a well performing lens. Still, I was hoping the lens would do better.

It could be that the Lumix X 35-100mm f/2.8 has a curved focus surface. That would explain the less than optimal sharpness in the corners. While traditional lens design states that the focus plane should be level, portrait lenses often have a spherical focus plane. That is helpful, since you can focus on the eyes, and retain the correct focus when reframing the image.

In my bokeh comparison, the Lumix X 35-100mm f/2.8 also comes out somewhat suboptimal.

In internet forums, you will often find people saying that for the best sharpness, you should get the lens with the largest aperture, and then stop it down. In my test, I think I show that this is not always true. The three less bright lenses are already quite sharp when wide open, an there is not much need to stop them down further for the best sharpness, contrary to what you would often read in online forums.





9 comments:

  1. Hello. I have been following your blog for a while now and find your reviews very useful. However, while I like the idea of this one very much - leaving aside how different the lenses are in use it can put some things into perspective - I am not sure I agree with the setup. Most of these lenses, especially the 100-300mm, would be used at longer distances, so to make this a comparison that is practically relevant you should have shot the test at least at 3-5m, better even 10 or more. Now, I know that this is not easy to do as you need the right space for that, but if you do such a test again and have a chance to add a little more distance it would be really useful. Either way, thanks for sharing!

    ReplyDelete
    Replies
    1. I think you are right that most people would use these lenses at longer focus distances. But in my previous comparison at 140mm, I used infinity, and I wanted to use a closer focus distance now to cover more example uses of the lenses.

      I know some people use the 100-300mm at fairly close focus distance to photograph large bugs like dragonflys. But then, they probably set the lens to 300mm, not 100mm.

      In the future, I will be doing more comparisons, especially with the 35-100mm f/2.8 lens. And then I will use longer focus distances.

      But this time, I decided to use the shorter focus distance and another type of subject for the sake of variation.

      Delete
  2. Ah, I missed that review. You are right to say that there will be people for whom the closer distance will be useful and for others it can still be interesting or perhaps even an indication of general performance. Overall I think that reviews should aim to use focus distances that are practically relevant - but then even in commercial operations reviews are often done at close focussing distances as it is logistically easier. So it is certainly not a criticism that would only apply to your post. The opposite in fact, I usually really appreciate your focus on issues that are practically relevant. Too many forum photographers get hung up on abstract numbers when they may not actually be able to tell the difference in a real photo.

    One review I'd particularly like to see would be about focussing speed of the m4/3 telezooms. If you want to have a chance to catch a bird in flight, is it better to use the 100-300 or the 75-300 etc. I hear many opinions but less informed information. As the system is, relatively, weak on tele lenses I certainly would appreciate such information. Not saying you have to do one though, just a suggestions.

    ReplyDelete
    Replies
    1. I agree with you that focus performance with long lenses, especially with moving targets, is very unclear. Unfortunately, it is also quite difficult to test. I suppose one could use different lenses over time, and then give a subjective view on which works the best.

      But to test focus speed with moving targets in a scientific way, is very difficult. At the moment, I don't that that is feasible for me.

      Delete
  3. Are you sure that you have a good copy of the 35-100? The pictures I take with it seem to be a lot sharper than the ones you have taken.

    ReplyDelete
    Replies
    1. I'm not sure of anything, that's one of the reasons I test it.

      Delete
  4. The area of sharpness with 35-100mm F/2.8 at 100mm F/2.8 is very thin (especially for close located objects). So the autofocus misses will cause the blur effect.
    Is the 35-100mm F/2.8 at 100mm F/2.8 is sharp? It's up to you to decide - http://radikal.ru/fp/d8274542e0f645f2b4069cb0d072ffd8

    ReplyDelete
  5. I should have read your review before buying 35-100 lens. It loses focus in even NOT so low light. The only way I can control this lens is to use inconvenient Aperture mode. The edge of this lens is f/2.8. But the blur in low light makes it totally unacceptable for the high price. Hardly any reviewers emphasize this major weakness. By the way, with new firmware v1.2, the micro vibrations in video is still there, hardly any improved. If you are interested in this lens, I have tons of failure footage for your reference.

    ReplyDelete
    Replies
    1. I still think the Lumix X 35-100mm f/2.8 is a good lens. It is just not very sharp at 100mm f/2.8. At shorter focal lengths, it does much better.

      Delete