Zoom And Not Refocus? Don't Do It With The Olympus 12-100 f/4 PRO Lens

Cape Blanco Lighthouse, Oregon (click to enlarge)
Olympus E-M5, Panasonic 12-35mm f/2.8 lens @ 23mm; 1/1000th sec. @ f/6.3; ISO 200

There are prime lenses, consisting of only one focal length and then there are zoom lenses, which allow you to change and use many focal lengths without changing lenses. (Leica makes some multi-focal length lenses that allow you to switch between three predetermined focal lengths but nothing in between, but that is another story)  Zoom lenses go back to about 1902 when the first was patented, according to Wikipedia.

In the past, zoom lenses did not have a good reputation due to the many compromises in image quality that had to be made in order to facilitate the ability to offer a variety of focal lengths in one lens.  However, that has changed dramatically as with the advent of computer lens design and modern manufacturing of aspherical lens elements, zoom lenses have largely (not exactly, but pretty close in most cases) equaled most prime lenses.

However, there still is an issue with zoom lenses that most don't realize or to which they don't pay much attention.  The question arises, "Do you need to refocus your lens after you zoom?"  The answer is "depends," of course. 

There are two types of zoom lenses.  A parfocal lens theoretically maintains accurate focus while changing focal lengths, in other words, when you focus at one focal length then change to another, the focus stays accurate.  The other type of multi-focal length lens is called a vari-focal.  This type of lens has to be refocused each time after changing focal lengths as it is not designed to maintain accurate focus.  Zoom lens, in today's culture, is the name often times given to both.

This had me thinking that I have never tested any of my current zoom lenses, either for my Fuji system or my Olympus system.  Are they parfocal or vari-focal?  If I don't set out to test them, I really won't know.  However, my habit has always been to refocus after zooming.  I am almost obsessive about this due to using zooms since the 1970s and how they almost never maintained accurate focus after zooming.

The first two variable focal length lenses I owned were in the late 1970s.  The two are sort of legendary today as they were about the best your could buy at a price anyone who used personal funds (not a corporate bank account) and was not wealthy could afford. The first was the Vivitar Series 1 35-85mm f/2.8 lens.  The second was the Vivitar 70-210mm f/3.5 lens.  Both were excellent for their time.  In fact, you can still find them today in the used lens market.  However, I quickly learned that they did not maintain accurate focus after zooming.

Lets jump to today and modern lenses.

In this post, I tested my excellent Olympus 12-100mm f/4 PRO lens.  I tried to set up a testing protocol to remove any variables other than focus accuracy after zooming.  Here is how I set it up:

(Disclaimer:  this test may or may not be scientific in nature.  I did my best to take into account all variables and eliminate any variable except focus accuracy in the testing procedure.  Your results with your lenses may or may not vary from my results.)

Before I placed my Olympus E-M1 Mark II on a very large and sturdy tripod, and due to the difference in temperature and humidity indoors to outdoors this time of year where I reside, before I began I acclimated the camera and lens to the weather for about 30 minutes. The camera was leveled and squared to the target, a brick wall.  Only the center of the image, where the focus point landed, was considered when gauging accurate focus.

The test was basically set up to cause focus to be more critical rather than less critical. In other words, widest aperture, relatively close distance.  Less depth of field influence

The camera was attached to the tripod via a secure L-Plate.
The camera was set in Aperture Priority at the lens widest aperture, f/4.  (It is sharp at this aperture)
The camera was set on single frame advance using the anti-shock reduction setting to eliminate any potential shutter shock (electronic first curtain)
The camera was set for a 2-second delay to remove any camera movement
S-AF
Single focus point selected
Facial Recognition turned off
Daylight white balance
Base ISO—ISO 200
Image stabilization in camera and lens turned off

Here is how I made the exposures.

For the first test set, I set the lens' focal length at 12mm and focused the camera on a flat brick wall with good detail and contrast (direct sun at an angle). Remember, I'm only going to consider focus accuracy from the small area under which the focusing spot highlighted in the very center of the frame.  Corners and edges were not considered.  I then zoomed to 50mm and exposed a frame.  For direct comparison with that exposure, I then focused at 50mm and made a second exposure. I repeated this, and all test subsequent exposures, 5 times to remove any anomalies or sample variation.

Here is a 100% crop of one of the exposures focused at 12mm, then the lens zoomed to 50mm, then the shutter pressed.

100% crop. Lens focused at 12mm, then zoomed to 50mm, then exposure made without further focusing (click to enlarge)

Here is a 100% crop of one of the exposures focused at 50mm directly.

100% crop. Lens at 50mm, focused and exposure made (click to enlarge)

For this next set, I set the lens' focal length at 50mm, focused, then zoomed to 100mm and made an exposure without further focus.   I then focused at 100mm and made a second exposure. Again, I repeated this 5 times to remove any sample variation.

100% crop. This exposure shows focusing at 50mm, then zooming to 100mm and exposing the frame.
No refocus (click to enlarge)
When compared to the image below, the focus is slightly off, but better than the 12 to 50 images above.

I'm not sure how much you can see the difference between this image and the one directly above while viewing on
the web, but this image is sharply focused but the one above just misses sharp focus. 100% crop. (click to enlarge)

For this next set, I wanted to see the difference between 12mm all the way to 100mm. I set the lens' focal length at 12mm, focused, then zoomed to 100mm and made an exposure without further focus.   I then focused at 100mm and made a second exposure. Again, I repeated this 5 times to remove any sample variation.

100% crop. It can't get much worse than this.  The lens was set to 12mm, zoomed to 100mm then an
exposure was made.  (click to enlarge)

100% crop. For comparison, this was refocused at 100mm.  Nice and crisp.  (click to enlarge)

For this last set, I wanted to see the difference when zooming the other way, from 100mm to 12mm.  I set the lens' focal length at 100mm, focused, then zoomed to 12mm and made an exposure without further focus.   I then focused at 12mm and made a second exposure. Again, I repeated this 5 times to remove any sample variation.

This exposure was made after focusing at 100mm, then zooming to 12mm.  No refocus
200% crop as it the exposure was made at 12mm which give a better view of the details (click to enlarge)
Even at a 200% magnification, I did not see a difference between not refocusing and refocusing.

This exposure was made focusing at 12mm directly.
200% crop as it the exposure was made at 12mm (click to enlarge)

To sum it up.  Good news and bad news.  If you focus wide then zoom in to a longer focal length, it wasn't even close!  If you don't refocus at the longer focal length and if your lens is like mine, your resulting image will be out of focus.  That was consistent in every frame.  The bigger the difference between the focal length at which you focus and the longer one at which you make your exposure, the more your image will be out of focus.

However, if you focus at 100mm, then zoom out to 12mm, you should be okay for accurate focus, which makes sense if you think about it.  Depth of field should compensate for any moderate focusing error that may exist.  Even at 200% magnification, I could not see a difference between the images that was not refocused and the ones that were.

Is this lens a "parfocal lens?"  To me it technically seems to be a vari-focal lens.  Just a word to the wise.

You may want to do one of two things.  You may want to develop the habit of always focusing after zooming.  You may also want to test your own lenses of any manufacturer to better understand their characteristics.  Again, another example of better knowing your gear for more success.

Next, I'll run the same tests with my three Fuji zoom lenses.

Thanks for looking. Enjoy! 

Dennis A. Mook 

All content on this blog is © 2013-2017 Dennis A. Mook. All Rights Reserved. Feel free to point to this blog from your website with full attribution. Permission may be granted for commercial use. Please contact Mr. Mook to discuss permission to reproduce the blog posts and/or images.