home > photography
Review the quality of the Minolta 100-300mm xi lens as well as determine the optimal f-stop. This lens can be had cheaply on the used market, but is it worth using?
I took a small tripod, and braced it against my chest. (Ideally, I should have taken the time to get my large tripod.) Sometimes, Super Steady Shot (SSS) would produce blurry results when the camera was braced, so some photos were retaken with SSS off (as recommended in the manual). A more rigorous test in better light would be preferable, but this will have to do for now.
These photos were shot hand-held in bright midday sun. I used "A" mode and Auto ISO, allowing the camera to adjust the other settings as it saw fit. There are signs of heat wave distortion, particularly near the bottom of the orange building, where the light is passing over hot railroad tracks. So, wavy lines are not the fault of the lens, but a natural artifact of a 91 degree day.
A 100% crop is pasted in the lower-left corner.
While this looks a tad soft, I think you'd have to print fairly large before you could notice problems.
Note that a higher ISO was chosen, but the f11 photo still has slightly more detail than the 6.3, and slightly less "purple fringe" effect. Even at f11, there is some purple fringe, but it would probably not be seen unless enlarged, and that can often be fixed in photo editing software.
Just for reference, here is the same scene at a still not-so-wide angle of 105mm (taken with the 28-105xi from about the same location):
And the same scene at 28mm:
It seems amazing to look at this photo, then look back and read the posters in the window (in the earlier photos).
One word: bokeh.
The following photo of the speedboat is a 100% crop, but I did a bit of sharpening, PF removal, and noise removal, to try to get optimal results. (Then, because it's so large, I used an aggressive level of JPEG, so you'll see the blocky artifacts in areas. It doesn't fundamentally change the overall look, but it does mess up some detail, so I want to point out that the original does have a bit more detail.)
I expected it to be sharper. Is the focus off? I think the heat distortion is the main problem, with f8 being a tad soft anyway. The f11 pics came out better, but f8 would be an easier aperture to work with. It's a bit rough-looking zoomed in this far, but you'd have to print very large (or print this crop) to notice problems. There's still a bit more detail here than if you used, say, a 5 megapixel camera to capture the same scene (which you can see if you downsize this photo). In practical use, it's not quite as bad as it looks, in other words.
This is what it looks like when you aren't obsessing over the 100% crop:
This 100% crop used f11, and is unmodified except for resaving:
You can see the wavy lines on the fence. There is a lot of heat distortion in the air, and this is probably causing a great deal of the quality problems with the photo.
Here is the full view:
This was taken with similar early evening lighting as the photos taken in the teleconverter test.
While this isn't a back-to-back test against the lens with a teleconverter, it does appear that the 100-300xi can resolve as much if not more detail and do so with fewer artifacts. (While I used a shaky tripod, I believe the Super Steady Shot was off.)
Other test photos at longer distances produced similar results -- soft at 5.6, and pretty good at f8, and much better at f11.
Best to only use the lens in bright light. I am not sure how it compares to other 300mm lenses, but at least this one is very inexpensive. What would an expensive lens add? Less CA and other distortion, but you still might not want the shallow depth of field by having a wider aperture. (Although, it would be nice to have something a bit better at 5.6 or 6.3.)
I love the bokeh -- nice and soft, not with harsh edges like it is with many lenses.
Noticeable purple fringe and CA distortion, but not too severe, IMHO. Highlights tend to set off the PF.
Gives a lot of detail, just not as sharp as one may want when viewing at 100%. What do you want for $72?
Last updated: June 8, 2008.