Photozone reviews the 100 STF GM

Started Aug 22, 2017 | Discussions thread
Jerry Fusselman
Jerry Fusselman Contributing Member • Posts: 849
Lots of mistakes in the review. Still appreciated!

I very much appreciate the review, though it has quite a few comments that belie an incomplete understanding of the lens. Here are some of its problems (according to my understanding, of course, meaning I'm always happy to hear comments and corrections):

  1. "To make things even more complicated regarding the f- and T-values, they range from f/2.8 to f/20 and T/5.6 to T/22 respectively." This seems confused and pointlessly confusing. The apodization effect has no measurable effect above f/8. All lenses have a bit of light loss due from the elements. We don't even know where the reviewer's Tstop values come from. Until we know something specific, I think it is safer and more descriptive to go by the well known schedule for this lens: f/2.8 = T/5.6, f/4 = T/6.3, f/5.6 = T/7.1, f/8 = T/8, and they are equal above there. That is, above f/8, any difference between T-stop and f-stop matches ordinary nonzoom lenses.
  2. "The camera will always display the T-value which feels a little unfortunate because such a lens is primarily about the depth-of-field rather than the effective speed so it's a little difficult to estimate the effect on the depth-of-field when stopping down." This is nonsense. The depth of field won't change like a standard lens would when you switch from f/2.8 to f/4, so f/stop doesn't convey the information the reviewer claims it would convey. f/stop doesn't mean what he thinks it means with STF lenses. However, T-stop is useful, for it tells you how much light you are getting. Since quantity of light can be measured well with an STF lens (using T-stops), but depth of field cannot (using f-stops), it seems clear to me that T/stop is more useful.
  3. "The latter gives you access to a maximum object magnification of 1:4 which may not quite touch the realms of a true macro lens but which may be "good enough" for many as far as close-focus applications are concerned." This is true, but the part about "good enough" is hugely misleading in my view. My issue is the reverse: "When is my Sony 90mm macro good enough compared to the dreamy results I get with the 100mm STF?" I should probably test it to be sure, but I am happy to predict that few image-evaluation juries looking at macro images of flowers, for example, will prefer the best I can get with my Sony 90mm macro to what I can do with my Sony 100mm STF. The 90 macro is more convenient, yes, because it is easier to get past 1:4, I'll concede. But I've been using the Canon 500D close-focusing lens with delightful results to around 1:2.
  4. "One of the "oddities" of the lens is the vignetting characteristic. Either the lens is over-designed or Sony is applying some tricky under the hood. There is no increased vignetting to speak of at f/2.8 (T/5.6) neither in RAW or auto-corrected mode. The remaining light falloff is negligible." The reviewer imagines that the only great-bokeh trick Sony employed is the apodization element. Wrong. Sony also used an unusually large front element to help eliminate any visible cat's eye effect at the periphery. This is not "tricky under the hood" or any other form of cheating. The lens simply provides less periphery-light loss than probably any short telephoto lens you can name. So the complete lack of cat's-eye OOF results is exactly what makes for no increased vignetting when wide open. It should be obvious that cat-eye bokeh means that a good deal of light is not reaching the corners.
  5. "Please note that we tried to estimate the f-stops of the corresponding T-values. Sony doesn't provide any guidance for this. The reason why the Sony lens shows this variable T-value to f-stop behaviour isn't completely clear. We didn't observe this in other STF lenses that we tested so far." Huh? All STF lenses with a variable aperture have variable conversions between T-values and f-values. I'm not sure what guidance you could want from Sony if you're estimating it yourself.
  6. They write that f/4.5 = T/8. I doubt that, and it corresponds neither to the lens markings nor the EXIF data. It is really confusing for the reviewer to have his own schedule of mapping f-values to T-values, especially when he says that he wishes he had guidance from Sony to do it. (Sony has done it already, as I explained in #1 above.) I can't tell how the reviewer is setting the lens for his bokeh examples, but I'd like to know.
  7. "However, what you can also observe is that the effect is pretty much only visible in full at T/5.6. If you stop down by just 1/3 of an f-stop the effect is almost gone showing the conventional highlight discs again. " By the reviewer's captions, I am almost sure he means stopping down 1/3 of a T-stop, not f-stop, which is quite a difference. The lens doesn't even have a marking for 1/3 of an f-stop down from wide open. It does have one for 1/3 of a T-stop, which is one full stop change in the f-stop.
  8. "[Y]ou will pretty much always prefer to shoot at f/2.8 (T/5.6)." Going from f/2.8 up to f/8, you gain a bit of overall sharpness, you gain more depth of field (but not as much as you likely expect), but your bokeh quality gradually changes to a more common look. Sometimes, you will want this tradeoff part way or all the way to f/8. It is therefore quite silly to say that "you will pretty much always prefer" to shoot wide open with this lens.
  9. "I'd rather state that the lens is aligned to still life and nature photography where you are often facing very busy and contrasty fore- and backgrounds and this STF lens can kill hard edges like no other." This is inaccurate for two reasons: First, a wide open f/1.4 lens is likely to be able to "kill" a background more than the 100mm STF. More importantly, this lens is more about loving the background than killing it. It helps you make a beautiful (fore- or) back-ground, not a "killed" background. If you hate where your subject is located, maybe you do need a fast lens. But if you love (or at least like) your location's background/foreground/etc., then this STF lens can make your image sing.
  10. "Strangely vignetting is not an issue even at f/2.8 - something that we haven't seen so far in any other lens (which may raise the question whether Sony applied some trickery under the hood)." No. No trickery. Sony designed the lens for this real optical effect, as explained in #4 above.
-- hide signature --

Jerry Fusselman

 Jerry Fusselman's gear list:Jerry Fusselman's gear list
Sony a7R II Sony a9 Canon TS-E 17mm f/4L Canon TS-E 90mm f/2.8 Canon MP-E 65mm f/2.5 1-5x Macro +21 more
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow