I have it and it is outstanding optically. Surprisingly so, even at the 20 end. It's also relatively compact, about the same size and weight as the fixed 23. At that wide, you will have head distortion/elongation issues at the edges and corners of the frame though.
Great to know. I know somebody on here said it's even smaller than the XF 8-16 2.8...which is kind of mind boggling. Then again the 50 1.0 is similar in size to the GF 1.7 lenses so anything is possible at this point. gear-nut wrote:
I have it and it is outstanding. Surprisingly so, even at the 20 end. It's also relatively compact, about the same size and weight as the fixed 23.
The GF20-35 is quite a nice lens overall, but one of the things that drives me crazy about all GF lenses, at least all of the half a dozen that I own, is the lack of a focusing scale on the barrel of the lens so that you know for a fact when you're at the near end of the scale, which would be very important for doing environmental portrait work.
The digital focusing scale that you can show in the viewfinder and rear LCD is useless and distracting and more importantly, not even close accurate.
There's a good bit of barrel distortion inherent in the lens that is automatically and very effectively dealt with in Capture One but can subtly alter your crop if you're not aware of it and if your precise crop is important.
If lacking an on lens focus scale is the biggest complaint about this lens then I think I have nothing to worry about. Peter Figen wrote:
The GF20-35 is quite a nice lens overall, but one of the things that drives me crazy about all GF lenses, at least all of the half a dozen that I own, is the lack of a focusing scale on the barrel of the lens so that you know for a fact when you're at the near end of the scale, which would be very important for doing environmental portrait work.
The digital focusing scale that you can show in the viewfinder and rear LCD is useless and distracting and more importantly, not even close accurate.
There's a good bit of barrel distortion inherent in the lens that is automatically and very effectively dealt with in Capture One but can subtly alter your crop if you're not aware of it and if your precise crop is important. ...Show more →
You can look in my flickr account for some recent portraits made with this lens. I don't have much experience with it yet because I just got it recently [for mostly "quirky " portrait work]
Yeah I actually saw some of these earlier this week and almost DM'd you about your experience with it.
For context I intend to use it for stuff like this, which I shot on a 24-70 and the most zoomed in I am is 30mm. Felt the 20-35 is better for these than the 32-64 would be and there are times when I'm asked to be wider, since fisheye is popular, but I don't want the fisheye look though. Z09_4750 by Marcus Beasley, on Flickr Z09_4714 by Marcus Beasley, on Flickr Z09_4625 by Marcus Beasley, on Flickr Z09_4607 by Marcus Beasley, on Flickr Z09_4814 by Marcus Beasley, on Flickr
olegkin wrote:
You can look in my flickr account some recent portraits made with this lens. I don't have much experience with it yet because I just got it recently [for mostly "quirky " portrait work]
gear-nut wrote:
One more thing: with people, it’s actually better to not correct barrel distortions as it leaves heads looking more natural.
That's generally the opposite of my experience. Not correctly the barrel distortion leaves the person's head bulging out even more so correcting it almost always leaves it more natural looking, but if that's the effect you want, then move the slider to zero or somewhere in between.
I saw a fashion portraits shoot with wide-angle a couple years ago and I liked them a lot. Since then I was trying to take a few wide-angle shots with gf30. It worked out so great almost every time, but pictures looked too normal to my surprise. So I wanted to get a bit wider, and after watching the same video above, it convinced me gf20-35 is the way to go. And as a bonus lens doubles as wide angle landscape lens. Only wins, no negatives.
One thing I ask models to do to reduce unpleasant distortion is to move their hips back slightly. Additionally, I ask them to lean forward when I shoot from a lower angle. This reduces the difference in distances from the head, hips, and legs to the sensor, resulting in better-looking images.
Wonderful lens and a “standard” lens on my 100II. I didn't use it for portraits yet. So far, I’ve only used my GF110 for that. Video is however inspiring, so I’ll definitely give it a try.
Peter Figen wrote:
That's generally the opposite of my experience. Not correctly the barrel distortion leaves the person's head bulging out even more so correcting it almost always leaves it more natural looking, but if that's the effect you want, then move the slider to zero or somewhere in between.
It will with pincushion but not barrel.
gear-nut wrote:
It will with pincushion but not barrel.
You've got it backwards. Uncorrected barrel distortion will cause a head to bulge in general but placement in the frame will affect as well as how much inherent distortion there is in the lens. A lot of pincushion can cause this if it's at the edge of the frame but then it can also do fun house effects as well. The easiest way to see this is to just move the slider in C1 from 100 to zero and back on any lens that has inherent barrel distortion like the 20-35 or the 32-64.
You can also just shoot someone with a non corrected fisheye and really see a pronounced effect. Generally with both the 20-35 and the 32-64, the correction is mild but it is enough to, in my opinion, make sure that the correction is on for people. But everyone will make their own decision with their own images. Just another thing to be aware of.
I'm not fond of the 20-35's vignetting in its uncorrected state. If you shoot mostly at low iso though then the sensor has sufficient latitude to correct it without issue.
Peter Figen wrote:
You've got it backwards. Uncorrected barrel distortion will cause a head to bulge in general but placement in the frame will affect as well as how much inherent distortion there is in the lens. A lot of pincushion can cause this if it's at the edge of the frame but then it can also do fun house effects as well. The easiest way to see this is to just move the slider in C1 from 100 to zero and back on any lens that has inherent barrel distortion like the 20-35 or the 32-64.
No I don’t. Again, I am talking heads near edges and corners, central can be different. With a fisheye lens, almost 100% barrel distortion, heads near the edge or corners look relatively “normal.” Correct the barrel and those heads elongate, stretch and distort significantly as they linearize.
Center frame is different. While relatively normal shape, they can and usually will look a little bulbous or protruding. Correcting the barrel can improve those by flattening them a little, but then any at the edges would stretch and distort irregularly, looking “wrong.” IME the edge elongation effect starts getting visibly distracting around ⅔ the way out from center.
