The case for the 16mm

For discussion of the E and FE mount mirrorless system
User avatar
bossel
Viceroy
Posts: 1031
Joined: Mon Jul 21, 2008 2:49 pm
Location: France, Côte d'Azur

Re: The case for the 16mm

Unread post by bossel »

When I got my NEX 5 I got the double kit, but never used the 16mm. Sold it 2 weeks ago.
David Kilpatrick
Site Admin
Posts: 5985
Joined: Sat May 19, 2007 1:14 pm
Location: Kelso, Scotland
Contact:

Re: The case for the 16mm

Unread post by David Kilpatrick »

I've just run tests all day mainly on 16mm/24mm full frame equivalents. Tired and a bit depressed. The NEX pancake 16mm is probably the best of the bunch except for the 24mm f/2 Zeiss on full frame. It is a much better lens than, for example, the new 16-50mm f/2. SSM lens used at 16mm. But I'll bet very few owners of the 16-50mm will be willing to see that or admit it.

David
agorabasta
Viceroy
Posts: 1198
Joined: Wed Oct 22, 2008 7:41 pm

Re: The case for the 16mm

Unread post by agorabasta »

David,

That 16mm Nex pancake really can produce some good results, especially now if coupled with newer NexC3. Most of my previous problems with that lens got solved in that particular toy body, and only there near-completely.

The only problem that is there to stay is the loss of contrast, both local and global, with bright light sources in the frame. That's what you often get shooting indoors and/or at night and that is very obviously ruining the overall impression from the results. Even in the good bright sunlit scenes the brighter objects appear more blurred against the rest of the scene - it may really kill the subjective perception of quality.

So the objective assessment of that lens quality would really depend on the choice of criteria. And with a regular set of measurements it may appear better than it really performs.
User avatar
artington
Imperial Ambassador
Posts: 553
Joined: Sun Jan 11, 2009 5:22 pm

Re: The case for the 16mm

Unread post by artington »

David Kilpatrick wrote:I've just run tests all day mainly on 16mm/24mm full frame equivalents. Tired and a bit depressed. The NEX pancake 16mm is probably the best of the bunch except for the 24mm f/2 Zeiss on full frame. It is a much better lens than, for example, the new 16-50mm f/2. SSM lens used at 16mm. But I'll bet very few owners of the 16-50mm will be willing to see that or admit it.

David
David

Did you include the Voigtlander 15/4.5 VM in these tests? A bit slow, perhaps, but I find Voigtlander lenses to be pretty good. My VM 28/2 spends a lot of time on the NEX-5 (via the excellent Metabones adapter) and the results have been excellent to my eyes.
Vidgamer
Imperial Ambassador
Posts: 527
Joined: Wed Jun 11, 2008 3:37 am

Re: The case for the 16mm

Unread post by Vidgamer »

David Kilpatrick wrote:I've just run tests all day mainly on 16mm/24mm full frame equivalents. Tired and a bit depressed. The NEX pancake 16mm is probably the best of the bunch except for the 24mm f/2 Zeiss on full frame. It is a much better lens than, for example, the new 16-50mm f/2. SSM lens used at 16mm. But I'll bet very few owners of the 16-50mm will be willing to see that or admit it.

David
Depressed because the fancy, expensive gear doesn't do better? Well, I guess that makes me less depressed to pay full price for the 16/2.8! So, it evens out. :mrgreen:

But seriously, I'm sorry if I distracted you on your Saturday with my thoughts on the 16mm. I recall that you had tested this early-on and had more positive thoughts that just about anyone on the lens. Thanks for the efforts! I think it's worthwhile to get to the "truth" on this lens, as it continuously is used as an example of what is wrong with Nex.

By the way, the meme on DPR now seems to be "lens variation" -- that explains why some hate the lens and some love it. (Partially due to a generic article on lens variation just appearing on DPR.) I don't think it's variation except in a few cases. For the most part, I suspect all of the copies are soft in the corners and tighten up by f8.

Part of the problem is that real photos seem to look better than test photos. A series of test shots in the backyard struck me as a bit depressing. I often test hand-held, which is generally how I'm going to use the camera (while on the move). The hazard of doing this is that occasionally, f8 will end up looking like f2.8 -- must have had too much motion with the reduced light! At any rate, yes, the corners are a bit of a mess, and yes, with light that can wash things out a bit, but as I took photos, even low-light at f2.8, they looked pleasant. No,the lens is not "tack sharp" at 2.8, and details suffer, but at ISO 1600, details suffer. But indoor photos looked fun and interesting and the corners did not stand out as particularly objectionable.

I do have a slight concern that maybe my copy is poorer than other copies, but at the moment, I'm assuming not. Not sure how to tell.

Say, is it worth using the lens hood (borrowed from the 18-55)? Unfortunately, there's not enough room to reverse it for storage. Was considering buying an extra and modifying it, but that may defeat the purpose....
Vidgamer
Imperial Ambassador
Posts: 527
Joined: Wed Jun 11, 2008 3:37 am

Re: The case for the 16mm

Unread post by Vidgamer »

agorabasta wrote:David,

That 16mm Nex pancake really can produce some good results, especially now if coupled with newer NexC3. Most of my previous problems with that lens got solved in that particular toy body, and only there near-completely.

The only problem that is there to stay is the loss of contrast, both local and global, with bright light sources in the frame. That's what you often get shooting indoors and/or at night and that is very obviously ruining the overall impression from the results. Even in the good bright sunlit scenes the brighter objects appear more blurred against the rest of the scene - it may really kill the subjective perception of quality.
Someone posted to DPR a shot with the sun in the frame and said it was an advantage that the lens handled it as well as it did. Seemed pretty good... In my testing, PF was kind of bad at wider apertures with the old leaves-against-the-sky trick, but it's hard for me to get too worked up over that. (I can always rescue that for the most part in software.)
So the objective assessment of that lens quality would really depend on the choice of criteria. And with a regular set of measurements it may appear better than it really performs.
I've been thinking the other way. Measurements show the blurry corners, but in normal use, it may not be noticeable. In most of my photos, it just looked like the rest of the oof bokeh areas. Only when I was in brighter light in the backyard did I see how bad the corners looked!

I still don't quite have a handle on this lens, but I guess I'll give it a try.
David Kilpatrick
Site Admin
Posts: 5985
Joined: Sat May 19, 2007 1:14 pm
Location: Kelso, Scotland
Contact:

Re: The case for the 16mm

Unread post by David Kilpatrick »

The focus field of the 16mm NEX is cup-shaped. The corners are focused further away than the centre. Many people test by shooting a landscape or a flat surface. With the cup-shaped focus field, the corners of a grass lawn (for example) will seem to go much more out of focus but if you have trees in the distance at a top corner, they will appear sharper. Cup-shaped focus field goes with pincushion distortion. It is not very friendly for room interiors, as the focus shift is directly opposed to the angle of walls, floor and ceiling and the centre (focus point) is often a far wall. So to get round this you need to stop down, and be sure to focus midway into the room, not on the far wall.

The 16-50mm focus field is cap-shaped, but the 'cap' is a sudden shift into the corners. Tested on a flat surface it will look much worse than the 16mm NEX, but with typical subjects like the landscape or the room interior the focus field tends to follow the planes of the subject. Except, of course, for trees on the horizon. Again, the best answer is to stop down and adjust the focus point to get the best depth of field taking into account the curvature of plane.

This is where the 24mm f/2 scores most highly. It has a perfectly flat focus field and almost no distortion at normal working distances - I have done a review for the British Journal, and when I started, I intended to create an Adobe Lens Profile for it. But the lens simply does not need a profile; there's nothing to correct! Now if the 16mm f/2.8 had been made to that standard, it would be twice the price and not a pancake.

David
Vidgamer
Imperial Ambassador
Posts: 527
Joined: Wed Jun 11, 2008 3:37 am

Re: The case for the 16mm

Unread post by Vidgamer »

David,

That makes sense, and explains a lot of the corner results. The grass in the foreground was noticeably worse in my tests, for example.

Indoors, that makes sense, to try to focus other than on the far wall. Usually that will be the case unless attempting to photograph a big empty space. None of my photos would have been focused (intentionally) on the far wall.

I do wonder if I have a slight "variation" problem in that the blurring appears to be slightly more on one side than the other. Not obvious unless really looking for it.

Contrast and sharpness seems better than the 18-55 once at about f5.

Thanks for the tips!

Gary
User avatar
Birma
Tower of Babel
Posts: 6585
Joined: Sun Jun 29, 2008 3:10 pm

Re: The case for the 16mm

Unread post by Birma »

I agree with Vidgamer that the SEL 16/2.8 is a bit disppointing in tests, but then turns out far more appealing in real use. I am still amazed at how little 'glass' there is in the tiny thing.
Nex 5, Nex 6 (IR), A7M2, A99 and a bunch of lenses.
Vidgamer
Imperial Ambassador
Posts: 527
Joined: Wed Jun 11, 2008 3:37 am

Re: The case for the 16mm

Unread post by Vidgamer »

I have not had much opportunity for photos, but today I took a couple of snapshots using p mode, and it used f4. Corners are a bit of a mess, but the rest of it is pretty sharp I think. CA can be helped with software.

I understand the discussion of field curvature, but there must be more to it than that, as it doesn't seem to matter if the subject in the corners are farther away (as tree branches are still a mess in the corner). Maybe it's the subject matter - bottom corners look not too bad when it's pavement. It's weird though - even at the same aperture, it's better sometimes than others. Hmm.

Regardless, with or without a slight crop, I can probably live with it. But i can see where this would bother one, if too obsessive.

At f5, it seems much better. Maybe I need to try a mode, but with the overcast or dim light, I like the extra speed... Decisions, decisions.
Vidgamer
Imperial Ambassador
Posts: 527
Joined: Wed Jun 11, 2008 3:37 am

Re: The case for the 16mm

Unread post by Vidgamer »

I finally got around to posting some results (had something more interesting to share). I think the lens is sharp enough. Yeah, the corners are sometimes messy, but this is not often a big deal, and not often that you're that wide-open. I took it on this trip as my only lens, and rarely had a desire for a different focal length. Weird.

http://www.flickr.com/photos/avidgamefa ... otostream/

More threads on DPR are going on and on about the lens and how bad it is, while others will respond with perfectly fine photos taken from it. I'll take "poor" lenses like this at affordable prices. Thanks Sony. And thanks David for giving it a closer look.
Vidgamer
Imperial Ambassador
Posts: 527
Joined: Wed Jun 11, 2008 3:37 am

Re: The case for the 16mm

Unread post by Vidgamer »

OK, I like replying to my own replies. :mrgreen:

I tried taking some comparison photos between my 18-55 and 16, and decided that the 16's photos were much better. But I forgot to turn off OSS, so I think the test may be invalid. I'll try again as time permits!
David Kilpatrick
Site Admin
Posts: 5985
Joined: Sat May 19, 2007 1:14 pm
Location: Kelso, Scotland
Contact:

Re: The case for the 16mm

Unread post by David Kilpatrick »

I took a look, but the largest size didn't help me see whether the 16mm was doing well. It certainly seemed to be. I think I looked just before departing off to do something - Dec 23rd is that sort of date...

David
sparaxis
Initiate
Posts: 42
Joined: Fri Nov 07, 2008 10:30 pm
Location: Baltimore USA

Re: The case for the 16mm

Unread post by sparaxis »

If you look at the MTF curves of the lens, it is quite obvious that the extreme corners are poor at full aperture:
SEL16 MTF.gif
SEL16 MTF.gif (14.83 KiB) Viewed 5464 times
The shape of the curves is indicative of field curvature. (Zeiss has some interesting articles on MTF curves on their web site.)

However, in the real world, it is not as big a deal as it seems. Firstly, I tend to crop my images from 2:3 to the more traditional 3:4 or 4:5 proportions for printing, and the corners miraculously disappear! Secondly, I use f8 when I need sharp corners. (Much better than having to use f16 or F22 on an elderly Zeiss folding camera with 3 element Novar lens - I have taken great pictures with those sorts of cameras)

I suspect that the 16mm could have had better corners, but that SONY had a different agenda. Try the 0.75X converter and see what I mean. Suddenly you have a very neat 12mm f2.8 lens about the same size as the kit zoom. And it is no worse than the 16mm by itself in my experience.

The fact that the two converters were announced at the same time as the lens is a clue. I assume that Sony designed the three together, perhaps degrading the base 16mm performance in ways that optimised the performance of the lens plus converters.
Vidgamer
Imperial Ambassador
Posts: 527
Joined: Wed Jun 11, 2008 3:37 am

Re: The case for the 16mm

Unread post by Vidgamer »

I ran another test in the backyard, but I think focus wasn't exact between the lenses... Or perhaps the 18-55 is better in the center but worse elsewhere. Then clouds rolled by as I swapped lenses making things inconsistent.

Overall, I think the 16mm looked better, except in the very extreme corners and the center, where the18-55 did a good job of focusing. Everything else looked marginal on the 18-55, particularly wide open. F8 is probably a wash. The 16mm seemed sharper but I'm not sure it was more pleasant on branches. But overall, color and contrast seemed better.

I probably need a better test subject.
Post Reply

Who is online

Users browsing this forum: No registered users and 18 guests