[Class A]

You never struck me as a person to be so affected by cognitive dissonance.

I'm not affected by cognitive dissonance.

I simply deplore the fact that Pentax often prioritises size and build quality over optical quality and lens speed. I'd rather carry around a big plastic-fantastic that produces fantastic images than a jewel that is beaten by zooms (given that the difference between a zoom and a prime is not just that the latter can be less flare-prone).

[Digitalis]

I simply deplore the fact that Pentax often prioritises size and build quality over optical quality and lens speed. I'd rather carry around a big plastic-fantastic that produces fantastic images than a jewel that is beaten by zooms

I have said before, the differences between these specialised lenses are in the end, not that big. The Sigma 30mm f/1.4 has worse corners than the DA15mm f/4. A modern 30mm f/1.4 aspherical lens has NO excuse for such an optical deficiency. I would still buy the DA15mm f/4 ASPH because of its ability to use filters, unobtrusive form factor, and build superb quality. I would also buy the Sigma 8-16mm again* because of the High IQ for Landscapes, and Architecture photography.

[Class A]

A modern 30mm f/1.4 aspherical lens has NO excuse for such an optical deficiency.

I'm sure the designers of that lens wanted it to perform exactly the way it does.

The 15/4 needs good corner performance because it is a wide-angle lens. The 30/1.4 does not need good corner performance because it is a specialist normal that is often used for portraits.

I believe the properties of a lens should be optimised for its intended use and I applaud Sigma that they were brave enough to neglect the 30/1.4's corners in favour of its centre sharpness and bokeh. Too many lenses these days aim for good bench performance and hence fail to have some desirable character.

[civiletti]

I'm sure the designers of that lens wanted it to perform exactly the way it does.

The 15/4 needs good corner performance because it is a wide-angle lens. The 30/1.4 does not need good corner performance because it is a specialist normal that is often used for portraits.

I believe the properties of a lens should be optimised for its intended use and I applaud Sigma that they were brave enough to neglect the 30/1.4's corners in favour of its centre sharpness and bokeh. Too many lenses these days aim for good bench performance and hence fail to have some desirable character.

You consider 30mm lens a portrait lens?

[Digitalis]

I believe the properties of a lens should be optimised for its intended use and I applaud Sigma that they were brave enough to neglect the 30/1.4's corners in favour of its centre sharpness and bokeh.

Well that is where we differ: I don't applaud sigma for the 30mm f/1.4 ASPH and I don't think Sigma is brave, lackadaisical would a more accurate term to describe their crass attempt to make a general purpose "Normal" lens for APS-C format. You remember what normal lenses used to be like? they set the standard for a lensmakers entire lens line up. A good example is the classic 1967 Takumar 50mm f/1.4 - sharp from corner to corner at f/4 and it has excellent bokeh - by contrast the sigma 30mm f/1.4 isn't nearly as sharp as an old takumar in the corners by f/11, Coma interferes with the Bokeh at the corners as well forming hard edged "wings". There isn't a single thing I find appealing about the Sigma 30mm f/1.4.

[Class A]

There isn't a single thing I find appealing about the Sigma 30mm f/1.4.

That's fine.

I personally find the images the Sigma 30/1.4 produces to be very appealing.

If one can spend multiple grands on a Leica lens then one can probably do better, but that's pretty irrelevant for most of us.

[Class A]

You consider 30mm lens a portrait lens?

On APS-C, 30mm is practically a "normal" lens, i.e., fulfils the role a so-called "fast fifty" has played for FF format.

A normal lens is not a "classic" focal length for portraits but "fast fifties" have been used for portraits in the past and I find the example portraits done with the Sigma 30/1.4 I've seen very convincing.

N.B., it is harder to get great performance out of a fast 30mm K-mount lens compared to a fast 50mm lens K-mount lens. That's part of the reason why the standard normal focal length in film days was 50mm as opposed to the theoretically more appropriate 43mm (the diagonal of the film format). Oscar Barnack partially chose 50mm because it was (and is) easier to get good performance out of a fast lens if its focal length is great than the registration distance of the mount.

[Digitalis]

I really don't get the "me too" criticism regarding the 9 aperture blades and why the Sigma lens is supposedly given an advantage.

Sorry I didn't reply to this comment earlier - but the reason I picked on the Sigma lens for this is because an 8-16mm f/4.5-5.6 isn't really a lens designed for bokeh, 9 bladed apertures make sense on a 85mm f/1.4 lens not on a UWA zoom lens. Pentax understands this, which is why the DA15mm f/4 has a 7 bladed aperture in it so that highlights are rendered as stars - some people don't like this effect and in some circumstances it can be visually distracting in an image. But 9 bladed apertures don't really create this star effect anywhere near as well as a 7 bladed aperture does.

[Class A]

Sorry I didn't reply to this comment earlier - but the reason I picked on the Sigma lens for this is because an 8-16mm f/4.5-5.6 isn't really a lens designed for bokeh, 9 bladed apertures make sense on a 85mm f/1.4 lens not on a UWA zoom lens.

The bokeh of a lens is determined by its PSF, not the number of aperture blades.
In some circumstances the shape of the aperture and hence the number of blades can be seen in images (OOF highlights) but I personally don't think the number of blades is terribly important in these cases. Whether or not they look asymmetrical or jagged is of more concern to me, personally (something Pentax is not particularly strong in).

Pentax understands this, which is why the DA15mm f/4 has a 7 bladed aperture in it so that highlights are rendered as stars...

I believe Sigma equips their wide-angle lenses with nine blades because they want users to have 18 rays in star burst shots.
Do you not think that 18 rays look better than 14?

I have no idea why you believe that seven blades create better star burst shots than nine. Many users specifically recommend the f/3.5 version of the Sigma 10-20 over the variable aperture one, because the former has nine blades (yielding 18 rays) whereas the latter only has six (yielding only six rays).

[Digitalis]

The bokeh of a lens is determined by its PSF, not the number of aperture blades. In some circumstances the shape of the aperture and hence the number of blades can be seen in images (OOF highlights) but I personally don't think the number of blades is terribly important in these cases. Whether or not they look asymmetrical or jagged is of more concern to me, personally (something Pentax is not particularly strong in).

I agree up to a point, the shape of the aperture blades is of paramount importance, not the number number of blades. Focus distance has a significant impact on Bokeh. On the subject of aperture blades I have a 9 bladed Sigma 100-300mm f/4 APO EX and those blades aren't anywhere near circular so you get nonagon bokeh from f/5.6 up, I personally find this characteristic visually jarring.

Do you not think that 18 rays look better than 14?

This is a matter of personal preference but I think the 14 point stars of the DA15 look more interesting, if you have multiple light sources in the scene 18 point stars often will be visually distracting where 14 point stars would be somewhat less distracting, less is more(up to a point).

FA31 - 9 bladed aperture:


DA15mm - 7 bladed aperture:

[Digitalis]

I have been mucking around with the Sigma 8-16mm, like the DA15mm f/4 ASPH Limited it is a lens that encourages experimentation because the results can be...interesting:


Pentax K5IIs - Sigma 8-16mm f/4.5-5.6 ASPH @ 8mm f/11 1/125th ISO 80 - shot at ground level.

[johnc]

8mm HDR (hopefully subtle-ish!)

[johnc]

Two more from the 8-16 - black and white HDRs this time (desparate attempt to get some photos despite the very dull weather):

[dadipentak]

desparate attempt to get some photos despite the very dull weather):

Very successful attempt, I'd say!

[normhead]

I had my first real trip out with this lens yesterday. I have to say, I'm very impressed. First, because I had my DA 18-135, Sigma 70 macro and A-400 along on the walk, and never took the 8-16 from the camera.

On shots like the one below, which were taken right into the sun, there was no noticeable CA or purple fringing. I'm not sure if you had told me this before, I would have believed it. The photozone CA numbers are terrific, but this was still unexpected.



Another surprise, 16mm is long enough for walking candids... who would have thought?



And it focusses close enough for some isolation images.



All in all great lens, I expected a lot from it, and it's more than I expected.





I'm already planning to reshoot a lot of images from the last few years, where distortion ruined the image. My images are spread equally between 8, 12, and 16mm, every setting on this zoom is useful. I might have to take it off the camera someday... but not just yet.