[Doundounba]

Samyang 12mm F/2? Just 245g. Not available in Pentax mount though, but Fuji-X, Sony-E, Samsung NX...

[ChristianRock]

In both 15mm and 12mm versions,
it was also made for mirror-lockup use on Nikon F DSLRs:

https://www.cameraquest.com/VCSL1215.htm

I'd toyed with the idea of snagging one to try on a K-01,
until the 08 on the Q7 scratched that itch completely.

So? If you get the mirror out of the way you can stick any mirrorless lens in there. But that's not a viable option, I would say, unless you don't compose your shots.

Samyang 12mm F/2? Just 245g. Not available in Pentax mount though, but Fuji-X, Sony-E, Samsung NX...

Register distance doesn't allow this to be implemented for SLRs. This is exactly what we were discussing regarding the Voigtlander also.

Laws of optics and physics, guys and gals...

[lytrytyr]

So? If you get the mirror out of the way you can stick any mirrorless lens in there. But that's not a viable option, I would say, unless you don't compose your shots.

Check out the link to see how its done on (D)SLRs.

On the K-01, you compose with the rear screen.

[ChristianRock]

Check out the link to see how its done on (D)SLRs.

On the K-01, you compose with the rear screen.

Because it's mirrorless.

I suppose you could use Live View in modern DSLRs, and somehow find a way to keep the mirror up all the time between shots, but it defeats the purpose of having a DSLR. Why not just get a mirrorless? There's no way an SLR will win the size battle against the mirrorless cameras.

[johnyates]

Design of the perfect lens would prioritize wideness and sharpness and compactness,

Wide, sharp and compact. Pick two.

[ChristianRock]

Wide, sharp and compact. Pick two.

It would be even more impossible you add "cheap" to the list of requirements

[Nicolas06]

It's not quite in fisheye territory, but that lens has massively curvilinear characteristics. Curvilinearity is a gradient, not a shorthand for fisheye/notfisheye, and it definitely has much more pronounced distortion than lenses typically described as rectilinear. Take a look at the spires in the photo above,

Or this tree and building in the corners of this shot :


Or this skyline:


This horizon:


More bendyness:


These are all significantly worse than the Pentax-DA 14mm, which I own and would not consider to be truly rectilinear either.

Look more to me you just forgot to enable auto correction of optical distorsion on your software... DA15 or DA21 or many other have distorsion.

[Nicolas06]

DA21 has 2% distorsion... Could produce the kind of issue you showed for the horizon... DA15 is a bit better at 1.5% and the 16-50 as more than 3% distorsion.

As for having good UWA, Samyang produce quite a few they are big but sharp corners to corner and also come with wide apperture like 14mm f/2.8, 16mm f/2...

Sigma has a really nice 8-16, Pentax a good 12-24 there no impossibility at all... Things tend to be bigger, that for sure, but a slow prime lens wouldn't need to be that big.

To me look more lenses maker think there no market:
- high end is for FF not APSC and FF doesn't need such short focal lengths anyway.
- expensive lenses kind of justify their price partically by being faster.

[Edgar_in_Indy]

Back to the OP's original questionof pocketability, I find my Sigma 8-16mm to be much more pocketable than my old Sigma 10-20mm. Not that the 8-16mm is small, but it is narrow and strait, whereas the 10-20mm was wider and flared out, so it was not pocket friendly. So my 8-16mm fits in a pocket on the side of my camera bag, while my 10-20mm would not.

The 8-16mm would be heavy in a jacket pocket, but it should not have trouble fitting, and seems to be the only K-mount lens in existence that meets the requirements of the OP.

[Nicolas06]

Did you even bother to read the posts above in this thread? I've already covered the Samyangs and the Sigma and their deficiencies.

A fast, sharp, reasonably priced, and small hypothetical "12mm Limited" is an impossibility in K mount.

You seem to be entirely missing the point of this thread.

I said that there zooms that are big, but not necessarily that big. Make a prime with same apperture it would be far smaller. I said there are prime that do exist but with wider apperture... Reduce the apperture and you are done.

As if an DA12 f/5.6 could be small enough I have no doubt. Would it manage to sell well, that's another story. They could even manage 12mm f/.4 and stay under FA31 ltd size.

I don't think there a big enoug market for that:
- 12mm is not that far 15mm and we already have something
- 12mm start to be pretty extreme and thus less are interrested.
- this would be expensive and slow. Who would want to pay for that?

I understood the subject and when many say "that impossible" I just think they are not innovators or guys I would want in my company.

[lytrytyr]

- 12mm is not that far 15mm and we already have something

Those two focal lengths are far enough apart
that Voigtlaender saw fit to offer both.

- 12mm start to be pretty extreme and thus less are interrested.

!2mm is not that extreme on APS-C.

It's about the same angle as the wide end
of the 08 zoom on the Q7.

[Nicolas06]

That is one of the biggest rhetorical cop-outs I've ever seen. I have 600 years of research on the physics of light and optical engineering on my side. You have ... your opinion.

Not wanting ot be offensive but many things change, and they change quite fast now. Far less than 600 years ago making a wide angle was really difficult to make and the guys at that time would have say that with more than 500 years of history one would have already managed to do it already if it was possible. For sure some of the best optician would have said such lenses would break the laws of physics. The laws of physics didn't really change but now wide angle are everywhere and UWA are common. And so called "laws" are just a model we make to simplify and understand reality... Such models evolve with time.

I do not expect Pentax engineers to make a breakthrough in physics... But I do hope that an optician somewhere is open minded enough to try new things and try to approach the problem from a different angle and find a way to get the result in an innovative way that would solve the issue.

Here seems to me we need far less... There already 10 & 14mm primes, there already zooms. They are big but are zooms or provide "fast" appertures. Designing a smaller lens that would cover 12mm (likely a prime) is not research, this is just an engineering job. It might be bigger than DA15, but it will be smaller than Samyang 10mm, DA12-24 or Sigma 8-16 zooms. This might come with going for f/5.6 or why not, f/8 max apperture but you'll get something, that for sure.

But the first step is to try.

[Nicolas06]

So you are backpedaling and saying 12mm now? The discussion was whether there could be a compact, rectilinear Limited Series-quality 10mm lens in K mount. And oh, it has to be AF. I said in my very first post that 12mm is about as far as you'd likely be able to go. 10mm is out of the question if you want something that isn't extremely large (given the necessary size and curvature of the obligatory front element) , unacceptably slow, not rectilinear, or makes severe sacrifices in edge performances (i.e. not Limited quality). The closest candidate is the Canon, which has major light falloff in the corners and is a large lens. The Samyang? Horrible distortion, falloff, and manual focus. The Sigma Zoom? Horrible distortion, large size, and major light falloff. The existing Pentax zoom? I's a fisheye at the short end. The voigtlanders? Very bad curvilinear distortion, falloff, manual focus, AND IT IS FOR A VERY MUCH SHORTER REGISTRATION DISTANCE THAN THE K MOUNT, SO IT IS NOT RELEVANT AT ALL, SINCE IT WORKS WITHIN A COMPLETELY DIFFERENT PHYSICAL SET OF PARAMETERS.

Saying that it's merely an engineering problem is like saying that a man giving birth is only an engineering problem. The constraints you are working under simply do not allow it in any way that meaningfully resembles what you want to achieve. The laws of physics relevant to building a camera lens are by this point as well-understood as the laws of Newtonian motion. To try to wave away this objection by saying that I am being closed minded because "laws are just models" we "might understand reality better" in the future is akin to saying I might jump headfirst out my 4th floor window and not break my neck because we don't understand exactly what causes gravitation, so maybe just maybe I will float away instead of falling to the ground.

I take it you'd be willing to go first?

The intial poster did say on the first page:

"These are all possible solutions. I like the (so far non-existent) 12 mm Ltd best. "

So:
- You argue 10mm is impossible when I speak of 12mm.
- you are shouting (caps)
- you are complaining that subject is 10mm, not 12 while poster of the thread clearly said in his second post of first page that 12mm is a solution.
- you are happy to conclude showing a long solved problem. People I know solve the issue to go to the ground from 4th floor by using the nearest elevator or stairs...

I don't see the point anyway to spend 100 more post on the topic. You think the design is impossible. That a point of view. I don't think so, that's another point of view. Let's go on.

[narual]

Hey, DC! Suggestion: chill out. You're being rude to people who are trying to help.

[normhead]

The Sigma 8-16 is only rectilinear down to 12mm... but 12 should be doable/ as in Sigma has already done it.