[kenyee]

But I know there was a design philosophy that apparently cost a lot of Pentax engineers their jobs, when Hoya wanted to make them Canon, Nikon, Sigma clones.
....
[B]I have a Sigma 8-16, sharp edge to edge, distortion corrected etc. but my wife hates the picture taken with it. If she's not right dead centre, they actually distort how she looks. She looks much more natural with an uncorrected lens. .

That's the first time I heard of Hoya firing the lens designers for not cloning Canikon lenses. Where's your source for this? I always thought they were layed off because Hoya was cutting costs and new lens development wasn't high on their agenda for selling the camera division

As for the Sigma 8-16, it's an ultrawide. You get craploads of perspective distortion from it if you put anything near the edge. Even w/ my Sigma 10-20, you get this. Don't put people near the edges (or even near the edges of a 17-70) because it'll make them look short and fat
They're mostly landscape lenses because landscapes don't complain about being stretched out and abnormal looking

[Clavius]

That's the first time I heard of Hoya firing the lens designers for not cloning Canikon lenses. Where's your source for this? I always thought they were layed off because Hoya was cutting costs and new lens development wasn't high on their agenda for selling the camera division

With Hoya being an optics company, I always thought that was a strange theory. You'd think would boast about their optics through lens designs. No I think those engineeres were layed off because they refused to design colourful FA ltds.

[ElJamoquio]

Thank god for FF. The corners are always sharper than APS-C corners.

[normhead]

With Hoya being an optics company, I always thought that was a strange theory. You'd think would boast about their optics through lens designs. No I think those engineeres were layed off because they refused to design colourful FA ltds.

Well then, they deserved to be laid off, what am I supposed to put on my white k-01? Those guys left me hanging.

[MMVIII]

Vignetting is easy to fix in post or, even automatically with in camera lens corrections. This actually seems to be the "modern" way -- just look at Olympus and Panasonic for examples of companies that "fix" their lenses flaws in camera.
!

Just a sidenote - fixing in post, yes, but nevertheless you just push the corners, for example with the Sigma 35mm f/1.4 and a vignetting of 2,05EV wide open you are photographing with iso400 in the corners while the center is at iso100, or, worse - because low light where you need to open - iso6400 while 1600 in center. Just bringing this to awareness, there is no PP-magic...

[Rondec]

Just a sidenote - fixing in post, yes, but nevertheless you just push the corners, for example with the Sigma 35mm f/1.4 and a vignetting of 2,05EV wide open you are photographing with iso400 in the corners while the center is at iso100, or, worse - because low light where you need to open - iso6400 while 1600 in center. Just bringing this to awareness, there is no PP-magic...

You are correct. However for landscape photography, shot at low iso on a tripod, it should be feasible to fix pretty easily.

[monochrome]

That's the first time I heard of Hoya firing the lens designers for not cloning Canikon lenses. Where's your source for this? I always thought they were layed off because Hoya was cutting costs and new lens development wasn't high on their agenda for selling the camera division

Either normhead and I read the same obscure internet rumor article in 2010 or that was what Hoya wanted - and the Pentax engineers pushed back. That was what I always heard.

Surely, when the market share drops to the level it did ANY lens engineer is excess compensation to be reduced. But a lens engineer who refuses to design what the owners ordered (design lenses to score high on DXO test charts) becomes excess compensation with a bad attitude to boot. Such engineers are now rumored to be contracted as consultants to Sigma and Tamron. Try as they might, Ricoh is said to be having trouble hiring them back. As in, "I wouldn't work for you if my only other alternative was in Fukushima."

And I believe we've heard James Malcolm say it took him longer to harmoniously integrate the Ricoh and Pentax engineering teams (due to issues of differing philosophy) than he originally imagined it would.

When (not if) Ricoh / Pentax release their FF camera and the obligatory introductory suite of 2 zooms and 3 primes to complement it, there will be an interesting period of evaluation and comment as we discover which design philosophy was the winner.

[Kunzite]

But... those engineers are making now lenses to score high on DXO test charts Why are they doing it for Sigma and Tamron, but refused to make them for Pentax? It doesn't make sense.
IMO, cost cutting is a more plausible scenario.
I would not let my imagination go wild from simple statements which most likely aren't meant to say much besides what the actual words are saying.

[normhead]

But... those engineers are making now lenses to score high on DXO test charts Why are they doing it for Sigma and Tamron, but refused to make them for Pentax? It doesn't make sense.
IMO, cost cutting is a more plausible scenario.
I would not let my imagination go wild from simple statements which most likely aren't meant to say much besides what the actual words are saying.

We are telling the un-official Pentax story, what the truth is, we'll probably never know so it's irrelevant. We have no idea what design principles those lens designers are doing now. Maybe those companies wanted those designers because of pixie dust concepts learned at Pentax. Mr. HiraKawa may have been hired to bring the old Pentax magic with him.

Jun Hirakawa technical report…

This paper introduces features the titled lenses on the following items: 1) Standard and middle range telephoto lenses for a 35 mm single-lens reflex cameras sold by Pentax, 2) design concept and performance; unique expression with high quality picture, 3) exterior; machined aluminum body to provide high grade sense, 4) focal length; 43 mm (diagonal length of 35 mm size film) and 77 mm (lucky number among 70-80 mm), respectively, 5) both AF and MF operation possible, 6) helicoid screw at focus part provides torque feeling, 7) realization of smooth AF operation by a little reduced gear ratio of drive unit and 8) lens aberration compensation; priority to practical photographs expression than numerical evaluation.

Not one of the numerical evaluators, be it photozone.de, DxO or Imaging Resources have a clue what he was talking about, yet they pass judgement on his lenses. You have to ask, if they have no idea what he was trying to do, how do they know if he was successful? They don't, they use their own criteria and apply them to all lenses, relevant or not.

Reading between the lines, is way better than ignoring all the lines and assuming the worst.

[Cannikin]

I have a Sigma 8-16, sharp edge to edge, distortion corrected etc. but my wife hates the picture taken with it. If she's not right dead centre, they actually distort how she looks. She looks much more natural with an uncorrected lens. Your eye is not used to seeing uncorrected images. They don't look natural. The human eye is one lens, and the brain is used to seeing a certain amount of field curvature and softness at the edges. Only in this day and age of technology that ignores human form would you even have to make this statement.

This distortion along the edges of the frame is an inherent property of rectilinear projection (gnomonic projection), where straight lines are projected as straight on the flat sensor plane. The greater the angle between the center and the object, the more sensor area it will take up, even for the same object at the same distance from the camera. The distortion increases the wider the angle, until at 180 degrees it becomes infinite (which is why it is impossible to have a 180 degree rectilinear lens). This is an unavoidable consequence of projecting a spherical field of view onto a flat plane (the sensor/film). It is present in all rectilinear lenses, even telephoto lenses. It is merely much more noticeable in wide angles, because there is a greater angular separation between the center and the edge.

This is similar to how a map with mercator projection, where all longitude and latitude lines are staight, will cause land near the poles to become stretched and take up more area relative to land near the equator. Greenland appears bigger than all of South America, despite the fact that the actual area of Greenland is a little over 1/4 that of Brazil. The actual poles are always cut off because they would appear infinitely large, and be impossible to render.

The human eye does not have this problem, because it renders a spherical field of view onto a (relatively) spherical surface (the retina). Thus an object on the edge of your vision would have the same size and shape as the same object in the center, because their projected images take up equal areas on the spherical retina.

[kenyee]

But... those engineers are making now lenses to score high on DXO test charts Why are they doing it for Sigma and Tamron, but refused to make them for Pentax? It doesn't make sense.

That's what puzzles me too...and do the lenses that were released under Hoya's reign have better DXO test charts than previous lenses? I didn't get the impression they were better than the 31Ltd which is (probably?) the sharpest FA lens.
Sigma has really been kicking out excellent primes lately (and I mean full frame sharpness)..

[vimbuza]

Then why don't they just design a spherical sensor?
Actually just make it round and spherical and let us decide how to crop it.

[Cannikin]

Then why don't they just design a spherical sensor?
Actually just make it round and spherical and let us decide how to crop it.

You would still end up with the same projection problem when it came to displaying the image. Most people have flat displays and display all images at more or less the same size, regardless of AOV.

Unless you have a spherical display, and are willing to display the image at sizes directly proportional to the original AOV (i.e. longer shots would be smaller, wide angle shots would be bigger), or a display that could dynamically alter its curvature for different AOV, it would still always appear distorted to some degree.

[Kunzite]

We are telling the un-official Pentax story, what the truth is, we'll probably never know so it's irrelevant. We have no idea what design principles those lens designers are doing now. Maybe those companies wanted those designers because of pixie dust concepts learned at Pentax. Mr. HiraKawa may have been hired to bring the old Pentax magic with him.

What un-official Pentax story? Any sources, something to point in that direction? Where did you get that Pentax engineers were fired because of their "design philosophy"? (as opposed to cost cutting, something we know Hoya did).
And how come the truth is irrelevant? I'm afraid we can't replace it with fiction.

[Class A]

Any sources, something to point in that direction?

<SARCASM>
Sure.

Monochrome said

"Either normhead and I read the same obscure internet rumor article in 2010 or that was what Hoya wanted - and the Pentax engineers pushed back."

There are apparently no other explanations for why the same story can be told by two different people.

The second alternative he mentioned refers to the truth and is thus irrelevant (as we are instructed by normhead).

So the first alternative -- there was an "obscure internet rumor article in 2010" -- must apply.

That's good enough as a source, isn't it.
</SARCASM>