[texandrews]

*unless you stack images. I'd really love for Pentax to make a UWA tilt-shift lens or better yet a camera that has that capability like the Hasselblad arcbody.

Yeah, would be great. Also, a 645 that featured focus stacking like the Focus Stack Tool in Phase One XF....

[lytrytyr]

I don't understand your point. I just walked down the Pentax products page, normal angle section and pulled these. Pentax gave the same charts.

"MTF calculated without referencing the effects of light diffraction phenomenon"

"Calculated", not "measured". Diffraction happens!

They had to know the MTF sweet spot to put it in the lens ROM for the camera's program line to work.

I would be interested to know what they use in their "program line" program. Anybody hacked it?

[lytrytyr]

However Zeiss doesn't do MTF testing with full spectrum light (for legal reasons) I can't go into detail... but I can tell you they use a pretty sharp filtration in their lens testing.

They are quite upfront:

"The MTF charts are valid for .. white light"

and

"Data ... [n]ot for IR version".

[lmd91343]

"MTF calculated without referencing the effects of light diffraction phenomenon"

"Calculated", not "measured". Diffraction happens!

At the wider f-stops diffraction is an ever decreasing component until it is almost neglible. The published charts show both wide and narrow f-stops. As the circle of the circumference of the lens opening grows, the area within that circle grows much faster. Diffraction only occurs at that edge of the that circle (aperture blades edge).

MTF is a ratio of contrasts and therefore must be calculated from two observations. It cannot be directly observed like line pairs per millimeter.

[lytrytyr]

At the wider f-stops diffraction is an ever decreasing component until it is almost neglible.

This thread is about an "ultra-wide lens". For such lenses, diffraction kicks in early when stopping down from fully open.

MTF is a ratio of contrasts

Nope. As the distance from the image center varies, the MTF charts plot the single contrast (as a percentage ratio between the brightnesses of adjacent black lines and white lines) for each given resolution (listed in terms of how many of those line pairs are packed in each millimeter of image). When lens faults smear the image, the black lines bleed in to the white lines, reducing the percentage of contrast between the two. There are two MTFs for each lp/mm, one for line pairs arranged radially out from the center (like a star chart), and the other for concentric circular line pairs.

The Imatest charts you see on sites like Optical Limits and Optyczne present a limited version of the MTF information differently, identifying at how many lp/mm the MTF contrast hits 50%.

[lmd91343]

This thread is about an "ultra-wide lens". For such lenses, diffraction kicks in early when stopping down from fully open.

Diffraction occurs when light passes over an edge. Such as an iris edge.
As an aperture widens, the length (circumfrence) grows increasing the diffracted light. However, the AREA of the undiffracted light increases by a squared value, increasing much faster.

It is present at all apertures, just very low at wide ones as a percentage.

Nope. As the distance from the image center varies, the MTF charts plot the single contrast (as a percentage ratio between the brightnesses of adjacent black lines and white lines) for each given resolution (listed in terms of how many of those line pairs are packed in each millimeter of image). When lens faults smear the image, the black lines bleed in to the white lines, reducing the percentage of contrast between the two. There are two MTFs for each lp/mm, one for line pairs arranged radially out from the center (like a star chart), and the other for concentric circular line pairs.

The Imatest charts you see on sites like Optical Limits and Optyczne present a limited version of the MTF information differently, identifying at how many lp/mm the MTF contrast hits 50%.

Good explanation of the overall process. The formula where you have the derived 50% is MTF(u) = C’(u) / C(u) where u is the frequency( line pairs per mm)

[RobA_Oz]

This thread is about an "ultra-wide lens". For such lenses, diffraction kicks in early when stopping down from fully open.

I assume that diffraction is the same at any point around the aperture edge, regardless of aperture size, simply because it is an edge phenomenon. Following that logic, it occupies an increasing proportion of the aperture area as the iris is stopped down, becoming more noticeable in the process until it affects the entire image. I’m struggling to see why a wider angle of view would change that. Perhaps I need to look at some ray-tracing.

[Digitalis]

Is the second one anywhere near Victor Harbour? I lived in Adelaide for 6 months from the the middle to the end of 2010 and took a couple of trips down to Victor Harbour...

That was taken on the coast of Sellicks hill which, if you take main south road, it does eventually lead to Victor Harbor.

[TwoUptons]

I assume that diffraction is the same at any point around the aperture edge, regardless of aperture size, simply because it is an edge phenomenon. Following that logic, it occupies an increasing proportion of the aperture area as the iris is stopped down, becoming more noticeable in the process until it affects the entire image. I’m struggling to see why a wider angle of view would change that. Perhaps I need to look at some ray-tracing.

My understanding is that diffraction impacts the image as a function of the aperture size, but we usually deal with aperture in terms of ratio...

So a 50 mm f2.0 lens with a 25mm aperture would have about as much impact from diffraction as a 100mm f4, since both would have the same 25mm physical aperture, right?

-Eric

[Mistral75]

My understanding is that diffraction impacts the image as a function of the aperture size, but we usually deal with aperture in terms of ratio...

So a 50 mm f2.0 lens with a 25mm aperture would have about as much impact from diffraction as a 100mm f4, since both would have the same 25mm physical aperture, right?

-Eric

They would not. They would have the same entrance pupil, i.e. the same optical image of the physical aperture stop, as 'seen' through the front (the object side) of the lens system, but not necessarily the same physical aperture stop.

[biz-engineer]

Still making new DSLR lenses? Interesting move. I can see how that manual focus DSLR lens can be used with pretty much any camera directly or with an adapter, covering a large part of the market.

[Mistral75]

Still making new DSLR lenses? Interesting move. (...)

Irix begun by making SLR lenses (15mm f/2.4 then 11mm f/4 then 150mm f/2.8 Macro), then they derived cine versions from them, then they made additional cine lenses (45mm T/1.5 then 30mm T/1.5 then 21mm T/1.5) and derived still versions from them.

[TwoUptons]

They would not. They would have the same entrance pupil, i.e. the same optical image of the physical aperture stop, as 'seen' through the front (the object side) of the lens system, but not necessarily the same physical aperture stop.

Right... I was over-simplified...
The same would be true for things like retrofocus wide angles, right? where the actual aperture is probably larger than would be indicated by the focal length and F-stop.

But the diffraction is still based on the actual size of the aperture, not the entrance pupil, right?
So if I put a 2x teleconverter on a lens, the percentage of the image impacted by diffraction is the same.
If I use a 300mm f4 lens that then is a 600mm f8, diffraction will have the same impact because the aperture is the same, even if the focal length and f-stop are not.

-Eric

[Wheatfield]

Cutting through the drivel, I just think it's nice that there is a third party that thinks enough of Pentax to make lenses for the mount.

[Serkevan]

Cutting through the drivel, I just think it's nice that there is a third party that thinks enough of Pentax to make lenses for the mount.

Well in this particular case I'd argue it's more making K-mounts for the lens
Jokes aside, I'm also happy that it's a thing.