[mikeodial]

I have read a several comments that the K-3 will out resolve certain lenses. Does anyone have an experience with their K-3 as a result of upgrading.


I am not sure what the technical characteristics that cause this are, but wondering if I can expect some problems from my current non DA lenses listed below as I make a move to the K-3.


100mm f4 M series Macro, 100mm f4 A Series, 200 mm K series f4, 35-105mm A series, Rokinon 14mm (Not worried about this one).


Are they lenses to avoid that someone has experience of in the forum for this issue.


Or is this even an issue?


Thanks,

[robthebloke]

Non-issue. Those lenses will be fine.

[dadipentak]

Or is this even an issue?

I can't imagine losing any sleep over it.

[jatrax]

I have read a several comments

Don't believe everything you read. Even if true it only means that the k-3 is getting everything possible from a particular lens, not that the image will be any worse.

wondering if I can expect some problems from my current non DA lenses listed below as I make a move to the K-3.

Out resolve means that the limiting factor is the glass not the sensor. So the image will be as good as it gets from that particular lens.
If you have a lousy sensor and great glass then the limiting factor is the sensor, in this case you have a great sensor and good glass, you should be just fine.

Or is this even an issue?

Only if you would rather argue about minutia on an internet forum instead of taking pictures

[stevebrot]

Or is this even an issue?

No, it is not really an issue. I have a pretty broad range of lenses on my shelf and I can think of only one circumstance from that group of a lens performing adequately to the limits of the sensor on my K10D, but not on the K-3*. Even in that case, the image quality when down-sampled to 10 Megapickles was equivalent to the same lens on the K10D.

What has truly surprised me is that many lenses that should have been flunkies on the K-3 were actually more than up to the challenge of the high resolution sensor and are better in some ways than more highly-regarded options. A good example would be my Helios-44M 58/2. On the K-3 its center resolution is a little better than the highly regarded Pentax-K 55/1.8. Who would have thought, eh?

As Jatrax notes above, if you were happy with a lens on your K-5IIs or K-01, you can expect similar or better results with the same lens on the K-3.


Steve

* That would be my Tamron 70-150/3.5 at 150mm.

[lytrytyr]

I have read a several comments that the K-3 will out resolve certain lenses. Does anyone have an experience with their K-3 as a result of upgrading.

Don't worry about the K-3.

The really demanding Pentax cameras are the Q and Q10,
at least when it comes to central performance.

[Imageman]

In film days the best monochrome film would out resolve just about any lens made, and still can out resolve almost any lens made, most monochrome film in general use however was only slightly better than the lenses in general use. Only colour film was out-resolved by the lens.


It was never a problem for the film to out-resolve the lens the only thing that matters is the lens and you want the sensor to out-resolve the lens, you get all the resolving power of the lens in the image. Only in digital has this been raised as an issue and is total nonsense.


Anyone who says its an issue needs his head feeling.

[ak_kiwi]

What has truly surprised me is that many lenses that should have been flunkies on the K-3 were actually more than up to the challenge of the high resolution sensor and are better in some ways than more highly-regarded options. A good example would be my Helios-44M 58/2. On the K-3 its center resolution is a little better than the highly regarded Pentax-K 55/1.8. Who would have thought, eh?

I found the exact same thing yesterday with the very same lense. I was very surprised when I looked at the Iso (6400) - this K-3 certainly keeps surprising me
(Raw sooc converted with default light-room sharpening)

[dadipentak]

Only if you would rather argue about minutia on an internet forum instead of taking pictures

Ha, that's not even a complete sentence--what kind of idiot are you?

:~)

[jatrax]

Ha, that's not even a complete sentence--what kind of idiot are you? :~)

Of course it is a complete sentence. And I also used the word minutia which counts double. And that's just a straw man argument, it does not matter whether the sentence is complete. The minutia is the big thing here. You probably don't even have a dictionary.

[stevebrot]

It was never a problem for the film to out-resolve the lens the only thing that matters is the lens and you want the sensor to out-resolve the lens, you get all the resolving power of the lens in the image. Only in digital has this been raised as an issue and is total nonsense.

Ha! Ha! You went directly to the core of the matter. A high resolution sensor allows your lenses to perform to their full capabilities.

Historically, the challenge of "miniature" (35mm) film photography was to design lenses that were capable of supporting the small format. Most lens designs in the early 1920s were not up to the challenge. The films were not that great either, but the better lenses were intended for much larger formats . Once optical design and manufacturing responded to the need, film makers responded with more uniform and better performing product. I think it is safe to say that only in very special circumstances (e.g. document reproduction, astrophotography, electron microscopy, large prints, etc.) was the absolute resolving power of available films a major consideration or a limiting factor in practice*. If you wanted a larger reproduction, you simply used a larger format


Steve

* The actual resolving power of modern non-specialty films remains adequate (60-150 lp/mm) for most purposes. The best modern lenses resolve at the upper end of that range.

[monochrome]

Of course it is a complete sentence. And I also used the word minutia which counts double. And that's just a straw man argument, it does not matter whether the sentence is complete. The minutia is the big thing here. You probably don't even have a dictionary.

Unless you are caviling about a single detail the correct use is minutiae.

But that is a minutia.

[RobA_Oz]

Kodachrome's resolution was said to be around 20MP equivalent. Obviously, film grains aren't as neatly arrayed as a sensor's elements, but the K-3's sensor clearly exceeds that figure, and the Nikon D810 does so by a wide margin. Nonetheless, the K-3's sensel density is actually higher because its 24MP are packed into an APS-C format, and the "20MP" Kodachrome was available in a 35mm film format, so the K-3 may well be out-resolving some lenses. Any lens on the resolution limit with Kodachrome would be out-resolved by the K-3's sensor.

Does it matter? That's a question each of us has to answer in terms of our individual requirements.

[stevebrot]

Kodachrome's resolution was said to be around 20MP equivalent. Obviously, film grains aren't as neatly arrayed as a sensor's elements, but the K-3's sensor clearly exceeds that figure, and the Nikon D810 does so by a wide margin.

Strangely, that is not the case regarding the digital sensors. Translation of MTF data to lp/mm yields values of about 170 lp/mm for today's best lenses on a 36Mpx FF sensor. Current TMax 100 film attain base resolution of about 150 lp/mm and Mahn in Germany claims 300 lp/mm for their ATP 1.1 product. The late and much lamented Kodak Technical Pan was capable of 200 lp/mm. I threw those numbers in not for comparison, but to indicate that film resolution is of the same order of magnitude as current sensor technology.

As for the 20 Mpx estimate for scanned Kodachrome, that is an Internet legend based on the pixel count for a 24x36 slide scanned at 4000 dpi. A 4000 dpi scan of drug store film yields the same 20 Mpx. Optical resolution is measured in lp/mm not megapixels. The resolution of Kodachrome was tested by Modern Photography in 1978. Kodachrome II's best result was measured at 86 lp/mm when paired with the Leitz Summicron 50/2. The last product insert for Kodachrome 64 includes a MTF graph of base resolution (no lens) indicating similar performance for that product as for KII. Apparently the resolution of Kodachrome was the limiting factor the tested lenses in 1978. Contrary to conventional wisdom, Kodachrome was never the paragon of film resolution and no lens was designed around it as a performance target. Failure to perform on Kodachrome was inexcusable in a professional grade lens and in many ways it was the lowest indicator of acceptable performance, but most monochrome films did and still do present a stiffer optical challenge.

I have oversimplified the film resolution story as well as the challenge of extracting maximum resolution from a film + lens combination, but the truth is that most film era lenses perform quite well on modern digital sensors and also on modern high resolution films. The "designed for film and therefore lacking" argument is patently invalid. So is the talk about sensel density as a point of comparison. Yes, there is a tendency to oversimplify the digital resolution story as well.

Back to a lens being out-resolved by a sensor. There are good lenses and there are crummy lenses. Having a competent sensor or an incredibly good technical film does not make a good lens worse. Neither does a crummy sensor or nasty film even the playing field between lenses. As noted above, a great sensor allows most decent lenses to attain their best performance and nothing more. A decent analogy would be a race car. The optics are analogous to the engine, drive chain, and suspension. The sensor/film is analogous to the tires.** Crummy tires and you can't go fast regardless of the excellence of car and driver. Excellent tires and your speed is determined by engineering and driver skill.


Steve

(...still shoots film and has scanned a ton of Kodachrome as well as the better B&W films currently available...currently shoots the same lenses on both film and the K-3...gets similar results from both, at least for monochrome...color films are not so good... )

* Traditional film resolution testing is done using a microscope to evaluate the target image on the slide/negative. There are only the lines from the target, no pixels or image processor. More recently it is done using digital processors to generate MTF graphs similar to those used in lens tests.

** Yes, despite the current techie obsession over sensors they truly are the minor part of the puzzle...critical , but minor. The biggest challenge in digital photography is to provide sensors that are truly up to the capability of available lenses.

[RobA_Oz]

A helpful dissertation on the resolving power of film, thanks Steve. I wonder why we still miss Kodachrome? Personally, I preferred Agfachrome for skin tones...