[UncleVanya]

Something to remember, lenses tested at higher resolution test as higher. Like dxomark, optical limits (formerly photozone.de) has a history of testing Pentax on lagging sensor technology.

All Tests / Reviews

Notice that the resolution of these lenses goes up between the 10mp and 16mp tests. 24 and 36 results don't exist. The tests are at best useful to understand relative sharpness within a lens family.

[MossyRocks]

right now i'm looking for a second len can make me "pleasure myself" the maximum quality out of 24mp for my "private times"

Going strictly by the numbers the HD Pentax-D FA* 50mm F1.4 SDM AW may be what you are after. That said you should really be asking if you need the maximum resolving power your sensor can give you? Others have stated that you likely do not and a lot of people just get wrapped around the axle on this. For a while I was like that too and chased after maximum resolution. What cured me was I read somewhere on here where someone said that you should go and get a large print (24"x36") of one of your images and really look at it. This also works for curing people who focus on noise too much as well. Try focusing on picture quality (the aesthetics and subjective qualities) instead of image quality (objective qualitities) provided that you have the focus correct and exposure correct for what you were going for.

I use a lot of old glass with my K-3, much of it not considered all that great resolution wise by modern standards. For example here are a couple of shots I took with my K-3 and my old beat up (actually well loved) S-M-C 28mm f/3.5 Takumar. Keep in mind that these are macro/pseudo macro shots with extension tubes or reverse mount and that lens so not even a real proper macro lens which would have probably produced better results.

A pre civil war watch
Lens: S-M-C 28mm F/3.5 Takumar on the #1 (9mm) extension tube
ISO 100
F/8
4s


A microchip from the early 70s still on wafer at a 2:1 macro
This is also a pretty heavy crop to show a single chip that is 4mm across on the long edge so is not down sampled much
Lens: S-M-C 28mm f/3.5 Takumar reverse mounted
ISO 100
F/8
30s

[normhead]

Something to remember, lenses tested at higher resolution test as higher. Like dxomark, optical limits (formerly photozone.de) has a history of testing Pentax on lagging sensor technology.

All Tests / Reviews

Notice that the resolution of these lenses goes up between the 10mp and 16mp tests. 24 and 36 results don't exist. The tests are at best useful to understand relative sharpness within a lens family.

I used to preach, well if a lens is excellent at 10MP it should be equally excellent at 16 MP. If you go through the charts however, you quickly realize changing sensors can significantly change the characteristics, either that or the tester used different copies of the same lens and got very different results. Since most testers only use one copy of each lens for each test, their results are at best anecdotal. There's just no real science to that. Scientific equipment can be used to produce non-scientific results, and Internet testers are notorious for that.

Any test that ignores sample variation, and cannot produce an accurate standard deviation is pretty much worthless. what one copy does, is irrelevant. What's needed is a test of at least 20 copies, measurements from the worst lens, measurements from the best lens to establish the range, a curve to show you what you are likely to get if you buy the lens with a median and a norm. That would be science. What gets posted on the internet is not really scientific.

Tests can give you ideas what to look at and help narrow things down a bit, but you still have to look at the images and see if they are for you. Despite the fact that I've some great images taken with point and shoots and kit lenses, these tests somehow always tell you to buy the most expensive, and that's just not necessary.

So my motto of the day is "Buy any lens more than $400 or $250 used, and buy the focal length that most appeals to you, with the rendering that most appeals to you." In this case the smooth background bokeh thing probably means 70 mm or more and ƒ2,8 or wider. 70mm 2.8 out of focus areas are much smoother than 35mm 2.8 images. 200mm 2.8 is even better. And hence the appeal of the DA* 50-135 2.8 as a portrait lens. It covers all the bases. Too bad it costs so darn much and isn't full frame. What a great FL range it would be for a full frame portrait shooter.

[UncleVanya]

I used to preach, well if a lens is excellent at 10MP it should be equally excellent at 16 MP. If you go through the charts however, you quickly realize changing sensors can significantly change the characteristics, either that or the tester used different copies of the same lens and got very different results. Since most testers only use one copy of each lens for each test, their results are at best anecdotal. There's just no real science to that. Scientific equipment can be used to produce non-scientific results, and Internet testers are notorious for that.

Any test that ignores sample variation, and cannot produce an accurate standard deviation is pretty much worthless. what one copy does, is irrelevant. What's needed is a test of at least 20 copies, measurements from the worst lens, measurements from the best lens to establish the range, a curve to show you what you are likely to get if you buy the lens with a median and a norm. That would be science. What gets posted on the internet is not really scientific.

Tests can give you ideas what to look at and help narrow things down a bit, but you still have to look at the images and see if they are for you. Despite the fact that I've some great images taken with post and shoots and kit lenses, these tests somehow always tell you to buy the most expensive, and that's just not necessary.

All true and I agree that more rigorous tests with multiple copies would be useful.

[northcoastgreg]

Measuring the sharpness of a lens generally involves testing how well the lens resolves fine lines on a piece of paper. If you plan on primarily photographing fine lines on flat surfaces, perhaps these tests have some level of relevance. But if you're planning on photographing objects in three dimensional space, well, that's a significantly different challenge for a lens, requiring separate design goals. Lenses that focus on sharpness and acutance of fine lines over everything else are often too sharp for their own good. They exhibit a clinical sharpness that looks unnatural. Our eyes don't render the world with hyper-acutance everywhere. Naturally rendered detail, among other things, includes an element of smoothness around edges against backgrounds, which gives those objects dimensionality. Well designed lenses are designed to take images of a three dimensional world, not lines on a two dimensional surface. And so it's typical lens resolution tests of limited relevance for evaluating lens performance, and should be approached with due caution.

[ZombieArmy]

Near 100% crop of the moon with the Takumar bayonet 135mm f2.5, and the 24 MP KP.



I think this is most certainly resolving higher than my 16 MP K-01, and this is ancient glass.

[Papa_Joe]

Since Pentax is producing digital cameras their "low" resolution is complained in nearly every test.

In most cases the honest explanation for this phenomenom is also given: Modern resolution tests measure the contrast or even only the net size of a jpeg picture. Since Pentax cameras do a very restrained processing, they produce relatively low contrast jpegs while on factory settings. This is actually an advantage if you want to process a jpeg, since there is still something to process. The more agressive processing of "resolution optimized" jpegs gives no headroom for any improvements. But because of this preserving processing the cameras produced by pentax have lower resultion results then those of other manufacturers in every test. Take it easy, it's not a bug, it's a feature - really!
The german digitalkamera.de recently tested the K-1 mark II. They found, that at 50% contrast the sensor resolution of the K-1 was in the range of a 16 to 20 MP camera. At 10% contrast the resolution of the K-1 was at the theoretical possible maximum value. They also stated, that in comparison to a Nikon D810 there was no visible difference when reading the small fonts in the test chart! In other words the resolution of the camera is as good as those of any other brand, the jpegs are just not designed to meet the test criteria. That Pentax still sticks to this principle after 15 years of "bad" test results has my deepest respect.

The standard conclusion is, that you have to use Pentax cameras in RAW format, as their jpegs are not optimal. No need for me, as I am quite happy with the jpegs of my K10D. But if you want higher resolutuion for your jpegs, just turn the settings for contrast and sharpness in the menu a bit to the right and you should have jpegs with "more" resolution, i.e. contrast.
Since Pentax cameras have such a "bad" resolution, the lenses tested with them, of corse do not get high marks. The testing is done with standard jpeg settings.

[normhead]

Since Pentax is producing digital cameras their "low" resolution is complained in nearly every test.

In most cases the honest explanation for this phenomenom is also given: Modern resolution tests measure the contrast or even only the net size of a jpeg picture. Since Pentax cameras do a very restrained processing, they produce relatively low contrast jpegs while on factory settings. This is actually an advantage if you want to process a jpeg, since there is still something to process. The more agressive processing of "resolution optimized" jpegs gives no headroom for any improvements. But because of this preserving processing the cameras produced by pentax have lower resultion results then those of other manufacturers in every test. Take it easy, it's not a bug, it's a feature - really!
The german digitalkamera.de recently tested the K-1 mark II. They found, that at 50% contrast the sensor resolution of the K-1 was in the range of a 16 to 20 MP camera. At 10% contrast the resolution of the K-1 was at the theoretical possible maximum value. They also stated, that in comparison to a Nikon D810 there was no visible difference when reading the small fonts in the test chart! In other words the resolution of the camera is as good as those of any other brand, the jpegs are just not designed to meet the test criteria. That Pentax still sticks to this principle after 15 years of "bad" test results has my deepest respect.

The standard conclusion is, that you have to use Pentax cameras in RAW format, as their jpegs are not optimal. No need for me, as I am quite happy with the jpegs of my K10D. But if you want higher resolutuion for your jpegs, just turn the settings for contrast and sharpness in the menu a bit to the right and you should have jpegs with "more" resolution, i.e. contrast.
Since Pentax cameras have such a "bad" resolution, the lenses tested with them, of corse do not get high marks. The testing is done with standard jpeg settings.

In my recent poll, I tried to match Pentax jpeg using raw and my PP software. It actually take s fair bit of work to match the jpeg, although the raw gave considerable more latitude for changing the colour balance to something more artistic. Something 90% of the poll respondents didn't even like. MY current conclusion, you have to be pretty good with your PP software to do better than the K-1 jpeg engine, and it's going to take more time than I usually spend on all but the best images. My guess is 90% of the forum would be happier saving time and shooting jpeg, as they don't believe in doing the things raw lets you do.

[Docrwm]

In my recent poll, I tried to match Pentax jpeg using raw and my PP software. It actually take s fair bit of work to match the jpeg, although the raw gave considerable more latitude for changing the colour balance to something more artistic. Something 90% of the poll respondents didn't even like. MY current conclusion, you have to be pretty good with your PP software to do better than the K-1 jpeg engine, and it's going to take more time than I usually spend on all but the best images. My guess is 90% of the forum would be happier saving time and shooting jpeg, as they don't believe in doing the things raw lets you do.

Bingo, we have a winner!

---------- Post added 01-03-19 at 01:21 PM ----------

Going strictly by the numbers the HD Pentax-D FA* 50mm F1.4 SDM AW may be what you are after. That said you should really be asking if you need the maximum resolving power your sensor can give you? Others have stated that you likely do not and a lot of people just get wrapped around the axle on this. For a while I was like that too and chased after maximum resolution. What cured me was I read somewhere on here where someone said that you should go and get a large print (24"x36") of one of your images and really look at it. This also works for curing people who focus on noise too much as well. Try focusing on picture quality (the aesthetics and subjective qualities) instead of image quality (objective qualitities) provided that you have the focus correct and exposure correct for what you were going for.

I use a lot of old glass with my K-3, much of it not considered all that great resolution wise by modern standards. For example here are a couple of shots I took with my K-3 and my old beat up (actually well loved) S-M-C 28mm f/3.5 Takumar. Keep in mind that these are macro/pseudo macro shots with extension tubes or reverse mount and that lens so not even a real proper macro lens which would have probably produced better results.

A pre civil war watch
Lens: S-M-C 28mm F/3.5 Takumar on the #1 (9mm) extension tube
ISO 100
F/8
4s


A microchip from the early 70s still on wafer at a 2:1 macro
This is also a pretty heavy crop to show a single chip that is 4mm across on the long edge so is not down sampled much
Lens: S-M-C 28mm f/3.5 Takumar reverse mounted
ISO 100
F/8
30s

As you said, tubes and reverse mounts are great, inexpensive, ways to do macro but try the M100/4 Macro or the F50/2.8 Macro and you'll start to appreciate the strengths of lenses that were designed from the ground up as macro lenses. They do resolve better but they also typically have far less distortion of the image in any dimension.

[MossyRocks]

but try the M100/4 Macro or the F50/2.8 Macro and you'll start to appreciate the strengths of lenses that were designed from the ground up as macro lenses.

Some day I will get one of them they are on my list of things I want but have a hard time justifying actually getting given how little macro I do. I have been contemplating getting the FA 50/2.8 macro as the camera store near me has one for a close to stole it price. The only reason I haven't is that I just bought a FA 77 and don't want to get in trouble with my wife like one of my friends did with his.

I mostly included those 2 shots to show that even using an old lens in a way that is only going to magnify lack of sharpness still produces what most people would consider sharp good images, especially the 2:1 macro shot with the lens set at f/8 giving an effective f-stop of f/24 so diffraction is rearing it's ugly head taking away from the sharpness since I am also magnifying the diameter of those airy discs by a factor of 3.

[Docrwm]

Some day I will get one of them they are on my list of things I want but have a hard time justifying actually getting given how little macro I do. I have been contemplating getting the FA 50/2.8 macro as the camera store near me has one for a close to stole it price. The only reason I haven't is that I just bought a FA 77 and don't want to get in trouble with my wife like one of my friends did with his.

I mostly included those 2 shots to show that even using an old lens in a way that is only going to magnify lack of sharpness still produces what most people would consider sharp good images, especially the 2:1 macro shot with the lens set at f/8 giving an effective f-stop of f/24 so diffraction is rearing it's ugly head taking away from the sharpness since I am also magnifying the diameter of those airy discs by a factor of 3.

Those shots are very nice and well illustrated your point. My point was that different purposes necessitate different choices in lens design. Specialty lenses are great at their specialty (hopefully) and less suited to uses dissimilar to their intended purpose. All-around lenses like the 18-135 are fair-to-good at a lot of things but not excellent at anything specifically. Comparing an M100/4 Macro to the 18-135 is simply as ludicrous as comparing a Prius to a Porsche.

[mikesbike]

I'm using wonderful K70 and 18-135, the versatility of 18-135 is the reason i'm happy walking around and enjoying any moment
right now i'm looking for a second len can make me "pleasure myself" the maximum quality out of 24mp for my "private times"

i'm checking Dxomark, and i but really feel kinda...off....none limited lens i see capable resolve 24mp or at least 15mp! the only best is Sigma one can resolve 16mp , the DA* can't hold up as well

what you guy think? should i bother too much?

Don't let it bother you at all. This is a machine test, subject to many variables, and from all I have seen very often does not compute in the real world. As far as a 100mm macro lens is concerned, it is a specialty lens good for extreme closeups, and other 100mm uses, but your 18-135mm lens can do many things and do them well that this lens cannot do.

Go by visual comparisons- they are most reliable. Your DA 18-135mm is capable of producing very sharp images. Learn at what settings it is at its best, which are many. The only reason it is "rated" in the mid-level of Pentax lenses is there are numerous factors commonly used to rate lenses. One is available aperture- so a zoom lens having a constant aperture capability throughout its zoom range, like f/4, means if you are shooting in Manual mode and are zooming to different framing, you won't have to constantly reset your aperture if you are shooting at wide open aperture- practical advantage for these conditions. And it can stay at f/4 at focal lengths where other lenses would go no wider than f/5.6 or so. Therefore it gets a higher rating. The DA 18-135mm lens will have a variable maximum aperture when zoomed. If the constant aperture lens is also a fast zoom lens, like f/2.8, it can provide a shallower depth of field (DOF) at the same focal length (FL) meaning the background will be blurrier to make a subject stand out more. It can also provide faster shutter speeds at the same light level than a lens having a less wide aperture (higher number). So it gets a higher rating yet. There are a lot of other aspects. This does not mean your DA 18-135mm is not capable of producing excellent images- it certainly can and does. Many are posted in threads in these forums.

I have some of the highest-rated lenses, but I still value having my wonderful DA 18-135mm!

---------- Post added 01-03-19 at 01:45 PM ----------

In my recent poll, I tried to match Pentax jpeg using raw and my PP software. It actually take s fair bit of work to match the jpeg, although the raw gave considerable more latitude for changing the colour balance to something more artistic. Something 90% of the poll respondents didn't even like. MY current conclusion, you have to be pretty good with your PP software to do better than the K-1 jpeg engine, and it's going to take more time than I usually spend on all but the best images. My guess is 90% of the forum would be happier saving time and shooting jpeg, as they don't believe in doing the things raw lets you do.

I also agree! I haven't seen a true statement of this kind in a long time. I have found it is often troublesome to bring meaningful improvement over images from a really good camera jpeg processor. That of the KP is awesome! It is a great timesaver. Shooting both gives one the option in situations where it "might" present an advantage.

As to a second lens, in your case it would best be a prime lens (no zoom) of a certain focal length. These can provide the absolute top level quality, but to do that they have to be only one focal length. So, the first thing to determine is- what will be the focal length? Do you need more wide angle? If so, forget what I said about having to be a prime lens- although the very small DA 15mm f/4 Limited is delightful in its top-quality build (another rating factor), fine imaging, and its extreme compactness. However, the excellent extra-wide angle DA 12-24mm f/4 will give prime lenses real competition for quality images- very sharp and low distortion (another rating factor). Excellent edge-to-edge performance. It would also provide a lot of additional range to your present lens.

If you need more aperture for low light situations, so you can get adequate shutter speed without having to increase ISO as much, you need a "fast" prime. The DA 50mm f/1.8 is a good one and has a short telephoto effect- good for portraits. For more versatility, the FA 35mm f/2 is excellent for edge to edge sharpness, and its good build.

[normhead]

I have some of the highest-rated lenses, but I still value having my wonderful DA 18-135mm!

The design doesn't make sense to the test chart analysts, but when you get a great result something, you get something better than what you'd get with a "superior" test chart lens. It's not all about reproducing black and white lines on flat surfaces. There are no flat surfaces in nature, there are no straight lines in nature. Sharp edge to edge on flat surfaces is almost never going to give you the best result out in the real world, unless you are in love with images of brick walls.

[StiffLegged]

Are you making photos or shooting test results? If you're already happy with the images you're getting, then you're getting results, rather than hitting technowonk scores. If there's another lens you fancy for your camera, you could do worse than read the lens reviews at this site.

[tvdtvdtvd]

what you guy think? should i bother too much?

My images look good, (or bad), regardless of DXO scores. I am always surprised this company manages to carry so
much weight and respect. Then again, if you only take pictures of test charts then I suppose DXO may have some merit....