[normhead]

The Photozone rating

The Pentax SMC-DA 21mm f/3.2 AL Limited is a very good lens but it is not as special as its longer cousins (specifically the Pentax FA 31mm f/1.8 Limited) nor is this really possible because of the increasing design problems when pushing things towards wide and ultra-wide angles. The center performance is exceptionally high whereas the borders require f/5.6 to achieve very good levels. The amount of barrel distortions and vignetting are Ok but nothing to rave about for a new fix-focal lens. CAs are quite well controlled. The coolest aspects of the lens are naturally the extremely small size, the very low weight and a relatively moderate price tag of around 500€/US$. Combined with the "longer" pancakes lenses, the DA 40mm f/2.8 Limited and DA 70mm f/2.4 Limited, you'd have a complete yet very unobtrusive fix-focal lens setup for street photography.

And rated at 3 out of 5. I look at that and I think, this is someone who hasn't used this lens. I bought the lens largely on the advice of forum members, couple with the fact that I found one second hand at a very reasonable price a 2 hour drive away from me.

You look at the specs for this lens and they are nothing spectacular. Until you see this...



Or this



There is a dreamy quality about these images that no other lens I own achieves, although I've seen evidence that the 31 ltd is similar. You don't want that quality in every shot you take, but when conditions are appropriate, there is no other lens I own that is going to give me that. Photozone says "good for street photography" to which I respond "huh?". So what I'm saying here, is in my experience, what the people say about a lens that use a lens is more informative than what you read at the test sites. In most cases, they haven't used the lens enough to know what it's even good at. Photozone doesn't have a clue about what the best quality of this lens is, and they don't have a test for it. It's nice to have the numbers when arguing sharpness etc. but based on the Photozone ratings, I wouldn't have bought this lens. And I'm really happy I didn't look up the numbers until after I bought it. I would have missed out on an awesome lens, and some awesome pictures. Maybe not for everyone... but, I'm not everyone and neither is anyone else. Photozone simply cannot define in numbers what it is this lens has.

[liukaitc]

Here is a more laboratorial test of this lens by DxO...
DxOMark - SMC DA 21mm F3.2 ED AL Limited

[normhead]

Isn't that funny, nothing in either review anticipates why anyone would use this lens. I bet DxO would actually try and "correct" what I think is the strength of this lens. I'm beginning to think the numbers guys really don't understand art. Before working with this lens I thought this kind of analysis would tell me whether I would like the pictures from a given lens. Now I'm wondering if these numbers are even relevant. At all. Maybe this is like measuring the outside dimensions of a car, and looking at it's horsepower rating, to try and tell you if you'd enjoy driving it. With a car, you know you have to sit behind the wheel. Do people really think they can make decisions on lenses from the information of these testing sites?

How about a site where they take a landscape with each of a series of lenses and then use a focus group to rate the images. Why would that not be more valuable than these technical tests. After all, images aren't rated on how technically good they are in terms of sharpness. They should be rated on how good the images look to the human eye. There is a tacit assumption here that sharpness etc produces better images. My DA*60-250 produces sharp images, sometimes over sharp. There is an assumption here that if you buy the best lenses rated by these testers, one will produce more pleasing images. I just wonder how that plays out in real life. My guess is, it's not a one to one correlation, or even close.

[mattt]

Compare and contrast - the OP is talking about ART, the subsequent DxO assessment is engineering... the divide will never be crossed. That's the difference between pointing it at sunsets, and pointing it at grids of text and lines you find on test paper. Dreamy and sharp are on opposite ends of a spectrum I'd say.

For me - the function of a pancake is irrefutable - and 21mm has a great story telling FoV.



EDIT: Heh... you beat me to it.

[Bob from Aus]

That's why in the end I listen you you guys on the forum rather than the "official" reviews. the craziest review I read was a guy testing different slr digital cameras including a K7 in cold weather with gloves on. He marked the K7 camera down because the buttons were small and he confused the number of focus points with the number of metering points/areas. The K7 was the only camera he tested designed to work at minus 10 and was weather sealed. This guy couldn't even read the K7 specifications properly.

As they are much more familiar with other brands I thank many of the reviewers don't get it.

[riff]

I bet DxO would actually try and "correct" what I think is the strength of this lens. I'm beginning to think the numbers guys really don't understand art.

Other then fit and finish I'm not taking much from these type of len.s review.s anymore. The numbers well they are numbers but are they the relevant numbers? I think no, so I look at photos to make decisions.

I have a particular fondness for old (pre-computer) lenses. These lenses were optimized by sensitive eyes not computer algorithms and are probably impossible to make today. The results speak volumes to me, the numbers are perhaps not so good. Not to say that there aren't great new lenses as well.

[luftfluss]

Isn't that funny, nothing in either review anticipates why anyone would use this lens. I bet DxO would actually try and "correct" what I think is the strength of this lens. I'm beginning to think the numbers guys really don't understand art. Before working with this lens I thought this kind of analysis would tell me whether I would like the pictures from a given lens. Now I'm wondering if these numbers are even relevant. At all. Maybe this is like measuring the outside dimensions of a car, and looking at it's horsepower rating, to try and tell you if you'd enjoy driving it. With a car, you know you have to sit behind the wheel. Do people really think they can make decisions on lenses from the information of these testing sites?

How about a site where they take a landscape with each of a series of lenses and then use a focus group to rate the images. Why would that not be more valuable than these technical tests. After all, images aren't rated on how technically good they are in terms of sharpness. They should be rated on how good the images look to the human eye. There is a tacit assumption here that sharpness etc produces better images. My DA*60-250 produces sharp images, sometimes over sharp. There is an assumption here that if you buy the best lenses rated by these testers, one will produce more pleasing images. I just wonder how that plays out in real life. My guess is, it's not a one to one correlation, or even close.

Numbers - measurements - are of course relevant, they just might not be relevant to you. "Testing sites" should be just one more source of information to be used in deciding on a purchase.

Photozone says "good for street photography" to which I respond "huh?". So what I'm saying here, is in my experience, what the people say about a lens that use a lens is more informative than what you read at the test sites. In most cases, they haven't used the lens enough to know what it's even good at.

Many DA 21 users feel that the lens excels at street photography.

[normhead]

Many DA 21 users feel that the lens excels at street photography.

If I lived anywhere near a street, I might know about that.

These lenses were optimized by sensitive eyes not computer algorithms and are probably impossible to make today.

I really feel both the DA 18-135 and DA*60-250 are Hoya lenses and done to have good mtf numbers... not necessarily to take pleasing looking pictures. I and many others suspect Mr. Hirakawa and others were fired by Hoya because they refused to design to technical specs.

This is because currently the subject plane is the target of this numerical evaluation, and thought not to be a suitable evaluation of the depiction of solid objects. Certainly, we think evaluation of object depiction by the numerical value method should be established urgently, but for the time being, that comes after the design.

In other words, he felt the current system of using flat plane evaluation for lenses that were to be used for photographing 3d objects was inappropriate. Obviously Hoya did not agree with his opinion and decided to go with the flat plane model. Hirakawa obviously felt a model could be developed for lenses photographing 3 d objects, but noted that at that time one was not available. There's no indication that this has changed. The question here is if Pentax has purged all who thought like Mr. Hirakawa, and if there's anyone left to start making great lenses again. I'd characterize both my DA 18-135 and DA* 60-250 very good lenses, but for myself, they aren't what I expect from Pentax, whereas my 21 ltd is.

[luftfluss]

If I lived anywhere near a street, I might know about that.

Ha, lucky you!

I have no interest in street photography myself, but the DA 21 is on my short list, along with the DA 70 and Tammy 70-200/2.8. Personally, I don't care if those lenses are the sharpest in their respective classes or not, I know they are all "sharp enough" and I like the photos I've seen taken with them.

Your previous post reminded me... a few years ago I was researching television for a friend. At the time, there was the huge debate as to whether plasma or LCD was the superior technology. During my research, I came across the results of a survey about what people most noticed regarding picture quality. I don't recall all of the choices, but I do remember that the most important aspects of picture quality to most consumers was color and contrast. Those are more visceral qualities, and that's what more people respond to (outside of subject matter).

[Bob from Aus]

I use my 21 Ltd as my secret weapon. My clients love the photos.

[normhead]

Ya, I'm kind of rating photozone 3 out of 5, and rating my 21 ltd. 4.8 out of 5. I definitely know which one helps my photography more.

[monochrome]

In other words, he felt the current system of using flat plane evaluation for lenses that were to be used for photographing 3d objects was inappropriate. Obviously Hoya did not agree with his opinion and decided to go with the flat plane model. Hirakawa obviously felt a model could be developed for lenses photographing 3 d objects, but noted that at that time one was not available. There's no indication that this has changed. The question here is if Pentax has purged all who thought like Mr. Hirakawa, and if there's anyone left to start making great lenses again. I'd characterize both my DA 18-135 and DA* 60-250 very good lenses, but for myself, they aren't what I expect from Pentax, whereas my 21 ltd is.

Wasn't there a runor in January (or a statement in a Ricoh interview) that Ricoh is trying to rehire many of these engineers, including Mr. Hirakawa?

[interested_observer]

Here is a more laboratorial test of this lens by DxO...
DxOMark - SMC DA 21mm F3.2 ED AL Limited

If you scroll down in the DXO analysis they compare the 21Ltd to the CZ 21 which is rated 2 points higher (11 to 13). Then take a look at a few reviews of the CZ21 and its legendary, etc.(at $1,800)

• Zeiss Distagon T* 21mm f/2.8 ZF (ZE) (on Canon EOS) - Review / Lab Test

... you are comparing engineering specs against an excellent photographer is able to do in a wonderful location.

... I like the images VERY much!!!!

[luftfluss]

I really feel both the DA 18-135 and DA*60-250 are Hoya lenses and done to have good mtf numbers... not necessarily to take pleasing looking pictures. I and many others suspect Mr. Hirakawa and others were fired by Hoya because they refused to design to technical specs.

In other words, he felt the current system of using flat plane evaluation for lenses that were to be used for photographing 3d objects was inappropriate. Obviously Hoya did not agree with his opinion and decided to go with the flat plane model. Hirakawa obviously felt a model could be developed for lenses photographing 3 d objects, but noted that at that time one was not available. There's no indication that this has changed. The question here is if Pentax has purged all who thought like Mr. Hirakawa, and if there's anyone left to start making great lenses again. I'd characterize both my DA 18-135 and DA* 60-250 very good lenses, but for myself, they aren't what I expect from Pentax, whereas my 21 ltd is.

You don't think the 18-135 takes really nice pictures? From what I've seen here on PF, I think it far outperforms what one might expect from Photozone's rating. Images from it look very nice, and have a certain something to them that seems more refined (or something) than I would expect, just looking at specs. I find this with my F 35-70, too, which produces photos with that "traditional" Pentax look.

[luftfluss]

Wasn't there a runor in January (or a statement in a Ricoh interview) that Ricoh is trying to rehire many of these engineers, including Mr. Hirakawa?

Yup: https://www.pentaxforums.com/forums/pentax-news-rumors/173978-pentax-re-hirin...t-go-hoya.html