[normhead]

But this can't be the design philosophy if the "same" lens gets different results.

It is somewhat mind boggling...

After all, why would someone who posted these results on the DA*60-250 produce so much worse results on the DFA 70-210? They clearly have the talent to do better.


That's why I went immediately to same lens tested by both testers and sample variation. Some may criticize that but, bottom line, it's the only thing that makes sense right now.

I once asked Klaus over at Optical limits why he didn't test the lenses independent of the systems they go on for a test of lens against lens in exactly the same conditions same camera system,same sensor, eliminate all the variables but the lens used.
He said, the system to do that would cost $150,000, are you willing to contribute?
There in lies the problem, inquiring minds want to know, but not enough to give a competent researcher the money he needs to run the tests we want performed.

And really , apart from curiosity what difference dose it make? You can decide "I like this lens " without seeing any numbers. And you can buy a lens you don't ultimately don't like, even though the numbers said you'd like it. You can see from forum posts, selecting a lens you'll like is a very inexact science.

When I got my 21 ltd. and started using it I didn't really relate to it. Some one commented I was using an "inferior" lens. It handled high contrast poorly with CA and fringing. I was ready to give up on it, but one day on a whim, I tried it on a shaded midday dam and spillway. In a low contrast pastel environment, that lens is phenomenal creating a smoothness I can't match with any other lens. When I see criticism of that lens now, I think "the guy just never figured out how to use it."

---

Last edited by normhead; 03-29-2020 at 12:02 PM.

[BigMackCam]

What we need to do here is ship the same lens to someone with a Sony and Pentax and Nikon adapters, and test the different incarnations on the same camera, then the Pentax lens on the Sony Camera and k-1. We have way too many variables to know what's going on. Are the sensors used by Pentax more prone to reproducing CA, is it some component of the Pentax implementation of the design? Was the same lens used for both of the tests, and could that lens be defective? We don't know anything.

Absolutely, Norm. Camera differences need to be taken out of the equation by using a common platform (whatever that may be) to capture results from both the Pentax and Tamron versions of the lens. I'm not sure how practical that will be, since the Pentax lens incorporates KAF4 electronic aperture control, and there isn't an adapter with that capability. Perhaps it would be possible to set the aperture on a Pentax camera, remove the lens while the camera is still powered up, then transfer the lens to the adapter... I seem to recall that works with the HD DA55-300 PLM, but that may be a false memory.

There was a post some weeks back regarding a review by CameraVille guy (and so to be taken with a large pinch - perhaps a big handful - of salt) comparing the HD FA35/2 refresh vs the older SMC model. The reviewer noted a greater incidence of CA on the HD version of the lens. I was quick to argue against the suggestion that coatings were responsible for this. I'm still doubtful... yet I wonder whether there's the faintest possibility it might be a factor.

[RGlasel]

the recent Pentax D FA 70-210's resolution falls quite badly at the longer end, and the CA is significantly worse than the Tamron tested on the D810. Or the Tokina Opera, having much less CA than the original Pentax.

I wonder if the bodies used to test the same lens with different name plates aren't baking in different results.

The highlighted part would explain some of the discrepancies, the argument about "pure" RAW is a waste of time, no camera produces images without processing the data from the sensor. The thing is there are too many variables to draw any firm conclusions; different lenses on different bodies from different manufacturers tested at different times by people who don't have the capacity to isolate the effect of these variables. No review should be viewed as authoritative.

[Wheatfield]

The highlighted part would explain some of the discrepancies, the argument about "pure" RAW is a waste of time, no camera produces images without processing the data from the sensor. The thing is there are too many variables to draw any firm conclusions; different lenses on different bodies from different manufacturers tested at different times by people who don't have the capacity to isolate the effect of these variables. No review should be viewed as authoritative.

I wonder how many "tests" are jpeg using the manufacturers' default settings, or using a raw conversion using the manufacturer's default.
The ONLY way to accurately test lenses is to do all tests on the same camera body with identical conversion settings, something I highly doubt is happening.

[lesmore49]

I'm no optical engineer, just a photo enthusiast....so I can't and don't promote myself as an optical expert. I know my strengths and weaknesses and we all have them.

But getting to the meat of the matter. I do remember Popular Photography's tests of lenses on two lenses...the DF A 100 macro WR F2.8...which they concluded was 'superb' and the Pentax 12-24 was referred to by PP as the 'best of the bunch' for ASP-C wild angles in this class. The tests, etc. were over a decade ago.

I have both these lenses and bought them after I did my due diligence/ research, etc. (such as it is) and they have been two of my most favoured, go to lenses.

I think that sometimes we become too focused on 'number's provided by analysis. Case in point my 18-135 and my 40 Limited I bought after much thought. I had read opinions on both these lenses and some of the stuff wasn't very positive...some said the numbers weren't too good...soft in the corners...etc...so I hesitated. Eventually I said I'll buy them as I like those focal lengths.

I did and let me tell you...I don't know what all the crabbing was about. These are great lenses and I've blown up some pix to 11 x 14 with them and frankly and I do think I'm picky...I can't find any issues .

[zapp]

We are talking about 2 or 3 current lenses, tested on different cameras, looking at only two parameters. Sample size is one. Original Pentax designs make the difference on Pentax cameras with Pentax cameras.
The DFA*50 and the upcoming 85 are not available for D800 :-)

[normhead]

We are talking about 2 or 3 current lenses, tested on different cameras, looking at only two parameters. Sample size is one. Original Pentax designs make the difference on Pentax cameras with Pentax cameras.
The DFA*50 and the upcoming 85 are not available for D800 :-)

I believe the Tokina Opera is, which is a licensed DFA* 50 1.4.
If I'm wrong, I'm sure someone will correct me.

[Kunzite]

I believe the Tokina Opera is, which is a licensed DFA* 50 1.4.
If I'm wrong, I'm sure someone will correct me.

Of course; and I already mentioned it as testing "better" in terms of CA and maximum resolution.

[Madaboutpix]

While, admittedly, a certain curiosity remains on my side why Pentax lenses seemingly tend to fare less well than some of their bigger-brand cousins in laboratory tests, I no longer obsess about test results and rely more on actual high-res images taken with the lenses in question as they are accessible on, say, Flickr or SmugMug. "Our" own tests here on PentaxForums are also quite useful in deciding whether a lens is optically compelling or less so. At any rate, I'm more interested in real-world images than in brick walls, newspapers, and test charts. Even testing guru Roger Cicala over at LensRentals doesn't seem to be averse to using Pentax gear when he wants to have some innocent private photography fun.

In fact, that deeply-ingrained insistence of the Old School Pentax engineers on a "humanized", applied photography-based, and more holistic assessment of image quality strikes me as a bit of a brand differentiator and an enduring reason to keep me solidly in the camp. It is good to know that most of them are passionate photographers themselves and develop gear for photographers rather than measurebaters, and I'm proud to be using gear designed by people thus motivated.

[luftfluss]

Using DPR's "Studio scene" tool, it is pretty apparent that the K-1 is out-resolving the Nikon D800E. Of course, the lens used and focus accuracy are critical in making this comparison worthwhile.

[normhead]

Using DPR's "Studio scene" tool, it is pretty apparent that the K-1 is out-resolving the Nikon D800E. Of course, the lens used and focus accuracy are critical in making this comparison worthwhile.

Apparently they test every camera they can with a Sigma 70 macro, which they have in many different mounts. But they have on occasion had different focus points for different lenses.

[Digitalis]

particularly those where there is an other brand version - be it Tamron or Tokina - I've noticed how Pentax lenses tend to measure worse.

In which metric? As far as MTF is concerned Pentax lenses perform on par with most lenses. I think the big mistake people make with reviews is there are no error bars as they don't test multiple samples - personally, I do this because sample variation can in some cases make a substantial difference.

There was a paper published by hasselblad many years ago ( circa mid 1980s) stating that a mis-focus can knock off a good 20% of the MTF score of a lens - I take that as a good starting point when viewing graphs of an individual lenses performance, you can expect to see performance that is within 20% of the sample tested.

And what about the CA?

Pentax and their improved SMC coatings enhance contrast - this is a double edged sword as far as LoCa and first order planar aberrations as it tends to amplify them. It isn't anything inherent in the optical design - its just Pentax being...uh, Pentax.

Being measured wrong is just one possible cause; and it could be as simple as using a slightly different processing

Using cameras to measure lenses is the first thing popular lens reviews get wrong. Cameras introduce a lot of variables it can be a real headache to eliminate them.

Apparently they test every camera they can with a Sigma 70 macro, which they have in many different mounts. But they have on occasion had different focus points for different lenses.

It would be better if they used exactly the same lens. It is possible to get industrial lenses that can be adapted to pretty much any lens mount and be optically engineered to very exacting standards of resolving power, Alas, such a thing would cost a lot of money.

[biz-engineer]

Camera differences need to be taken out of the equation by using a common platform (whatever that may be) to capture results from both the Pentax and Tamron versions of the lens.

It would be mostly an academic exercise to know the lens only contribution to resolving power between brands, since what does actually matter to us photographers is the image we get out of the lens + camera of the same mount. If Pentax camera + lens deliver less detailed relative to a Nikon or Sony kit, that's what users should be concerned about. That concern, in addition to the accelerator chip in the K1II, would actually be good to motivate the design and commercialization of a new Pentax full frame model featuring a 45Mpixels or 61Mpixels sensor.

[BigMackCam]

It would be mostly an academic exercise to know the lens only contribution to resolving power between brands since what does actually matter to us photographers is the image we get out of the lens + camera of the same mount. If Pentax camera + lens deliver less detailed relative to a Nikon or Sony kit, that's what users should be concerned about. That concern, in addition to the accelerator chip in the K1II, would actually be good to motivate the design and commercialization of a new Pentax full frame model featuring a 45Mpixels or 61Mpixels sensor.

Yes, but most people have enquiring minds. We don't just want to know if one combo is better than another, but why - especially since the contributing factors (lens design, sensor resolution etc.) would appear to be so similar. Answering this might not have any practical benefit to us as photographers, but it appeals to our thirst for knowledge.

In fact, what actually matters is whether we can take the photos we want with the equipment we own or have available to us. That's what we should be concerned about - not differences between one brand's combo and another brand's equivalent which, for most practical applications, are irrelevant...

[Kunzite]

In which metric? As far as MTF is concerned Pentax lenses perform on par with most lenses. I think the big mistake people make with reviews is there are no error bars as they don't test multiple samples - personally, I do this because sample variation can in some cases make a substantial difference.

There was a paper published by hasselblad many years ago ( circa mid 1980s) stating that a mis-focus can knock off a good 20% of the MTF score of a lens - I take that as a good starting point when viewing graphs of an individual lenses performance, you can expect to see performance that is within 20% of the sample tested.



Pentax and their improved SMC coatings enhance contrast - this is a double edged sword as far as LoCa and first order planar aberrations as it tends to amplify them. It isn't anything inherent in the optical design - its just Pentax being...uh, Pentax.

MTF50 and no, Pentax doesn't reach the same numbers (I won't say performance) as the competition - at least not according to ephotozine and OpticalLimits (the latter admits to using different sharpening, but also warns against directly comparing systems).
IMO this isn't something we can attribute to sample variation; it's too uniform. Although in case of the 70-210, the resolution drop at the longer end could be sample variation, or a slight misfocus.

About CA (lateral CA) we're talking about a measured difference from under a quarter of a pixel to almost a pixel (for the Pentax, respectively Tokina 50mm).