[beholder3]

Honestly, all "lens test" by soccer mom Lara or Photozone or DxO are not worth anything more than your average one liner Amazon user review.

The only lens testing which is out there and has some merit is what lensrental does.

1)
Especially lensrental results do show that copy-to-copy variation is huge for all lenses and makers (especially bad for Sigma and such 3rd party producers). So any test done with only one or three copies is just a subjective user review of his own lens, not the model.

2)
Cheap hobby testing done with Imatest or similar tools is crap. Only optical benches do test the lens as such and at infinity.
a) the distortion tests in imagest with extra close test chart are mostly irrelevant as you tend to use non macro lenses at infinity. And due to focus systems the distortion at close focusing and infinity can change massively.
b) Photozone and similar testers test JPG. So a lot of JPG settings influence results.
c) such software does depend on sensor resolution and you can NOT just multiply results for current sensors
d) AA filter or not is not taken into account
e) poor versions (and photozone is one sample) do even refocus corner shots, so hiding any field curvature
f) as also proven by lensrentals: the sensor stack can lead to different results for the same lens even if sensor size and MPx are the same.

[normhead]

OK, you might have gone a bit too far comparing testers to soccer moms. But at least we can say for whatever lens Klaus owned or borrowed at Photozone he actually tried to come up with something and had some kind of testing regimen. He does come up with one real comparison even if it's only good for one lens.

I'm not going to argue every point. but I'm definitely not in agreement with your opinion of Klaus' test of field curvature. He doesn't refocus for field curvature measurements.

He did refocus the DA 18-135 for the edges, because the edges and centre are in sharpest focus at different settings in the long end. But it's not a common practice. The DA 18-135 is definitely a "not for the test charts " lens, and as far as I can tell Klaus completely didn't get it.

[JamesDCrockett]

No kidding, lw/ph should be a neutral cross platform measurement. The differences from site to site and the evolution of testing methods over the years, with changes in the software used for evaluation etc. make it all worthless.

For example Imaging resources changed their lighting on test subjects a few years ago. Results before the change are not directly comparable to results after the lighting change. There just is no reliable test site. Part of that is sample deviation, part of that is different people interpreting the test standards different ways.

All you can do is look for clues. The above posted material is speculation based on perhaps faulty data. I personally think Klaus dropped his 31 ltd. after he tested it at 10 MP before he test it on the K-5. But Can't prove that.

Yet, I still prefer speculation based on data that might be faulty compared to speculation based on no data at all. "I like this lens better" could be based on so many different things, it's kind of a last resort if I can't find anything else.

Based on the clues I'm expecting a big price, a huge resolution increase, and a more modern interpretation of whatever formula Pentax uses to compromise between absolute sharpness and other rendering properties.

Hopefully ephotozine or another site will have up a review/test by the end of April. I will wait until then. I do have high expectations for this lens.

[robjmitchell]

Given that this is the website that originally trashed my DA 18-135, and then watched every other website including this one join in the charge, I don't take them very seriously.

The problem being there are no other websites I take more seriously.

Which is why I said . This is a case of too little information. But I have to say, the fact that camera manufacturers release lw/ph numbers based on the lens designs, not actual performance and that there is no standard test lab where these kinds of things can be confirmed. I'd argue that at the moment it's the best we have.

That being said, I've always been suspicious of the 16 MP 31ltd. test. I just can't prove it's wrong. Sometimes you go with what you know. What put me on to this was Kenspo's statement about needing modern glass for modern sensors. As a good spokesman for the brand, I've been trying to unpack that. Pentax isn't going to say the limiteds suck, and they don't. Their rendering is fantastic.But if you need really high res, maybe they aren't the correct choice.

I'm just trying to make sense of the old "Lenses for the way people take pictures, not for the test charts" as opposed to the new "modern glass for modern sensors" The only way that makes sense to me is if they are going for both rendering and test charts, as if they might not be mutually exclusive as the old motto would suggest.

I don't know what the DFA 28-105 would test as on the charts, but compare to my 7 or 8 other non DA+ zooms, it's the sharpest, it has 3d rendering, and overall, it renders better than even my beloved 18-135. Personally I think Pentax has turned a page, and it will be good for those who can afford it. The big question now is , will the clearly heavy investment in development of probably pretty slow selling lenses, kill the company, before they see the fruits of their labours?

I wish the first wasn't a 50. And I wish they could show me, it's similar to the Nikon 24, in terms of giving APS-c better resolution than I can currently get on my K-1. And also confirm what the resolution will be like on the K-1"

Well between Ricoh announcing a f4 zoom (with presumably more to come) and licenseing designs like the 50 to Tokina (and potentially getting new designs in return) I would say their future looks a lot more rosy than a few weeks ago!
A agree that pentax will be aiming for both sharpness and rendering, with their awesome coatings still allowing for 3D pop even in complex designs. The DFA*70-200 showed it can be done.

[kjames5]

I did assign the ProPhoto RGB to the Jpegs, and so not sure why they rendered so poorly. Thanks for the note though!

Kerrick

A bit OT here, but I suppose you are working in ProPhoto RGB colour space. The images you posted are untagged so they are rendered in sRGB. If you assign a ProPhoto RGB profile they look much nicer and less dark.

[clackers]

I did assign the ProPhoto RGB to the Jpegs, and so not sure why they rendered so poorly. Thanks for the note though!

Kerrick

Guys, it should be sRGB on your calibrated monitor or all bets are off as far as websites and browsers go.

You can *try* embedding ICC profiles and all that sort of thing, but really, you have to play it safe and accurate, especially in a mobile age where a tablet or phone might be very primitive in the way it handles a colour space.

[kjames5]

Guys, it should be sRGB on your calibrated monitor or all bets are off as far as websites and browsers go.

You can *try* embedding ICC profiles and all that sort of thing, but really, you have to play it safe and accurate, especially in a mobile age where a tablet or phone might be very primitive in the way it handles a colour space.

I will write this once and once only, to avoid confusion. I upload and post hundreds of images a year on Facebook, Instagram, etc., using my ProPhoto RGB Tiffs as a source to create the uploaded jpegs, and have never experienced this degree of dark Jpeg imagery. I create images for publication in print and on many websites, so what happened with these two K-1 MKII images is not something I quite grasp, but if my image posts create nasty comments I have far better endeavours for my time. Write me directly if you want to see cleaner color and proper
image density, but I won't post my impressions beyond now. That's all folks. Kerrick James

[beachgardener]

it should be sRGB on your calibrated monitor or all bets are off as far as websites and browsers go.

absolutely sRGB, yes. Do these other sites do any conversions to srgb when uploading jpeg photos in the prophotorgb colour space, it has been said here that on this site uploaded images are not influenced by a website upload editing process, maybe the others are.
Just a thought, hope you work it out.

[rawr]

I will write this once and once only, to avoid confusion. I upload and post hundreds of images a year on Facebook, Instagram, etc., using my ProPhoto RGB Tiffs as a source to create the uploaded jpegs, and have never experienced this degree of dark Jpeg imagery.

It's still conventional (and wise) to make sure all your JPEG exports destined for the web are set as sRGB. It's not a personal swipe at you by any means.

Just the best means possible to ensure some display consistency across all the myriad platforms likely to be displaying your images, from tablets to phones to office PC's, and across the many types of web browsers out there. Most which are not colour managed, and so will default to displaying sRGB, which may have unpredictable consequences for JPEG images created in other color spaces.

[davidsladek]

It's still conventional (and wise) to make sure all your JPEG exports destined for the web are set as sRGB. It's not a personal swipe at you by any means.

Just the best means possible to ensure some display consistency across all the myriad platforms likely to be displaying your images, from tablets to phones to office PC's, and across the many types of web browsers out there. Most which are not colour managed, and so will default to displaying sRGB, which may have unpredictable consequences for JPEG images created in other color spaces.

And this is consistent with what my university teacher (at Institute of Creative Photography) and a master printer keeps telling us. Plus one of his favourite phrases: laptop monitor is not a monitor but a laptop cover

@kjames5

Apparently, as per this article main browsers are now able to display Pro Photo RGB correctly (I'd question that, even different variations of monitors/browsers play role and vast majority of people have uncalibrated monitors, or they check pictures on mobile devices where nothing matters), but if, for some reason Facebook upload had stripped the profile from the pictures than you stand no change and we all here understand it and do not blame you for it.

Looking forward to seeing more of your pictures

Keep up the good work

[RonHendriks1966]

According to the nation’s revered cherry blossom prophets (aka the sakura forecasters at Weather Map sakura.weathermap.jp), this year’s flowers are due to arrive around March 19 in Fukuoka; March 22 in Tokyo and Osaka; April 8 in northern Sendai. They’re also expected to bloom in Hiroshima from March 20 and in Kyoto from March 22, according to the Japanese Meteorological Corporation.

So I guess the lens is for next years cherry blossom season.

[BigMackCam]

So I guess the lens is for next years cherry blossom season.

I think that's highly unlikely, Ron, given that the Tokina version - using Pentax's design - is due for release this Summer. I could be wrong, but I just don't see Ricoh waiting another nine months after that...

[Rondec]

I will write this once and once only, to avoid confusion. I upload and post hundreds of images a year on Facebook, Instagram, etc., using my ProPhoto RGB Tiffs as a source to create the uploaded jpegs, and have never experienced this degree of dark Jpeg imagery. I create images for publication in print and on many websites, so what happened with these two K-1 MKII images is not something I quite grasp, but if my image posts create nasty comments I have far better endeavours for my time. Write me directly if you want to see cleaner color and proper
image density, but I won't post my impressions beyond now. That's all folks. Kerrick James

I don't think anyone is trying to upset you and I for one appreciate the photos and your sharing any impressions you might have -- both of the DFA *50 and the K1 II.

[Kunzite]

I don't think anyone is trying to upset you and I for one appreciate the photos and your sharing any impressions you might have -- both of the DFA *50 and the K1 II.

Indeed.
I appreciate Kerrick James' contribution here. It doesn't look to me the comments were in any way intended as critical to him - if anything, they were critical about the mess of various software and devices all behaving in different ways, that we have to deal with.
This being said, I'd really love to see a (physical) gallery with his prints! Too bad I live on another continent.

[beholder3]

He doesn't refocus for field curvature measurements.

I said they tend to refocus for normal "resolution" tests which then look much better than real results, which do suffer from field curvature.
Refocussing between shots is lying to the audience.