[StephenHampshire]

Photozone is a tool that is meant to help you make a decision. Klaus is very honest about WHAT he tests, HOW he tests it. It's as good as it gets. The numbers serve as a basis for comparison, but of course they don't tell the whole story.

He does also describe each lens toroughly, almost always giving Pentax bonus points for manufacturing, build quality, etc.

Read a few reviews to know the basics of the lens you are considering (and read more than one review), then look at sample pictures to decide if the lens delivers the look you're after.

Excellent advice, I don't t think there is any conspiracy behind the mixed reveiws of Pentax lenses from photozone but my experience of lenses I have that they have tested is mixed. i.e
DA 55-300 - their review is much better than my experience with this lens, I think I have a not-so good copy ( eben though I bought it at a B and M store and tested it first!)
DA-70 f2.4 - my copy replicates the Photozone test very well
Sigma 30mm f1.4, again concurs pretty much with the review
DA 18-135- my copy seems a lot better than the one Klaus tested, or at least the good microcontrast hides the dodgy MTF performance or something!
Note pz always puts the caveat about tests from one camera not necessarily reading over to another, my lenses have been on a K20 and K5, and now a K30.
The reviews here are pretty comprehensive and, if you average them out probably give a good representaion of real life experience with the lens.
Out of interest, what do other system users think of the Photozine reviews of their lenses??

[gofour3]

Just remember there is the real world and then there is 'the Photozone.' LOL. I appreciate the work that goes into testing, etc. But, I know what I see when I use the 43 limited. There is no doubt that the 43 and the 77 were optimized for portrait work and corrections were left out of the design to optimize field performance at the sacrifice of bench testing. The photos speak for themselves.

Don’t forget the FA31, FA43 & FA77 were optimized for a FF film camera as a wide angle, normal & a portrait lens.

Any Photozone test of these lenses on a DSLR have nothing to do with the lens design, the test only shows how a film lens fares on a APS-C body.

Phil.

[bdery]

Unless I'm missing something, or Klaus specifically states so,
it's not always clear to me whether the border MTF readings
are taken at the same focus setting as the center or not.

It would make no sense at all if it wasn't so. I know Klaus uses central focus (or central focus confirmation with MF). And in any case, absolute numbers don't mean much, but in relative (compared to one another) the method is constant.

[lytrytyr]

It would make no sense at all if it wasn't so. I know Klaus uses central focus (or central focus confirmation with MF).

An issue certainly arises with lenses that have a strong field curvature.
Here is what Klaus says:

"If a lens suffers from field curvature and/or residual aberrations . . . this is taken into account - in this case the corners are measured independently from the center using different reference images."

Photozone Lens Test FAQ

And in any case, absolute numbers don't mean much, but in relative (compared to one another) the method is constant.

I'm not quite sure what you mean here.
The method is not constant
if some lenses are given border readings corresponding to relatively widely differing central focus points,
while others are not.

[bdery]

Here is what Klaus says:

"If a lens suffers from field curvature and/or residual aberrations . . . this is taken into account - in this case the corners are measured independently from the center using different reference images."

I am surprised by this. that,s not how I would do it.

I'm not quite sure what you mean here.
The method is not constant
if some lenses are given border readings corresponding to relatively widely differing central focus points,
while others are not.

Indeed in that case there is an additional bias.

What I meant is that, in theory, comparing lenses together has meaning, but saying "lens X has 2143 sharpness" is useless.

[cali92rs]

I don't understand the premise of the thread.
"Conspiracy" implies that photozone is deliberately skewing data to make Pentax look bad. What could possibly be the reason for them to do this? What would they have to gain?

A more likely explanation is that your ego has been hurt, for whatever reason, by some website saying that your lens isn't as good as someone else's. You take pictures, you see the results. Why does it matter what a website says?

[lytrytyr]

What I meant is that, in theory, comparing lenses together has meaning, but saying "lens X has 2143 sharpness" is useless.

Thanks for the clarification.

[lytrytyr]

"Conspiracy" implies that photozone is deliberately skewing data to make Pentax look bad. What could possibly be the reason for them to do this?

The C word may just be a headline to get your attention,
the real question being why the FA 43 on the K5
got fewer "stars" than a Canon kit lens,
or indeed the DA 18-55 WR on the K5.

Kind of like the parallel thread
https://www.pentaxforums.com/forums/pentax-slr-lens-discussion/206592-lens-ra...ml#post2183809
about the user reviews on this forum.

[Blue]

I don't understand the premise of the thread.
"Conspiracy" implies that photozone is deliberately skewing data to make Pentax look bad. What could possibly be the reason for them to do this? What would they have to gain?

A more likely explanation is that your ego has been hurt, for whatever reason, by some website saying that your lens isn't as good as someone else's. You take pictures, you see the results. Why does it matter what a website says?

While the Photozone crew aren't a conspiracy, there are plenty of Pentax bashers that like to use the mtf data from them to criticize Pentax lenses. The FA Ltd series is particularly hammered in numerous threads here at PF alone. It doesn't help that Photozone heaps praise on the DA 70 Ltd and is critical of the FA 77 Ltd. This is in part because of the torture test the FA 77 was subjected to but the 70 was not. In addition, the designer stated in a scientific publication that the design of the FA ltd lens wasn't done in a way to optimize mtf on a flat surface.

[Blue]

The C word may just be a headline to get your attention,
the real question being why the FA 43 on the K5
got fewer "stars" than a Canon kit lens,
or indeed the DA 18-55 WR on the K5.

Kind of like the parallel thread
https://www.pentaxforums.com/forums/pentax-slr-lens-discussion/206592-lens-ra...ml#post2183809
about the user reviews on this forum.

I can't imagine anyone would believe that the DA 18-55 WR would be as good at 43mm as the FA 43/1/9 optically. For one thing, the 43 has a wider aperture and Ghostless coating. In fact, it would have to be stopped down to even have a chance against the FA 43. Plus, Photozone cautions about extrapolating across systems and even cautions against comparing the K10d to the K-5 within Pentax lenses.

[lytrytyr]

I can't imagine anyone would believe that the DA 18-55 WR would be as good at 43mm as the FA 43/1/9 optically. For one thing, the 43 has a wider aperture and Ghostless coating. In fact, it would have to be stopped down to even have a chance against the FA 43. Plus, Photozone cautions about extrapolating across systems and even cautions against comparing the K10d to the K-5 within Pentax lenses.

Both the FA 43 test with 2.5 stars,
and the DA 18-55 test with 3 stars,
were done on the K5,
so there are no cross-platform issues there.

[lytrytyr]

In addition, the designer stated in a scientific publication that the design of the FA ltd lens wasn't done in a way to optimize mtf on a flat surface.

As bdery and I were discussing a little earlier,
Klaus' methodology may compensate for a curved image field.

[lytrytyr]

I can't imagine anyone would believe that the DA 18-55 WR would be as good at 43mm as the FA 43/1/9 optically.

Klaus' tests on the K5 do indicate that at f/8,
the kit zoom between 28 and 55mm
resolves the borders better than the FA43.

Lateral CAs also show up worse on the FA 43.

[Blue]

Klaus' tests on the K5 do indicate that at f/8,
the kit zoom between 28 and 55mm
resolves the borders better than the FA43.

Lateral CAs also show up worse on the FA 43.

The "star" system is relatively subjective. Plus, the 18-55 is a zoom and the rating is for its overall fl. He also states that its corner performance is its weakness and it needs stopped down to f8, especially at 18mm. He also factors price into the subjectivity rating and this has lead to bias in the evaluation of the FA LTD lenses, especially the 43 and 77. He favored the DA ltd because of their initial price advantage. However, the price advantage has evaporated with the MAP.

The rating of these 2 lenses giving the 18-55 WR the advantage should be a red flag.

[Blue]

As bdery and I were discussing a little earlier,
Klaus' methodology may compensate for a curved image field.

May is the operative word because he may not have compensated.