[hcc]

The FA31mm f1.8 Ltd is a legendary Pentax lens, one among the three FA Ltd. It was highly praised by PF in-depth review:
Pentax-FA 31mm Limited F1.8 Review - Introduction - PentaxForums.com
with the highest score (9.3/10) of any lens, And it is very well liked by PF users:
SMC Pentax-FA 31mm F1.8 AL Limited Reviews - FA Prime Lenses - Pentax Lens Reviews & Lens Database

Welcome to Photozone! reviewed first the lens with a Pentax K10d:
Pentax SMC-FA 31mm f/1.8 AL Limited - Review / Test Report
and it was Highly Recommended then.

Today Photozone.de released its 2nd review (or re-review) of the FA31mm based upon a K-5 camera body:
Pentax SMC FA 31mm f/1.8 AL Limited - Retest

Again the lens is Highly Recommended. The verdict:
"The Pentax SMC-FA 31mm f/1.8 AL Limited is an excellent lens. It is already sharp at f/1.8 and razor sharp from f/2.8 to f/11. Distortions and CAs are very well controlled. On the downside it vignettes quite a bit at f/1.8 so the ultra-large max. aperture takes its toll here. The bokeh (out-of-focus blur) is very smooth and buttery which is fairly unusual for a lens with aspherical elements. Typical for all Limited lenses the build quality is exceptional and it is a joy to use the lens. It also shares another, rather unfortunate, characteristic with its cousins - a high price tag. Quality comes at a price but 1200€/1000US$ seems a bit over the edge but then it is probably just another example of diminishing returns upon higher investments. Nonetheless it is refreshing to see that Pentax focuses on quality rather than costs - there're enough plastic lenses out there after all."

[chesebert]

The new review reduced the quality rating by half a star. Also the MTF figures are measured based on different scales and the graphs look inconsistent between the 2 reviews. I believe the MTF done by www.ephotozine.com is more representative (could be wrong).

[stevebrot]

The new review reduced the quality rating by half a star. Also the MTF figures are measured based on different scales and the graphs look inconsistent between the 2 reviews. I believe the MTF done by Digital camera reviews, photography techniques, photography gallery and photography forums is more representative (could be wrong).

Different sensors == different MTF

Revised rating is for K-5...


Steve

[chesebert]

Different sensors == different MTF

Revised rating is for K-5...


Steve

But the shape of the MTF are not the same. I am guessing the reviewer used AF?

[VisualDarkness]

Well, isn't it true that different sensor can give totally different readings with the same lens? Just look at NEX 5n vs (the brutal borders of) NEX 7.

[vrrattko]

Anyway....rather dissapointing in terms of resolution - the cheap DA35/2.4 beats FA31 in resolution figures.....now i never tried DA35/2.4, but i find it hard to believe, because FA31 i had was very sharp.

[VisualDarkness]

Anyway....rather dissapointing in terms of resolution - the cheap DA35/2.4 beats FA31 in resolution figures.....now i never tried DA35/2.4, but i find it hard to believe, because FA31 i had was very sharp.

Well, it's good right from the start but maybe sacrifices some top peaks instead. They are praising the Bokeh as unusually good and that's an accomplishment.

A lens that puts everything in the biggest apertures and sacrifices the rest: Leica D Summilux 25mm f/1.4 - Review / Test Report - Analysis

[chesebert]

Anyway....rather dissapointing in terms of resolution - the cheap DA35/2.4 beats FA31 in resolution figures.....now i never tried DA35/2.4, but i find it hard to believe, because FA31 i had was very sharp.

I still think the they messed up the test. If you just compare the MTF figure for FA31 against DA 35 2.8, the FA31 is sharper from f2.8. Actually, FA31 is about as sharp as it gets across all Pentax lens tested on the K10.

It makes no sense that putting FA31 on K5 somehow "degrades" FA31 compared to everything else.

[Trapezoid645]

Opinions, opinions, the latest one (in our PF dbase) from big A island puts the FA*31mm equal to a beer bottle - rated 3/10. My 5 years old black MIJ for $725 works beautifully for me. Don't see how any rating can change that....

[Rondec]

The difference in the tests seems to be that Photozone has added an "extreme" border test. Not certain how that differs from the border test, but it seems like the FA 31 is a little weaker way out there. Advantage for the DA limiteds has always been more edge to edge consistency in sharpness.

For most applications, the FA 31 is going to be excellent.

The other thing to note is that the FA 31 really does peak by f2.8, although it keeps good sharpness up to the f8 tested. Bokeh is definitely a notch above other lenses, even if sharpness is fairly equivalent.

[jsherman999]

It makes no sense that putting FA31 on K5 somehow "degrades" FA31 compared to everything else.

I wonder if it's the exact same copy of the lens that was used in the K10D test?

[lytrytyr]

The difference in the tests seems to be that Photozone has added an "extreme" border test. Not certain how that differences from the border test.

See the fourth FAQ on Photozone Lens Test FAQ

[Rondec]

Sure. It just seems like that probably isn't that big a deal. At least on APS-C. Maybe means that this lens is a bigger question on a full frame sensor?

[JinDesu]

Sure. It just seems like that probably isn't that big a deal. At least on APS-C. Maybe means that this lens is a bigger question on a full frame sensor?

Well, they tested it on the k-5 (or at least, on APS-C).

[VisualDarkness]

The (most often) higher pixel density of APS-C can lower the resolution score compared to FF and considering the pretty good border score here the 31mm could get really nice result if tested on a bigger sensor.