[selar]

Why does the emoticon work in the reply you posted but not in my original post?

[Blue]

Why does the emoticon work in the reply you posted but not in my original post?

I am not sure.

[Class A]

Many prime lens designs, except retrofocus wide angles,
have their roots going back for a century
(more like glass plates than film).

Some traditional lens designs, such as the double Gauss design, are hard to improve upon as the basis for an optical design, unless you involve expensive exotic glass.

Nikon's latest 85mm lens just copies the optical design from the predecessor. It appears they could not improve it with reasonable means.

In Adam's comparison the FA 77/1.8 is sharper @ f/1.8 than the DA 70/2.4 is at f/2.4.
Disclaimer: I have not checked whether this could be due to a problem with Adam's test.

While the FA 77/1.8 has PF issues I wonder to what extend these could be addressed with a "for digital"-redesign without destroying the FA 77/1.8's rendering quality that many love.

I do not believe that the FA 77/1.8 design/technology is outdated. Maybe I'm wrong and tweaks could improve it without causing the "pixie dust" to trickle out, but I'm skeptical.

[selar]

[Asha]

The newer DA designs use more aspheric elements than the FA. Not saying FA has none (since I don't know, definitively), just that the compact DA designs necessarily include them.

The "old" designs are still highly relevant, IMO, and I do not see how MTF can be degraded from one camera to the next if all we're talking about is the lens itself. There is something about the MTF test that introduces sensor performance as a factor. In this case, the "rating" really should be a lens/camera pair rating. This is subtly stated on PZ.

[lytrytyr]

I do not see how MTF can be degraded from one camera to the next if all we're talking about is the lens itself. There is something about the MTF test that introduces sensor performance as a factor. In this case, the "rating" really should be a lens/camera pair rating. This is subtly stated on PZ.

Right. A moderately good lens could appear excellent on a low-density sensor,
but show up less well on a more dense sensor
whose closer pixel spacing challenges the lp/mm resolution of that lens.

[westmill]

An easy way to see it is high resolution cameras magnifies everything.
It shows and allows you to get the highest definition possable but then it
does exactly the same thing with any flaws of any given lens. You gotta take the rough with the smooth.

[Class A]

... a more dense sensor whose closer pixel spacing challenges the lp/mm resolution of that lens.

Ideally, lens tests specifying results with the unit "lp/mm" (or based on "lp/mm") should yield the same results independently of the camera that was used.

This unit should refer to the lens resolution only and should hence be independent of the pixel pitch of the sensor. Only where a low pixel-pitch sensor is outresolved by the lens, a higher pixel pitch sensor should yield higher resolution numbers for the lens, but everywhere else (corners, higher f-ratios) the two measurements should ideally agree.

[westmill]

What I seem to have noticed over the years, is that I remember most of photozones nikon test were carried out using the 10milion pixel D200. There were a couple of lenses that actualy outperformed the camera and went past the resolution limits. These same lenses when tested on the D300 where scratching the surface of the resolution limits.
This is the point I decided That DSLRs are now outperforming the lenses with ease.
Ive yet to see a lens get even close to the resolution limits of a 16milion pixel camera.
The more pixels though, does seem to have a strange efect I dont understand as yet.
Because they often do perform better slightly and show more resolution than on the lower pixel count. Its still not enough to reach the resolution limits though. Im 100% convinced you dont get more quality from more pixels. just bigger files. And that quality itself lies soley on the lens now.

[bossa]

When the airy disc of a lens has scale approaching the pixel size on the sensor there can be interference etc and 16million pixels on APS-C seems to be about the maximum regardless of how good the sensors are due to optical considerations.

Diffraction Limited Photography: Pixel Size, Aperture and Airy Disks

[lytrytyr]

Ideally, lens tests specifying results with the unit "lp/mm" should yield the same results independently of the camera that was used.

Imatest, as used by Photozone, is far from that ideal, although higher pixel-count sensors do bring it closer.

[HT Tan]

It sounds like a pretty good lens, but too expensive for me!

[lytrytyr]

When the airy disc of a lens has scale approaching the pixel size on the sensor there can be interference etc and 16million pixels on APS-C seems to be about the maximum regardless of how good the sensors are due to optical considerations.

As the link you posted shows, that depends on the aperture being used.

[Edgar_in_Indy]

Also consider that 31mm is 4mm shorter than the 35/2.4, without loosing MTF. The 31mm is more comparable to Sigma 30/1.4 and well it's not even a close game: Sigma AF 30mm f/1.4 EX DC (Pentax K) Review / Test Report - Analysis

Not to take anything away from the mighty 31mm, but it seems like you're comparing the Sigma 30mm tested on the K10D vs the Pentax 31mm tested on the K-5.

If you look at both the 30 and 31 tested on the K10D, you'll see that at even wide open at f1.4, the 30mm is already a tiny bit sharper in the center than the 31mm at f1.8. Very impressive sharpness for f1.4. The 30mm maintains its center advantage, and edges out the 31mm for maximum sharpness at f4.

Now the 31mm has a clear advantage at the borders, so it's all a matter of trade offs. I'm glad that both lenses are out there, since some people will value maximum center sharpness at wide apertures, while others will prefer more even sharpness across the frame.

[Blue]

The tone of this thread is odd. It got a review of 4.5/5 (optical quality), 4.5/5 (mechanical Quality), 3/5 (price/performance) and but many are acting like it got rated as a Coke bottle glass.