[normhead]

What's the highest LW/PH on APS-c you can find for any lens on Photozone, for a lens available on K-mount.. and what is the CA @ f4 for that lens. I'm curious, anyone feel like helping out?

D7000 (16 MP APS-c) SIgma 85 1.4- 2867 lw/ph @F4 CA .53
------------------------Tamron 60 macro 2821 lw/ph @F4 CA 1.13

---

Last edited by normhead; 02-12-2013 at 09:06 AM.

[robbiec]

Probably the Tamron 60 Macro on the D7000, 2800+, Sigma 85 also shows well in the centre or the Zeiss 85 on Sony Nex :-), 3000+, 0.7 @ f/4 CA

And you can still get hold of the Zeiss 85 f/1.4 new if you know where to look :-)


If I ventured to guess, a Zeiss Makro Planar T* 50 f/2 on a K-5 IIs would be the ultimate at f/4-f/5.6 @ around 2850ish with CAs at F/4 measured at 0.1

[normhead]

Darn, no numbers for the Sigma 70 or Tamron 90...

[JinDesu]

Zeiss Distagon T* 28 F2 on D7000 - 2635 LW/PH CA 0.93 @ F4. Max was at F5.6, 2782.

Amusingly: Tamron 17-50 on D7000 - 2825LW/PH CA 1.73 @ F4 and 17mm. It's sharpest at 17mm and F4.

Tokina AT-X 35 F2.8 Macro on D7000 - 2783 LW/PH CA 0.65 @ F4. Max was at F5.6, 2810.

Pentax DA 35 F2.4 on K-5 - 2723 LW/PH CA 0.59 @ F4.

That's all I could find. The rest of the lenses on the k-5 were below 2700 LW/PH on Photozone.

[normhead]

I wonder why the numbers on the D7000 on photozone top out at 2850, when the numbers for the K-5 top out at 2750, when they use the same sensor.

[robbiec]

Different filters in front of the actual sensor. The K-5 IIs should measure the same as if not better than the D7000.

[normhead]

Amusingly: Tamron 17-50 on D7000 - 2825LW/PH CA 1.73 @ F4 and 17mm. It's sharpest at 17mm and F4.

But 1.73 CA, that's brutal, that's why people use primes.

[baro-nite]

Darn, no numbers for the Sigma 70

He's tested the Sigma 70 on both Canon APS-C and Canon 135.

[normhead]

Yes the 70s numbers are excellent , as are the Tamron 90's tested on other systems, but the Canon EOS test was done at 15 MP.

I also left out the Samyang lenses, they seem to be top of the line whatever they are tested on... if you can do MF

[maxfield_photo]

Numbers are just numbers. For my money there are three types of lenses: Very sharp, sharp enough, and not sharp enough.

“There is nothing worse than a sharp image of a fuzzy concept.” -Ansel Adams

[normhead]

"Sharpness is a bourgeois concept" Henri Cartier-Bresson


While I appreciate that an oversize image my capitalize on sharpness, I've also seen some small scale Samyang images that are so sharp it hurts to look at them. I've also had criticism of "over-sharpenned" 60-250 images that hadn't had any sharpening applied. Maybe we should have a discussion of which lenses are too sharp as well.

But from what I can tell, the CA numbers are just as important as the LW/PH.

[normhead]

Numbers are just numbers. For my money there are three types of lenses: Very sharp, sharp enough, and not sharp enough.

“There is nothing worse than a sharp image of a fuzzy concept.” -Ansel Adams

"Sharp is a bourgeois concept" is right up there too...

It is interesting to use the numbers to predict what lenses you might like. Having pretty much completed my aquisition of lenses, I'm guessing it's about to become a lot less important.

But I have to say... when we went from the Sigma 70-300 to the Tamron 90, there was an incredible difference in micro contrast, colour, colour contrast and apparent depth to the images. The images looked 3D from the Tamron, and flat by comparison form the 70-300. There is a lot more than just being sharp, but there's no guarantee that if a lens isn't sharp it has them, and certainly no guarantee that if it is sharp, it doesn't have them.

[robbiec]

Anyone know if you can divide the max LW/PW for a lens by the max for the system to get a score?
So for example, the K-5 max = 2750, take the DA35 f/2.4 @ 2723 and you get 0.99 (i.e. touching sensor limits) @ F/4.

[normhead]

Anyone know if you can divide the max LW/PW for a lens by the max for the system to get a score?
So for example, the K-5 max = 2750, take the DA35 f/2.4 @ 2723 and you get 0.99 (i.e. touching sensor limits) @ F/4.

The theoretical max for a K-5, if you artificially created white and black lines all the way down the sensor would be 3200 lw/ph in that there would be 3200 distinct lines.