[gazonk]

DA70 patent's crop factor: 1.43x


The bad news is: I'll have to recreate most of its content as the free image hoster with unlimited archival time deleted all of my content ...

Yes, I noticed most of your images were gone...

The good news is: Meanwhile I learned how to run numeric resolution tests.

MTF50 DA70, f/5.6
center: 3200 LW/PH
short FF edge 1950 LW/PH
extreme FF edge 1100 LW/PH
extreme FF corner 680 LW/PH

I wonder if it still could be quite useful for shots were border sharpness is irrelevant, e.g. portraits?

[falconeye]

I wonder if it still could be quite useful for shots were border sharpness is irrelevant, e.g. portraits?

I do think so. It carries a "+" in my list

Portraits may not be the best example though as you will probably want to use it wide open. On the other hand, at low ISO the corner vignetting should be correctable. It's probably an excellent FF cityscape and people shots lens.

The lack of sharpness in the corners may be less that it appears if it is due to lateral CA (what I assume) and one applies a true lateral CA correction. A true lateral CA correction is one where the red and blue images are rescaled with respect to the green one. Not one where seam colors are desaturated. Lens correction modules do this.

[24X36NOW]

I wonder if it still could be quite useful for shots were border sharpness is irrelevant, e.g. portraits?

I do think so. It carries a "+" in my list

Portraits may not be the best example though as you will probably want to use it wide open. On the other hand, at low ISO the corner vignetting should be correctable. It's probably an excellent FF cityscape and people shots lens.

The lack of sharpness in the corners may be less that it appears if it is due to lateral CA (what I assume) and one applies a true lateral CA correction. A true lateral CA correction is one where the red and blue images are rescaled with respect to the green one. Not one where seam colors are desaturated. Lens correction modules do this.

Technical issues aside, I for one believe that the "problems" of "corner sharpness" and "edge sharpness" and "vignetting" are in large part only really "issues" for "lens tests" and those trying to convince you of the supposed "need" for new lenses. If you choose a random sample of 100 actual photographs, I'd be shocked if you could find five that were spoiled as images (i.e., not because someone pixel-peeped and "showed" you that some "imperfection" exists in a much-more-than-likely completely unimportant portion of the image) because of any of those overblown "issues."

[thibs]

Technical issues aside, I for one believe that the "problems" of "corner sharpness" and "edge sharpness" and "vignetting" are in large part only really "issues" for "lens tests" and those trying to convince you of the supposed "need" for new lenses. If you choose a random sample of 100 actual photographs, I'd be shocked if you could find five that were spoiled as images (i.e., not because someone pixel-peeped and "showed" you that some "imperfection" exists in a much-more-than-likely completely unimportant portion of the image) because of any of those overblown "issues."

Agreed. Some lenses really push those issues too far IMO and then it is visible. But in the vast majority of cases, I agree with you.

[RMabo]

The thing is that customers reads lens tests. They buy lenses after the performance in tests. Most of them do. Even if difference between lenses is small, the reader gets the impression that the difference is big and noticeable.

The fact about sample variation and that the lens the customer is buying can have a slightly different performance, is strangely seldom communicated.

It is important for lens makers to get good results in lens tests. Sigma is an expert of this. They knows about it. Good results in tests and you get many awards and you sell a lot. Fascinating stuff.

Then the actual samples coming out from the factory can be different... but the average user doesn't know about it. Thinking he has such a good lens... because he is not spotting the errors.

[thibs]

Beleive me bunch of customers from my retailers learn pretty quickly what Sigma means wrt compatibility/quality control.

Thank you for this very interesting Patent.
But why do you say "for APS-C"? The only mention about APS-C is about the equivalent range starting at 45mm.
I guess it could also be the "new" 18-250mm equivalent for a FF camera!

I will agree with you on it could be a FF, and the more logical replacement mentioned else where in the Forums was talked a 18-270 as a replacement for APS-C Sencors

---

Last edited by music_lover; 07-03-2010 at 02:12 PM.