[slip]

Lens Sharpness

a lot of this makes total sense, please let me know what you think as every point of view is valuable to me

very interesting...read about Nikon using Phase one for their D3 promotions

thanks

randy

[Boker]

Most agree with the story but some points either I don't fully understand or disagree with like this

Depth of Field

There is no such thing as depth of field Depth of field works presuming you'll permit a certain level of softness. If you're seeking the absolute maximum sharpness, there is almost no depth of field See also Selecting the Optimum Aperture, since you can't stop down recklessly due to the limits of diffraction.

[Just1MoreDave]

I agree with a lot of the points. After doing some of my own lens testing, I realized how easy it is to miss the perfect focus just enough to make a big sharpness difference. Or at least something you can see in a 100% crop. There is no way that I am that careful using a lens for a real photo. That especially applies to the sharpness difference between something like my Pentax-A 50mm f1.4 and Pentax-F 50mm f1.7. I might be able to demonstrate it in a test but in real photos, it's all about where I focused.

Still, I wouldn't ignore sharpness as one part of performance. It helps to know where your lenses are not that great. It's useful as a sorting tool, for those of us with high LBA levels. I think Rockwell overstates the quality of modern lenses if you include the consumer zoom category. Those are compromise designs, and the designers may have compromised sharpness for something else. Also Rockwell gets to handle all the lenses, so he can talk about handling differences, while today the typical person gets a lens on Amazon or whatever. Should the typical consumer order each 17-70 just to see how they feel, or maybe look at sharpness to get an idea of overall quality?

Recently I started to like a few unsharp lenses. Here is a long post about one of them: Sears 28mm f2.8

[lightbulb]

I was complaining with a friend about shooting in high humidity and motion blur. These are my two biggest problems.

[mikeSF]

ugh, i clicked the link...it is the photog equivalent of getting "Rick rolled"...Rockwelled...ugh!
close window, close window.

[timo]

Everybody hates Rockwell, but there is a core of common sense in quite a lot that he says. On lens sharpness, he is right up to a point, but to say you shouldn't worry about sharpness is a bit silly - if something is not sharp it might be helpful to find out why. For example, if (using K5 fine adjustment) I adjust my DA21 to focus properly at medium to short distances, it front focuses at infinity. I was wondering why landscapes weren't sharp with this lens - and I found out why. Now I override the AF, quickshifting back to infinity for distant shots, with much better results.

I agree that sharpness or the lack of it is overrated as an inherent lens characteristic, but it's worth a bit of effort to investigate it sometimes.
(A decentred lens is much more annoying than one that's uniformly soft. Which is why I've gone off the economy zooms.)

[nater]

I think he has some good points, but I also think he stretches his argument a bit farther than he should. I also couldn't help but think of Lomography, where sharpness matters very much (you don't want too much of it, typically):

?What the hell is Lomography?? - Lomography

I did appreciate Ken Rockwell pointing out that Nikon used a Phase One P45 medium format digital back to produce the 'sample' photos for the D3.

[tromboads]

Long story short, anything that gets photographers away from faping on about kit and actually outside taking photos has to be a good thing.

Sharpness and issues around have to be the biggest fap of all. Long story short, with the quality materials we are dealing with today. the 0.3% difference isnt gong to magically make anyone's photos any more interesting.

There, I said it. Unless you are photographing bomb sites at 10,000 feet, sharpness be dammed. I'm much more interested in actual composition then any level of "sharpness".

[bdery]

Most agree with the story but some points either I don't fully understand or disagree with like this

Depth of Field

The explanation is that sharpness decreases when you move away from the razor-thin sharpest point. In absolute terms, there is no such thing as depth of field, there is only the focus point.

BUT... the camera has a resolution limit, so over a certain range, everything will look uniformly sharp. Plus, your eyes have their own limit. this means you can calculate a depth of field over which everything will look acceptably sharp, and that's what we call DOF. But it's not a lens thing, per se, it's a sensor thing.

Another example of Rockwell using provocative sentences to build credibility (he must know stuff, since he's implying that what we think we know is wrong).

[Boker]

Pretty sure my lens is limited by depth of field as with most any optic,sensor has nothing to do with depth of field no more than film does

[Lowell Goudge]

Most agree with the story but some points either I don't fully understand or disagree with like this

Depth of Field

There is no such thing as depth of field Depth of field works presuming you'll permit a certain level of softness. If you're seeking the absolute maximum sharpness, there is almost no depth of field See also Selecting the Optimum Aperture, since you can't stop down recklessly due to the limits of diffraction.

All he is saying is that there is only one absolute point of focus. depth of field is a concept based upon acceptable sharpness, usually defined by being able to display a point as something less than 1/100 of an inch at the final print magnification.

all that depth of field really means is the range of focus distance that a point will appear as something smaller than that 1/100th of an inch circle

[Lowell Goudge]

Everybody hates Rockwell, but there is a core of common sense in quite a lot that he says. On lens sharpness, he is right up to a point, but to say you shouldn't worry about sharpness is a bit silly - if something is not sharp it might be helpful to find out why. For example, if (using K5 fine adjustment) I adjust my DA21 to focus properly at medium to short distances, it front focuses at infinity. I was wondering why landscapes weren't sharp with this lens - and I found out why. Now I override the AF, quickshifting back to infinity for distant shots, with much better results.

I agree that sharpness or the lack of it is overrated as an inherent lens characteristic, but it's worth a bit of effort to investigate it sometimes.
(A decentred lens is much more annoying than one that's uniformly soft. Which is why I've gone off the economy zooms.)

but you come back to kens original point

all the charts and tests are done at a distance that you don;t normally use the lens at. If you are going to test or make adjustments (for lets say focus) then make them at the distance you use them at.

why would you calibrate a landscape lens at 2-3 feet when you spend all day at infinity? that is not correct use.

[northcoastgreg]

Although Rockwell's general point seems fairly compelling, as usual he over-states his case and then defends it with poor arguments. Perhaps the biggest flaw in his argument is that he confuses visual sharpness (i.e., how sharp images from a lens appear to the eye) with measured resolution. It would be more accurate to contend that measured resolution is over-rated. While there exists a correlation between measured resolution and visual resolution, that correlation is not perfect. Other factors contribute to visual sharpness, such as contrast, acutance, and even color rendition (i.e., a lens with particularly bright, vivid color rendition, like the Zeiss lenses, can give an illusion of sharpness).

One problem with Rockwell is that he has a private agenda in reserve. He is notoriously fond of superzooms and kit lenses. Now some of these lenses can have problems with visual sharpness that can be detected even at screen resolution. The lenses with the worst resolution are the superzooms at the long end of the zoom. And kit lenses can have problems with severe field curvature and reduced edge to edge sharpness at the wide end of the zoom and at infinity focus; plus they are prone to decentering issues which can reduce the overall crispness of the image, even when viewed at 8 by 12 enlargements. Not all photographic lenses, even when used "properly," are in fact "sharp," although, outside of the kit lenses, the superzooms, and the consumer tele-zooms, most have sufficient resolution for 95% of photographers. And even kit glass, superzooms, and tele-zooms are sharp enough in much of their focal range (assuming they don't have decentering problems), and may be entirely adequate for certain types of photography.

The implicit conclusion of Rockwell's article is that there is no significance difference between cheap glass and expensive glass, After all, nearly all glass is plenty sharp, and soft images are caused by "improper" use. But this is simply not true. There is more to a lens than just visual (or measured) sharpness. The bokeh, the quality of the rendering, and particularly the color rendition all have a significant impact on the final image. A kit lens simply cannot compete with a fast fifty in portrait work. The differences are noticeable even in web resolution jpegs. I briefly owned a cheap Tamron 75-300 zoom. At 300mm, that lens not only suffered from less than stellar resolution, but it produced images with low contrast and mediocre color rendition. All the post processing tricks in the world could not make the output of that lens come even close to what I can get out of the DA* 300. The differences are significant, and noticeable, even by non-photographers, in 5 by 7 prints.

[Lowell Goudge]

i have no intention in taking either side in this debate other than observing that like all generalizations, there will always be flaws in it.

Greg puts up some of these flaws, in absolute terms, but also in his rebuttal actually reinforces some of the underlying point, that almost every lens out there is sufficiently sharp, by stating that all kit lenses and superzooms have sufficient resolution for 95% of photography.

If this is the case, does that mean that all the fancy primes expensive tele's etc... are filling the 5% . Wow, what a marketing prize. sell a camera with adequate kit lens, or super zoom, and then sell another 10 lenses to the user he does not need and don't provide any additional benefit.

Actually there is not much difference between Greg and Ken Rockwell, only where the dividing line in the last 5% actually sits.

[Boker]

Think somewhere he stated you needed to spend at least $500 or so to get a sharp lens.I have 50mm 1.7 Sears lens $20 that is probably as sharp as some zooms costing over the $500 and there are a lot of lens out there in this category