[GeneV]

Look at the far upper-right corner, where that little house is. (Well, it looks little in the picture.)

What's the funky stuff happening along the top edges of the roof? Is that background clouds or hills, or is something weird going on? It looks pixelated.

If you look closely at anything this size, it will look pixelated, because you get to the pixels pretty quickly. There may be some reduction artifacts there, but really it is just hard to tell what is going on in a reduced format like this.

I do notice that on one of the roofs, it appears there is a less sharp area just inside the edge, but that it is sharper right at the edge. Vapor from the heat on the roof or some other artifact? Steven just needs to upload a larger version.

[Marc Sabatella]

Look at the far upper-right corner, where that little house is. (Well, it looks little in the picture.)

What's the funky stuff happening along the top edges of the roof? Is that background clouds or hills, or is something weird going on? It looks pixelated.

Posterization and/or some similar JPEG compression artifact.

The posted image at the posted size is definitely not a good example to determine anything about sharpness from. For one thing, it's far too small. For another, there is just no way the corners are in focus if the middle building is. I don't care *what* some DOF calculator says about hyperfocal distance or whatever - DOF formulas are based based on what would make for an acceptable sharp print viewed from a typical distance for that size. They do *not* guarantee pixel-level sharpness when viewing at 100%. So even if we could see 100% crops of the corners, it would be entirely unreasonable to expect them to be sharp unless you manually focused one of those corners yourself using live view at 100% magnification.

BTW, the forum doesn't let you attach larger images for good reason - threads would load too slowly. Instead, you should make a crop of a small section and post that without resizing (ie, a "100% crop"). But again, not really relevant here - there's no way those corners are in focus.

[kuau]

Is there a way to upload a .pef of the image?
making a jpeg and getting the size under 700K is a bit challenging for me.

Steven

[Marc Sabatella]

No, there's no way to upload a PEF file and have anyone else see it. Like I said, don't try posting the whole photo - just crop out an area you want to show. If you crop out and area of, say, 600x600 pixels, the file size should end up *way* under 700K. But again, you'd need a different example if you want to claim soft corners - you'd need an example where the corners were actually supposed to be in focus.

[GeneV]

For another, there is just no way the corners are in focus if the middle building is. I don't care *what* some DOF calculator says about hyperfocal distance or whatever - DOF formulas are based based on what would make for an acceptable sharp print viewed from a typical distance for that size.

When I use the DOF marks, I get sharper corners than when I let the camera focus. Whether those corners survive a pixel peep depends upon what one expects.

However, I don't understand how that applies to the buildings in the photo Steven has posted. I don't know where his camera focus was, but I'm pretty sure that those houses are all at a distance that is way past infinity for a 15mm lens. In theory, why shouldn't a building that is at infinity both in the middle and at the sides be in the same focus across its width?

[Marc Sabatella]

I'm not sure what you mean by "way past infinity" in this context. I'd just observe buildings in the corner are much further away than the building in the middle. If the lens was focused on the building in the middle, the buildings in the distance might officially fall with the DOF as calculated by standard DOF equations that assume standard print sizes videwed from typical distances, but that does *not* mean the distnt buildings would stand up to a 100% view as well as if one had actually focused on them. I guess you're implying that even the near building is already at infinity, so there would be no difference between focusing on it versus focusing in the distance, but I'd just say that has not been my experience. It might be only a slight rotation of the focus ring, but it does make a difference if one is going to pixel peep.

[GeneV]

I guess you're implying that even the near building is already at infinity, so there would be no difference between focusing on it versus focusing in the distance, but I'd just say that has not been my experience. It might be only a slight rotation of the focus ring, but it does make a difference if one is going to pixel peep.

Actually, I thought I was stating that directly. The infinity stop on this lens (which is probably past infinity) is no more than 4 feet, and these buildings are several times farther away than four feet even at their nearest. Building codes would usually require the nearest building to be a good deal farther than four feet from the retaining wall, and the retaining wall is already at some distance from the camera.

However, if you see differences in focus in objects farther away than infinity, it raises an interesting question about what infinity focus really means.

[Marc Sabatella]

What makes you say "the infinity stop on this lens (which is probably past infinity) is no more than 4 feet". As far as I can tell, both statements are quite untrue.

[kuau]

I did some more testing today with my DA15mm Limited, I was shooting mostly vertical at f8
what I noticed was image was sharp on the left hand side of the image and on the right it starts to get blurry. I am sending back the lens to B&H and having them ship me another one.
Thanks for all the help everyone. I will try to post another image
fyi, For focusing I just did hyperfocal at f8.

Steven

[GeneV]

What makes you say "the infinity stop on this lens (which is probably past infinity) is no more than 4 feet". As far as I can tell, both statements are quite untrue.

It has been stated many times on this board that an AF lens turned all the way toward infinity to the point that the lens stops mechanically when turned by hand is past the point of infinity focus. The last number on the lens focus scale on the DA15 is 3 feet, and there does not appear to be sufficient turn between three feet and infinity to make up another foot, given the scale provided. Perhaps that is meaningless. If there is something about the way I stated that concept earlier which is unclear, I apologize.

I understand the point of infinity focus to be something different from hyperfocus. From the point of infinity focus, the circle of confusion remains the same regardless of distance, whereas at a hyperfocal point, the circles of confusion actually increase with distance from the point of focus, but the circles remain acceptably small (which may vary in the eyes of the beholder). Thus a fine adjustment in the hyperfocal range may sharpen the image, but at infinity it does not. Is this understanding wrong?

Marc, I respect your knowledge, and I enjoy learning. I would sincerely appreciate it if you would explain, as I invited in my last statement in the previous post, a bit more about the meaning of "infinity focus" or correct whatever incorrect or imprecise terminology I have used rather than just stating that my sentence is "quite wrong." If you don't have the time or the interest to explain what you mean, that is completely cool, and I will just drop it and read a book. We are all volunteers here.

[aerodave]

The last number on the lens focus scale on the DA15 is 3 feet, and there does not appear to be sufficient turn between three feet and infinity to make up another foot, given the scale provided.

No, infinity is infinity...as in all the way out to forever. The distance scale on a lens is not linear, and rapidly approaches infinity as you get close to that end. Just look at the difference between the relatively evenly-spaced blue marks on the DA 15 (any lens does the same thing)...starting with 0.7 feet, the same amount of twist in the barrel increases your focal distance in the following steps:

0.8 ft (a 0.1 ft increase)
1.0 ft (+0.2 ft)
1.5 ft (+0.5 ft)
3.0 ft (+1.5 ft)

This is a rapidly increasing function that quickly becomes asymptotic and brings you to infinity focus. All the distances between 3 ft and infinity are scrunched into that last little twist of the barrel in a highly non-linear way. So the true focus plane becomes quite sensitive in this range, but luckily wide-angle lenses have deep DOF to cover you.

[Marc Sabatella]

Exactly. That short rnage on the lens between the "3" marking and the infinity marking actually covers quite a huge range in term of actual focus distance. The lens is actually focusing at 4 feet, at 5 feet, a 6 feet, ... at 400 feet, at 500 feet, at 600 feet. The focus ring doesn't give you a ton of control because you don't normally *need* a ton of control once you get far enough away, but you *do* need enough control to account for the difference between that near building and that far building. At least if you intend to pixel peep. As noted, at the posted size, no one could really see a problem.

And while some lenses do indeed go past infinity in terms of where they stop turning, many of these lenses actually have infinity *marked* reasonably. They just allow for turning past that, mostly to give the AF motors room to play with, and potentially to deal with thermal expansion/contraction. At least, that's my understanding.

[GeneV]

No, infinity is infinity...as in all the way out to forever. The distance scale on a lens is not linear, and rapidly approaches infinity as you get close to that end. Just look at the difference between the relatively evenly-spaced blue marks on the DA 15 (any lens does the same thing)...starting with 0.7 feet, the same amount of twist in the barrel increases your focal distance in the following steps:

0.8 ft (a 0.1 ft increase)
1.0 ft (+0.2 ft)
1.5 ft (+0.5 ft)
3.0 ft (+1.5 ft)

This is a rapidly increasing function that quickly becomes asymptotic and brings you to infinity focus. All the distances between 3 ft and infinity are scrunched into that last little twist of the barrel in a highly non-linear way. So the true focus plane becomes quite sensitive in this range, but luckily wide-angle lenses have deep DOF to cover you.

Thank you. That is what I was getting at with asking about the understanding of "infinity." I sometimes see "infinity" given as beginning at a specific distance from a camera with a certain lens, especially in documentation for pocket cameras which do not offer fine focus adjustments. However, that appears to be inaccurate terminology.

Now, can someone tell me what this person is saying, when distinguishing between hyperfocus and infinity focus. Hyperfocusing vs infinity focusing - Photo.net Minox Forum ( I have read this article before), because based upon what I am hearing, as long as we are talking about photographing discrete points on this planet, all focus which results from setting the lens at "infinity" is really just a form of hyperfocus.

[Alex00]

I just got a new DA15mm limited from B&H for my newly acquired K7 and I also picked up a used used FA31mm Limited and a also got the 20mm Voigtlander manual focus lens.
So far I am very pleased with the FA31mm tack sharp and also the 20mm Voigtlander,
but when I tried my DA 15mm limited, results not so good, very soft corners at all apertures f8-f16 some very noticeable CA which I can correct in post, but I can't fix soft corners. So I am not sure if I have a bad copy of the lens or should I be looking at the DA14mm lens, I know it's much bigger but I am not sure about the performance..

Thanks
Steven

I had the same issue with my da15. I could not get good focus in general, I got a replacement and the new copy i received is perfect.

Although CA is minimal compared to other lenses in it's range, it is sometimes unavoidable to get CA in high contrast area. I wouldn't worry too much about this as it is very minimal and only is some situations.

[Alex00]

Thank you. That is what I was getting at with asking about the understanding of "infinity." I sometimes see "infinity" given as beginning at a specific distance from a camera with a certain lens, especially in documentation for pocket cameras which do not offer fine focus adjustments. However, that appears to be inaccurate terminology.

Now, can someone tell me what this person is saying, when distinguishing between hyperfocus and infinity focus. Hyperfocusing vs infinity focusing - Photo.net Minox Forum ( I have read this article before), because based upon what I am hearing, as long as we are talking about photographing discrete points on this planet, all focus which results from setting the lens at "infinity" is really just a form of hyperfocus.

Marc,

I have to agree with GeneV on this. hyperfocal will get everything in the background in focus and doesn't matter what aperature you are on. Higher Aperture will only get you more foreground in focus. The very slight turn to achieve hyperfocal will get the job done. Example, if you are on hyperfocal f5.6 on the 15mm, then everything from about 3ft to infinity will be in focus, where as if you directly focused on the object that is 3 feet away then you will only get a certain distance in focus and not the entire background to infinity. The same goes if you were trying to focus on a far away object, although this is OK, using hyperfocal is the way to go and you do not have to worry about where to focus.

I did a quick test on this and you can see from the images below. Both are done at f5.6. The first image is focused on the front plate which is about 2 ft away not using hyperfocal. This means we get focus from about 1.5ft to 3ft.

using hyperfocal and as you can see in image 2, we get focus from 3ft to infinity compared to getting focus from 1.5ft to 3ft. Although we did not get the 2 feet target in focus, to achieve this, we would simply move 1 ft away and take the shot to get everything from 2ft to infinity in focus.

Of course jumping to f8 means we can get focus from about 2.5ft to infinity. The higher the aperture the closer of a foreground focus we can get.

image 1 at 5.6 - Focused at 2 ft no hyperfocal


image 2 at 5.6 - using hyperfocal