ben egbert wrote:
Better eyes than mine. You are correct. I need to talk to that optometrist. At this size I can't see the difference even with my glasses removed. Fact is, I had to go back to the originals to check which was which.
I feel lucky to have good visual acuity. But actually, I carefully compared features in the mountain to see which one had more information.
Not the answer you are looking for but my own solution to this kind of problem is to use an old school Nikkor 28mm f3.5 Perspective control lens on my 5D2. You can get this lens cheaply secondhand, mount it on your full frame Canon with an adapter, stop it down and enjoy sharp, vertically corrected images. The advantages of this lens are cost and ease of use, it doesn't offer tilt (which I personally find difficult to use), just good old straight verticals. Its horses for courses, and I am not convinced myself that a zoom is necessarily the best tool for your task.
Well had you read the entire post you would know I have a TSE lens at my disposal. Yes, tilt is hard to use and for WA seldom required, but I use shift a lot. I could have used it for this image but this was intentionally taken to learn about tilted lens images.
By the way my Zeiss 50 is the last adaptor lens for me.
Monito wrote:
I feel lucky to have good visual acuity. But actually, I carefully compared features in the mountain to see which one had more information.
I guess this should be a good demonstration of my inability to manual focus. I still don't see any difference at this resolution. But I do at 100%. The way I test is to copy one image onto the other and turn the top layer off and on while viewing at 100%. I might be able to see the difference here side by side, but not for sure.
It only matters to me when printed. This would be trivial for a screen saver and even less so for web presentation. Especially after sharpening.
These two images were taken at different focus distances. The less sharp image was taken with focus on the near center tree, the better one with focus on the mountain. The first would be a good example of hyperfocal.
Its possible that the first is sharper in the grass, but if so, I can't see it even at 100%. No matter, mountains have priority for sharpness in my world.
Ben,
Of both picts you posted the second seems a little better at long distance, but it is also darker/contrastier. If this is decernable at the (500? px) size it should be clear at 100% on the monitor.
I use TS-E lenses for landscape and most often it is T rather than S, but occassionally both. These lenses are not easy to use since the small size of the LCD and ltd. 10x magnification.
It is clear on my monitor when overlaid full size. Not at all here at 900pixels and not at all on my camera even after taken and chimping at 10x.
Both images are within acceptable sharpness as defined by hyperfocal standards. But those standards are sort of outdated.
When I visit an art gallery of noted photographers, I also see soft corners on their prints. It probably does not matter to most people. I have to point it out to my wife, but not to my son in law.
Well I hope not, landscapes is all I do and I bought this to cover the 24-70 landscape range.
I had a pretty good prime kit and have spent the last two years shooting with primes and TSE lenses, the 17 and 24. But I wanted coverage finer than 24-35-50-70.
I am pretty sure I will get some good landscapes with this and will not really have a leaning tree problem because I will figure out the best solution.
Cropping a 35mm shot to 28mm is not optimal either.
Hyperfocal focusing is a compromise at best. It is based on several assumptions that are not as valid these days. First and foremost is that print size (and by extension, displayed on the screen size) is only 8x10 inches. People print much larger more often or want that option.
Can't crop 35 mm to 28 mm, but you can crop a 28 mm shot.
Monito wrote:
Hyperfocal focusing is a compromise at best. It is based on several assumptions that are not as valid these days. First and foremost is that print size (and by extension, displayed on the screen size) is only 8x10 inches. People print much larger more often or want that option.
Can't crop 35 mm to 28 mm, but you can crop a 28 mm shot.
I agree on hyperfocal, when I do my calculations, I use a CoC of .015. But it hardly matters because the same focus issues apply to a prime or zoom. I almost never missed focus with my 35L using af. Its when I am 24 or below that I have trouble.
Not all scenes work with tilt. If you have any tall near subjects, tilt does not work. And if you can't manual focus in the LCD, tilt is a real problem. Also, most tilt solutions for 24mm are 1 degree and only 1/2 degree for 17. Almost not worth the effort.
My method has been to use AF confirm for my distant subject and then tilt and use live view to check the near subject. Works sometimes, other times hyperfocal or focus stacking works just as well.
More recently I have relied on the focus scale for distance. The lcd is ok for near stuff, but not far stuff.
Now my 50D has Live view af. I never use the 50D for landscapes but tried it for focus and it looks to be a much better way to focus.
I expect future cameras will have large remote viewing screens with peaking.
As far as your 28mm comment, I would need a 24 28 35, 50 then what, 60 in addition to the UWA primes I already carry? Lots of glass to tote and lots of lens changing. I have been doing this for two years with fewer primes but no zoom.
For those using Photoshop CS6 you might take a look at DXO's new viewpoint plugin. Its' about the easiest kerning correction program I've ever used.
Regards
Paul
