Thanks for the images cyra. Now I know it was the right thing to do to get the 25mm. Learning a lot from your compositions. I really struggle to capture the beauty of woods. Several of these are excellent to my eyes!
thank you very much! It seems that some are oversharpened though. Need to work on that.
I spent some time in the woods this early spring and I find myself better getting the hang of this lens.
Earlier on I did have some trouble as to how to use it. It is not so good at mid range as I said before - producing some harsh oof areas sometimes at f 4 or f5,6.
One interesting thing I noticed about this lens is, that at infinity it almost doesn't matter what aperture I use, as you can see looking at the last two. It matters more where exactly the focus is, which can be tricky when you focus by viewfinder (not life view).
Hyperfocal distance works very well with this lens. Set focus on 2m and everything will be sharp from 1 m to infinity (at f11 theoretically, but it seems that this is almost true also at most smaller apertures).
I have not much experience of my 25 yet, but placing the absolute field of focus exactly right appears to be of great importance for the impact of the image.
Hyperfocal distance is somewhat dubious, and always sparcles some debate. Without dvindling into the scientific basis at this point, I would suggest that one might find the concept ok on large pixel sensors as the mighty D700 as well as the 5D ofcourse. But when using denser sensor like the 5D2 and whorst case the 7D, just forget it! I brought home a few hundred documentary images, shot with the 7D+21ZE at a conservative hyperfocal distance setting: Total mush, NEVER more EVER. (Stupid, I know. But sometimes a kick in the Glut max is of essence.) Iīm not saying it canīt work, just that it must be scaled to the pixel desity of the sensor. To further blur the situation, diffraction starts to show up at f/8-f/10 on the 7D.
placing the absolute field of focus exactly right appears to be of great importance for the impact of the image.
thats certainly true. Where the light is, there should also be some focus, I guess. The difficult part is, that you can hardly judge from the viewfinder where your "best" focus is. I often do a few pics in a row and shift the focus slightly. There is often a better one within the bunch
Also I want to do a test shoot, writing down exactly where I focus, since the best focus point seems to be a bit off from the theoretical hyperfocal distance. ( a set of 3: viewfinder focus, hyperfocal distance, life view focus) - at least for infinity shots that should be easy to work out.
Interesting what you say about the pixel size of the sensor making a difference in whether hyperfocal distance works. I guess with small pixels you need to just focus more carefully for starters. I will get a D7000 soon, I hope this is still usable with the Zeiss lenses.
thanks Philippe!
@ dj dunzie: I think you will be happy with the 35. If you want to go wider you still have your other lens, and at some point you might replace that with a 21mm Zeiss.
I think I also need another wide lens, with 3 "tele" lenses (50, 85, 100) just a 25 as a wide won't do Zeiss lenses really make you crazy for more...
Cyra... I think you're right re the 35. Placed the order today and should have my grubby paws on it by mid-week!
I OFTEN wonder if I wouldn't have been better off with the 21 than the 14-24 (pricing is nearly the same either way). I went for the 14-24 because it added the AF (which for a super wide really isn't anything I should have cared as much about), to complete the "trinity" with the 24-70 and 70-200, and because the thing has a really good reputation. Truth is, I've been second guessing myself ever since. However, I'll enjoy the 35 in the meantime while contemplating a trade there...
Thanks again everybody for the really detailed and helpful opinions and feedback... much appreciated!
Gary - I thank you again as always for the assistance, and great images to compare.
As one of the cornerstones in the hyperfocal distance model is the circle of confusion, it will per logic be mathematically tied to pixel size. If the accepted CoC is significantly larger than the pixel, as presented in the RAW-file, the result will be less than satisfactory, going for mushy at some level. The old HFD CoC is too relaxed for the imaging systems of today, of course determined by personal demand and expectations.
As a result of the critical field of focus I always use live view x10 for focusing, and have many times thought of, but seldom done the focus bracketing.
dj dunzie wrote:
I OFTEN wonder if I wouldn't have been better off with the 21 than the 14-24 (pricing is nearly the same either way). I went for the 14-24 because it added the AF (which for a super wide really isn't anything I should have cared as much about), to complete the "trinity" with the 24-70 and 70-200, and because the thing has a really good reputation. Truth is, I've been second guessing myself ever since.
I have also been second-guessing myself about the 14-24 since the day I bought it. It is large and heavy and has seen less action becaue of this, sports that naked front element I always fear for, and always read endlessly of the Zeiss 21's "pop" even at f/2.8 to further taunt me. Also, the Nikkor demonstrates focus shift at apertures smaller than f/4 as was well documented by Lloyd Chamber. I "console" myself with FL 14-20 & 22-24 I would otherwise not have and pretty stunning imagery to boot but having never shot the Zeiss, I cannot say which I would prefer. It also does not distort in a moustache/wave fashion like the 21 and is therefore more suitable for architectural subjects as well.
I also debated long and hard about the Nikkor 24/1.4 vs ZF.2 21 and the extra 2 stops + AF won out.
Yes, you are sounding like me. I guess that's the "fun" of buying gear... second guessing yourself when you aren't out using it. I really love some of the images I get to see on this forum with the 21, but the truth is there are dozens of great threads with stunning 14-24 images too. Having said that, I've yet to lay my hands on a CZ lens I didn't love! Truth is it's hard to go too wrong with any high end gear if it fits a need... but I'm glad I've got a brand new 35 coming my way to enjoy! Cheers!
Z250SA wrote:
I have not much experience of my 25 yet, but placing the absolute field of focus exactly right appears to be of great importance for the impact of the image.
Hyperfocal distance is somewhat dubious, and always sparcles some debate. Without dvindling into the scientific basis at this point, I would suggest that one might find the concept ok on large pixel sensors as the mighty D700 as well as the 5D ofcourse. But when using denser sensor like the 5D2 and whorst case the 7D, just forget it! I brought home a few hundred documentary images, shot with the 7D+21ZE at a conservative hyperfocal distance setting: Total mush, NEVER more EVER. (Stupid, I know. But sometimes a kick in the Glut max is of essence.) Iīm not saying it canīt work, just that it must be scaled to the pixel desity of the sensor. To further blur the situation, diffraction starts to show up at f/8-f/10 on the 7D....Show more →
I totally agree as I found out today using the ZE 35 on a 5D2. Played it conservatively with the hyperfocal setting and the infinity distance is very soft. I thought the conservative setting would allow a much greater depth of field, and at f/11. Oh well, I learned something. Now if I can just remember it, not likely.
Z250SA wrote:
As one of the cornerstones in the hyperfocal distance model is the circle of confusion, it will per logic be mathematically tied to pixel size. If the accepted CoC is significantly larger than the pixel, as presented in the RAW-file, the result will be less than satisfactory, going for mushy at some level. The old HFD CoC is too relaxed for the imaging systems of today, of course determined by personal demand and expectations.
according to dofmaster
a crop sensor needs to be stoped down one stop more than a full frame in order to reach focus all the way to infinity at HfD.
They are not exaclty taking pixel size into account though, but using some "standard" value for circle of confusion. I have been reading that their calculation is off, that might be the reason why.
Is there any other way to find out a hyperfocus value for a camera-lens combination that works rather than by trial and error?
As a result of the critical field of focus I always use live view x10 for focusing, and have many times thought of, but seldom done the focus bracketing.
lifeview only works for me when using a tripod which I find more cumbersome than useful from 85 mm on down. Also I actually do like to focus manually.
one more observation: if you focus at about hyperfocal distance, it is a good idea to do focus bracketing towards infinity. The only reason why you should get mush is that the focus was set to close. When you set it further to infinity the only risk is that at the near end things are a blurred a bit earlier, but thats not a big deal.
example with theoretical hyperfocal distance according to DOF-master. ( I need to test this!)
25 mm @ f 11 on full frame (D700): if you focus at HfD at 1,87m everything from 93 cm to infinity should be in focus.
if you get it wrong and focus at 1,5m your focus is from 1,83 m to 7,54 m and everything further away is mushy.
but if you get it wrong towards infinity and your focus is at 2 m everything from 0,96 to infinity should be sharp, if you focus at 3m everything from 1,15 to infinity is sharp, at 10m from 1,56 m to inf. and so on, focus all the way at infinity and everything from HfD =1,87m to infinity should be still in focus.
same numbers apply for a 21 @5,6 on full frame.
WIth a 7D HfD is about 3m and things are sharp from 1,47 m to infinity - theoretically - according to DOF-master
if you focus a crop camera to the full frame HfD at 1,87 you get sharp from 0,75 m to 1,5 m - and infintiy is totally mushed obviously.
(not sure about whether the applied circle of confusion is correct at DOF-master, but it gives you an idea about how it works).
I did make that mistake with the 25 at the beginning. I focused on the foreground object, and focus was off behind it giving a harsh bokeh and a bad picture
rattymouse wrote:
Wow...and the 25mm is supposed to be not good? These examples are fantastic!
That's an internet myth that probably originates from people who hate vignetting shooting test charts at 1m wide open and not taking field curvature into account. I haven't seen anyone who has actually used the lens say that it sucks.
As there is something wrong with the hyperfocal as presented here, according to my experience, which with the 21mm is solid enough, I had to take a look at the DOF-masters pages.
As expected their calculations are based on flawed assumptions. Take a look at their list of Circles of Confusion for Digital Cameras. They use the same CoC for all Canon croppers (sorry I donīt know the Nikons good enough to judge, but the same applies). They use the same CoC for the 12,8Mp 5D and the 21 Mp 5D2 as well. For example the 10D has roughly double the pixel size compared to the 7D. But they use the same CoC. Hillarious! And whorse than I expected, though I have seen much rubbish surrounding the hyperfocal distance theory through the years.
The CoC is the basis for the hyperfocal distance calculations. Use the wrong CoC and the answer will be just as wrong. I rest my case.
i shoot most of my ZF 25/2.8 landscapes at f111 or f13 but use the f8 markings for hyperfocal shooting. that delivers exceptional results on my D3X, better than the ZF 21/2.8.
Herb...
Z250SA wrote:
As a result of the critical field of focus I always use live view x10 for focusing, and have many times thought of, but seldom done the focus bracketing.
Zeiss lenses really make you crazy for more... 