Is it me or is this Rokinon 14mm soft?

GeoKras1989 wrote in post #17267060
Methinks folks are over-complicating this lens. Unless your subject less than 2 feet from the lens AND you are shooting at f/2.8, there is no reason on God's Green Earth to focus it, ever. (OK, astrophoto is a big exception; infinity has to work there.) Set this lens at 3' and f/11 and everything from 18 inches to Denver will be in focus. If you have to go to f/5.6, focus at 5 or 6 feet and everything form 3 feet to Albuquerque will look just fine. Focusing this lens past 6 feet, when NOT shooting astro, is wasting potential. I spent over an hour with this lens and never touched the focus ring. I got about 20 shots of everything from flower beds at 2' to lanscapes half a mile deep. For 95% of what I will ever do with this lens, 3' focus and f/8 is all I need.

I just bought one off the forum (still in transit) and this is exactly what I was going to do with it. Set it to F11/3' and fuggeddaboutit! I hope it works

travisvwright wrote in post #17267071
If this is the case then mine is for sure bad. It does not work that way.

with your sample shot, some of the background is very very close and some far. Like the fence may exhibit corner smearing at that distance. It's a common problem with my lens as well. Stop it down to F11, and forget about DOF calculators, I'm pretty sure this lens has some field curvature that doesnt follow those guidelines.

Generally, I like to focus on something kinda close, wide open, to take account for close corners, then stop it down to F11. Basically, if you have something close, you can get out of focus quite easily with this lens and the ridiculously wide FOV.

If one part of the frame is 1 feet away from the focal plane, AND the lens has field curvature, you've got to stop it down quite a bit to overcome that issue (if even possible). I do have an issue of smearing at the corners on occasion, however I am generally happy with the results the lens provides. Take it on the field and give it a whirl, home test shots might drive you mad

travisvwright wrote in post #17267071
If this is the case then mine is for sure bad. It does not work that way.

Here is a shot that I called a total fail. Not the lens. This is my fault. Focused at 2 feet, f/2.8, flash bounced off the ceiling, front element about 1 foot from her nose. Her nose is OOF, but the ears look great. This lens, if not defective, is quite capable of L-quality resolution.

GeoKras1989 wrote in post #17267060
Methinks folks are over-complicating this lens. Unless your subject less than 2 feet from the lens AND you are shooting at f/2.8, there is no reason on God's Green Earth to focus it, ever. (OK, astrophoto is a big exception; infinity has to work there.) Set this lens at 3' and f/11 and everything from 18 inches to Denver will be in focus. If you have to go to f/5.6, focus at 5 or 6 feet and everything form 3 feet to Albuquerque will look just fine. Focusing this lens past 6 feet, when NOT shooting astro, is wasting potential. I spent over an hour with this lens and never touched the focus ring. I got about 20 shots of everything from flower beds at 2' to lanscapes half a mile deep. For 95% of what I will ever do with this lens, 3' focus and f/8 is all I need.

That is also my experience with this lens. You can use hyperfocal focusing (see the dofmaster calculator on how to set it) and never touch the focus ring again. This is by far the more easy to use lens that I have.

Miguel

In the interest of consistency, I mounted my Rok 14 on my 60D. OP is shooting a crop, too. I tried shooting with LV @ 10x, f/2.8 and a camera subject distance of 7 feet, and tripod mounted. The best I can do with LV really sucks in comparison to my method. I couldn't get a properly focused shot to save my @$$. Without changing anything else, I go to f/8, and focus at 3' (object still at 7'). The results are astounding. The fuzziness at f/2.8 is NOT the lens. Everyone who has ever used this lens agrees it is phenomenal wide open. I can back that up with examples, if required.

The first shot is my best of three attempts at 10x LV. The second is the only shot I made using f/8 and focused at 3' (remember, the object is 7' from the sensor). Both are 100% crops. Focus is on largest canister. In the LV shot, the second one is in better focus. That is a testament to how hard it is to use LV, sometimes, and to thin DOF at f/2.8, even at 14mm.

IMHO, LV is a waste of time with this lens. When you can get everything from 16" to 10,000' in focus without touching the focus ring, why bother? Oh, that would be about f/11 and focused at 3'. I understand you youngsters who grew up on AF being addicted to it. Once you get used to MF, especially with this lens, you'll wonder why AF was even included on the Canon version. It is truly a hindrance with such a lens.

GeoKras1989 wrote in post #17267793
IMHO, LV is a waste of time with this lens. When you can get everything from 16" to 10,000' in focus without touching the focus ring, why bother? Oh, that would be about f/11 and focused at 3'. I understand you youngsters who grew up on AF being addicted to it. Once you get used to MF, especially with this lens, you'll wonder why AF was even included on the Canon version. It is truly a hindrance with such a lens.

this is a falsehood perpetuated about hyperfocus and what is actually in focus. there is no such thing as being in focus for long distances. there is only an infinitely thin plane of truly being in focus, and everything else is some degree less than that.

now, putting the semantics of this aside, yeah, for some distances the difference may be undetectable, but in reality many people really do not know or care whether something is critically in focus. but if you really want to make sure if something is in the very best focus possible, no matter your f settings and subject distance, LV is the best and only way to do it.

also, i will note, the markings on this lens are off to some degree. i've used distances calculated via one of those hyperfocus calculators and still got noticeably out-of-focus shots. i don't trust those calculations at all nor this lens for doing it that way. just use LV and at least you know you got the shot as best as possible.

as for the comment of "you youngsters" i would also retort that i think many people who have used manual focus their whole lives are probably noticing that many of their old shots were actually out of focus but never noticed, or if not coming to that conclusion, just aren't looking close enough at their new images.

AND, i don't get your picture comparison... yeah, you're right, it's hard to focus at 2.8, even with LV. i've had my share of screw ups. however, it's even harder without LV. and then you compare with a smaller aperture image. yeah, of course the latter is better. and if you used LV to set the latter photo instead of calculations, it might be even sharper, though admittedly probably not by a lot or even perceivably. That's an apples to oranges comparison. if you wanna get really anal, i would say that f11 is not optimal either cause you start to get diffraction softness. It is better to go with a bigger aperture--and that's hard to do if you're not using LV.

Xyclopx wrote in post #17267818
this is a falsehood perpetuated about hyperfocus and what is actually in focus. there is no such thing as being in focus for long distances. there is only an infinitely thin plane of truly being in focus, and everything else is some degree less than that.

now, putting the semantics of this aside, yeah, for some distances the difference may be undetectable, but in reality many people really do not know or care whether something is critically in focus. but if you really want to make sure if something is in the very best focus possible, no matter your f settings and subject distance, LV is the best and only way to do it.

also, i will note, the markings on this lens are off to some degree. i've used distances calculated via one of those hyperfocus calculators and still got noticeably out-of-focus shots. i don't trust those calculations at all nor this lens for doing it that way. just use LV and at least you know you got the shot as best as possible.

as for the comment of "you youngsters" i would also retort that i think many people who have used manual focus their whole lives are probably noticing that many of their old shots were actually out of focus but never noticed, or if not coming to that conclusion, just aren't looking close enough at their new images.

AND, i don't get your picture comparison... yeah, you're right, it's hard to focus at 2.8, even with LV. i've had my share of screw ups. however, it's even harder without LV. and then you compare with a smaller aperture image. yeah, of course the latter is better. and if you used LV to set the latter photo instead of calculations, it might be even sharper, though admittedly probably not by a lot or even perceivably. That's an apples to oranges comparison. if you wanna get really anal, i would say that f11 is not optimal either cause you start to get diffraction softness. It is better to go with a bigger aperture--and that's hard to do if you're not using LV.

First paragraph is nit-picky garbage. At least we agree on that.

Second paragraph is just wrong. LV depends on your eye-sight and judgment to make a decision. My method does not.

Third paragraph is a hodge-podge. I do not use, nor do I advocate the use of hyperfocal distance calculators. Nobody is any good at GUESSING where in the hell 43 feet from the lens is. Those calculations are useless. We do agree that the lens markings are off. So what. My method still hold. Focus at the 3' mark, set f/11 and everything from 16" to 10,000' will be in acceptable focus in any print you decide to make from that frame. Period. And again, LV is A method, hardly the best. Try using that with a moving target.

An area we agree on. Lots of my old football (soccer) photos from the '70s would be judged OOF at today's pixel peeping resolutions. What was the alternative before AF. Not taking the shot. That was the only alternative.

Your last paragraph is gibberish. Focusing with LV at f/2.8 is still dependent on your eyes and your brain. Neither of mine work well anymore. And again, I NEVER SUGGESTED THE USE OF ANY CALCULATIONS. In my method, NO CALCULATIONS ARE NECESSARY. Turn the focus ring to 3', who cares how accurate it is. Set the aperture to f/8-11. Let electromagnetic physics take charge of the sharpness and press the shutter button. Diffraction BEGINS to set in when the light focused on a pixel exceeds the width of that pixel. For a 6D, that is about f/10.3. To notice it in a real photo, you'd have to go several stops beyond that. Diffraction may be a very slight problem for crop cameras with lenses that go to f/32. On a full frame body with pixel pitch like the 6D, you'll never notice it at f/22. Been there, done that. F/16 on any lens I own is just as crisp as f/8. Diffraction is rarely, if ever, relevant in a real photograph. That is laboratory mumbo-jumbo best left to those more interested in tech-specs than photography.

GeoKras1989 wrote in post #17267929
First paragraph is nit-picky garbage. At least we agree on that.

yup agreed.

i guess your method if not using calculations is just using an arbitrary number? i guess... but the point is which is going to produce the better picture. and well, if your eyes aren't good enough for LV, I think this whole discussion is moot.

in the real world, say doing a landscape, i think you're gonna see a difference between the true focus point and say the mountains in the far back. not "everything is in focus". whether you care about the small differences is a different subject.

i'm not sure what your distinction is between "calculation" and "your method." you're giving a value. it's either a calculation or it's arbitrary. i don't get your argument or the point. but whatever it is, i'm fine with dropping this one. sure... it's not a calculation then.

this i am not sure whether you are correct or not. but in respect to this argument, i can change the wording a little differently then for the same point: there will be a f-stop where your lens is sharpest. this is not anywhere close to f11 for most lenses. for most lenses by f11 there will be a noticeable degredation of iq.

this is severely inaccurate. there is a humongous difference in iq between f8 and f22. if you do not see the difference then that explains much of this discussion. i even have a video of a sea anemone that i started shooting at f8. when i noticed that that the deepest parts of the creature were not in focus, i changed it gradually to f22, the minimum fstop of my lens. you can clearly see that though more was in focus, the picture got dramatically softer. i didn't copyright the video yet so i can't post it, but you can trust me that this is true.

i personally can see the difference between pictures taken at f8 and f11 and above. it's enough of a difference that now i am going to start using a different technique i saw in effect on the FM forums--the guy shoots all his landscapes at f5.6 and focus stacks a zillion frames. it produces the sharpest pictures possible. if you care about critical sharpness, then you definitely don't want fstops too high.

here is the thread for reference--he talks about his method in detail:
http://www.fredmiranda​.com/forum/topic/13181​73

i think f5.6 is overkill for me cause more frames mean more chance for problems, but i probably would do f8 or so.

Xyclopx,
Shall we agree that my eyes aren't what they used to be, and let it go? I present only a method that works for me, with this particular lens. I don't expect my methods to work for everyone, but they may work for someone. Certainly, more acute vision would help me in the effective use of LV. And now, back to regularly scheduled programming, please.

GeoKras1989 wrote in post #17268250
Xyclopx,
Shall we agree that my eyes aren't what they used to be, and let it go? I present only a method that works for me, with this particular lens. I don't expect my methods to work for everyone, but they may work for someone. Certainly, more acute vision would help me in the effective use of LV. And now, back to regularly scheduled programming, please.

Sure. Sorry for the diversion.

Using GeoKras method I've got a lot better results, but still not blown away.

travisvwright wrote in post #17268385
Using GeoKras method I've got a lot better results, but still not blown away.

here's the bottom line travis, so you can move on... this lens is nearly as sharp in the center as my 24-70 2.8 II, one of the sharpest standard lenses out there. if it is significantly less sharp than your other lenses @ ~f8, in the center, and you know for sure you correctly focused it, then you have a bad copy.

as many have said the copy variation in this model is extreme. i know of no other lens out there capable of such good performance, yet have so many copies that are far on the other end of the spectrum.

order 3 copies, return 2. move on. (i don't normally condone this strategy, but for this lens it's almost necessary.)

also, i STRONGLY suggest you to check for centering. it's a must on this lens. (or don't check and be happy in blissful ignorance, whatever works.)

this is a $300 lens. canon makes a 14mm 2.8 too, and it's ~$2000. obviously there were huge compromises made. btw, there's a 10-22mm in the for sale forum for $380 or so. you'd be much happier with that one on your crop camera.

First shot doesn't have much detail, but the right side looks better than the left, to me. The second shot looks fine from here. The third one is just killed by flare and glare. The distance atmospheric haze also degrades the third one. The trees look fine in #4, and the left-side softness is not apparent, but the distant haze is killing your background detail. Can you provide EXIF for these shots. If that were the best I could get, I wouldn't be pleased. It is also possible you just need a bit more practice with UWA lens. EXIF, if you can, please.

EDIT: Agree with Xyclopx. The 10-22 is a much better lens for a crop camera. That price he mentions is quite good. I recently sold mine for $375.

Re: Is it me or is this Rokinon 14mm soft?

So after 43 replies, did you get your answer?

totally missed the crop factor. Might be better off with a more fitting lens, like a tokina F2.8 if you need 2.8, or canon 10-18..... 10mm is significantly wider.