Discussion in 'Nikon' started by photo5, May 16, 2010.
Yeah, short story... He thinks DX guys should just buy the Nikkor f1.8.
Typically f1.4 lenses don't perform all that great @ f1.4 or even near f1.4, so that is no big surprise. And depth of field is very shallow so that a little bit of focusing error will be an issue.
If you shoot Nikon DX, yes, the 35mm/f1.8 AF-S DX is a good alternative. That lens has its own drawbacks (chromatic aberration), but at least the price is affordable. But if you shoot other brands ....
In several full resolution, uncorrected test photos I've seen online, the 35mm/f1.8 AF-S DX shows pretty significant chromatic aberration in high contrast situations. The purplish fringing in sample photos such as trees against bright skies is about as bad as anything I've seen in my 8-year-old P&S digicam. I haven't seen this with the Sigma 30/1.4 in comparable full rez test photos. That alone was enough to give me doubts about the Nikkor. And I've seen enough recommendations for the Sigma from folks I consider credible to persuade me to consider it over the Nikkor.
These are lenses that, like the 24-120 VR, have attracted so many wildly contradictory opinions I'd want to test both myself first and buy the one specific lens that proved satisfactory in a test - not one *like* it, but that actual lens I'd tested myself.
thom can be a little hard on lenses sometimes. he's attached a lot of weight to corner performance, which is understandable since he's mainly a landscape and wildlife shooter, and in all honesty, the 30/1.4 is not that good in the corners no matter how far you stop it down.
but here's the thing: i dont shoot landscape with that lens, i shoot events and live music, and i've done some street shooting with it as well. it's great for candids and, obviously, for low-light. i normally use the 30 in situations where i'm wide open to 2.8, and there are definitely shots with that i've gotten i couldnt have gotten with any other lens in my bag.
and, when i tested it informally against my friend's 35/1.8, we both noticed it clearly outperformed the 35 both in terms of center and border sharpness at wide apertures and quality of bokeh. the nikon's high CA was plainly evident as well, and the sigma overall seemed to have more natural colors at 2.8, which is interesting, considering sigma has a rep for warmer, yellowish tones. the sigma was definitely warmer, but in a natural way, while the nikon seemed too cool and much paler, if that makes any sense.
sample variation is likely an issue with both lenses, but i've had the sigma for 3 years now and havent felt the need to sell it or get it recalibrated.
if you're a landscape shooter, you might want to avoid both lenses and go for the ZF 35. if you're on a budget, go for the 35/1.8. but if you shoot low-light events frequently or professionally, and/or like creamy bokeh, the sigma is the clear choice.
Well for me, I never need f1.4, so I wouldn't consider the Sigma for more than twice the cost of the Nikon 35mm DX. Also, the Nikon performs very very well at f1.8, which is a huge plus for me. I personally have not seen CA problems with this lens at all. It could be dependent on which body you use, and perhaps dependent on the subject matter. The Tokina 12-24mm f4 zoom had CA issues that were readily noticeable when I used it with my D80. But the Nikon 35mm f1.8 performs very well on the D300 and I have no regrets buying it. It is a better value overall than the Sigma.
dave, just as an fyi, i tested the 30 against the 35 on a d300. the sigma handled high-contrast scenes better. i also have a tokina 12-24 and d80, so i will agree that combo will show CA.
frankly, i had thought the nikon would do better, and my friend was a bit shocked when i zoomed in at pixel level at how much more detail there was in the sigma shot, plus the colors were more realistic. i dont think the 35 is a bad lens at all--for the price, it's great--and i might have scooped it had i not had the 30/1.4 first. i'm just saying that my real-life experience with the sigma is different than hogan's test charts.
Thom seems to say that the 30mm does not excel at anything. This lens excels at center sharpness and it excels in its creamy buttery (don't you just love those words!) bokeh. For people shooting wide open, edge sharpness is mostly a non-issue due to narrow DOF. You chose your lens for what you want it to do.
I would ordinarily think that an article should be judged on its own merits but sometimes it really helps to check what kind of images the author of the review has published to see the background against which the review is written. If you're looking for a people lens, then read reviews from photographers who do people images. Hogan's optics rating seems to be a "flaw-o-meter" type of thing - if a lens has optical flaws at some settings he will lower his rating irrespective of what the lens is capable of when used for its intended applications i.e. at its best. I think his optical rating of the 35/2D AF of "**" while the 35 DX gets a "****" is just hilarious. I just re-acquired the 35 D as I realized that 1) for some situations I really need a small wide angle with autofocus and 2) when Nikon finally decides to make a new 35mm prime, it will most likely be closer to the 24/1.4 AF-S than the 35/2 AF in pricing, to which I respond "forget it, Nikon". Make something reasonable for practical use rather than a sports car type thing.
Indeed. Missing from that sort of write-up is a simple, qualitative sense of how the lens works on certain types of subject matter, in real life. Though Nikon's 35/1.8 wasn't available at the time I purchased the 30/1.4 in question, I've had a chance since to give it a whirl. No thanks! It's not horrible or anything, but the CA is definitely a distraction, the extra 5mm longer actually matters (to me, it's just 5mm too many for why I'd use a "normal" lens on DX), and ... ugh, that harsh stuff it does in OoF areas. Some people are allergic to it, and some aren't. I'm intermittently so.
I simply like the way the 30/1.4 looks on the sort of subjects/situations for which I use it. Is that bit of hair-splitting worth the extra $200? Ehh... depends which week you ask me. I'll almost always say yes. As for the build? I have an embarassing anecdote from just last weekend:
I had the 30/1.4 capped (front and back), with the hood reversed, in my bag. I was shuffling things around at the end of a long day, transferring stuff to a Peli case. Caveman that I am, I kicked the bag over, and the 30/1.4 bounced out, and started rolling down the slanted concrete deck of a parking garage. I had to sprint to catch it, and it traveled about 75 feet, jauntily bouncing along the pavement with its hood acting like a springy tire. To my astonishment, there wasn't a mark on it, except for the rear cap, which caught a small scuff. And the lens is working perfectly. Whee! Product testing, in the field.
Wow! Not impressive at all.
I'll stick with the 35/2 Nikkor, thank you.
I was aware of the CA issue but I still like my 35/1.8, and have no intention to replace it with something heavier, larger, and more expensive to get better IQ. I think this lens is intended to be used for people who already have a slow kit or a f2.8 type of zoom lens that cover the 35mm focal length and need a cheap, light, prime lens to shoot under low light. Most people would use this for candid shots of families, pets, etc. For this type of application, CA under high contrast lighting and soft corners are really not issues that one need to worry about. Certainly not for the price I paid. I look at all my pictures taken with this lens, CA was never present since I rarely use it under broad day light. The lens renders skin tone in a lovely fashion and the bokeh is also quite good, certainly much better than the Nikon 50/1.8.
The problem with the Nikon is the busy Bokeh. The cat image doesn't show high contrast areas nor OOF specular highlights. The Sigma is know for better Bokeh.
Thom just doesn't like it because it doesn't have "Nikon" on the brand tag. Every other review I've seen, and my copy, all point to it being a superb lens. The odd thing about Thom's "test" is that the edge test at f/1.4 is sharper than the centre image at f/1.4. I get tired of poor test methods being blamed on equipment. The centre is far sharper than the edge at f/1.4.....on this, and every lens made. If Thom's test shows different, then he's borked the focus on the centre test.
Honestly, what does one expect while shooting all the way open at f/1.4 on a $400 lens? Here are a couple of grab-shots from a few minutes ago. In this first example (wide open), I just used a corner focus point to get the light stand grip head, in the lower-right corner, in focus. DoF is, of course, razor thin when wide open at that distance. Point is: the area in question is in the least-sharp part of the lens's coverage.
And here's a look at what it's doing to the far corner of the frame in such a case, when something that's in the far corner happens to be in the in-focus plane.
Needless to say, if you're shooting at f/1.4, bitty details in the corners are rarely an issue. Stop this same shot down, and it gets very, very sharp very quickly.
And how about CA? I grabbed a couple artifacts off the shelf. I wanted something with a dark, well-defined edge, and I wanted to completely over-expose from behind it so that we could see the expected color fringes where those high-contrast transitions occur. This is always at its worst when you're shooting wide open. So, this is at f/1.4. You're seeing a bit of contrast reduction, as one would expect when shooting with that huge aperture.
And here's a crop. If you look carefully, you'll see a bit of green CA along that wooden grip and hammer, where the blown highlights will always cause such trouble. In my similar tests, the 35/1.8 was really quite awful, by comparison (but then, so was Nikon's 50/1.8, and 50/1.4 - regardless of generation).
Needless to say: stop things down when you can.
That whiff of CA is long gone, of course. Things are very sharp (here, in real life) across the frame, when you only somewhat stop down. But I didn't buy this lens to use it at f/9.
Thanks Matt. Your quick test is more valid than Thom's!!!
@Matt. thanks for posting. i'd say you have defended the 30/1.4's honor quite chivalrously.
@Dave. i like the composition of your shot. it's plenty sharp. but i dont like the bokeh. thankfully, you kept the OOF elements way in the background and filled most of the frame with the kitty. but in a shot with a subject that didnt fill as much of the frame, i think the 35's shortcoming's would be more apparent.
Hey, I like Dave's cat picture. That's what a photograph is supposed to do - be interesting. The only reason I bothered to pull out my ad-hoc hairsplitting examples was because we're talking about the picky little things that Thom brought up. I'm not going to pick on him, either - these are opinions, obviously. But it seems far more reasonable to look for actual examples of the lens in real-life use, and qualitatively decide if you like it or not, and if it has earned its price tag. For me the 30/1.4 passes that test. It's subjective territory, that's for sure.
But I didn't want casual references to "awful" corner performance to go completely unanswered, lest a lot people come to Dan Brown's immediate conclusion (above). More of what I shoot involves Nikon lenses than anything else. I like Nikon lenses, and don't think that most people would ever regret their newer 35/1.8 DX ... unless they're using it (as I do) for specific reasons, in which case Sigma's 30/1.4 just seems like a better fit. The extra $200 doesn't bother me because I keep lenses for many years.
I've read Thom's pieces too, with a bit amazement that the Sigma got pounded on so heavily. The "only 2/3 stops difference"... really, I do not get it. Fast primes are fast primes, and fast can't often be fast enough. It's not only about speed, sometimes one wants this hyper-shallow DoF. Yes, it costs extra money, but the Sigma can shoot at f/1.4, the other 2 tested simply don't. Only alternative: a MF 35 f/1.4, and I was lucky with the price I paid, but that would probably still get me 2 Sigma's Sigma deserves some praise for being early to recognise that DX/APS-C shooters also want a standard prime lens, and a well built fast one.
I do agree quite a bit, though, with the story on the AF-D 35 f/2. I have that one too and I've always found it quite funny how this lens cannot AF with the outer AF points on my D300... none of my other AF primes have an issue with that, but the 35 f/2 keeps hunting and hunting. At f/2, it's simply quite far from good on DX (can't judge for FX, though), from f/2.8 on, it's a very nice (and light-weight) lens, though.
I've always found it quite funny how this lens cannot AF with the outer AF points on my D300... none of my other AF primes have an issue with thatThat was exactly what happened to my 17-55mm/f2.8 AF-S DX after I dropped it inside a padded camera bag. The lower AF points on my D2X wouldn't work with that lens, and I tried it on two other D2X and had the same issue.
It turns out that the lens mount was bent a tiny bit. It is not visible from the outside, but one side of all images from that lens was out of focus. The AF point issue is a strong indication. Get your lens caliberated.
Here's a shot with the 35/1.8, wide open, that shows its somewhat harsh bokeh. But still quite acceptable.
It's also helpful to consider the "foreground bokeh" of a lens.
hmm, i never said i didnt like dave's cat. it's a great shot. but i think that shot illustrates two things: 1) that the 35's bokeh can be a little bit jittery/nervous/distracting (just like the 50/1.8) and 2) that a skilled shooter will know the strengths and weaknesses of their lenses and compose/frame/adjust accordingly, as dave has done.
jim's first shot is another example. the 35 is very sharp where it needs to be sharp. and the woman holding the glass is within the acceptable 'bokeh zone' for that lens. but if you look at the wine bottles on the wall behind her, they're a bit blotchy. in his second shot, the depth of the OOF is reduced considerably, so the harshness of the bokeh never really interferes. this correlates with my experience with the 50/1.8, that if you keep the OOF plane relatively close to the in-focus subject, it's not that bad. but if you have too much OOF area and/or more distance between focused subject in foreground and defocused backgrounds, the less-than-stellar characteristic of the bokeh rears its head.
What is all this talk about chromatic aberration and lateral color fringing? I thought recent Nikon bodies corrected for this -- or am I just a sissy for shooting jpeg?
Toby, nope, no problem shooting JPEG only if it suits your needs. I've done that quite a bit myself, especially when I need the extra shots more than I needed raw files.
Not all Nikon dSLRs automagically correct for CA. My D2H doesn't, gotta do it the hard way. Another reason I'm a fan of Corel PSP X - very handy, easy to use filters for correcting both CA and purplish fringing.
I have a 35mm f2 and for some things it's awesome and others not so great. The CA does really bug me. Still I'm pretty happy with it generally speaking.
That said, these guys at least used to use the Sigma a lot and they had some amazing stuff:
Shun, thanks for the tip; Since the outer edges of the lens are not terribly sharp wide open, I figured it lined up that the AF module would also have issues. Center is never an issue indeed, did not consider some misalignment. It does happen on all 4 sides, though, with this lens. I'll give it a better look some day, but at present I enjoy the 35 f/1.4 too much to bother
That said, these guys at least used to use the Sigma a lot and they had some amazing stuff:
http://meninenuotrauka.lt/that slideshow is a pretty good example of why sharp corners dont matter with a fast prime.
that said, i've kind of been toying with the idea of the tokina 35mm/2.8 macro. i have a hunch it's sharper than both the sigma 30 and the nikon 35. anyone out there used it?
The 30mm f1.4 Sigma excels at something besides outstanding center sharpness and creamy bokeh.
It excels at being a 30mm lens. The 35mm Nikkor makes a lousy 30mm lens: it's 5mm too long!
I like a "real" normal, a lens with a focal length about the same as the image diagonal. My favorite normal on 35mm film was always a 45mm f2.8 Tessar, not a 50mm f1.8 or 1.4 double Gauss. That 11% difference in focal length was always enough to make a difference to me. I was also rather fond of 40mm lenses.
35mm? Plagh! That's a 52.5mm equivalent, even worse than 50mm. Sigma got the focal length perfect.
I just saw this thread, and Matt said pretty much the same things I would have said about this lens (and the 24-120 VR). ;-)
I have owned the Sigma for a couple of years now and am very pleased with it. I use it for the same uses as I would use a 50/1.4 on a FX system. However, wide open, on a DX body, it beats the pants off the Nikkor 50/1.4 (latest version) on an FX system in terms of sharpness, CA, loss of contrast, etc.
I have not used the new Nikon 35/1.8 G, but have owned the 35/2 AFD for years. It's a very nice lens, and I've taken many fine pix with it, but it's got really severe sagittal aberrations wide open as you get about 75% of the way out to the corners. For example, look at http://static.photo.net/attachments/bboard/00V/00VYrU-212333684.jpg in the thread, "35 mm lens - Which one?" ( http://www.photo.net/nikon-camera-forum/00VYeY).
I don't think the edge AF issues mean the lens is necessarily out of alignment. These lenses were designed long ago and peripheral point AF was not available to test at that time - all cameras had just one sensor in the center. Perhaps the aberrations of the 35/2 are simply such that it makes it difficult for phase-detect AF to work with the furthest off-center points.
While I have grown older, but not necessarily wiser, I have reached a conclusion. I think there is great merit in always buying your camera brand. Though Nikon gear is horribly overpriced, In the long run, go ahead and buy all Nikon. It's a karma thing.
John, "horribly overpriced"? LOL... compared to Tamron, perhaps, but compared to Leica...???
We hear complaints, sometimes, about choosing a lens because of its bokeh. That should stop, now that we're choosing based on karma.
Real karma is fine, but instant karma will get you every time.
i cherish my 35mm but i will have to admit my wanting to have the sigma 30mm f/1.4. there is just no comparison on the bokeh. it's creamy against nikon's "busy" character. of course the latter can be avoided by choosing your background.
i am encouraging my granddaughter to take photography class (she's interested) so i can justify the purchase of the sigma 30mm ----- i will let her use the nikon 35mm.
i don't see much CA in my 35mm. or maybe i just don't concern myself too much with it.
Wasn't "Instant Karma" a title for one of the Rockford Files episodes? If so, I rest my case!
john the nikon 35 costs less than the sigma 30. i'm not sure how a $200 fast prime with AF-S could be "horribly overpriced," since that one's a pretty sweet deal. the karma of course is learning to live with CA and nervous bokeh.
John Lennings ("J" looks like "L", backwards, and it's close in the alphabet)
John Lenngnis (swap a couple of letters)
John Lennonis (knock a piece off the "g")
John Lennon's (remodel the "i")
John Lennon's "Instant Karma"...
Separate names with a comma.