« Need Some Fuji XF Lenses? | Main | Examples »

Friday, 15 March 2013

Comments

If you want DxO to do wide angle lenses, have them work on this-

http://m.wimp.com/nextsurveillance/

Not a Nikon system, but a billion more pixels than your D800.
The concept of a multi-lens in a lens camera has been discussed at TOP in the past, but here it is. Too bad it's not for sale through your links.

With this you could find your Kindle

As someone who has stuck with the D700 - and actually shoots with an OMD most of the time - I really think that I may have bought my last DSLR, at least as long as the D700 still works and maybe forever. What seems to be the case from the DxO tests is that only my 24-70 G and 50 1.8G are currently rated as anywhere near up to scratch for use on a D800.

As the owner of an 85 1.8D, a 14-24 and a 70-200 "VR1" plus a selection of fisheyes, DX lenses (I still have a D200 - hah!) and old manual Nikkors, I have nothing tested so far that's likely to give adequate results (ie meet the resolution capability) on bodies over about 16 Mp. Maybe the 14-24G? It used to be such a great lens.

The way I see this is that the manufacturers are all trying to create a situation of "permanent revolution" which will meet even Trotsky's requirements. Sell 'em all and buy a new set of super-lenses, probably at staggering prices. How much difference will all this mean for most people's photography?

Of course in many respects the situation with photographic hardware just reflects the overall economic philosophy that rules, and is inexorably destroying, our entire world.

Amongst the small selection of snaps that I'm really happy with, taken over a timescale measured in decades, it's clear to me that "I.Q." has little to do with the quality of these images which include shots taken with a Minox B and an early Zenith SLR. Maybe this is a hackneyed observation but it's increasingly obvious, particularly as I steadily slide down the affluence scale.

M4/3 is the first camera format I've owned that meets the size / performance ratio that I've always looked for. Meanwhile I await my OMD's return from its third trip to Olympus' Portuguese service centre. It went in originally for a sensor clean. It has been seriously faulty ever since...
Roy

Mike, now that you have been using your D800 for several months, please give us a summary of your feelings and opinions about the camera? I have one and really like the images. It is big and bulky and not a carry all day camera, but on balance, I do think Nikon did well with this camera. I am also interested in hearing from the other TOP readers as to their opinions about the camera as I am sure we will after you post your views.

[Hi Albert, I'm not ready yet. After the weather gets warm I will. --Mike]

Thanks, like I needed another reason to buy that 100mm Zeiss Makro Planar.

You're not helping ! See I always like 40mm on film (Minolta 7sII) and tried out both the Minolta 28/2 and 35/2 on APS-C (kept the 28/2 and it was one of my most used lenses for a long while). When I switched to Nikon, I went with the no-brainer 35/1.8 (fast, sharp, small & cheap) but kind of miss the 28mm focal length. With the current instant rebate, it's $600 ... still 3X the price of the 35 ... tempted me last month when I bought the 85/1.8. Now you're tempting me again. (The full frame upgrade argument doesn't hold any water, though, as I have no interest in the FL if I were to ever move to FF).

The Nikon D800 is a freak of nature, someone recently said. When it works well it works really, really, well.

I've only two wishes for the D800 - I wish it focused better, and I wish it focussed better. Ok, three wishes, I wished it focussed better without having to invest a ton of time into lens tuning and experimentation. Firmware update please?

If I were granted another wish, it'd be great if I could buy this very sensor in a light weight compact mirrorless camera with a new line up of smaller made-for-purpose lenses and an optional adapter to sling my D800 glass on. For this new small marvel (please Nikon do not try to make it as small as the Sony RX1... just "reasonably" small) I'd like a simple trifecta of decently but not silly fast lenses: 24/2, 35/2, and 85/2. Oh and all available by July this year please? Thanks Nikon!

In the meantime I'm not minding too much the D800 with a 35/1.4 and 85/1.8 and usually nothing else, it just doesn't pack down small to allow food in my backpack.

I will be interested in how(if) they rate the Nikkor 14mm f2.8 - if you want wide, this is the real crack. I am totally addicted to it.

The angle is what I see through my glasses, so it's easy to compose. Changing your placement by inches makes a difference. You need to be as close as possible - preferably inside the subject...

I can zoom to 100% on my D800 images and read details - who needs a zoom? It is heavy but balances toward the camera. I think I prefer it to the 14-24 zoom, but I've always read that the zoom is "better".

I shot the entire Udvar-Hazy Air&Space Museum with it - see http://www.flickr.com/photos/brucebordner/sets/72157629613910706/

Agree with most all this stuff, I can certainly say that every 85mm I ever owned was a pretty decent lens, with the 85mm Carl Zeiss being stellar (as I'm sure Mike will attest, except for the close focusing dealio).

I'm primarily a prime shooter, and I still think, even based on my eye, that primes are better than most any zoom I could ever afford. BUT, the thing that is the best about primes is that I don't want to shoot pictures with a coffee can on the front of a credit card. Any zoom I ever owned fit this profile, and it was just annoying to me, I like my lenses smaller than my camera bodies.

Just saw a Nex7 with some zoom lens on it that was just ridiculous. Sheesh, who's holding that thing....?

So both Mike and DxO endorse my love of prime lenses. I may have some sane photographic principles after all.

Lenses, mean nothing.
Scientific lens tests also mean nothing.
Anything able to be purchased should have some value to you, as a photographer, if not, the tests reviewed and the lenses themselves are of no value.

And yes, Mike, please do a detailed review on your D800, not from a technical viewpoint rather from a how do you feel aspect.

To me the latest round of full frame (or 135mm film frame size)Nikon cameras
are but a point in photographic time, which too shall pass as something else different appears.

In the film days we (I) kept my cameras until they become uneconomical to repair. Still use my F100 and an F3.
90 percent of the time. Obtained from an online classified a Nikon D3200 for C$300.00. It satisfies my DSLR needs.

Mike, I have Michael's lens align set up. If you want to come down some afternoon we can run your 28 through it's paces. One lens won't take much time....

And the issues I had with shot to shot focus have been resolved with the focus "flick" we discussed. Annoying but accuracy is much improved.

I always thought DxO's lens testing method was ridiculous because it is essentially based on an encoding of "information content" transmitted by the lens. In practice this seems to mean that any lens with a wider aperture than the next one will automatically get better scores on a given camera because, well, you can use it at a wider aperture. And when you do that, for a given lighting situation, more information gets transmitted (less sensor noise). Which translates to better scores.

I quote from http://www.dxomark.com/index.php/About/Lens-scores/DxOMark-Score , ***emphasis*** mine:

"The DxOMark Score corresponds to an average of the optimal quantity of information that the camera can capture for each focal.The quantity of information is calculated for each focal length/aperture combination, ***and the highest values for each focal are weighted to compute the DxOMark Score***.
DxOMark Score is based on low-light conditions (150 lux and 1/60s exposure time). We chose these conditions because we believe low-light performance is very important in photography today, and because photographers need to know how well lenses perform at their widest aperture. Lenses with a high f-number are usually more expensive, so photographers want to know if the performance is worth the expense. The score does not account for depth of field, and only considers performance at best focus."

So, an optically better f/2.8 lens gets rated lower than say, an f/1.4 lens even if that lens performs worse at equal f/2.8 by conventional standards.

Do you have an opinion on this method's, erm, logic?

"The Nikon D800 is the current undisputed king of DxOMark, with results that eclipse every other camera from all other manufacturers."

They didn't tested yet DP3 Merill...

The DxOMark article is practically useless because it simplifies too much. Ranking lenses based on a single artificial number derived through some formula? Doesn't sound right for anyone who actually shoots real scenes with their lenses. Sharpness is distilled into a single number derived by their own formula, several aspects of a lens are not measured or the results of the measurements are not shown. Now I do have a science degree and I value good measurements as useful predictors of lens performance, but in this case the numbers are watered down so much that I think it's best to just go out and shoot with the lenses and see how they work out.

It's not as if the lenses mentioned in the article would be bad, but by selecting different data points, different weightings or doing different measurements the rankings can be entirely different. I'm also finding in my own photography that technically well performing lenses have imaging properties that make them less than optimally suitable for certain situations, something which is not revealed by simplified rankings.

What is the focus "flick"?

For everyone bitching about the Nikon focussing on here, I'm glad it just isn't me. I've said on here many times that my Nikon stuff seems possessed compared to the Canon stuff I used.

When I was a studio director for a big retailer, we decided to get into 35mm big time for fashion location shoots, up 'til then it had been all 120. My background was all 120 and sheet film, so I did an extensive test of both Nikon and Canon systems, and not to make this a long boring blog, but the Canon system won out by a yard, not least of which was their decision to totally redesign their lens mount for the future, while Nikon was trying to duct-tape everything together to make their old mount work; when it really didn't even work between old and new bodies in their own system, without buying certain bodies that had the "flip up flag". And don't even start me about how inelegant that shaft-drive auto-focus solution was.

Anyway, when I got down to buying a Nikon digital myself, after renting and borrowing Canon stuff up 'til then; I did it solely for the fact that you could shoot native tiff. Even today, I deliver my work to clients, and 70% of them don't even call me to work with the RAW file on their selects, they think the JPEGS look fine, so you can see where this would be a boon to guys like me that had been "nailing" transparencies for years. So I bought a Nikon.

And that's where the trouble began.

The first problem surfaced when I was photographing a large group of people out of doors in great light, and they covered almost the entire frame, and the camera kept trying to focus in the far corner of the frame on a bridge abutment four blocks away! The second time was when I was doing a simple portrait with strobe lighting, and even with strong modeling lights, the camera would not focus anywhere on the subject, even when I set the focus point I wanted. It just hunted back and forth. I tried to focus manually, but I have one of those ridiculous screens where you can't tell if anything is in or out of focus. I did focus by "eye", and 3/4's of the shoot was out of focus! Let me tell you, with the right screen, I have no focus problems at all, and with film generally everything is where I want it. I even sent the body and lens back to Nikon, where they said everything was normal.

Anyway, all of this has led to a vote of "no confidence" on the Nikon stuff, I'm afraid to take it on a job for fear it won't focus and i'll end up trying to eyeball it. It's so bad, I bought a M4/3rd's, and can't believe how accurate the face focus control is, and these "contrast control" focus systems are supposed to be less accurate and slower than a normal DSLR, but you can't tell it by me, it's far better than my Nikon.

This problem alone, along with the fact that the files do not look very good with face tones, which I've sort of gotten around by going back to using filters; makes me think that either people are either not very discerning about this type of stuff, or have a boat load of legacy lenses they still want to use when they can get them to fit (or they're doing a hell of a lot of Photoshop work, which I don't get paid for).

I for some reason my M/3rd's stopped delivering the goods, I'd have to say the new and elegant Canon 24 and 35mm lenses would drive me right back into the arms of Canon.

I do not understand how Nikon could release a monster camera like this with clear auto focus issues in the final release..

Stumps the hell out of me..I've been bloated by serious acquisition fever on this beast and Im pissed to hear that such a basic attribute is bad on it. Any sign of Nikon trying to resolve this..acknowledgment?

The second time was when I was doing a simple portrait with strobe lighting, and even with strong modeling lights, the camera would not focus anywhere on the subject, even when I set the focus point I wanted. It just hunted back and forth.

"even when"? Should you not be setting the focus point every time? Anything else is 'operator error'.

The hunting could also be due to confusing the horizontally-sensitive / vertically-sensitive AF points.

It may not be the camera.

David...

To your point, my problems are with the Nikon D90, introduced in 2008, and the Nikon D300, introduced in 2007, so to hear that people are still having similar problems to mine with the D800, is disconcerting to say the least!

On another blog on here, I mentioned that I was told by a rep that Nikon was a tiny company, whereas Canon was far larger and had many more resources to apply to development and production. I can only think Nikon's stumbling around, from the lens mount and auto focus fiasco (which they've finally rectified with the "G" mount stuff today), to this seemingly inherent problems with their auto focus program, must be a result of their company size and assets.

Back in the day, camera companies had a methodology of developing a full system approach for their cameras. They introduced more and more lenses of various speed, price point, and abilities, more metering heads with greater capabilities, etc. Today, Nikon has NOT developed a complete line of lenses for their APS-C series cameras, while they're already starting the "1" series, and other weird camera with various non-complete capabilities. Nikon also seems to now be and "also ran" company, seeming to develop different systems as a response to what others are introducing instead of working on full development of capabilities for their APS-C and full frame systems.

Arn...

...the point being, I never had any of these problems when using Canon equipment the same way, with exactly the same type of subjects. Setting the exact focus point you want, which I do frequently with the Nikons, is also problematic when I found out that not all the focus points of the Nikon were of similar ability, and that only the center one was "best", with the abilities of the outboard points being marginal.

Interesting to note, when I used Canon equipment set with all the focus points "on", it almost always picks the face to focus on, and I've found this to be true with everything from the 20, 30, and 40D, to the 5D Mk1 and Mk2.

The joy of the M4/3rd's system, is that I can set it for "face recognition", and just leave it. Generally, ALL my pixs for money has a face it it that needs to be the focus point.

The Nikon D7000 has the same pixel density as the D800. Try finding a wide DX prime that lives up to the sensor. It does not exist. Good luck finding a zoom lens too. According to DXO mark the uber Nikon DX 17-55 f2.8 is worse than a third party lens costing 1/3 the price. All in all these new high-density, high-mega pixel sensors are significantly out-resolving most zoom lenses. [Ctein disagrees (see Myth #4). —Ed.] Sure spend $1500 or more on a 'pro' zoom. It ain't much better than the consumer grade on a 16mp DX / 36mp FX sensor. It could even perform worse! But self-deception works that way.

We're well beyond the point of diminishing returns unless one is prepared to spend a lot more money on lens performance suited to these new sensors. The new Zeiss 55mm f1.4 costing a mere $4K!

I settled for the Voigtlander Nokton 58mm f/1.4 SL II on my D7000. 1/8 the price of that new Zeiss, stands up to the 16mp sensor just fine and makes my 'old' Nikon lenses look like complete junk IQ wise. Perhaps the D3/3s sensor was the place to stop my own, Sisyphus-like, DSLR upgrade madness in pursuit of that perfect kit take to the grave. If only Nikon would take the D4 sensor and put it a D800 body I could perhaps quiet the voices of "endless possibilities" in my head.

Of course I am saving for a D800E and the Sigma 35mm f1.4 in hopes of achieving that perfect 35/85 two body two lens combination kit. The D7000/Voight 58mm f1.4 being the 85 side of the equation. But now there is the D7100! Sigh. Time to push that boulder of self-deception uphill again.

Matt,

I use the center spot, focus and recompose method in my work. For some strange reason after the first shot I would get a series of soft images, not completely out of focus but not tack sharp. Situation: the subject moving slightly in a pose (moving the violin bow up two inches) I would then refocus on the eyes, the camera thinks that it is still in focus, recompose and shoot. Result: soft image.

The "focus flick" is just manually moving the focus ring out so that the camera is forced to retain focus lock in between shots. Focus accuracy jumped to nearly 100% and is now something I regularly do to maintain the habit. Mike had mentioned that he had to do that with another camera system that he had (not sure if it was oly or pentax).

I do know the appearance is worse with the D800 with the resolution allowing such a high level of detail inspection... Heck I notice eye movement now that I never saw with my D3s.

Just as a counterpoint to Marek Fogiel, I owned the Zeiss 50mm f2.0 Makro Planar. I bought it because I was sick to death of my Canon 50mm f1.4, particularly its distortion, coma, and sunstars.

I spent a couple nights shooting with both the Canon 1.4 and the Zeiss 2.0. I went about my business as normal, and when I found a scene, I shot it with both lenses. For a couple scenes I did full aperture series from wide open to f8. I just shot the same things I always do, the same ways I always do, but with two lenses.

I returned the Zeiss. I was deeply unhappy with it:

Distortion seemed well controlled, but it had abysmal coma until f5.6 and it produced godawful sunstars. In terms of image quality, it flared much worse than my Canon, it had an ugly pee-yellow color cast (I guess that's the "Zeiss warmth" people go nuts about), and most surprising of all, the far corners of the image never got sharp, even at f11. It was really noticable too, and it worried me enough that I asked the guys on the FM alt forum about it...and they said it was normal for the 50 MP.

The Zeiss was undeniably sharper than the Canon from 2.0 to 4.0 and it drew images beautifully at 2.0 and 2.8. From 5.6 on, the Canon was the champ--less flare, sunstars that were merely bad, and razor sharp across the frame.

Physically, I found the Zeiss frustrating. The focus helicoid was quite stiff and seemed to be about a mile long--with 99.9999% of that travel being between MFD and 5', and the remaining fraction of a millimeter being reserved for 5' to infinity. The focus ring was smooth metal grooves, which are very hard to grip when wet, and are painful to touch in cold weather.

In the end, I kept the Canon 50mm f1.4 and bought the Canon 50mm f2.5 Compact Macro to supplement it. The compact macro flares like crazy, produces ugly sunstars, has bad coma, and shows the exact same corner softness problem as the Zeiss Makro-Planar. But it has a fast and well proportioned focus throw and costs $1000 less than the Zeiss.

Mike - there are lots of reports of focus shift with the 28mm f/1.8G, you may want to look into that. I've read one review where they don't find it (photozone), but most reviewers do mention it being a problem. I haven't noticed it in my copy, but I bought a Nikon refurb, and I rarely shoot below f/8 anyways (landscape).

I've been very happy with the combo, I'm able to get about 50% of the resolution of my 4x5/Provia/90mm Rodenstock combo (now on Ebay).

And so, we realise with 36 MP, how the limitations of phase detection manifest themselves in potential 60X40" prints from 35mm formats...

Phase detection is inherently inaccurate unless carefully calibrated for all apertures, focus distances (,zoom settings) and lighting conditions. Just that with print film and (up to) 12MP cameras we couldn't quite see the error much of the time.

Yes, you could build a camera and lens accurate enough but would you be able to afford it?

Thanks Mike for linking to Ctein's Myth #4 and I am actually in complete agreement on a lens by lens basis.

Perhaps Ctein could write an article on the DXO mark sharpness metric. Assuming I got this right, this metric seems to reflect his knowledgeable statement "until the lens resolution drops to only half that of the sensor, improving sensor resolution will produce an observable improvement in image resolution."

The pro Nikon DX 17-55 f2.8 I speak of tests at less than 1/2 the resolution of the Nikon DX7000 sensor. Whereas the comparable Sigma is much sharper but costs significantly less.

http://www.dxomark.com/index.php/Lenses/Camera-Lens-Database/Nikon/AF-S-DX-Zoom-Nikkor-17-55mm-f-2.8G-IF-ED/%28camera%29/680/%28cameraname%29/Nikon-D7000

http://www.dxomark.com/index.php/Lenses/Camera-Lens-Database/Sigma/17-50mm-f-2.8-EX-DC-OS-HSM-Nikon/%28camera%29/680/%28cameraname%29/NIKON-D7000

Actual use of both lenses indicated that this metric was correct and I specified the Sigma and a D7000 for my spouse to document her students' community service work. It helped that her copy of the lens was a keeper.

I think there is general confusion on what lenses work best on the newer sensors and your post certainly offers better guidance than Nikon. Perhaps Nikon should hire you and Ctein as a consulting team!

Tim Ashley wrote:
"Similarly, lenses with strong field curvature can be useful but are generally irritating. For example, the Nikon 28mm F1,8G lens, is really sharp - but its sharpness at mid-apertures is very very not planar: it is shaped like a runway, as shown below..."

You ca see an exemple by the end of this article
http://tashley1.zenfolio.com/blog/2013/3/dear-tim


Just a point for ponderation...I'm a long time Carl Zeiss lens lover, and have used Hasselblad (and Contax) for years because I adore the lenses, BUT, I love the lenses for their "look" as well as other attributes. A lot of that "look" is a result of proprietary coatings, or some other manufacturing secret of theirs. Believe me, I've had many a retro-focus wide angle Zeiss "dog", and it took them three tries to get it right with the CF and later 50mm for 'blad; so be fore-warned: if your looking to solve sharpness problems you think are inherent in a Nikon or Canon system lens with a Carl Zeiss purchase, it may pay to "try-before-you-buy", by renting or borrowing the one you want, and doing a comprehensive personal test.

Now so many cameras "auto-fix" lens abberations in their own lens line, an after-market lens without a chip in it attuned to these programs might make whatever your using look worse, or make the system lens look far better than it really is. Truth is, as much as I like the beauty of Carl Zeiss stuff, it's inherent "look" may not even make it through whatever the chip is doing to render the image, so I shy away from them and stick with the manufacturers stuff.

I love Zeiss, but it may not be the silver bullet you're looking for!

Jack,
cool link on what you can do with a "multi-lens in a lens camera"

Here is a link on how to do it
http://www.cs.columbia.edu/CAVE/projects/gigapixel/
or here
http://disp.duke.edu/projects/AWARE/

>>I can say, that both the Sigma 50mm ƒ/1.4 or Nikkor 50mm ƒ/1.4D are light years behind the Zeiss Makro Planar 50mm ƒ/2 in any aspect visible on the print

No doubt about the Nikon 50/1.4D. However, the bokeh of the Sigma 50/1.4 at the wider apertures has a certain appeal that my Zeiss 50/2 Makro-Planar canÄt match. At f/4 or smaller it's just the other way around...

Coming back to the Nikon 28/1.8G: a great lens, and it's definitely in the same league as the Nikon 35/1.4G. In fact it has smoother bokeh than the latter. At infinity the Zeiss 28/2 might have the edge,but we're really nitpicking here.

Carsten

My experience with Nikon FX gear up till I bought the 24mm f/1.4G, and then the D800 had been excellent. With those two products the AF bogies started to haunt me. I don't know what the engineers at Nikon are up to, but I don't think rigorous field testing is on the list.

Mike, if the D800 is overkill, so be it. Sometimes getting everything you want is only possible with an overkill solution.

Mike, your last entry as a response to Ed Richards sort of makes me smile. I just got done reading a few new articles about how magazine photo editors are now requesting work from people specifically using iPhones and Instagrams, and even Sports Illustrated had a whole very nice looking section of pre-season stuff done on an iPhone. The latest issue of American Photographer also covers all this.

I just went through a round of e-mails with colleagues talking about this "trend", and have to say some of the stuff we've seen has looked great, and I'm not sure I'd want to limit myself to those phone/cameras, but we all have to realize, we're looking, at best, to "software improved" original 8 megapixel jpegs with some sort of inexpensive wide-angle lens. All the people running on about spending hours on raw files to get what they want, and all these multiple step work flow situations, well, I guess it's whatever floats your boat, but all of a sudden there are a lot of people buying stories shot with iPhone cameras. Hate to have bought a $3-8K full frame camera to have some art director ask me next week how my Instagram "chops" are!

I've always said I could do everything I've shot in the last 5 years on a Canon G series, if I could get what I wanted out of the lens; it's hard to know if these iPhone articles are "fake-trends", as my journalism buddy Romenesko always says, or they're the real deal, in which case getting involved in M4/3rd's puts me on easy street!

The comments to this entry are closed.