7

I have a Nikon D5100 DSLR with a 18-55mm lens and a 55-300mm lens. I'm shooting with a tripod (and remote) over a table top where I've arranged flowers and other miscellaneous things, using mostly the 18-55mm lens. I photograph outdoors on overcast days.

After editing, I'm still not happy with the sharpness of my images. Is it the camera? The lens? Is there something else that I can do to improve my images? I want to be able to print them out to at least 16x20 inch size.

I shot in aperture priority and played around with adjusting aperture. I do think the stopped down ones gave the best quality (about F/4). I shoot in Raw.

I know I don't have high end equipment, and I'm hesitant to pour a ton of money into this, but can anyone share any suggestions to make my images better? Am I using the wrong lens? Is there a better one that I could buy for what I'm trying to do? Thank you so much!! flowerart

mattdm
  • 143,140
  • 52
  • 417
  • 741
Jill Pauli
  • 73
  • 5
  • You mentioned that you shoot outdoors on overcast days. Do shield your subject from any winds or breezes that might cause movement of the lightweight petals or fronds? – Stan Sep 10 '16 at 20:58
  • 1
    Welcome to stackexchange. Nice work, BTW. – Stan Sep 10 '16 at 21:00
  • Thanks! Yes, I've only really shot when it's been really calm, but a shield is a good idea. – Jill Pauli Sep 10 '16 at 21:03
  • 1
    What aperture are you using? The 18-55mm lens is terribly soft unless stopped down significantly, which at least you can since you are using a tripod. – Itai Sep 10 '16 at 21:04
  • Are you using RAW format or JPG? What resolution are you saving from the camera? – Aganju Sep 10 '16 at 21:07
  • The dsomark website has stats on many different lenses at various settings. Yiu can consult that to see how much better you can expect at what price. – JDługosz Sep 11 '16 at 07:29
  • 2
    Can you show a RAW sample of an image you consider not sharp enough? That will help figure out what is wrong in a way your low-res sample cannot. In my experience, it is not at all true that the 18-55 mm lens is "not sharp enough" and there will certainly be several things you can do to improve the result without spending money. Lots of people love to bash this lens online, I guess they feel the need to justify having spent 10-20 times more money on a better one ... Yes, a more expensive lens will often (not always) give sharper results, but not dramatically so. – Szabolcs Sep 11 '16 at 08:01
  • This really is a nice image – Janardan S Sep 18 '16 at 05:29
  • Actually stopping down means making the aperture smaller , like f16 you should say opening up if you are making the aperture bigger – Janardan S Sep 18 '16 at 05:34

4 Answers4

10

It's not you, it's the lens.

The kit lens is extremely soft wide open and remains noticeably soft until F/6.3 at least. Around F/8, it gives better results but never gets tack-sharp. Stopping down further only goes so far since you will already pass the diffraction limit at F/13.

For this type of work, it is best to get a macro lens which is designed to give uniform sharpness and little distortion. Nikon makes a few and so do third-party makers like Sigma and Tokina. Considering that you are shooting from a tripod, you can control framing and do not need a particularly bright lens, only one which is sharp. There are Nikkor 40mm and 60mm macro lens which Nikon calls Micro and are worth considering. Both these are F/2.8 lenses and will give very sharp results between F/4 and F/5.6.

Keep using the tripod, this is essential for getting maximum sharpness. Use a low ISO such as 100 to 400 to get the cleanest images and good dynamic range. Make sure your subjects are still. So shield them from the wind as much as possible.

Itai
  • 102,570
  • 12
  • 191
  • 423
  • Would you recommend a 40mm or 60mm macro over a 35mm lens, such as this one: Nikon AF-S DX NIKKOR 35mm f/1.8G Lens – Jill Pauli Sep 11 '16 at 01:43
  • 2
    While I have not tried the 35mm F/1.8G, I did try the F/1.4G version and it would be completely adequate. From the look of your composition, it does not seem like you need to get that close and the F/1.8G is said to focus at 25cm (10 inches about) which could be enough. The 40mm Micro performs impeccably well though, so in either case, I would recommend that one first. – Itai Sep 11 '16 at 02:16
  • 3
    -1 for suggesting that not much can be done without spending money. Do you actively use the 18-55 mm? I do. It is not extremely soft, and should be adequate for the task if focused accurately (live view), making sure that the optical axis is perpendicular to the tabletop. The 35 mm DX lens (which again I do use) on the other hand has high distortion and noticeable field curvature. I did try to use it to photograph flat things, so I am speaking from experience. If going out to buy a new lens, it wouldn't be a good choice over the 40 mm. – Szabolcs Sep 11 '16 at 08:21
  • 1
    @JillPauli The 60 mm may force you to stand on a ladder to get everything in the picture. The 40 mm is wider and a better choice if you do decide to buy a lens. – Szabolcs Sep 11 '16 at 08:22
  • @Szabolcs - Not only have I used an 18-55mm, I have had over 24 samples of this lens supplied as kits to various cameras! They were all extremely soft near wide-open and I have performed extensive tests to find their usable range which I found to be extremely limited consider the diffraction limit. – Itai Sep 11 '16 at 14:54
  • Sigma makes an amazingly sharp 70-300mm lens for about $100/€ which has a macro mode which works quite well, albeit you need some distance (about 1m) to your subject. – glenneroo Sep 12 '16 at 19:49
  • 1
    I ended up testing out Nikon's 40mm macro indoors, without a tripod, with minimal editing and the image was SOOOO MUCH better than what I got here. – Jill Pauli Sep 13 '16 at 05:03
  • This answer is all fine and dandy, but by my math, you can get a 35mm f1.8 prime AND a flash for the price of one dedicated macro lens. And better lighting + prime glass will have much better returns than just a macro lens. Remember that OP is shooting outside on overcast skies. Actually, just shooting inside with a flash pointing to the ceiling and 18-55 at f8 would already yield much better results. (source: had a similar setup, a couple years ago) – JoséNunoFerreira Oct 07 '16 at 16:51
4
  • Try shooting at f/8 or f/11. Or even f/16
  • Make sure your tripod and the flowers are perfectly steady during the exposure
  • With the anticipated print size and expectation of perfect sharpness, you should better use higher end camera with sufficient resolution (like 24MP), pretty good lens (like dedicated macro or high end zoom) and pretty good steady tripod with remote shutter release
  • Use lowest ISO
  • Take advantage of sharpening options in your editor. Use multipass approach with separate phase for print
  • Try both lenses at ~55mm and see which one gives sharper results
  • If your camera has live view, try focusing via live view
  • Make sure you are printing at the right resolution (should be around 300 ppi) and that your image quality (resolution, compression) does not get crippled on the way to the printer (some printing services with online upload like to do that)
MirekE
  • 5,205
  • 11
  • 10
  • Guess and test is fine, but it's a lot of work when the data is already available. There's usually no need to test a lens yourself - you can just look up where it performs the best. Comparing both lenses at 55mm is a terrible idea - neither lens would be expected to perform their best at those focal lengths, both at the extreme ends of their ranges. Focusing in live view is an awful idea - if using AF it won't matter and in MF the LCD is simply does not have the resolution to judge focus. – J... Sep 11 '16 at 12:09
  • @J... The "test" as you call it won't take more than 10 minutes. Less than studying your article. Best f stop in DXO does not mean that the lens will perform the same under your conditions. Besides veracity of the data, Resolution depends on shooting distance, required depth of field, rigidity of the tripod for different exposure times and other factors. 55 might not be ideal from DXO perspective, but you need to use focal length that best suits your subject and goal. Unless crippled, focusing in live view is the most accurate. – MirekE Sep 11 '16 at 15:54
  • Most of that is beside the point. Focus distance has little effect on sharpness delivered by the lens. Lens sharpness is dictated by the aberrations in the optics - both the aperture and the focal length (for zoom lenses) affect the area of the optics which are sampled to contribute to the final image. The DXO data can tell you where you will get the optimum performance from the lens. You naturally still have to take care of the rest. – J... Sep 11 '16 at 20:55
  • 55 might not be ideal from DXO perspective, but you need to use focal length that best suits your subject and goal. - yes, and OP's goal, in this instance, is to get a sharper image out of a mediocre lens. Especially in this case it is important to leverage everything you can, and that means finding the best part of the lens to use. It is the nature of a mediocre lens that you have to make compromises - either you want to shoot at F/4 or you want a sharper image. Ideally you have a lens that does not make you choose, but if sharpness is the priority then you give somewhere else. – J... Sep 11 '16 at 20:56
  • @J...: f-stop and focal length on a zoom are not the only things that influence final resolution of a 16x20 print. So the DXO data address just a little. Besides that, I found that many of my lenses behave differently than what sites like DXO say, so I apparently take the data less seriously than you. There are tens of factors that play a role, I give them some order of importance based on my own experience. Your experience and priorities are apparently different. Since this is not a discussion site and I did not come here to validate my view, why don't we just leave it like that. – MirekE Sep 11 '16 at 21:23
  • I'm just saying there's a better way of finding optimum lens settings than picking a few settings at random and testing them in ad-hoc conditions. This adds lots of time to a workflow and won't always produce the best results. I agree with everything else you're saying but this part doesn't need to be guesswork. The only time I've seen DXO data not entirely match a real lens is if there's a focus cal issue, either with the camera or lens, or a dud lens. Usually you don't notice this until you get to the high end of lenses and bodies. Good bodies let you calibrate and correct this, of course. – J... Sep 11 '16 at 22:52
  • +1 100% this. Technique and a bit of experimentation are free upgrades to image quality, way before kit upgrades. – JoséNunoFerreira Oct 07 '16 at 16:58
4

To add some data to the other answers, a great resource is DXOMark - it has test data on a huge number of lens/body combinations. The 18-55 has no test data for the D5100 but I'm using the data here from the D7000 which has the same sensor.

Here is the Perceptual Mpix map, which provides a general overview of the perceived sharpness of the lens sampled over its range. This is an absolute measure so this is the one you need to look at when comparing between lenses.

enter image description here

At a glance, we can see this is not exactly a stellar performer. To find the lens's "sweet spot", however, we need to dig deeper. It may not be the best lens out there, but if you're looking to make the best of what you've got let's see what we can do.

We can look at the sharpness global map, which can give us center-sharpness or 1/3, 2/3, and edge-sharpness maps. The acutance global and field (to follow) maps are now relative indicators - these compare the lens to itself so you can't compare these maps between lenses.

At the center we see something like this :

enter image description here

From this you might think F/8 is fine across the range, but thankfully there are also field maps which can show you the full image sharpness at each setting.

At 18mm, F/5.6 the field map looks like this :

enter image description here

Sharpish in the center, but falls off to fuzz in the corners. Maybe acceptable for portraiture, but probably not what we want for still life. You said you shot the example picture at F/4 - you can only get this aperture at 24mm and below, so let's look and see what that gives us :

enter image description here

Eeep! Not exactly what we're after.

We know from above we'll find our best performance around F/8, so let's compare a few focal lengths :

enter image description here

So it's clear here that field sharpness is slightly better fully wide at 18mm, dropping off a bit to 24mm, improving again at 35mm, and then falling off at full zoom to 55mm. So, if you didn't want a new lens but wanted the sharpest full field shot from this lens you'd probably want to be shooting at 18mm, F/8 or 35mm, F/8. 18mm is probably too wide for still life and you'd need to correct for the distortion so 35mm F/8 is probably where you want to be.

For your 55-300mm lens we can go through a similar exercise and find 100mm F/8 to be the sharpness sweet-spot of that lens - shooting there would give you the best performance between both lenses as the 55-300 can be a bit sharper than the 18-55. enter image description here

If you wanted a really cheap improvement and you like the 50mm focal length to work with, the Nikon 50mm F/1.8G is cheap and gives good performance at F/5.6. It's also less than half the cost of the Nikon 60mm macro and even cheaper than the 40mm macro.

enter image description here

It's important to note here that, when comparing lenses, the field maps are always relative indicators versus the lens's peak sharpness. The PMpix maps give a better point of comparison between lenses as they show an absolute level of sharpness. Here the field map of the 50mm and 18-55mm lens look similar, but the absolute sharpness of the 50mm is far superior when comparing the PMpix map in the first image above.

J...
  • 611
  • 4
  • 10
  • 1
    Worth noting that DxO's perceptual megapixels are an absolute measure for a lens-camera combination, so if you are using them to compare lenses make sure you're using graphs made with those lenses on the same body. – mattdm Sep 11 '16 at 16:00
  • @mattdm Yes, absolutely, and also important to note that sometimes the body will be the limiting factor if the lens sharpness is better than the sensor can resolve. – J... Sep 11 '16 at 17:23
  • I'd say that OP with a 2 lens basic kit is not really nitpicking between 2000$ lenses...imho, the better advice would be to upgrade the 200$ lenses he/she has, irrespectively of DxO ranking....as the kit lenses are already on the lowest end of the quality spectrum. :) – JoséNunoFerreira Oct 07 '16 at 16:56
  • @JoséNunoFerreira It's not about nitpicking - it's about making the best of what you have. OP said they didn't want to spend a lot of money so, if that's the case, then you can still find the settings to get the best shot you can out of the kit you've got to hand. – J... Oct 07 '16 at 17:00
  • @J... Oh, sorry. I didn't see that you were analyzing OPs two lenses; thought it was a comparison between kit 18-55 and another lens (that OP didn't have). I'll edit the comment, not very constructive. – JoséNunoFerreira Oct 07 '16 at 17:08
0

Others have already given good advice to get better results without new equipment. If you do end up going down the Nikon micro/macro lens route its worth noting two things.

  1. If you get a short focal length, it's difficult to get even lighting on the subject. 60mm is as short as I'd go personally.
  2. Second hand/old model lenses may well have the same optics and be considerably cheaper. the 60mm is a case in point. It might have noisier autofocus but that's irrelevant for your purpose. They tend to get light use too.
Olivier
  • 4,173
  • 2
  • 23
  • 35
kingswindsor
  • 1