Open

by damned truths

submit your photo


Hall of Fame
View past winners from this year

Please participate in Meta
and help us grow.

Take the 2-minute tour ×
Photography Stack Exchange is a question and answer site for professional, enthusiast and amateur photographers. It's 100% free, no registration required.

I have a Sony NEX-5R with two prime lenses: 19mm F2.8 Sigma, and 35mm F1.8 Sony. I'd like to compare these two lenses to see why the second lens cost me more than twice the first, what the differences are, whether they are noticeable, whether I'll be better served by buying a cheaper lens over a better lens in the future, or whether more expensive lenses make sense for me, etc.

I'm NOT looking for:

  • A quantitative test (like MFT or other numbers). I'm looking at a qualitative / subjective comparison for my kind of photography (landscapes and low-light).

  • A synthetic test like a resolution chart.

  • Reviews, which describe lenses in general terms like being soft or having chromatic aberration or what have you, which don't give me a concrete idea of what the actual differences will be for me in day-to-day shooting.

For example, I compared a point-and-shoot with an iPhone at Why does my point-and-shoot take hazy photos? which give me a good idea of the differences.

I'd like to do a similar test for my NEX. If I had a zoom lens and a prime lens for my NEX, I could zoom the zoom lens to match the focal length of the prime lens to compare apples with apples.

But when I have two prime lenses, they capture different fields of view, so I'm comparing apples to oranges. Here's a photo from the wide-angle lens:

Wide-angle

compared to a photo from the normal lens:

normal

I could crop the image produced by the wide-angle lens to match the photo produced by the normal lens, but that would in effect be a digital zoom and therefore be unfair to the wide-angle lens.

One technique I can think of is to pick a scene, shoot it with my normal lens, and then switch to the wide lens and walk closer to the center of the scene to match the photo taken earlier. Is this a good technique?

Obviously it won't work in cases like this one, where I took these photos from my balcony and can't walk towards or away from the scene. What are some other techniques to compare lenses? Thanks.

(I looked at How do I best compare lenses? but it doesn't address these questions.)

share|improve this question
1  
I guess you still can compare the image quality of the two images, since these 2 lenses are mainly made for 2 different purposes they shouldn't be compared as the same focal length as you mentioned. If you really want to compare more closely you could always try to zoom both photos at 100% and compare it there –  Yao Bo Lu Dec 26 '13 at 10:03
    
Zooming both photos 100% will still result in a different angle of view on each, won't it? In any case, zooming 100% is not a representative test, since I usually don't look at my photos at 100% :) –  Kartick Vaddadi Dec 26 '13 at 10:07
2  
What i'm trying to say is that since these 2 lenses are not the same focal length, I would probably never compare them in the same field of view. Since they are not meant to be compared that way. I would keep the field of view you get, and just compare the image on abbreviation, sharpness, color on the field of view you already have. Just keep the same Whitebalance, contrast and exposure on both pictures –  Yao Bo Lu Dec 26 '13 at 11:21
    
Hi Kartick, I hope my answer below helps at least a bit. Different focal lengths do have a qualitative impact on your images that you can anticipate without taking the photos. If you took two photos at different focal lengths, but moved to make sure the field of view was the same, the closeness of objects in the scene to each other would be different. At smaller focal lengths, objects within the scene seem further apart (front-back) and with higher focal lengths, they would seem closer together. The question is, what would you like in your landscape shots? –  Aleks Danger Dec 26 '13 at 13:39
    
Take a look at What makes a good lens good? My answer there should give you some key things to examine. –  mattdm Dec 26 '13 at 14:07

5 Answers 5

up vote 1 down vote accepted

There is more to comparing lenses than looking at the images. And you would want to compare lenses of the same focal length and max aperture. As the price also depends on focal length and max aperture.

However, for image comparison of your two very different lenses, at least you need to compare the same images, which means your idea of cropping is the way to go. If you walk into the same FOV you change perspective which is another image. However, then you give the wide angle lens an advantage, if you only look at the middle, so you need to place the subject in the corner and crop there, too. You need to include high contrast things in your scene, so you can compare Chromatic aberrations. And make sure that you use the same aperture value. And view them in the same physical size. Very important. You'll end up scaling both down for that fullimage viewing, so the crop is not digital zoom.

Next, you need to touch them, shake them, feel them, hear them, turn the focus ring, and MF/AF switch. How is the build quality?

Then you need to do focus exercises indoor in low light. Focus close and far fast after each other - what is the hit/miss rate?

And then you have to remember the differences between them, since it is not the same type of lenses. The 35mm can give you a better DOF effect for subject background separation (though still a bit short for that) and let in approx 1 stop more light and make prettier photos of people, while the wide angle can capture a group photo in tight spots.

Other factors to look for that adds to the price include coatings that increase contrast and reduce lens flare. And the focus mechanism, is it ultrasonic and does it have full time manual override and does it rotate and extend on AF. Sigma and Tamron often spare on this department. I have only one sigma with that and its a ultra wide angle zoom. And the quality of aperture blades, how many, and are they rounded? I have lenses with 6 through 15 blades. Typical nowadays is 6-straight to 9 rounded.

Canon makes one with only 5 (50mm F1.8 II) and it is their cheapest lens ever. It is also plastic mount, no FTM, slow micro AF that rotate the ring, fiddly manual focus ring. The next step up costs 3 times as much and has F1.4, 8 blades, metal mount, micro ultrasonic AF, FTM, a nice touch and useful manual focus, but it is not sharper.

The next step up again has 8 circular blades, better coatings, the top of the line ultrasonic focus which is more robust, metal barrel and mount, and it costs 10x as much as the F1.8, and reviews state the IQ is disappointing for an L series lens, but an advantage over F1.8. In my pixel peeping the benefit seems to be mostly in the contrast. Judge for yourself. I tend to go for the middle road. Usually the cheapest gear is not good enough, the high end is for people who can deduct the gear in taxes, but the next step up from the cheapest tend to give good bang for the buck. However, in the case of these 50mm primes, people tend to go for Sigma 50mm 1.4 instead of the canon, which have features like the 50mm 1.4 but is optically slightly better.

50mm comparisons from 100$ through 1600$ lenses

share|improve this answer
    
I'm unlikely to buy two lenses of the same focal length or aperture so I'm afraid that that test won't be useful to me in practice. But your other ideas all sound useful to me. Thanks. –  Kartick Vaddadi Dec 27 '13 at 8:25
    
I forgot some other factors. editting –  Michael Nielsen Dec 27 '13 at 8:39
    
So I added same focal comparison to show how much emphasis there is on the features other than optics in price appraisal, so you can decide which features are important for you. –  Michael Nielsen Dec 27 '13 at 9:12
    
Thanks for the link to the site showing the comparison of the $100 through $1600 lenses. The differences are negligible for me, so I think I'll buy cheap lenses in the future, when the aperture and focal length and optical stabilization are the same. –  Kartick Vaddadi Dec 28 '13 at 4:09
    
I don't have an opinion regarding metal vs plastic lens bodies, or the number and shape of the blades, or USM, or the coatings. I guess it's better to compare the results than the technology, when I don't know how much the technology matters in day-to-day use. But your tips about checking autofocus, checking with high-contrast subjects, depth-of-field, and low-light photography are useful. Thanks. –  Kartick Vaddadi Dec 28 '13 at 4:59

"How do I compare two prime lenses of different focal lengths? …I have two prime lenses, they capture different fields of view, so I'm comparing apples to oranges."

By looking at pictures. Ideally lots of them, taken under conditions that match what you expect to use the lenses for, and deciding what you personally like. But then that's also the best way to compare two different lenses of the same focal length.

"One technique I can think of is to pick a scene, shoot it with my normal lens, and then switch to the wide lens and walk closer to the center of the scene to match the photo taken earlier. Is this a good technique?"

No. These are such wildly different focal lengths that they simply cannot take the same image: all of the spatial relationships – perspective, size, shapes – will change dramatically. You might make a brick wall the same size in both images, but the shape of the building will be different, its prominence will be different, the foreground and background will be different, and so on. There have been few times in my own photography when the choice between a 28mm-e and 50mm-e lens has been inconsequential.

share|improve this answer
    
Thanks, mpr. When I compare photos taken under conditions that match what I use the lenses for, I find that I often prefer the wide lens simply because it has a wider field of view, which I prefer. But that doesn't give me any useful information. I meant to ask: what aspects of the photos should I compare? I'm afraid that simply saying "decide what you personally like" doesn't really answer the question :) If I knew what to look for, I'd have had my answer already, with no need to post this question :) –  Kartick Vaddadi Dec 27 '13 at 8:17
    
Respectfully, knowing what photos you like is the most important information there is. I frequently see people shop for lenses based on charts and tests without looking at images – I work in a camera store – and it's a thousand-dollar mistake that I've made myself. It's important to know what "quality" means in all of its various forms, but given a choice between using the 'right' tool or using the 'best' tool, I hope I always have the strength to choose the right one. It's a happy but rare moment when 'best' 'right' and 'affordable' all combine into one lens. :) –  mpr Dec 27 '13 at 17:55
    
When it comes to learning about lenses, I find lenstip.com invaluable. Not only do they look at a whole range of behaviours that other reviewers skip – the section on coma will be particularly relevant for your night photography – but also they frequently provide good context and don't disguise their opinions. And if you can get past the specificity of a lot of this site, lavidaleica.com/content/what-good-lens leads to some excellent information as well. –  mpr Dec 27 '13 at 17:56

I'm not sure this will be the answer you're looking for, but I'll give it a shot. I won't mention focal length anywhere, either, as you don't want to base any comparison on that. Forgive me if I slip up.

Short answer

If you want to compare lenses qualitatively/subjectively, you need to test them shooting the scenes you intend them for. Compare your results and make a subjective decision.

If you can't do that, you should use the 'numbers' with the context (your preferred scenes) to make a judgement on which lenses will most likely give you the results you desire. A simple rule I use, objective + context = subjective. The objective numbers vary from lens to lens, but the context stays the same. So the only thing impacting your subjective analysis of lenses is the context - what you prefer to shoot.

Long answer

First, about price:

An expensive lens isn't usually expensive because it produces better photographs - an expensive lens is expensive because it is more difficult to make, more costly to make and may produces better images.

Having said that, wide aperture (or fast) lenses are more difficult to manufacture and cost more to manufacture, so they'll cost more to buy. A super-massive aperture for landscapes is not necessary at all. Usually it's necessary for 1) capturing movement in low-light conditions and/or 2) having greater control of depth-of-field.

Second, about quantitative/objective comparison:

I think the numbers and charts you don't want are necessary to make some judgement about lenses you want to buy without ever having the chance to use them. If you have the opportunity to test competing lenses, you can undertake the same comparison as you had with your compact (as linked). If you don't have this luxury, you'd need to make a comparison based on numbers.

The numbers are often associated with qualitative outcomes in your photographs - the stuff you want to make some prediction about, so with some experience you can begin to understand what these numbers tell you for various scenes (such as landscapes or low-light, or both).

My favourite spot for objective numbers to help me decide on new lens purchase is http://www.dxomark.com - this site lets you compare tested lenses on the body you're using. The body is less relevant, though, because you can make the same objective comparison between lenses on any body, as long as it's the same for all lenses.

The numbers are objective, but the numbers + context = subjective (what you're looking for). The context in this case is 'low-light' or 'landscape' or both. It's like me saying "that's shit" or "that's the shit!" - one is negative, one is positive. The word is the same, and we all know that, but the context gives it different meaning.

The things I look at are mainly sharpness and distortion. Sharpness because that can vary with aperture - some lenses are sharper at wider apertures, whilst some are sharper at medium-narrow apertures. Distortion because you may or may not want images to be warped or bent out of shape around the edges - and that depends on what you're shooting. Most distortion can be corrected in post-processing, though.

I don't worry too much about chromatic aberration (CA) because DxO measure Laterial CA which can be fixed relatively easily in post-processing. I also don't look a great deal at vignetting for the same reason. DxO do not test for Longitudinal CA as far as I can tell.

I'll add that reading lens reviews can reveal a lot, particularly when you decision is context-driven. For instance, I have a Nikon AF-S 85mm f/1.8G which, when shooting high contrast edges at a narrow aperture (1.8 to 2.8) produces images with a decent amount of Longitudinal CA (LoCA) - purple and green fringing around the edges. Had I read these reviews before I bought the lens, I would have known what I can and can't use this lens for. When I am shooting in a controlled environment and have control over the light and high-contrast edges, the lens produces superb images, as intended.

I hope this helps.

share|improve this answer
2  
Great answer, although I would suggest replacing "and produces better images" in part about price with "and may produce better images". –  mattdm Dec 26 '13 at 14:19
    
I couldn't agree more, thanks mattdm! –  Aleks Danger Dec 27 '13 at 1:03
    
Thanks, Aleks. It seems like your answer boils down to "Compare on sharpness" (because geometric and chromatic aberration can be mostly corrected). –  Kartick Vaddadi Dec 27 '13 at 8:07
    
Even with respect to the sharpness comparison, isn't it biased in favor of the normal lens because objects there appear bigger and sharper, everything else being equal? –  Kartick Vaddadi Dec 27 '13 at 8:16
    
I'm sorry I'm not sure what you mean by normal lens. Could you please elaborate? In continuation of all that's been discussed, once you've made an appropriate choice for focal length and maximum aperture you need for your style of shooting, that's when it boils down to comparing sharpness and other things like distortion and CA. distortion and CA can be fixed, but it's always preferred to have better optics. I only ignore these things when I hit a financial ceiling! If you have the money to spend, these should be considered too. –  Aleks Danger Dec 27 '13 at 13:02

One possible reason for the price difference can be found without even comparing the performance of the lenses. An f1.8 lens would generally be more expensive than an f2.8 lens. It's harder to build a lens with a large aperture without losing image quality.

If you want to compare the lenses to each other, then you would concentrate on properties that aren't directly affected by the focal length, like build quality, distorsion, vignetting and edge sharpness.

Also, rather than shooting with the best aperture for each lens, you would try the smallest and largest apertures, and see how well the lenses work at the toughest conditions. The expected result would be that the cheaper lens would suffer more quality loss.

share|improve this answer
    
Aren't geometric and chromatic aberration and vignetting corrected automatically, either in camera or in Lightroom? Maybe I should exclude these factors, and look only at sharpness, edge sharpness, and the widest and narrowest apertures. –  Kartick Vaddadi Dec 27 '13 at 8:20
    
@KartickVaddadi: I don't know about Lightroom specifically, but in Adobe Camera Raw those can be corrected but aren't corrected automatically, so I expect Lightroom to work similarly. Anyhow, you might want to put more weight on the quality flaws that are harder to compensate for. Edge sharpness might for example be a bigger problem for you than vignetting. –  Guffa Dec 27 '13 at 15:24

The primary difference between your lenses is the maximum aperture or lens speed. There is a huge difference between 2.8 and 1.8 and it is generally expected that the price will double for such an increase in lens "speed".

The way to "test" the difference between the lenses is to look at the situations where the difference in speed makes a difference. This will show you if you need the extra lens speed or not. The primary advantages of a fast lens (a lens with a large aperture (small f number)) is that it can get a shallower depth of field, which means it can produce softer blurred backgrounds. You can compare these by taking photos with both lenses fully open (smallest f number) and look at the way the background looks. You can also try using the faster lens and stoping it down to f/2.8 to see what the lens would be like if it had an f/2.8 speed at that focal length.

Another area that makes a big difference is low light shooting. Try shooting in low light with both lenses and look at the difference in how the lenses perform. The f/1.8 lens is going to be able to use a faster shutter speed or a lower ISO than the f/2.8 lens which buys you a lot more leeway in low light. This is really where you are going to see the strongest difference between the lenses.

Other factors that are not as easily tested are things like build quality and optical quality. Not all lenses are created equally. Some lenses are made with weak, cheap materials and can't hold focus accurately or result in different distortions to the image such as chromatic aberrations (color fringes that are generally red on one side of a high contrast area and blue on the other), vignetting (darkening of the corners), low overall sharpness (the image appears fuzzy or lacks detail even in areas that are ideally focused) or lacks edge sharpness (the center may have more detail than the outer edges of the image). For a lot of these you just have to look closely at the photos to tell the difference and are much harder to tell without controlled tests though, at least if the lenses are remotely close.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.