Early Black Friday Starts NOW!

Your content will be up shortly. Please allow up to 5 seconds
Gear & Apps

Should You Use Full Frame Lens on Crop Bodies? Find Out With Tony Northrup

By Justin Heyes on December 13th 2014

When starting out, most photographers choose to go with a crop body and invest in nice glass. This is a great way to start as you can keep your lenses and swap out your body for something newer down the road, eventually leading to that coveted full frame sensor. Starting out with the full frame lens and crop bodies may not provide you with the sharpest images though, according to Tony Northrup.

Should You Use Full Frame Lenses on Crop Bodies? Yes and No…

Lenses are far from perfect. Light and resolution can be lost when traveling through each element. Sometimes, with less expensive lenses, we are left with chromatic aberration, ghosting, flare, distortion and loss of resolution. Even the most expensive lens will not match the the true megapixels of the sensor.

[REWIND: Why You Should Multiply Aperture By Crop Factor When Comparing Lenses]

Full-Frame-Lens-Crop-Body-Nikon

The perceived megapixel (P-MPix) are what we are left with after the exposure. Take a Nikon D610 and pair it with the Nikon 24-70 f/2.8, the resulting resolution will be 15 P-MPix. A combination of the Nikon D3300 and Sigma 18-35 f/1.8 will result in 17 P-MPix. Most of the lenses that are tested by DXO Mark will give you the P-MPix in the spec sheet. To find out what the perceived megapixels of your camera and lens combination use the formula: MP/CF2 (Megapixels divided by Crop Factor squared).

Full-Frame-Lens-Crop-Body-Formula

In my opinion, each lens has a character and that is why we use them. If a lens is near perfect like the Zeiss Otus 85mm f/1.4, it can be boring. One of the main reasons that I use vintage glass (besides it being inexpensive), is that it provides a look that cannot be achieved by the digital specific glass of today. If you have a crop body camera, don’t let the P-MPix prevent you from taking pictures; you can still achieve excellent results from the basic kit lens. Check out our newest workshop DVD, Photography 101 to see what we’ve created using entry level DSLR and lenses.

[Via Tony Northrup Youtube / Images Screen Captures]

About

Justin Heyes wants to live in a world where we have near misses and absolute hits; great love and small disasters. Starting his career as a gaffer, he has done work for QVC and The Rachel Ray Show, but quickly fell in love with photography. When he’s not building arcade machines, you can find him at local flea markets or attending car shows.

Explore his photographic endeavors here.

Website: Justin Heyes
Instagram: @jheyesphoto

Q&A Discussions

Please or register to post a comment.

  1. Alexander Delacruz

    I am a n00b here so please be kind. After reviewing the website Tony suggested showing the sharpness on APS-C Lens I decided to go out and purchase an L Lens. All the lens listed seem to be less sharp besides the ones you gave examples too. Also, I have a 18-135 & 50-250mm and there is so much chromatic aberration in my pictures. I just bought an L Lens and so far seems to be faster & smoother. I will be purchasing an 70-200 soon. My question to you is I plan on upgrading to Full Frame no more than 2 years. Do you think I made the right decision in purchasing the Full Frame lens’s?

    | |
    • Matthew Saville

      Alexander, in a word, yes. If you plan to upgrade to full-frame in about 2 years, trust me, prices will come down and you’ll cave in and upgrade in ~1 year. It just happens that way!

      Even so, kidding aside, the real question is, do your full-frame lenses match the focal ranges you need on your crop-sensor body? If so, then you’re OK. Personally, I bought crop-sensor lenses because I needed specific zoom ranges that I just couldn’t live without for certain things. So I bought a crop-sensor ultra-wide, for example, for my landscape photography, because there were no full-frame lenses that went to 10-12mm and I really wanted that.

      | |
    • Alexander Delacruz

      @Matthew- I have been using my L Lens on my crop however I see allot of noise now. People have been mentioning the Lens is not good in low light however last weekend I had the same issue in day light fully zoomed in. Is it the lens with crop sensor, or am i doing something incorrectly? I try to leave the ISO lower than 1600 in night light and at 100 during day in sunlight.

      | |
  2. Raoni Franco

    I just came here to salute you guys for the very usefull discussion on lenses and crop vs ff. It was a breath of fresh air in this slow monday morning. Thanks!!

    | |
  3. Matthew Saville

    I’m going to go out on a limb here and say that I’ve actually completely given up recommending that new photographers even worry about full-frame, period. APS-C has arrived, and dollar-for-dollar, ounce-for-ounce, it just makes more sense as an investment for 75-90% of photographers out there today. APS-C mirrorles sensors especially are starting to see f/1.2 primes left and right, which gives more than enough DOF for 90% of photographers, plus incredible ISO results at up to 3200, which is again more than enough for 90% of photographers.

    Is full-frame always going to be superior? Sure. But that didn’t stop 35mm film from “beating” medium format film in popularity, for the exact same reasons of compactness, weight, and affordability.

    The bottom line is that people just always want more, they want the best, they never want to feel like they’ve “settled” for something sub-par. Especially when “the best” is becoming so affordable. Well, BMW’s and Mercedes’ are becoming quite affordable these days too, but I’m still not interested, and never will be.

    So, despite the fact that full-frame will always be better, APS-C will always be smaller, lighter, more affordable, and in some cases, advantageous even. I’d go so far as to argue that the situations in which APS-C is advantageous are even more common for practical photography, (focus point spread, effective reach…) than the situations in which FF is advantageous. (extreme low-light, extreme shallow DOF…)

    I know this opinion isn’t going to be very popular, but I’ve personally always enjoyed doing less with more. And probably more importantly, I enjoy not spending money on things I don’t actually need.

    =Matt=

    | |
    • adam sanford

      Matt, as always, excellent comments. I always forget about AF point *spread* — by it’s very nature, crop’s AF points will cover a lot more viewfinder real estate than FF. I often am AF point ‘location handcuffed’ when shooting small DOF work off center on my 5D3. You don’t want to half-shutter AF and recompose when your plane of focus isn’t flat and your target is close (i.e. for your vanilla rule of thirds portraiture). That relatively limited AF spread occasionally forces me to go to LV (nuts, I know) or focus and recompose but after stopping down a bit.

      I would have felt fine staying in the crop world *if Canon supported EF-S with L quality glass*, which they have never done and will never ever happen. The very best EF-S lenses are almost as *sharp* as L lenses, but generally (but not always) lack nice creature comforts like weather-sealing, internal focusing, USM, solid focus rings, and so on.

      But your point is 100% sound: if you are a Nikon or Canon shooter and working in good light, I agree that crop bodies are only a very small bit behind the FF equivalents. But in *glass*, I really think it’s night and day and FF lenses should be tried out — it’s more than just about the resolution.

      But if your chosen camera company is *principally committed to m43 or APS-C*, I would absolutely advise to stay there as those manufacturers don’t have split loyalties. If I was (say) a Fuji X shooter, I’d stay put for sure — the parent company is committed to that sensor size and lens mount, and much nicer lenses will come out for that mount than, say, for Canon EF-S or Nikon DX shooters. (Unless Sigma keeps making more wowzers APS-C dedicated art lenses like that 18-35 f/1.8.)

      | |
    • Matthew Saville

      Another good point- it does depend highly on the maker’s commitment to the format, and in that respect Canon IMO ranks dead last. (Although the EF-S 24mm f/2.8 is nice! Stay tuned for our review of that!)

      Traditionally speaking, this has always been the reason I avoided Canon- they have the biggest sense of elitism about both bodies and lenses, and while this does produce some really good stuff, it also causes tactful market placement of all “lesser” items to leave users feeling a desire to upgrade.

      In other words, they don’t go as all-out with their affordable stuff nearly as often as Nikon and other makers do, and to me this business philosophy is a turn off, even if I have the money for the best gear.

      | |
  4. John Cavan

    I didn’t watch the video, so full disclosure there, but I am reacting to the headline here. It is a simple fact that the sharpness of a lens is best at the centre and decreases towards the edges. So a FF lens on a crop body is, effectively, discarding the softest part of the image circle that lens will cast. Now, it’s entirely possible that there is a crop lens with greater corner to corner sharpness than a FF variant on the cropped body when comparing a pro-grade lens for a crop sensor versus a consumer grade lens for FF, but a don’t think I’d buy the idea that a high quality FF lens is going to be outperformed here.

    In any event, the biggest reason I would advise a newbie to buy a FF lens is that FF camera bodies are now in the consumer price range. While you may shoot crop now, it’s entirely reasonable to consider that you may shoot FF in the not-too-distant future. If you want to upgrade, you’re effectively in full system replacement mode as a consequence of buying crop lenses.

    | |
    • adam sanford

      John, I agree completely. The declining price of FF bodies says that any enthusiast — unless (a) reach, (b) high burst rate or (b) a smaller overall size is vital to them — will inevitably go to FF. That’s why if you are a professional or someone clearly hooked on photography and planning on amassing a nice array of lenses, cutting the cord from crop glass early in your lens stockpiling phase is wiser.

      In my case, after ‘having’ to buy an EF-S 10-22 (as UWA zooms + crop effectively force you to buy a specialized crop ultrawide if you need one), my next purchase back in my crop days was the original 24-70 F/2.8L. After that, everything I bought was for the FF mount, and when I switched over to FF, I only had to sell one lens. I know other folks that moved to FF many years after amassing 3-4 pricier crop lenses, and the transition was far more expensive.

      | |
    • Matthew Saville

      John, check out my previous comments regarding how a sensor can out-resolve a lens, despite being able to “throw away” the softer corners of a full-frame lens.

      For the most part, I simply wouldn’t recommend a lens that had really soft corners, regardless of its crop factor OR which sensor size it’s used on. If you’re arguing that a 70-200 is sharper on a crop sensor because it gets rid of the full-frame corners, that’s just a reason for me to NEVER buy a full frame camera!

      In reality though, the latest generation of full-frame glass is extremely sharp corner to corner, and so this crop factor benefit is a moot point IMO. These days, what is more important is whether or not the lens can deliver a sensors’ full megapixels or not. And yes, a 70-200 on a full-frame 24 MP will appear sharper, corner to corner, than a 70-200 on a crop-sensor 24 MP, IF (and only if) that 70-200’s maximum resolution begins to be out-stripped by the resolving power of that crop sensor. Which for older mk1 versions, this might be the case, but maybe not so much with newer mk2 lenses…

      | |
    • Matthew Saville

      (Sorry I meant for this reply to be a reply to the whole article, not a reply to a reply)

      | |
    • John Cavan

      Not arguing that the FF lenses are necessarily corner soft, it’s just that the crop sensor user should consider future growth of their system and that a sharp lens on a FF body is very likely to be sharp on a crop. I stopped looking at the DxO numbers ages ago, I just don’t find real world value in the information.

      As an aside, I think corner to corner sharpness obsessiveness is funny. Unless you’re selling to other photographers or serious collectors, the average consumer isn’t going to see it or care about it.

      | |
    • Holger Foysi

      @Saville: “And yes, a 70-200 on a full-frame 24 MP will appear sharper, corner to corner, than a 70-200 on a crop-sensor 24 MP, IF (and only if) that 70-200’s maximum resolution begins to be out-stripped by the resolving power of that crop sensor.” Indeed, this will always happen at smaller apertures, as resolution drops strongly (the perfect lens will drop in resolution (MTF50) to about 86lp/mm at f8, for example; real life performance is much worse and far below the lp/mm of the crop sensor). Otherwise, you will theoretically get the exact same lp/ph for an equal number of MP.

      | |
    • Matthew Saville

      @Holger, actually, it will happen at EVERY aperture, if, like I said, the lens simply doesn’t have enough resolving power to out-resolve or full-resolve both sensors’ pixel densities. If the lens’ resolving power is out-stripped by the crop-sensor’s pixel density, then you might get the same numbers on a chart, but your images in the real world will appear softer at 100%.

      To be clear, as John also hinted, …for 90% of users, this is all a moot point. 50% of users aren’t even going to ever use their lenses / sensors properly, and will never get the full potential resolving power anyways. Another 40% are going to know HOW to use proper technique, but will never actually utilize that “flawlessness” in real world printing or display.

      I just like to pixel-peep for the fun of it, I try not to take myself seriously and if I ever cross paths with someone using a horribly soft lens in the real world, I’m not going to scold them or something, in fact I couldn’t care less, 90% of the time. ;-) It’s just fun to geek out, it’s what I do in my spare time…

      | |
    • Holger Foysi

      @Saville: Of course, that’s what I wrote previously. Most lenses don’t provide the high lp/mm required and in that case the lens lp/mm is multiplied by the sensor dimension for the maximum possible line pairs per picture height, higher on FF, usually. The lens resolving power at large apertures can exceed that of the sensor (e.g. that of 16MP APSC sensors) with acceptable contrast. Zeiss and Leica themselves write that in their documents. Not every lens is capable to do this, however. A resolving power of over 100lp/mm at acceptable contrast is reported for some top glass (diffraction limit goes even to 500lp/mm for MTF50 at certain apertures). The nice thing, though is that you still benefit from higher MP even if the lens is not able to fully exploit the sensor, due to the fact that the system MTF is a product of lens and sensor MTF function (see http://www.zeiss.com/content/dam/Photography/new/pdf/en/cln_archiv/cln31_en_web_special_mtf_02.pdf).

      | |
  5. adam sanford

    Comparing crop to FF is, of course, more than just knowing how to multiply things by 1.5 or 1.6.

    With the same lens, FF will get you better low light performance, smaller DOF, and a larger viewfinder. Further, most contemporary FF lenses are simply better built lenses on a number of fronts — they can better manage chromatic aberrations, flare, distortion, etc. and often are packing much faster and more reliable AF systems.

    And, as I’ve said many times before, I utterly disregard DXO for lens testing as their lens metrics are heavily weighted by sensor resolution — they will call a lens a winner on a high MP body and call it a dud on a low MP body. I much more prefer lens resolution testing that uses a standard test body and allows the only meaningful comparison you can hope for — comparing how two different lenses perform on the same body. The minute you throw an additional variable into the mix — FF vs. Crop, Canon vs. Nikon, etc. — the truths become murkier to interpret.

    I strongly recommend folks try out FF glass on crop bodies. They generally do a better job than kit crop class.

    | |
    • Matthew Saville

      Adam I think one of the things that you might not have picked up from this video is the resoliving power of a lens, and how it can be “out-stripped” by a high-megapixel crop sensor.

      It seemed like just a lot of numbers, but what Tony was trying to prove is actually quite easy to understand / explain in real world terms: Some lenses, whether crop sensor or full-frame, are out-resolved by some sensors. Take the old Canon 24-70 for example, it’s actually a very soft lens. This wasn’t very apparent back when Canon only had 11-12 megapixel full-frame sensors, because that “resolving power” is quite forgiving, in fact it’s about as forgiving as 35mm film was. However a 20 megapixel crop sensor is simply magnifying that lens’ glass by a whole lot more.

      Since this is true for all lenses though, the “use crop sensor lenses on crop sensor bodies” argument can’t be taken for granted. You do need to read reviews of lenses just like you always should. If a lens is soft on this or that camera, then it’s soft, plain and simple.

      There are, however, plenty of dedicated crop-sensor lenses that are as sharp or sharper than an equivalent full-frame lens. Take a Canon 16-35mm f/2.8 at 16mm on yoru 7D mk2, and compare it against the Rokinon 16mm f/2. The Rokinon will utterly destroy the Canon, despite the Canon having its corners cropped out!

      That’s only one example, but I hope you get my point- As Stan said above, discerning photographers should simply pick the lens that is right for the job. And if that lens is a crop-sensor lens, they shouldn’t be afraid to invest in it…

      | |
    • Holger Foysi

      @ M.S: Agree. Crop sensors have much higher lp/mm (m43 16MP: 136 vs. FF24MP: 84lp/mm). You need to use lenses being able to provide this resolution to be able to get the most out of the sensor. Additionally, this strongly depends on aperture, as with increasing aperture the theoretical resolution drops (see the luminous landscape article on sensor and lens resolution). Due to the larger sensor size, a mediocre lens with given lower 60 lp/mm, for example, will produce 24*60lp/mm compared to 15.6*60lp/mm in terms of lp/ph, higher on FF sensors.

      | |
  6. Stan Rogers

    You know, it really doesn’t matter a whit whether the Sigma 18-35mm/1.8 on a crop sensor body has more effective megapixels than a Nikon 24-70mm/2.8 on full frame; they’re not covering equivalent angle of view ranges, and it’s really hard to zoom the Sigma to either 16mm or 47mm to try to make the orange into an apple so that you can compare them. It’s even harder to zoom that 18-35 all the way to 70mm (105 equivalent on a crop body — unless you’re a Canonista, in which case it’s a trifle longer). Like, you know, an effective portrait/headshot length. I know that Tony’s going to show up here trying to justify this blather to the masses (he tends to do that whenever anybody, anywhere, calls him on his stuff), but this is the silliest nonsense I’ve run across in quite a while. Crop sensor users who are paying attention, like everybody else, buy the lenses they need to do the work they’re doing, and a fast-ish35-ish to 105-ish lens is a spectacularly versatile lens for people-shooting (much better in many ways than the same lens on a full-frame body). The tiny handful of dedicated crop sensor lenses of exceptional quality may be well worth having, but they are incredibly few, far between, and try (for some unknown reason) to cover equivalent focal length ranges from the 35mm format, which are largely a happenstance of legacy (the 35-70 that grew wider as optics improved, and the awful 70-210 that was first contracted to 80-200 and opened up to f/2.8 by Minolta, then Canon, then re-expanded down to 70mm at the wide end). It’s not like any committee of experts sat down and decided that these were the optimum zoom ranges; they just happened — and happen to work better in a lot of ways on crop bodies.

    | |
    • Jim Johnson

      Yeah, I agree completely.

      Personally I find Tony Northrup’s videos a lot like political punditry— it’s a lot jargon and noise thrown out at people that does nothing to clarify and everything to justify one’s own viewpoint. To me they are worse than useless, they actually confuse most people.

      | |
    • Matthew Saville

      Stan, thanks for already including your own counter-argument that I was going to say in response to your other points: “Crop sensor users who are paying attention, like everybody else, buy the lenses they need to do the work they’re doing…”

      In my opinion, there are more than just one or two crop-sensor lenses that really kick ass, and depending on what you shoot a 18-35 1.8 on a crop sensor might be much nicer to have than a 28-50 2.8 on full-frame. Or, for those who don’t care about DOF as much, such as landscape photographers, a Sigma 50-150 2.8 on a D5300 or D3300 is significantly lighter, smaller, and more affordable than a 70-200 2.8 on full-frame. Or lastly, two of my favorite examples are the Tokina 11-16 2.8 and 12-28 f/4, both of which are so small, lightweight, affordable, and yet as sharp or sharper than any Canon equivalent, and right up there with any Nikon equivalent.

      All in all, in my opinion full-frame is simply no longer even necessary for the majority of photographers, unless you’re utterly obsessed with extremely high ISO, (beyond 3200) …or absurdly shallow depth. (f/1.2 on full-frame) Which, let’s be honest, is 90% of photographers. Even though 50-75% probably think they NEED full-frame, that’s mainly just gear lust.

      | |
    • Stan Rogers

      You’re fer sure right about the bokeh addiction thing, Matt — as a pro, I only ventured more open than f/5.6 occasionally, and getting *enough* DoF was more often a struggle for most of what I did. I find it kind of funny, actually, that the problem then was getting small apertures with 25-speed film, and these days it’s opening up to f/1.nothing on cameras where ISO 3200 is just another setting. As for the focal lengths, well, I just don’t see wide and have found very few wide-angle photos compelling, so I’d naturally gravitate to something a little longer. (I’ve owned many WA lenses, including fisheyes, PC types and the wonderful Minolta 24mm VFC, over the years, and they’ve earned their keep — but just barely, and apart from the VFC have never made a single portfolio picture, just client stuff.) Heck, my 35mm walkin’-around lens was a 90mm Tamron. So the FF zoom ranges would be a much better fit *for me*. And it’s the *for me* part that matters more than anything else in this game: the tech is the same tech for everybody (physics doesn’t do opinion) but the art has to be your own.

      | |
    • Holger Foysi

      @S. Rogers: “You’re fer sure right about the bokeh addiction thing, Matt — as a pro, I only ventured more open than f/5.6 occasionally, and getting *enough* DoF was more often a struggle”. All this boils down to personal preference and shooting situation. My wife and I like shallow DOF, as do my wife’s clients. I like FF because we are very versatile, being able to use fast primes and zooms for shallow DOF, as well as lighter lenses and zooms if this isn’t a priority. Additionally, we know to be able to use high ISO if needed. If you shoot several formats, it is very nice to know about equivalence to choose the right lenses for the job. After some time it’s mostly experience, but why not doing it right from the beginning and understand what is going on?

      | |
    • Stan Rogers

      @Holger Foysi: Experience is the only guide. That said, shooting people in the wild is only a small segment of this thing we call “photography” (it’s kind of silly to have just one word to cover such a vast set of enterprises, really), and even then “environmental portraiture” that completely blurs away the environment might as well have been done in front of a muslin in a studio. And yes, sometimes that’s the point. But as is the case with gigantic softboxes, extremely shallow DoF and big bokeh is often the photographer taking the easy way out of a challenging situation; there are times when it is the right solution, and times when it’s just a cheap way of getting a decent picture when a *spectacular* picture is sitting six inches of subject or photographer movement away (and a stop and a half down). I have no problem at all with shallow DoF when it’s used appropriately; I just don’t like it when it’s used *instead* of good photography.

      Now, I was talking mostly about studio photography (product and people), landscape and architecture up above. Landscape and architecture require sufficient DoF to work at all, regardless of the era. In studio, you need light to support whatever you’re doing, and 19,200 joules (watt-seconds) of flash doesn’t go as far as you’d think it would when you’re shooting something that needs to be sharp front-to-back at an ISO of 20-25. (And enough tungsten or HMI to get the job done is both blindingly bright and HOT; we didn’t have full-spectrum fluorescent or LED as an option back then.)

      | |
  7. Brandon Dewey

    This video has a lot of great information for those who are using crop sensor bodies. For years before I upgraded to a full frame camera I didn’t know I need to also multiply the aperture by the crop factor.

    | |
    • adam sanford

      Yep, but that’s only for the purposes of DOF. So you aren’t getting as much of your DOF’s money worth with the aperture listed on the lens if you put it on a crop body. Bokeh fanatics should go FF to get the most out of their f/1.4 primes.

      But an f/1.4 lens on a crop body *does* allow in the appropriate amount of light for (say) exposure considerations. So if you are buying a lens for its ‘speed’ (i.e. quicker shutters as more light can be gathered), you are still getting your money’s worth on crop.

      | |
    • Matthew Saville

      Good points, Adam, and thanks for bringing up this point of clarification.

      A lot of people talk about a full-frame sensors’ “total light-gathering capability” as if it magically gathers more light that creates “brighter images” than a crop sensor. It doesn’t. A full-frame sensor has bigger pixels, so it does gather more photons, but the exposure, from a metering perspective, is still going to be the same.

      So if you want to gauge DOF, multiple by the crop factor. If you want to gauge angle of view, multiple by the crop factor. However for the most part, because sensor technology is getting so good these days, a whole 1.5x stops is not necessarily the exact difference in high ISO performance, especially at lower ISOs.

      =Matt=

      | |
    • adam sanford

      Completely agree, Matthew. *If you have good light*, forget about FF unless you are doing razor thin DOF work with large aperture glass — crop and FF will basically give you the same IQ otherwise.

      | |
    • Matthew Saville

      I would caveat, Adam, “if you have good light, AND if you need to hand-hold or freeze motion…” Because for some people, who shoot on a tripod slow-n-steady, “good light” happens any time of day, and high ISOs become far less necessary…

      =Matt=

      | |