Should you use Full Frame Lenses on Crop Bodies? Yes and No...
HTML-код
- Опубликовано: 6 сен 2024
- SUBSCRIBE and like Northrup...
Buy the #1 book with 14+ HOURS of video on Amazon: help.tc/s
Worldwide use 10% off coupon 'RUclips': sdp.io/sdpbook
Lightroom video book $10 on Amazon: help.tc/l
Photoshop video book $10 on Amazon: help.tc/p
Photography Buying Guide on Amazon: help.tc/b
Worldwide use 10% off coupon 'RUclips': sdp.io/buybg
STARTER CAMERAS:
Basic Starter Camera ($280 used at Amazon): Canon T3 help.tc/t3
Better Starter Camera ($500 at Amazon): Nikon D5300 help.tc/d5300
Better Travel Camera ($500 at Amazon): Olympus OM-D E-M10 Mark II help.tc/em10ii
LANDSCAPE CAMERAS:
Good ($550 at Amazon): Sony a6000 help.tc/a6000
Better ($1,400) at Amazon: Nikon D5500 help.tc/D5500 & Sigma 18-35mm f/1.8 help.tc/s35
Best ($3,150) at Amazon: Pentax K-1 help.tc/K1 & Pentax 24-70 f/2.8 help.tc/p24
PORTRAIT CAMERAS:
Beginner ($950 at Amazon): Canon T6i help.tc/t6i & Canon 50mm f/1.8 help.tc/c50
Better ($3,000 at Amazon): Nikon D610 help.tc/d610 & Tamron 70-200 f/2.8 help.tc/t200
Best ($5,300) at Amazon: Nikon D810 help.tc/d810 & Nikon 70-200 f/2.8E help.tc/n200e
WILDLIFE CAMERAS:
Starter ($1,100 at Amazon): Canon 7D help.tc/7D & Canon 400mm f/5.6 help.tc/c400
Great ($3,200 at Amazon): Nikon D500 help.tc/d500 & Nikon 200-500 f/5.6 help.tc/n500
VIDEO CAMERAS:
Beginner ($500 at Amazon): Panasonic G7 help.tc/g7 & Panasonic 14-42mm help.tc/p42
Better ($1,400 at Amazon): Panasonic GH4 amzn.to/2p5dAmD & Panasonic 14-140 f/3.5-5.6 help.tc/p140
Best ($4,300 at Amazon): Panasonic GH5 help.tc/gh5 & Metabones Speed Booster XL help.tc/mbxl & Sigma 18-35 f/1.8 help.tc/s35 & Sigma 50-100 f/1.8 help.tc/s100
DRONES:
Beginner ($400 at Amazon): DJI Phantom 3 help.tc/p3
Travel ($1,000 at Amazon): DJI Mavic Pro help.tc/Mavic
Better Image Quality ($1,500 at Amazon): DJI Phantom 4 Pro help.tc/p4p
![Run Away - Tzuyu(TWICE) ツウィ 쯔위 [Music Bank] | KBS WORLD TV 240906](http://i.ytimg.com/vi/KnHNNyAYiu0/mqdefault.jpg)
📚 Buy Our Books on Amazon! 📚
📕Stunning Digital Photography: help.tc/s
📘Lightroom 6 Book: help.tc/l
📙Photoshop Book: help.tc/p
📗Buying Guide: help.tc/b
I have been having a hard time figuring this out. My camera (Sony Fs5) has a super 35mm. Would you suggest getting lenses for apsc/ super35 or full frame? This is in regards to the Rokinon cine vs veydra cine lenses.
Tony Northrup I'd get a good full frame body and a relatively cheaper APS-C body, with full frame lenses if I'm doing wildlife.
My idea for a Canon wildlife setup:
Canon 6D mark II, and Canon T7i, and EF 100-400mm f/4.5-5.6L II.
Hi Tony, So if one is using an full frame lens on a crop sensor body, but metering with an incident light meter, what effect might one expect in the accuracy of the exposure obtained by the meter?
Thanks
Tony & Chelsea Northrup
If I use a Ef-s 2.8 (crop lens on a crop body) will the bokeh be the same quality of a Full frame lens on a full frame camera?
Or can u expect the quality of the bokeh to be the same as a 5.04 phot on a full frame?
What would give me the best bokeh for my apsc sensor if I want to take great background blown out portraits?
Alright got the answer from one of your other videos thank u so much for your content.
The xfactor applies to both crop and full frame lenses.
Took me two full days to find the answer and clarify this lmao and it was on a 15 min vid on RUclips. You guys Rock.
Dear Tony, there are no words big enough to express how grateful I am for all the knowledge you share over here! You really saved me from many failures I could have. Also, the way of your explaining is so charming! Bravo for you and Chelsea!
Thank you so much! I'm so glad we could help!
Yes, one of the advantages is you can use.any lens with your Nikon crop camera.There are some good DX lenses, like the 17-55 f2.8 or Tokina 11-16, but for the most part an FX lens works wonderfully. Nikon , unlike Fujifilm, doesn’t have a huge selection of high end DX lenses. You can use FX on your crop camera. “But it render as good pictures”, “You buying high end equipment for your crop camera?” Yep, and the pictures look great. 😎
Seems to be a lot of confusion in the comments.
While the DOF may appear to be at a smaller aperture, the amount of light is not changed.
f/stops are affected, but t/stops are not.
I wish Tony would stop with his multiplying aperture advice and simply explain it the way you do.
I can't comment on the T stop thing but the exposure IS affected, why? ISO
Aperture = focal length/iris diameter (the openning)
And since a crop sensor can't collect the whole light that enters through a FF lens Iris, An image taken on a full frame body with a given lens at given exposure settings, is ONE STOP (or more) brighter than the same image taken on the same lens with the same settings on a crop sensor.
But why do they still look even in brightness? ISO 100 on APSC = ISO 200 FF equivalent. The explanation of the noise difference.
+THEKMT6 That's incorrect. Exposure is not effected by a crop sensor for the same reason that exposure is not effected when you crop an image in post. It doesn't matter that part of your image has been trimmed off or not because the intensity of the light over any given portion of the sensor stays the same.
+NightHawkInLight I agree with you, sensor size doesn't change the final exposure, and yes the light intensity is the same, but if you hypothetically shrunk a FF to an APSC it will receive a more intense light for each mm2
Only if you also narrow the focus of the lens accordingly to match the shrunk sensor, otherwise it would be no different in exposure. That's how a speed booster works
Thank you Tony for all your hard work. I have learned more from you in 3 weeks watching and applying your videos than from all my classes/courses etc. I love that you are not trying to sell me anything and just report the facts. This is very rare. Thanks Again.
Janet Crossland Thanks, Janet! It's my pleasure.
Finally clear advice! I need to buy a 5D MkIII now.
You are a great professional in your field. I do not know other photographers who know a lot of material about photography and equipment. Bravo!
A f2.8 aperture lens on a full frame sensor is the same f2.8 on a crop sensor it doesn't get darker and doesn't change the depth of field. Your correct in saying that it changes the area of view.
James Melito he’s proven it mani times on camera
@@sebastianmorales6849 if you'd measure the width of stuff in focus on sensor, using millimetres, it will be exactly the same for a crop lens and for a FF lens
What worked very well for me was buying fast full-frame primes (14mm, 28mm, 50mm) and native APS-C zooms for my old D50. When I moved up to the full-frame D750, and then the D850, the wide primes were still great to have and I concentrated on buying full-frame trinity zooms and longer primes like the 85mm and 105mm..
It would be nice it the P-Mpix stuff was an open measurement standard, rather than a DXO proprietary secret.
Everyone who is into photography Must watch this first. I am from the film days and never really understood until today. I just ordered the Canon 90D and went back in the middle of Tonys presentation to make sure the lenses I purchased were in fact for ONLY APS-C sensors, they are.Thanks again Northrups.
The one thing I don't understand is how an apsc specific lens can't stay true to its focal length or aperture. Why do they even market it as a 18-35mm when its not going to give you that... You know what I mean?
The focal length for example is a fixed measurement. It just behaves differently on different sensor sizes. Which is why the crop factor is used to transfer it into a 35 mm equivalent. You always have to keep in mind that it's just an equivalent to ease comparison between aps-c and ff sensors. I think the practical use for this 35 mm equivalent is getting less and less significant. If you know how a 35 mm lens 'behaves' on your aps-c sensor... you're good to go.
I don't think that's what he was saying, but that a lens designed for a FF sensor will yield a different field of view, related to the crop factor. A lens designed specifically for APSC should yield results true to its focal length. I'm not totally sold on the aperture crop argument but perhaps it's because I just don't know enough.
This video has been prepared and presented with sheer geometric precision. Well done.
Another very useful and practical video for people. Cheers Tony
+Philip Bloom Have you ever notice how sexy your name is?
this video cleared almost all my doubts of using full-frame glasses on APS-C body.... and the way you explain that information, that's awesome. I wish if I had a professor the same as you in ma university.....
Mr. Northrup, perhaps revisit this video with a follow up on using full frame macro lenses (vast majority of them are) on crop bodies...
I’ll like to know too
Great explanation Tony. You have a marvelous way of elucidating complicated scenarios in a simple manner without talking down to us neophytes.
When I bought my 60D years ago, I only bought FF lenses for it and still to this day I do not own a single crop lens... That's because I knew I'll definitely move to full-frame later and I was glad when I managed to survive till the 5D Mark III came out!
That's what I personally did and that's obviously only because I knew I was going to upgrade anyway. And to be honest, the quality didn't bother me in the slightest and doesn't to this day. I still own the 60D, all the lenses work just fine on it and I'm pretty sure my 70-200 F/2.8L IS II is just as sharp on the 60D as it is on the 5D3, which is awesome when I need that tighter zoom.
I assume it's a safe bet to think that at least a few other people also do what I did because they know they'll move to FF sooner or later too and don't want the fuss of selling useless lenses later once the body is upgraded..?
That's exactly my point. I also use a FF lens on a crop sensor and it's way better than my kit lens in terms of sharpness, plus it's an wide angle lens. I will eventually move to a full frame body so there's no reason to buy crop lens and sell it and get back a new one. Whereas I can only sell the body and keep all my lenses.
@Eternal Peace No problem. Glad I could help. Have fun taking pictures.
Thank you Tony and Chelsea so much. I'm learning so much from your videos. A beginner with a Canon 5 D Mk iii and a good selection of L Glass. I took your advice having never
purchased second hand equipment and purchased a Canon 7D Mk ii camera and a Canon 100-400mm L Mk ii lens. When funds allow my next purchases are the grip as I have big hands and the 1.4 x telephoto extention. And of course make time to practice the art as often as is possible. Kind regards Brian in England.
By this logic, the Eos 5DSR with 70-200/2.8 vs Eos 80D with 70-200/2.8 would be the same perceived megapixels then. They both have the same pixel pitch. The light hitting the 5DSR sensor is the same hitting the APS sensor, you're just cutting off the sides. If you scaled up the 24MP APS-C to full frame it would be a 50MP sensor. If all these L lenses suck on APS-C, then they will all suck on the 50MP 5DSR.
1965ohio the crop sensor still has more pixels per inch, therefore with that denser pixel count the flaws in a lens become more obvious imagine if a 5ds mark 2 had a 61.44 mp sensor. The maths I did was 24mp x 1.6^2 the square accounts for the area of the sensor as they're two dimensional. The 5ds proved some older canon glass couldn't keep up a higher megapixel camera will do the same again
+Joe Addison "the crop sensor still has more pixels per inch" eeehhhh, by a little bit, doesn't make much difference between the 5DSR and the 7DII.
"If all these L lenses suck on APS-C, then they will all suck on the 50MP 5DSR." Nope. Hold your horses for a moment. They will look very similar in a 100% magnification viewing. But as the later is a crop and we need to scale both to a common display size to make the comparison. Think about it, it's like a literal crop, so take one image with a 5DSR, reframe so that when you take a second one, the crop of it has the same framing as the first image. Stretch out the later to fill the same area (or downsize the larger at your preference)... which do you expect is better?
Yes, cutting off the sides changes the apparent noise for the overall final image. The total light is the important determiner of photon shot noise, not the light/area (though indeed the light/area is the same, but not really in the final image once they are resized to a common viewing size, at which point light/area is more for the larger sensor).
Your comment makes no sense. They would produce same resolution if you cropped the 5DSR image. If you keep it as it is you get over twice the number of pixels in effective resolution. You do not seem to get the concept of the final image. The idea is not to view pixels in 1:1 size.
Nobody says all L lenses suck on crop. The issue is that in some cases one might not get what one expects.
in many many years I did not see something like your video, so informative so technical, I enjoyed every second, I think it is the golden gate to understand what is all about the endless debate of buying Full or Crop frame, in my vocabulary I know no better word rather than Thank You So Much
Sami Elias You're welcome, and thanks!
So in conclusion, if you HAVE TO CROP ANYWAYS in such situations like wildlife, portrait, or etc, it could be beneficial to shoot with CROP BODY with FULL FRAME LENS. If you DON'T NEED TO CROP, then you should shoot FULL FRAME body with FULL FRAME lens.
This is very true and informative. I went this route years ago but I don't regret it. Low light on a budget was my primary goal. Saved a lot of money. Back in the 10mpx days sharpness wasn't as big of an issue as it is now. All photographs have a sharpness relative to the detail they provide.
You only explain the mathematical figure. Show us actual image comparison with the varieties of lens and body combinations
With the exception of sport photography and night photography the purchase of a lens is more important than a camera body. The lenses and especially manual prime ones have very slow depreciation. Considerable technological advancements in lenses happen every ten years. In Cameras happen every three years.
Zoom autofocus lenses are updated more frequently than prime autofocus lenses and prime autofocus lenses are updated more frequently than manual prime lenses.
I recommend in case you are not lazy with autofocus and you don't prefer sport or wild life photography the purchase of a bright manual prime lens. Manual prime lenses are not as complex as zoom lenses and that makes them more reliable. Also the lack of electronics or the minimum use of electronics makes them preferable for buys from eBay.
Tony; as a physical scientist (Ph.D. in Chemistry (Physical Biochemistry), a photographer and having done Digital Multimedia Forensics, I must say that I deeply appreciate your (and Chelsea's) videos. I especially like videos like this one when you go full Geek Mode and dive into the science. I get frustrated with arm-wavers who bloviate opinions that are wholly subjective (and often wrong). This is not to say that a significant fraction of Photography is not subjective itself, but you know what I mean. And, although I have come across this video some time after you posted it, I do appreciate your tackling a topic that few have even considered; however slightly. I have seen a number of your videos so far and find them useful, well constructed, enjoyably delivered and mostly timely and relevant. I also really liked some of your drone videos; especially the one for preparing for the FAA exam. So, Tony (and Chelsea), keep up the good work! And, although I don't ALWAYS agree with every opinion, you are solid in your research and thought processes and seem quite diligent. So, thanks again!
Thanks!
Very good. I knew about crop factor for adjusting focal length but not for aperture. Nice to have it clearly explained.
I still think it`s a good idea to get the FF glass and not bother with DX lenses at all. Most photographers who start up and are really into it, plan to get a FF camera anyway at some point. You`ll get plenty of detail from the crop lenses anyway while you`re learning but then when you get a FF camera you got all this nice glass waiting for you :). Never noticed any loss in sharpness due to using a crop camera (d7200) even tho it`s really crazy with pixel density. The images i get (if they are in focus) are tack sharp. I asume they will be even better when i get my hands on a FF camera, but in real world use it seems fine to me. Perhaps it`s cause i use prime lenses? In any case i think that if you know that you will be switching to FF at some point it`s better to get FF glass. Just less wasteful. Although to be honnest...the sharpness i`m getting from d7200 and sigma 35 1.4 really kinda makes me not want to go to FF since i do a lot of macro. Also if you go by this logic...we should all use medium format!
we
+EvulDali i heard that using full frame lenses on crop sensor cameras can damage your sensor, is that true?
BigLou7711 No. That is not true.
+BigLou7711 No, using crop lenses on FF bodies can damage them*
+oliver24x No that is not true either. DX glass on FX body is harmless, there's just a few strange and fun image effects. *
Tony, I just watched this video and like your friend, I brought my Canon 7D and Canon 24-70 back to Canon and they never mentioned the reason for lack of sharpness. On staff member suggested that I need new classes. I will try out APS-C lens like the Canon EF-s 18-85.
You switched sides in the middle and kinda said exactly opposite things. First, you counted megapixels by the amount of light the Canon 7D APS-C sensor is capturing from a full-frame lense as fewer pixels. Then, you said because Canon 7DII APS-C sensor has more pixels than 5DIII in crop-body mode with the same lense. Sigh.
Dude... Seriously.. I was watching and then was like... Wait WTF?! Didn't he just say the exact opposite of this?! 🤔
I think Tony still believes that light exposure is converted for APS-C along with depth of field. That isn't the case, light per square mm is identical, and even if he's trying to use pixel density as some sort of measure of low-light performance, that went out the window with BSI sensors like the one on the D850.
Yeah, I like this guy but he gets several things wrong as can be expected by anyone who makes this many videos. Most people can figure out where he goes wrong. It's when he talks to the complete beginner when he probably is giving some costly advice. I'd suggest people don't rush into spending a bunch of money on equipment until they get some experience.
It’s because he is cropping away the extra pixels at the edges
@@gutenbird When I first got my camera(Canon T6i) a friend let me use a really expensive lens. He is a Pro photographer. The pics I was able to take with that lens on my cheap crop body were astonishing. The background was completely blown out, and my subject was tack sharp. And I have not been able to match that level of sharpness since with my kit lenses(18-55, 75-300). Tony says I can get sharper photos from the 18-55 kit lens, than I could with an expensive lens on my APS-C. I'm not sure why he would say that, but it's patently false. I have the images to prove it.
Not sure about all of that math he does, but I'm not sure it applies to the light the same way it does to the focal length. But better glass is just that... better. No matter what size sensor you have.
Hi Tony. I have specifically bought Canons EF (FF) lenses for that exact reason. That when I switch to a full frame body, I only need to discard my EF-S kit lens. I think that also is a consideration.
I think there is a lot of confusion about focal length and aperture on APS-C sensor. 50mm lens is 50 mm lens no matter what sensor size is. f1.4 is f1.4 - no matter what the sensor size is. 50mm lens on APS-C sensor will give you the angle of view equivalent to 80mm on the full body, but it will not be 85mm lens, it will not give you the compression or magnification of 85mm lens.
The aperture is focal length divided by the diameter of the hole in the lens. That doesn't change just because you put it on a camera with diff size sensor. When you multiply f stop by 1.6 it's only because in order for you to get the same angle of view that is on full frame body, you have to stay farther away from the subject. That's it. its still lets in the amount of light that f2.8 lens does. you can verify it by playing with depth of field calculator and compare to full frame body, you will have to adjust the distance to the subject to get the same result.
I checked the DXO marks for full frame lenses on canon and while its true that you loose a lot on Canon 7D, it looks much better when you compare to Canon 70D. If you compare Canon 24-70 f2.8 ii, on canon 70D and 5Dmark ii, the sharpness is much closer between those two cameras than between 5D mark iii and 70D. In fact it seems every time they release a new APS-C sensor the DXO marks for that lens will be better on that sensor than on previous. Wish they had Canon 7D mark ii there for comparison.
Yuliya Brown f/2.8 lets in the same light per square mm regardless of the sensor size, but it doesn't let in the same total light. Total light is what determines total image noise and depth of field. Check sdp.io/crop for more info (as well as the following videos).
Tony Northrup I watched your videos. I got really confused, so I went and read up on all this at diff place. I found articles on this site (www.clarkvision.com/articles/does.pixel.size.matter/index.html) helpful to understand. I am not sure that I will agree with you on the total amount of light. The amount of light that lens lets in depends on the size of the hole. The amount of light the sensor can capture, depends on the sensor. so I think the lens lets in the same amount of light because the hole doesnt change, but the sensor with smaller pixes doesn't capture it all. So I came to conclusion that on the cropped sensor your best bet is to have as faster lens as possible. I am not sure how useful f1.4 on the full body, I think there will be hardly anything in focus, but on the cropped sensor you let in more light and hopefully will reduce the noise and your depth of field will be actually a little more useful than on a full frame. I am still trying to figure it all out. But thats why I am not crazy about sigma 18-35 f1.8 . I think the zoom range is not very useful for portraits, and you dont shoot landscapes at f1.8. so unless you constantly shoot large group of people in low light, I am not sure its worth all that excitement:-) I must be missing something, I am just way too new to all this:-)
Yuliya Brown Provided the lenses are optically perfect in terms of distortion, you can take the EXACT same image with a 14mm lens as with a 200mm lens, all you have to do is stand in the exact same spot, take the pictures and then heavily crop the 14mm image to match the 200mm image. I guarantee the exact same compression minus the depth of field due to the physical size of the aperture (matching DoF is when you need to move). So yes, a 50 can produce the exact same compression as an 85...
Moving on... The f/1.4 aperture is useful on a full frame, low light, for example, or those artsy-fartsy bokeh-licious shots of food and things and YES, you most definitely can shoot landscapes at that f-stop. When things are far away (and obviously if you focus on those things), such as a mountain range, your dof gets REALLY long. If you don't have much going on in the foreground - there is no point in going at f/22 at all. I'm not saying f/1.4 will be sharp, I'd obviously go at f/4-f/8, but that's only for sharpness. Your mountain range would still be in focus.
Another thing not to confuse over would be the physical size of 'the hole', the aperture. Although long lenses have massive apertures and wide lenses have tiny ones - the light gather is the same at the same f/stop. That's because the wider the angle of view - the more light can hit the front element, the more focused the beam of light is because of those bulbous elements. That's why we have this standard and that's why it works.
Optics is an amazing little bugger that confuses a hell of a load of people, but it's not THAT hard to understand if you put your mind to it and TEST... :)
If Tony Northrup has anything to add to or correct me with - feel free to do so!
I can see what you are saying Yuliya, but here it is as simply as I can put it. Two cameras, both 20 MP, one a full frame and one a crop. You put the same 100mm prime lens on both cameras and take a picture of a rectangle painted on a wall, both shots are taken from identical distance. The distance chosen is the distance at which the rectangle perfectly fills the frame on the crop sensor camera, we will call this distance A. On the full frame camera the field of view is wider so the rectangle would not fill the frame, even though the shots are taken from exactly the same distance. On the crop sensor camera you now have a picture of the rectangle at a resolution of 20MP. To fill the frame and get a 20MP image of the painted rectangle on the full frame camera you have to either a) move closer to the rectangle or b) stay where you are and put a 150mm lens on it. So, to sum up, yes, the lens used was a 100mm focal length lens and the lens itself stays 100mm focal length regardless of the camera it's on, however the image it produces on the crop sensor is EQUIVALENT to the image produced by a 150mm lens on a full frame, or, to get the same 20MP image of the rectangle on the full frame camera, you would need to use a 150mm focal length lens when taking the shot from distance A. Hope this helps : )
This is the biggest misinformation on the internet about focal lengths, dof, and all that. Yuliya you are 100% right and i have failed every time at getting people to understand this. Using an apsc sensor with say a 50mm lens is comparatively equivalent to a 1.6x magnification of the fov seen from a full frame 50mm shot. Optically nothing has changed, the vessel to capture the photo is still a 50mm lens.
Ill make an analogy, lets say you have some bears to feed. You buy Kanon FF bear food which is made for Kanon FF bears which all weigh 160lbs. It says 'feed bear 50lbs daily'. Thats 32% of the bears body weight for a healthy bear. So then you have another bear you want to feed but its a smaller Kanon CF bear that only weighs 100lbs. You transfer over the same food but the bear gets much fatter, thats because what was 32% of the larger bears body weight is actually 50% of the smaller bears body weight. Then you say ' wow cool, how do i get the big bear to match the little bears fattness? So you then go back and calculate 50% of the Kanon FF bears body weight is equal to 80 lbs of food. Then you can make all kinds of converse associations including ' 50lbs of food on a smaller CF bear, is equivalent to 80lbs of food for the larger FF bear' OR 'if i like the figure of the FF bear when i feed him 50lbs a day, i can feed the CF bear about 31lbs of food a day to acheive the same result.'
So YES 50mm on an apsc is equivalent to an 80mm on a full frame but, they are not the same bear! The bears are equally fat or fit but not the same!!!!
i learned something today i used to multiply by 1,6 to know the focal lenght on my crop sensor (Canon 80D) but i never knew i also had to multiply the aperture by 1,6. Now i even more want a full frame sensor. Thks for the video.
I think you are completely missing the point of why people with a DX body might buy a 24-70 f2.8. It would probably be much more useful to people if you explained that this lens on a DX body:
1) Has a focal length equiv of e.g. 36-105
2) ACTS LIKE a minimum aperture e.g. f4.2 lens when it comes to depth of field
3) .. but still has exactly the same f2.8 low light capability. f2.8 is f2.8 is f2.8.
The reason to buy FX lenses is because that's where the better lenses are.. e.g. constant aperture 2.8. You buy constant aperture 2.8 lenses in large part to be able to deal with low light. You absolutely get these benefits with a DX body.
??
Great staff is that you also cut the corners (it's like zoomed already 1.6x) when you mount fx lens on a crop body. But loosing dept of field.. Crops are good nowadays and tele lens are very expensive so someone can benefit from it.. In other hand wide lenses are also pricy so anyone shooting wide should stick to full frame. Cheers
YOU ARE A DOCTOR IN PHOTOGRAPHY. YOUR VIDEOS HAVE HELPED ME ALOT. IM FROM BELGIUM AND YOU HAVE MADE ME A BETTER PHOTOGRAPHER. THANKS alot for all your videos
DXOMARK replied me that 15-20% difference in their PMpix results or in my case a 2 PMpix extra makes no any visually noticable difference in image quality! See yourself their data on Nikon 40mm macro AF lens on D3300 vs D5300 vs D7100. the differencies are huge in numbers (same sensors...) but they say it is non significant, means nothing in actual sharpness. So I do have doubts using dxomark results. better to double check what you buy.
That sounds funny! Dslrs upgraded from 8 to 10 megapixels or from 18 to 20 etc and DXO told you 2megapixels makes no difference...
I doubt about their seriousness now!
Who else could explain something so subtle so clearly. Thanks Tony!
Sure thing!
So what would be a APS-C lens equivalent of a 70-200 f/2.8L?
Depends on the camera brand. Sony, Canon, Nikon? Which one?
Fugi 50-140 f2.8, really it'd have to be f 2 but that doesn't exist, so it's this lens
I think if Tony was stranded on a deserted island he could figure out a way to build a camera with bamboo, sand, and palm leaves. By far my favorite youtuber to learn about the science of photography. I think it's time I buy his books.
So, basically.... I could sell my 70d, pick up a 6d body and my three primes would instantly be significantly sharper. But, I'd be missing my touchscreen dearly.
Samtagri Man, I use that thing all the time. Some things, like wifi set up and quickly focusing on desired points, are night and day easier with the touch screen. Want to make sure the eyes are in focus... simply touch them and the camera focuses there and takes the shot.... too easy!
Samtagri Exactly.... I post a lot of stuff on the fly. If the shot needs little editing, I can edit in Snapseed or VSCO on my iPad and post from anywhere. That, and using my iPhone as a shutter release is legit.
singletrack29349 Yeah, where are our full frames with touch screens?! People always mock me for wanting touch screens, but that's cuz they aren't using them regularly like we do...
Tony Northrup I was actually really disappointed that the 6D didn't have one.
Samtagri The 7D doesn't have wifi cause the signal could not get through the weathersealed alloy body. Same reason all pro models don't have it. And it IS gimmicky in a pro camera. I'll give you the touchscreen, but wifi is for phones and instagram.
Hi Tony interesting video but surely what you are missing out in your comparison is the quality of the glass in a full frame/pro lens compared to a plastic lens ( In some cases ) in the apsc / cheaper glass and also for wildlife and sport the AF motors are quicker so will track a moving subject better. Its alright the apsc glass being sharper but if the AF cannot keep up your subject will be out of focus.
True or false: f/2.8 lens behaves on a crop body like a f/4.2 in terms of depth of field and compression, but it still behaves like an f/2.8 lens in terms of exposure settings.
That is, a scene best exposed at 1/100, f/2.8 and ISO 100 using lens X on a full frame camera will also be best exposed for the same field of view using the same lense on a crop body at 1/100, f/2.8 and ISO 100 -- PURELY considering exposure, that is, lighting of the scene.
+Art Altman True. That's why it doesn't matter that much if the lense is actually f/4.2, unless you care about the background blur. A 50mm f/1.8 will still be a really fast lens on an APS-C body.
This can be tested easy with using the same lens on full frame and APC body, use the same setting and see that is exposed the same.
Obiviously you will not get same field of view. A 50mm on FF will have 50mm fov, while on an APS-C it will have 75 (or 80 for canon).
+Art Altman
Edit: P.S. in short: Higher Pixel-density requires the light to be even more focused in order to get the same signal strength (per pix) out of a physical smaller sensor.
(same technology assumed).
... but leaving my ramblings from before in for reference.
Hmm, does the following argument make sense ? The 2.8 on a FF lens is calculated with a 35mm sensor in mind. On a APS-C you use only the portion that is hitting the smaller sensor which is overall less then the amount hitting the FF.
Or this Argument:
Because a lens is designed to spread the light evenly over the sensors surface, it doesnt matter how big the sensor is. Every pixel get the same amount of light. Or lets rather say every square inch. The Problem beeing that an APS-C Pixel is much smaller in dimension then one Pixel of a FF sensor. This also means the light would need to be much more focused in order to get the same signal strength in the sensor readout to be equal to the readout of an FF chip.
... this obviously will not change when putting the same lens on different bodies.
Come to think of it the second one is much more likely ;)
+Jens Vielmann "The 2.8 on a FF lens is calculated with a 35mm sensor in mind"
No, because f-stop is the ratio of the aperture diameter to the focal length. The incidental light intensity at a given point, all else being equal (aperture, focal length, and transmittance) is the same regardless of the lens' intended image circle.
The thing to keep in mind is that a FF lens of a given focal length will be physically larger than one intended for a smaller image circle, thus the physical aperture, in absolute diameter, will be larger. This is what accounts for the theoretical difference in total light per mm^s at the image plane: the larger lens admits more photons, in absolute value. APS-C lenses, conversely, have a smaller physical aperture. However, the ratio of that aperture, it's f-stop, still indicates the amount of light admitted relative to the intensity at the image plane.
I routinely use medium format lenses adapted to smaller sensor areas, and the behaviour does not change. A 90mm Asahi 6x7 lens adapted down to EF and put on a 6D behaves just like you would expect a 90mm lens to behave. This is pretty easy to test just by putting a chart shot of the 90mm next to a chart shot using the kit 24-105mm L lens set to 90mm. At the same aperture you will see little more than slight exposure differences which are attributable to the transmittance of the lens itself.
+Daniel Nguyen "if a lens designed for crop sensor is f/1.8 would it produce the same amount of bokeh as a Full Frame lens on a crop body with the same focal length and aperture?"
Yes. Regardless of intended format the bokeh will be (more or less) the same when the lenses are applied the same. A 50mm f/1.8 FF lens and a 50mm f/1.8 APS-C lens will produce nearly identical images when used on an APS-C sensor. There will be some differences, as there are between all lenses, but the depth of field, bokeh, field of view, and exposure value will not change.
+Folding Ideas If using a Canon 70D will a crop lens at f/4 produce the same amount of bokeh or depth of field as a full frame lens at f/4 at the same focal lengths? By same focal lengths, I understand that the full frame lens will be a larger focal length, so let's say the crop lens is at 50mm and the full frame lens is at 31.25mm (exact same focal lengths on both lenses).
I was puzzled by that question for a long time. Thanks for providing an excellent answer
Hey Tony, do you have recommendations for APS-C lenses? I bought the 18-135mm with my t6s and I borrowed a 24-105 f4 from a friend who said gives the sharpest photos. I found that the 18-135 was much sharper then the 24-105 and primes I'd bought. I was going to spend my money on the full-frame lenses but it seems crazy to spend that much money when they're only giving low quality images. The only advantage I can see in buying for full-frame is that when you upgrade to full frame you don't have to spend too much more.
Aidan S please suggest me some best lenses for cannon 800d
s k Some really good lenses I recommend: 10-22mm, 50mm and 55-250mm
I continue to be impressed with the practical knowledge you convey so clearly.
Thank you very much.
Dal
How does one work out the perceptible megapixels? Links would be helpful
+Ian R This can help: ruclips.net/video/MRzONYmjH3M/видео.html
Bring this comment to the top people ! As DXO mark does seem far from flawless
HA! With a divining rod or a Ouija board.
hmmm I would progressively shrink then expand the image by slowly increasing factors until the reconstituted image looks softer than the starting image. e.g. if you shrink a 6000x4000 image to 3000x2000 then expand it back again, and it looks identical to how it started, then the perceptual resolution of the starting image was no greater than 3000x2000
I worked out the sharpness angle years ago, using a Canon 5D4 and a 7D2 with the same lens the 7D2 image was softer. I looked into this and most sources told me I was dreaming. One however explained in detail about pixel width, as you said, magnifying the flaws. So, if a get a 50mp [ or 61mp on Sony] will the image be softer than my 5D4 at 30 odd.
There is some MAJOR flaw in the P-Mpix calculation. I have both fullframe and DX cameras and I can set up tests that I can't tell a difference (if anything sometimes the DX looks better/sharper) and I guarantee the people at DxOmark that came up with this couldn't tell the difference either. It's not just me either, most everyone I've seen talk about is skeptical because it contradicts what their eyes see. Also bullshit with the F/stops. I can't put my 70-200 f/2.8 on my full frame and DX cameras and they BOTH meter the same exposure.
Thank you so much Tony for concluding this topic for me. You are the best photography teacher out here on RUclips! I love your content with all my heart. Thank you so much for your time and vast knowledge. You are my photography hero, God bless you!
I think the video is misleading. When I viewed it, it seemed to say that if I shoot a full-frame lens on a non-full-frame camera, my maximum aperture would be affected. What IS affected is the depth-of-field at that aperture. A f/2.8 lens still collects the same light per area (although the sensor is smaller) whether full-frame or not. A better explanation is given at: petapixel.com/2014/03/28/concise-explanation-crop-factor-affects-focal-length-aperature.
Dave Newby hi Tony
Thanks for coming back to me
I was just so disappointed after getting the much sought-after 400mm 5.6 prime (not the zoom)
And putting it on my 70D so much noise,
After seeing what the 6D could do I immediately put the 70D up for sale,
After your constructive comments I am going to take it off sale and persevere with it
I know you must think have got more money than sense, the first know the latter yes, I really do love the 70 D and now I feel able to keep it many thanks again Tony for your great work Dave Newby
Tony, so what about crop lenses on a full frame body? there are plenty of lenses (like the 35mm 1.8 dx of Nikkor) that give only a little bit of vignetting on full frame. Following the logic of this video, these should be even sharper on an FF camera, right?
two years ago this would have saved me a lot of time and money, this is great work you are doing, thanks
Tony, I love your videos, but I can't agree that a crop sensor body raises the minimum f stop number of a lens for the following reason. The lens will be gathering the same amount of light at a given focal length/aperture setting and producing the same image circle regardless of which body it's on, its just that the crop sensor records a smaller part of that image circle than the full frame. This would not reduce the exposure of the shot on the crop sensor camera. Put another way, if you take a photo on a full frame camera and apply a 1.5 crop to it, the photo doesn't suddenly become under exposed from the f stop number increasing, does it? This is a different situation to a teleconverter which physically changes the amount of light a lens gathers and thus increases its min f stop number. Am I missing something?
the reason he multiplies the aperture by the crop factor is to get a similar depth of field but you are right, the exposure (or perceived brightness) is not to be multiplied by the crop factor it remains the same, but you must also multiply the iso by the crop factor squared to get a similar noise level (assuming they have a comparable pixel count because more precisely it is the ratio of individual pixel surface squared ).
hey schumifan78!
of course, the f stop does not change. the maximum opening of the lens will be the same, never mind onto which body you mout it. BUT: if you mout a FF lens on a crop body, a lot of light goes not onto the sensor but "shines" next to the sensor, on that area where the rest of the FF sensor would be, if you had a FF sensor. so the ammount of light hitting the crop sensor is less than the ammout of light collected by a an aps-c sensor. that is the reason tony says that the maximum f stop is reduced. maybe he should have said "the ammount of light hitting the sensor is reduced on a crop body by the factor of the crop".
The exposure over the image circle doesn't change but all things being equal the crop sensor obviously captures less total light, right? That's in addition to the effects on field of view and depth of field. These people who confidently, even truculently say the aperture doesn't effectively change are nuts IMHO.
This is probably the 5th time I've watched this in the past few months and I think I finally got it. And I LIKE math and stuff. What a confusing subject. Thanks Tony!
tony
i watch all your video and comments. thanks for the brilliant summations for all of us camera dummies. great comments. very practical.
thanks for your inspiration.
How is "Perceptual Megapixel" calculated? Without that information I don't know what the term means. I can't find this information on their website.
Revisting this video, because it's so helpful. Thank you for posting such high quality content, and... potentially saving me a lot of money. :)
Tony
Thanks for you valuable video. I am an amateur with a Nikon D7200 and would like a clarification as I got somehow confused with some of your comparisons.
I would actually look at the use I need to do and what is available, relative to what I can capture by looking through an APS-C sensor. I give you 3 scenarios that would probably summarize as many use ranges.
1 - wide to normal: say up to 100mm
2 - medium: say between 100 to 300
3 - long: over 300mm (I continue to refer to aps-c focal not ff)
I think we are in full agreement on scenario 1, the wide to normal range. It is better and cheaper to stick to aps-c lenses for aps-c bodies.
Also we are in agreement on the long side, here we simply have no choise, there are no aps-c lens available above the 300mm (native) range. You "have to go ff".
But on the medium range it sounds to me that ff lenses can also be better. Let's assume I need to shoot 300mm.
On DxO I compared a Nikon DX 18-300 with a DX 55-300 and an FX 70-200 f4, all measured on D7100 (the 7200 is not available on DxO but the pixel density is the same).
Honestly there seems to be no comparison, the 70-200 f4 is far better also if you consider value for money (the lens cost about 30/40% more but its performance can exceed this proportion). The f2.8 is slightly better but value for money suffers on many aspects.
Am I right ?.
Also if this is the case wouldn’t it be easier to say that the rule “aps-c lenses for aps-c bodies” is limited exclusively to the wide to normal range ?
Thanks
This is great advice! Bought some expensive glass for crop body with poor results. Researched lens and camera combos on dxomark, and got much better results
Now I get it why DX sensors using a full frame lens and the same ISO, Shutter speed and Aperture settings have lower exposures. It's because of the hidden Aperture x1.5 factor that invariably increases the aperture of the lens (but doesn't show on the camera settings). That also explains why the depth of field degrades when you use full-frame sensors on DX bodies. Selling the D7100 and getting the D610 as backup to my D750. THANK YOU!
What I think most critics of +Tony Northrup fail to take into account is:
* With the exception of wildlife photographers, the rest of us don't use cropped sensors to get a cropped version (tighter FoV) of the FF image; we want the same resulting image/composition, thus to get the same framing, keeping subject-to-camera distance the same, we would need to use a (CF times) shorter lens. Shorter lenses have bigger DoF for the same aperture setting and subject-to-camera distance.
* The "lens doesn't care what sensor is behind it" argument is meaningful only if we don't care to produce the same image as the FF case. Given the same aperture and subject-to-camera distance, the DoF or focal length of the lens does indeed not change, they remain as they were. But you are not getting the same framing (FoV) as the FF image, thus this becomes largely irrelevant and equivalent Aperture and focal length come into play.
* We want to see images, regardless of their initial capture size, projected to the same output size, be that a screen or photo paper. A cropped sensor image needs to be blown up more to reach the same target output size, so it's only natural that there is a loss of IQ / sharpness due to upscaling. An APS-C camera might benefit from discarding the softer FF corners and its (potentially) higher pixel density, but this is mitigated by the fact that you still need to upscale this image to the same target output medium area. In that sense, it's a lot like a digital zoom.
* Total light captured matters, otherwise we would be able to capture a 1mm^2 crop (same light intensity / area as the whole sensor) and expect it to yield the same IQ as the full FF frame area, but that's obviously not possible. That's why a blown up image from a crop sensor behaves as if it is noisier than it's FF equivalent, for the same ISO setting, even though the noise per pixel might be the same (e.g. FF: D800, APS-C: D7000).
* A lens does not have infinite optical resolution. If you discard a (greater) portion of the (huge yet) finite number of rays entering the front element by placing a smaller sensor behind the lens, you cannot expect to get the same total optical information.
* Wildlife photographers do have a motivation to use crop bodies for the extra reach, exactly for the use case that an FF body does not get them close enough. Provided that a FF can't get them close enough, the APS-C image may have better IQ than a crop of the FF image (crop mode on FF bodies) if the APS-C body has higher pixel density. For identical pixel densities (D800, D7000) I would expect the image from the crop body to be comparable to a cropped version of the FF image.
BINGO! Exactly answered all my current issues. I'm moving from semi to pro and want more from my images. I only have crop bodies for now and assumed that renting FF lenses would give me much better results and that's not the case. Not a great thing to learn when you get home from a job and everything is SOFT! Thanks for the info, some technical stuff but really great information!
whaatt???? r u kidding? crop factor have nothing to do with aperture? nothing with dof? only with fov? Take any nikon ff take a photo with a 50mm and then take the same scene in crop mode and pls let us know if the exposure changes and the bokeh changes... nooooooo no noooo noo. False False.
ONLY with FOV
This exact same topic is in every second comment on this video. Check sdp.io/crop
I use a full frame lens on a crop sensor, cause eventually I will move to a full frame camera. I can tell you that it works perfectly fine, the sharpness is way better than my kit lens and it's also wide angle.
excellent video, I'm currently looking for a new lens for my 70D and this really helped me a lot in narrowing down witch lens i should get for my camera.
One of stupidest video after jpg vs raw. F stays the same and I’m taking photos in raw.
Be nicer to people. We're all just doing our best.
Very nice Tony. I'm glad your viewers also mentioned sports as a good use for full-frame telephoto lenses on APS-C bodies.
A 24-70 on aps-c body doesn't behave "exactly" like a 36-1.5 on a full frame body. The field of view is the same, but the compression is still the same no matter how big the sensor is.
+harut82 I've researched compression and experimented with it, and it seems to be a factor of only the effective focal length. If you can find a photographic example showing how compression is a factor of sensor size but not effective focal length, please send it to me (tony@northrup.org).
I'm not arguing that compression is a factor of sensor size. I'm arguing that compression is a factor of "focal length" not "effective focal length"
I'm arguing that compression of a 50mm lens is always the same no matter what the sensor size is behind the camera.
If compression doesn't change when you crop a photo in lightroom why should it change if using a crop sensor?
35mm on crop body will not behave like a 52.5mm on full frame. The field of view will be the same, But on the crop body with 35mm lens you'll have the compression of a 35mm lens while on the full frame body you'll have the compression of a 52.5mm lens.
facebook.com/IFLphotos/posts/1309117285769267
+harut82 That article doesn't compare different sensor sizes. Seriously, an equivalent 50mm has the same compression, regardless of sensor size or absolute mm.
I'll be chilled and roaming RUclips, and this man comes with stuff I never knew before, only to disrupt my original plans, in a really positive way.
In other words, he's saving me from doing dumb stuff... Like buying full frame lenses for a Canon EOS 90D.
For real though, what Sigma Lenses, apart from the 18-35 and 50-100, should I get for a Canon 90D?
Wrong... or not wrong but your missing some points: If you were to take a full frame camera and a micro 4/3's WITHOUT moving the tripod take the exact same shot with the exact same lens. Then you were to crop the finished photo on the full frame to match the m4/3's... you would have the exact shot with the exact depth of field. So there's NO need to use the 2x crop factor applied to the aperture. But sure, if you throw on a longer lens on the full frame to match the micro 4/3's of course you'll get shallower depth of field with a longer lens... throwing on a longer lens while keeping the tripod where it is, is the same as moving closer to the subject. When you move closer to the subject, of course you get shallower depth of field. Basically, the aperture at 2.8 is ALWAYS 2.8... it's when you throw on longer lenses or move closer to the subject, you get shallower depth of field. And there's going to be a point where you can't just throw on a longer lens to match field of view because of minimum focusing distances. F/2.8 is ALWAYS f/2.8. Doesn't matter the size of your sensor.
Size always matters dude
I understood that this way: Lens have certain ability to draw details :resolution (from the square area,that is big as possible to fit into that round area a lens produces).If you f.ex. crop that area to half , decreases also that lens resolution about to half(altrough corner resolution increaces).But in wild life fotographing with tele lens that resolution falling compensates by more pixels (on that croped area),when no need to crop after vs ff body...
I have issues with this video. The technical explanation of why you should NOT use a full frame lens on a crop sensor camera is missing. Tony is just using DxO's proprietary term, perceptual megapixels, without ever explaining what that is. He is just repeating DxO's findings. The recommendation of using a frame lens on a crop sensor camera for wildlife is incorrect and conflicts with the recommendation of not using full frame lenses on crop sensor cameras. The correct answer is that an uncropped image from a crop sensor camera will he sharper due to increased pixel density than the same image cropped from a full frame camera. Therefore some are recommending that you bring along a crop sensor camera on shooting wildlife. However, if you don;'t crop, and you are able to get the entire image in the frame, the full frame will render a sharper image than the crop sensor.
I like Tony but this video is lacking.
+maghi “Magi” cat If you want more information about perceptual megapixels, here ya go: ruclips.net/video/MRzONYmjH3M/видео.html
As I mentioned in the video, I used DxOMark as an objective source, however, the root of this comes from actual experimentation and the frustration of many, many photographers who've contacted me because they were getting unsharp results with full frame lenses on crop bodies.
I believe I explained why the scenario works differently for wildlife; you have to crop anyway, and the higher pixel density helps extract more detail from the lens. It doesn't help as much as getting closer and using the full sensor; if you didn't have to crop, you'd get better results with the full-frame body.
Tony & Chelsea Northrup. Thanks for your great videos, tutorials and reviews. I´m kind of new on advanced photography, but i think i understand your video and explanation. What many people seems to be missing on this video is the most important key word: EQUIVALENT. if i understand well, It´s obvious that the aperture of the lens number cannot change by magic, nor any other physical parameter by itself. It is basically just a way to explain EQUIVALENCE comparing Fulll frame and Crop sensors. In practical life i think i understand the consequences. With a cropped sensor one have to move further away to have the same fOV, and when one moves away, the depth of field changes, because the light converges in a diferent way from the subject to the lenses and then to the sensor according to the distance, hence, in that sense, it has a smilar effect as if one increases the f-stop number, without actually seeing any diference in the exposure in the final photo comparison, because the equivalent light between the 2 cameras with it´s own native lens about is the same, but not the TOTAL light, due the to bigger Bucket for light that the Full frame sensor has, which is why i read that the full frame tend to have less noise in the same conditions, at higher ISOs, generally speaking. I know this is deeper, but to be an amateur starting to learn, i hope i´m close enough to get a simplier, but not so incorrect understanding of what you are saying, and if so, this maight be useful for others amateurs like me. Thanks again from latinamerica. Hugs !
My general rule of thumb is if you ever think you might go full frame just get the full frame lenses so you don't have to buy twice
Buy twice??
@@aight365 I think he means you have to change from dx lens to fx lens if you change to full frame
Tony thankyou why cant more reviews be like yours, explained exactly like it needed to be explained. Growing up on film i sometimes find the digital word very confusing but when it is explained like that it makes perfect sense. Thankyou
i;ve been wondering that for a long time. i've been researching and researching for 4 months. After seeing all the math equations in this video, i decided to spend the $ on a new smartphone.
yikes
Fantastic explanation. Answered my question and saved me money and heartache.
I have a Linhof Technika 5 x 7 inch and a 4 x 5 inch reduction back. Making my test shots on Polaroid 4 x 5 inch film. So you think the aperture and the sharpness of my Meyer Rodenstock, Schneider and Zeis optics change. Forget it !!!
I think you are telling the ( digital) tale of The Emperor's New Clothes by Hans Christian Andersen. It is not fair telling such tales to ignorant amateur photographers.
Great useful illustration....What I was searching.....you have synthesized it beautifully....
Very misleading video.
The focal length *does not* change.
The aperture *does not* change.
The bokeh *does not* change.
The equivalent focal length *does* change.
The equivalent depth of field *does* change.
+Roxy-Jane Stone Right, obviously crop factor is a mathematical conversion and it doesn't alter physical objects when you decide to use it, just like converting miles per hour to miles per kilometer doesn't make you go any faster. Nonetheless it's useful for understanding the results you get with your gear.
If you or someone else reading this is interested in learning more, visit sdp.io/crop.
very true, well explained, not misleading.
so regarding light 2.8 is 2.8 on FF as well as Crop sensor and regarding DOF it is 2.8 x 1.6 (on canon crop sensor), correct ?
Well despite what Tony says I agree with Roxy… And anyway if there is any difference between full frame and Crop sensor it's as Zack Arias said : NEGLIGEABLE! ruclips.net/video/PHYidejT3KY/видео.html
Same no sens with pixels war. ruclips.net/video/TFDPSPn1zbc/видео.html
Miles per hour to kilometer per hour… Not sure you understand fully what you're stating...
Thank you, I'm a beginner photographer and I've watched quite a few of your videos now. You'll be happy to know I just bought SDP and am looking forwards to giving that a read through. Looking forwards to continuing my journey in DSLR photography. Got my Canon 450d which seems more than good enough for my level at the moment!
Mr and Mrs Know it all
Hi Tony & Chelsea. I;d just like to say many thanks for putting all these videos on here for free. It really has helped me of the past couple of years. I also just got your SDP book and I love it. Everything explained in simple terms ( the way I like it ) Thats all really just thanks buddy.
CHRIS HENRY You're welcome, and thanks for your support!
1:35 nope... wrong again. It behaves EXACTLY like a 24-70 f/2.8 no matter what the light hits... you could put that lens on a cereal box camera and it would behave exactly the same.
Putting that lens on a cropped sensor camera, images are going to be cropped smaller than what a full frame would produce, that's the only thing. The lens doesn't know or care how big the sensor is, and does it's job exactly the same everytime but will be like cropping an image in photoshop.
BUT, when you start moving around or zooming in to achieve the same framing as fullframe, of course you'll get different results. Moving closer to a subject or zooming in gives you shallower depth of field. But cropping a full frame image to mimic cropped sensors gives you the EXACT same image. It's moving the camera around or zooming in to achieve similar results is why you're confused about this issue.
I'm afraid you are incorrect sir. It is exactly as Tony says in this video. I have tried it for myself. I used the same 24-70 2.8 lens on both a 6D and 60D. I shot the same scene, at about 44mm & f4.5 on the 6D and 70mm & f2.8 on the 60D and the 2 images are IDENTICAL. They are totally indistinguishable. Try it for yourself.
+Charles Ingraham Photography You shot: 44mm on a full frame, and then 70mm on a cropped sensor... therefor the equivalent would be on the 60D which would equal 112mm lens... so how is a 44mm lens and a 112mm (equivalent) the EXACT same shot? I think your math is off... Crop sensors make you already long lens even longer.
+Charles J Gartner I apologize. I shot the 70mm @ 4.5 on the full frame and approximately 44mm @ 2.8 on the crop.
+Charles Ingraham Photography my only point is: When you get closer to a subject you get shallower DOF... When you stop down, you get shallower DOF... the sensor size does not matter, unless you start compensating for those factors by moving closer or switching lenses. .... a lens does not know or care what your sensor size is.
+Charles J Gartner You must be making a different argument from what Tony is even talking about. The lens "behaves" differently on different sized sensors. That goes for both the focal length and the depth of field. That's a fact. I beg you to try it for yourself. Photograph the same scene from the same position using a full frame camera and a crop camera. Use the same zoom lens, and use the equivalent focal length and Aperature from full frame to crop and you will get IDENTICAL images. He is simply correct about this.
You have a talent with explaining such complicated mechanics in such a meaningful way, thanks Tony!
Great video!
What about ff vintage lenses on aps-c?
What evidence do you show to support loss of light?
John Gardiner he never claims a „loss of light“
You are right about sharpness. But I have a 50 f1.8 and 70-200 f4 IS canon lenses on the rebel t2i. I find the color rendition is better than my 15-85mm walk around lens on both these. Both give me fabulous potraits as well
First of all, I rarely comment on videos... But Tony Northrup is absolutely spot on. I did this research quite awhile ago, and tested/verified results from DxOMark. I really appreciate somebody that actually does research and provides REAL quantifiable results. Please keep up the good work!
Chase Martin Thanks, Chase!
@@TonyAndChelsea this seems like advice that was good, but had long outlived it's usefulness. My 10d and bodies from a couple decades ago benefited more from the fancy glass as they weren't able to out resolve the lenses.
I think the problem with a full frame lens vs a crop sensor that most people don't understand is that..... A LOT of light is lost on the sides of the sensor or even reflected past the sides of the sensor and camera body parts... AKA, A lot of reflected light jumps around in the camera body in areas the body was never intended to compassionate for .... Drive into fog with with a bunch of light and its harder to see the road because light it being reflected off the environment.
This is true but I have no idea how it relates to the sensor size
for wildlife I agree - if you gonna crop anyway (and you will - or else you must be ninja)... but that part about that crop sensor looses light full stop and more (that "fstop x crop factor = light lose") - it is straight WRONG - crop bodies do not lose light - you do NOT have to bump iso (not even a 1/3) using same f-stops and shutterspeeds compared to same scenario withCrop vs FF body (no matter what lens you use - for both bodies shutterspeeds will be the same at same f-stops and same isos)....
please Tony - make a fix for that - just ad some text bar in your video - because many people still finds and watches video - and this video is good with valid and dense info (except that terrible mistake)
+Alberts Kuks Please see sdp.io/crop for detailed information about this.
Obviously, you're right that the camera settings don't change--but the total amount of light the sensor is getting is very different.
+Tony Northrup
_"Obviously, you're right that the camera settings don't change--but the total amount of light the sensor is getting is very different."_
But you neglect that the total light per square mm is exactly the same.
The perceptual changes exist only because you have to stand back from your subject by the crop factor to get the same FOV as FF.
Light intensity is the same for any given Fstop from any given lens no matter the size of the sensor, the aperture variable isn't real, it's derived from the focal distance to achieve the same FOV.
Now here's something you might not have considered, if you take a picture with the D7000 and the D810 at the same distance from the subject and all else equal, your depth of field, your aperture, your focal distance and bokeh will be exactly the same. the only difference will be the framing, the field of view.
So when you talk about these settings changing, they are not, they are exactly the same, the variable comes in when you want to frame the shot the same, your relative distance from the subject moves back by a factor of 1.5x, this is what makes the perception changes from crop sensor to full frame. Your depth of field at the different focal distance is what has changed, not your aperture, because the exact same amount of light per square millimetre falls onto the sensor. Your focal length changes, not because of the lens, but because of the size of the sensor and its relation to the circle of illumination.
Your reliance on the perceptual mega pixels as a measure of the actual lens performance is flawed.
Let me prove it.
The D810 36 MP 24 x 36mm FF sensor
The D7000 16.3mp 15.6 X 23.6 Crop sensor
The D5300 24MP 15.6 X 23.6 Crop sensor
Lens Nikon AF-S Nikkor 50mm f/1.4G
Pixel area
D810 23.72um squared
D7000 22.75 um squared
D5300 15.13 um squared
DXO sharpness
D810 21 MP
D7000 10MP
D5300 14MP
Relative percentage
D810 21/36 = 58.33% of perfect
D7000 10/16.3 = 61.35% of perfect
D5300 14/24 = 58.33% of perfect
Perceptual megapixels are a guide, nothing more they are actually measuring sensor performance to pixel density rather than the lens itself.
Just look at the different scores of the lens from the D3200, 5200, 5300, the only difference is the sensor, not size or pixel density (minimal) yet there is 3 perceptual megapixels difference between them (12, 11, and 14 respectively)
One thing you forget about crop factor lenses is that because of their smaller circle of illumination, vignetting, barrel distortion, focus and CA at edges are more pronounced than using a FF lens, because you are only using the centre 67% of the FF lens on a crop sensor. They don't just cut out the centre of a FF lens to make the crop lens version, they design them within acceptable limits.
chuck90504
I would suggest that you rethink how you are actually considering the image circle and what the actual sensor is doing.
For any given circle of illuminations image, nothing changes just because you are using a different size sensor.
How much information from that circle of illumination you get is what changes.
The only variable that the sensor does is that it is a crop of the available information. Which is why I used the D7000 vs the D810 sensors as the pixel density is near identical and so does not get the effects relating to the circle of confusion of the DOF as a variable of pixel area.
The whole DOF arguments relate to framing, to frame on a crop sensor the same as on a FF sensor is to move back the corresponding distance to the subject by the crop factor ratio. This focal distance change is what effects the DOF for any given lens.
This can be proven by taking a photo with the same lens at the same distance from the subject so that the focal distance is the same for both shots and crop the FF D810 to the D7000 crop area in post production.
Of course you cannot crop the other way around, so you need to move back 1.5x the distance from the subject with a D7000 to achieve the same framing as the D810, and as soon as you do that then everything else changes with it including DOF.
chuck90504
_"When you have to bring in movement to achieve the same look, you've got an issue with perspective distortion"_
You get that using a different lens anyway.
A 75mm lens on a FF is a 50mm lens on a crop.
Given that f2 is different on both lenses (50mm 25mm aperture and 75mm is a 37.5mm arpeture) , you get a different perspective anyway.
Much is explained from a relative perspective of the different formats, most explanations use framing relative to a FF sensor and from that we get the variations in DOF and perspective, because with the same lens, the FOV of a crop sensor is the same as a longer lens on FF.
But given the equivalent lens, the f2 while still f2 and 50 mm on either camera, due to framing the f2 becomes f3 on a crop for DOF comparisons, which is kinda bullshit because it doesn't affect light intensity, because when you frame with that lens you are standing back 50% further from your subject, but still at f2.
It is the distance to subject that is the variable here, the Fstop recalculation is just a virtualization of the actual.
Consider the minimum focus distance of a lens, in both cases let's say it's 1 metre, the crop sensor while not actually allowing you to get closer to your subject, will in effect magnify the frame compared to a FF sensor.
Where you would crop a FF at minimum focal distance for a small subject, the crop sensor appears 1.5x closer as it's already cropped.
Higher pixel densities means more detail on a crop sensor generally, exceptions being the D7000 vs the D810, where the crop and the FF have the same amount of detail in terms of number of pixels of the subject at minimum focus distance. On both the same lens and the same F stop produce the same results.
chuck90504
_"if you and I both shoot a 100mm lens and I'm standing 1.5x closer than you, 9 times out of 10, I'm going to get more detail."_
That is going to depend on the ability of the lens to resolve to the pixel density.
_"it's pretty obvious that a simple concept of distance continues to escape you. "_
*"It is the distance to subject that is the variable here"*
Distance to the subject is the very basis of everything I've been saying.
The change in FOV has nothing to do with the lens, as it is still producing the same circle of illumination, but a crop sensor can only gather a percentage of that FOV hence the equivalent is translated to the lens in the calculations, though the lens has nothing to do with the actual changes.
For the sensor to get the same framing from it's limited FOV, the crop factor tells us the distance change.
The way explanations revolve around lenses is a smoke and mirrors explanation that works, the actual explanation lies in the percentage of the circle of illumination that a crop sensor can see along the vertical and horizontal planes.
The only advantage of a crop sensor, is that more often than not it has a higher pixel density, therefore it translates the analogue information from the circle of illumination at a higher megapixel count, but with the loss of light intensity per pixel, because of the smaller pixel area, but not per square mm of sensor. Full frame sensors are better in low light, not because of the sensor size, but the area of the individual pixels.
_"Even in those specialized genres, the advantage of crop is not evident until you've exhausted your longest lenses. "_
The length of the lens isn't the critical factor, it's minimum focal distance that is. What happens when you use macro tubes, you're closest focus shrinks in distance.
Glad i came across this video. I just recently purchased a 80D and was looking to invest in some L lenses from canon. Boy am i glad i didnt. Was looking at the 24-70mm L lens but instead bought 3 EF S lenses for less then half the price.
Putting cheap glass on expensive bodies is no good either.
Awesome... Now this is some serious professional explanation to lenses and bodies, for months i was confuse about which body i should invest on.. crop or full frame. I think i'm gonna go with a full frame now for portraits, wedding and all. Thanks
since when does mp mean sharpness, I have had images sharp as a tack with my old 12 mp Nikon D80, don't you really mean resolution, detail is not sharpness it is resolution, over technical people make me crazy, and sometimes with all there brains, they are looking at things wrong, numbers are for mathematical engineers, not creative photographers, even though most of what he is saying is rite, in some ways it is wrong, and cameras like the 12mp Sony A7s prove my point, 12mp with amazing image quality, with sharp as a tack images, yes a Nikon D810 or Sony A7r has more resolution, but that is not sharpness, and many people have gotten unsharp images with a D800 and pro glass, and personally i feel sorry for anyone who makes there camera and lens buying decisions entirely based on a DXO mark ratings, and I don't mean any of this as a personal attack on Tony, because I like him, but he can drive me crazy with all his technical talk, I bet even his wife rolls her eyes some times when he is going on and on with his technical wisdom, she is like, common Tony shut up and lets go to sleep already haha, but I would like to buy his new Light Room book, hopefully it's not over technical
Detail is not sharpness? It's both resolution and sharpness, if your lens isn't sharp it's going to take "blurry" pictures and while you have a high resolution you have way less detail. And the higher the resolution the more detail is possible with a sharp lens.
Tony has said many times that dxomark doesn't tell the whole story. He is using it here to get the information necessary to make his points. It seems you are quibbling over semantics as well. If you would look at some of the other videos that go over these technical details you would look past whatever small ambiguity in words used and understand what Tony is saying better here. I don't feel sorry for his wife or anyone else that gets to listen to someone that not only knows his subject but can explain it so clearly and with a pleasant voice and mannerisms. Honestly, I can't think of anyone doing a better job for free on the internet yet you need to criticize him on a video that nearly everyone would agree is very good to excellent. Methinks the problem is with you and not with Tony. I'm not saying this as an insult. I wish you the best and happy holidays, sincerely.
Galactu5 what problem are you talking about, I have no problem with Tony or his videos, i am a subscriber to his channel, and I understand all he is saying in this video, but that doesn't mean I have to agree with everything he says, and I have the right to give my opinion in return, and do you not see that what I said about his wife was just a joke, anyway happy holidays to you also, I hope Santa brings you a new camera
photographerjonathan hello! You mention a few times that Tony is overly technical. That is the main problem I see in your comments and I find that opinion to be inaccurate. I'm aware you like Tony overall, but the fact remains that photography has technical aspects to it and it is nice that someone out there has a channel that explains the commonly asked questions and the minutiae with such clarity. It may seem overly technical to you, but Tony's explanations and formula empower us to understand the concepts rather than just have some rough familiarity. You mention you understand what he is saying here and elsewhere yet see it as overly technical. That is the main problem I was getting at and I just wanted to suggest that it didn't seem to mirror reality.
Wish you nothing but the best and may you get that lens you have always desired this holiday season! Thanks for not taking my comments as an attack and for not attacking me back which is sadly way too common. Peace!
photographerjonathan Them 'numbers' may be for mathematical engineers, but believe it, when you're on a professional commercial shoot - no amount of 'creative' will make up for noisy images in the eyes of the director/client...
Tony, was able to follow most of this and agree, Wondering about your slide comparing the Nikon 810, 610, and 7100. See how cropping the pic on the 810 and 610 in post affects what I will call the "effective" megapixels, because we crop off part of the image to make the subject larger... but when using the 7100 haven't we effectively already cropped the image in camera, which means our "effective" pixel count would remain what the camera is rated for, I believe 2400mp?
total bull shit. I use a 24-120 f4 on my 7200. If I was getting F6 out of it, then it would render useless. Not the case. This lens works great in moderate light. its not an f6 on my 7200
You are right, it only effects the Depth of Field, not the total gathered light.
Exactly, the total light gathered DOES not change, but the DOF does.
I get your point but why there's a noise difference?
+THEKMT6 The light falls on a smaller surface so there will be less light gathered in total. To have the same exposure as on a full frame sensor, the sensitivity of the sensor must be higher, wich brings more noise.
Vincent van der Pas Well said, sensor size determines Equivalent values of Aperture, focal length and iso Regardless of the ino displayed in camera.
I have a very basic rebel T3 and I just bought a used canon 24-105 f4L after trying it for a week. As Tony said, I did not see the difference in the sharpness but the advantage I have with this lens is fixed maximum aperture throughout the zoom range. Also I feel that the focusing is better with this lens compare to my kit lenses and the colors that come out are more vivid.
Ameya Hingane try the Sigma 18-35 ; it will blow your mind!
Tony Northrup I was thinking about this but my dilemma is that if and when I upgrade to a full frame, I will have to find somebody who is willing to buy this for aps-c.
P-Mpixels don't exist.
THANK YOU. At last, someone spoke the truth.
Exactly, Photons don't change resolution because they are falling on a "smaller" sensor ! They don't make lens glass according to the sensor resolution. Quality may be but not resolution...
You do not take money because you are an accomplished Author...KUDOS to you Tony......