/p/ - Photography » Thread #3431567

[20 / 4 / ?]

26KiB, 360x240, northrup.jpg

Resolution with FF lens on crop body

Anonymous Fri 08 Feb 2019 23:49:14 No.3431567 View View Reply Original Report

Quoted By: >>3431569 >>3431570 >>3431572 >>3431573 >>3431581 >>3431641 >>3431665 >>3431674

Pardon my ignorance, but there's this claim put forward by the Northrup couple that I can't wrap my head around.

They claim that that mounting a full frame lens on a crop camera, will result in significantly lower resolution than put out by the sensor. For instance, according to their calculations, a 24-70 2.8 Canon L lens will only have a resolution output of 7 MP on a 18 MP 7D body.

Is there any validity to this claim, and if so, how is this explained in terms of optical theory? I understand that the light beam generated by a full frame lens will have a larger diameter than the mount diameter on a crop camera, but do the amount of photons hitting the sensor really equate directly to resolution in pixels?

Cheers for any insight. Here's a photo of the Northrup couple in bed.

Anonymous

Anonymous Fri 08 Feb 2019 23:58:16 No.3431569 Report

Quoted By: >>3431571 >>3431576

>>3431567
>resolution is dependant on lens
any validity he had is lost by that implication
please stop educating yourself from youtubers that don't take and talk about photos

Anonymous

Anonymous Sat 09 Feb 2019 00:06:38 No.3431570 Report

Quoted By:

>>3431567
i think i know the video you're talking about and tony was describing EFFECTIVE megapickles and how different lens combinations affect that

you should probably watch it again

Anonymous

View Same Google ImgOps iqdb SauceNAO Full-Frame-vs-APS-C-vs-M43-vs-CX.jpg, 262KiB, 960x641

Anonymous Sat 09 Feb 2019 00:06:48 No.3431571 Report

Quoted By:

it is somewhat correct but you should put a full frame lens of a APS-C camera anyways because full frame lenses tend to be sharper than APS-C ones.
just imagine it like this, APS-C uses a smaller part of the lens while a full frame gets to use a big part of the lens which in turn produces a sharper image.
Phones these days have 50MP cameras but only produce images that are effectively 2MP because the lens is godawful and the sensor tiny, in short full frame produces more usable megapixels because you're using a much bigger area of the lens.
there are of course advantages to APS-C and smaller for sharpness but that's another topic.
>>3431569
the effective/usable resolution is dependent on your lens but you're correct, youtube e-celebs are a retarded source of information.

Anonymous

Anonymous Sat 09 Feb 2019 00:07:18 No.3431572 Report

Quoted By:

>>3431567
Relax, these claims go straight into the trashbin.

Anonymous

View Same Google ImgOps iqdb SauceNAO VZ1Q7096.jpg, 319KiB, 1600x1067

Anonymous Sat 09 Feb 2019 00:08:25 No.3431573 Report

Quoted By:

>>3431567
Ether utter stupidity, or you're mixing things up. Mounting crop lens on ff cam will effectively decrease resolution. You'll get picrelated and will need to crop out px. Mounting dedicated ff lens on crop body, may decrease sharpness, but resolution will obviously stay the same. Ether way, stop watching moronic youtoubers that compose themselves like teenagers.

Anonymous

Anonymous Sat 09 Feb 2019 00:30:37 No.3431576 Report

Quoted By: >>3431580

>>3431569
Resolution IS partially dependent on the lens.

Anonymous

Anonymous Sat 09 Feb 2019 00:58:20 No.3431580 Report

Quoted By:

>>3431576
I know I'm just arguing semantics here but resolution is resolution, effective resolution is dependent on the lens.
you can have 50MP on your phone and still only output an effective 2MP.

Anonymous

Anonymous Sat 09 Feb 2019 01:03:44 No.3431581 Report

Quoted By: >>3431627 >>3431680

>>3431567
>They claim that that mounting a full frame lens on a crop camera, will result in significantly lower resolution than put out by the sensor.
Okay, so, here's the point they're trying to make.

Sensors have a resolution. That's obvious, that's what camera makers market at you.

But lenses *also* have a resolution. There is a limit to the amount of details a given lens can resolve. It's a lot more complicated because it varies by aperture and it varies over the width of the frame and it can vary depending on what camera it's put on (e.g., if one sensor handles light rays coming in at a tighter angle than others). It's usually measured in line pairs per millimeter (lp/mm).

So if a lens outresolves the sensor, then you get all of the resolution that the sensor can give you. That's obvious.

If the sensor outresolves the *lens* though, while you're still going to get all of the pixels that the sensor has, you're not going to get all of the detail that the sensor is capable of resolving.

Now, the reason full frame and crop sensors come into play is if you have a bigger sensor with bigger photosites and the same resolution, the lens' resolution is less likely to be the limiting factor. So say you have a full frame camera with a 12MP sensor, and a lens that resolves juuuuust enough detail to feed that 12MP sensor, but then you put it on a 12MP *crop* sensor. The lens only resolves enough detail to deliver about 8MP worth of detail to that crop region, but there's a 12MP sensor there. So you're going to get a 12MP photo with about 8MP worth of detail.

That being said, it's not nearly as simple as "Full-frame lenses have lower resolution", especially since crop-specific lenses are often lower quality, so there's not a hard and fast rule. But it's true that for a given lens, and a given sensor resolution, you're more likely to outresolve the lens on crop than on full frame.

Anonymous

Anonymous Sat 09 Feb 2019 02:43:46 No.3431627 Report

Quoted By:

>>3431581
So I see the talk is about effective lens resolving power. That's different than cam res that op was talking about. There is some truth in it. It's complicated matter, but you'll never be able to calculate into pixels what you actually get out of the lens. I guess tony half read some article on it and quickly made a vid about it not knowing what it's all about.
One notable case, where it come into effect, is when you're measuring effective resolution gains of high res mode on ibis/pixel shit modes of cams that support it. On most lenses, you'll not see any gains besides simple px upscale, you'll need lenses with high resolving power to actually see the difference, and even on them, it'll be only on certain f's that you'll start seeing actual 'resolution' improvements.
In vast majority of cases this is something you should never have to worry about. A little bit of haze, wind, hand shake, will do more to the sharpness of a lens, than resolving power, and there are things like micro contrast, and colour rendition of the lens, that will have more visible and noticeable effect on picture output.

Anonymous

View Same Google ImgOps iqdb SauceNAO 1541369879604.png, 492KiB, 839x463

Anonymous Sat 09 Feb 2019 02:48:02 No.3431631 Report

Quoted By:

Please post more cuties

Anonymous

Anonymous Sat 09 Feb 2019 03:29:32 No.3431641 Report

Quoted By: >>3431680

>>3431567
>a 24-70 2.8 Canon L lens will only have a resolution output of 7 MP on a 18 MP 7D body.
sounds about right but not because of it being a FF lens on a crop body. A lot of zoom lenses are soft as fuck, the 24-70 included, and resolve far less than modern sensors are capable of.

Even when it comes to primes, the 60mm 2.8g macro is a fantastic Nikkor lens but is noticeably soft on my D800E. I also have a lot of Leica glass, including the Zeiss ZM 50 Planar 2.8 and a 35mm Voigtlander color-skopar that are barely good enough for my Ricoh GXR, and the Zeiss is passable on the A6000 but soft and the Voitlander is soft as my dick after watching the kardashians.

It really has little to do with FF vs Crop lenses and more to do with resolving power.

also sharpness of a lens is really over rated in most cases. I as well as many others looks for the more intangible attributes such as look and feel

giannis

giannis Sat 09 Feb 2019 04:37:46 No.3431665 Report

Quoted By:

>>3431567
>mounting a full frame lens on a crop camera, will result in significantly lower resolution than put out by the sensor.
Mounting *any* lens on a sensor results on lower resolution than what the sensor is capable of.
The system resolution (lens res + sensor res) asymptotically reaches the lower of the two resolutions when the other one approaches infinity.

In practice, for a camera system this means that the system resolution will be a bit smaller than the smaller of the two, sensor res and lens res.

Also, a sensor is a "system" on its own, comprising of the the photosites (pixels) + microlenses + Bayer array + AA filter.
The greatest offender by a huge margin is the Bayer filter of course, it reduces the true resolution up to 50%.

So anyway, imagine the imaging process like a chain: the system is as good as it's weakest link, and you'd need physically perfect components everywhere else along the chain, to reach the upper bound the weakest link's theoretical potential.

Anonymous

Anonymous Sat 09 Feb 2019 05:12:44 No.3431674 Report

Quoted By:

>>3431567
>They claim that that mounting a full frame lens on a crop camera, will result in significantly lower resolution than put out by the sensor. For instance, according to their calculations, a 24-70 2.8 Canon L lens will only have a resolution output of 7 MP on a 18 MP 7D body.

This is observably false. Head on over to Imaging Resource and lookup their resolution test results for a 24mp FF sensor and a 24mp APS-C sensor of the same generation and manufacturer. The LPH values will be roughly the same (within a couple percentage points) even if you pick a pair where IR used the same FF lens on both bodies. Actually, using the same lens will likely bring the LPH results even closer together.

Further, there is no such thing as "effective megapixels." MP describes sensor sampling frequency, not resolution. Sampling frequency doesn't change with a lens change.

Contrary to popular belief, most lenses have extinction resolutions well in excess of current pixel densities for m43 and larger sensors. What we notice between better and worse lenses is the difference in the contrast of coarser details, otherwise simply known as sharpness.

With any lens detail contrast drops as details become finer. With a smaller sensor you are taking the same scene details and projecting them to a smaller point on the sensor. That's why crop images aren't as sharp ooc all other factors being equal. This is also the source of the "crop sensors need sharper lenses " meme. While there's a kernel of truth to the meme, you don't actually need a sharper lens. In most cases you can simply bump the sharpening parameter in camera, or apply more sharpening in post processing.

It's sad that they fell for this meme. What's even worse is that the source of the meme..."effective megapixels"...comes from DxO, a psuedo science website if I've ever seen one.

Anonymous

Anonymous Sat 09 Feb 2019 05:24:54 No.3431680 Report

Quoted By: >>3432004

>>3431641
>>3431581
>So if a lens outresolves the sensor, then you get all of the resolution that the sensor can give you. That's obvious.
>If the sensor outresolves the *lens* though, while you're still going to get all of the pixels that the sensor has, you're not going to get all of the detail that the sensor is capable of resolving.
This isn't exactly right. Final resolution of any imaging system will always be less than its weakest component. But improving ANY component will bring you closer to that limit. So it's not a simple matter of a lens limiting a sensor.

Bob Atkins demonstrated this in his 5Ds review. He took the worst 1990s consumer zoom he had and put it on a 6D. Coke bottle output. He then put it on the 5Ds. It was actually borderline usable. In one sense yes, you're not getting the same sharpness (or in this case extinction resolution) that you would with a better lens. But in another sense putting a 50mp sensor behind a shit lens STILL improved the final result.

Anonymous

Anonymous Sat 09 Feb 2019 21:12:00 No.3432004 Report

Quoted By: >>3432032

>>3431680
Got a link for this? Could be he only got better noise performance and counted it in as sharpness gains. Noise is one place where resolution helps a lot.

Anonymous

Anonymous Sat 09 Feb 2019 22:44:17 No.3432032 Report

Quoted By:

>>3432004
http://www.bobatkins.com/photography/digital/canon_eos_5Ds_review.html

Noise didn't play any part in this. He took the shots at low ISO under plenty of light.

Anonymous

Anonymous Sat 09 Feb 2019 23:43:07 No.3432059 Report

Quoted By: >>3432063

Outside the scope of the resolutionpower debate. A full frame lens will always give a more consistent performance on a crop sensor as the crop sensor cuts the corners. All lenses are sharper in the center than at the edges. Thus the crop mainly uses the middle sharp part of the lens.

Anonymous

Anonymous Sat 09 Feb 2019 23:47:42 No.3432063 Report

Quoted By: >>3432093

>>3432059
that's a 50/50 situation but it's like what you're saying, APS-C is way more consistent with quality, but at times a FF will outperform an APS-C camera in the corners, especially with good lenses, but even then the lens has to be really good.

Anonymous

Anonymous Sun 10 Feb 2019 00:56:13 No.3432093 Report

Quoted By:

>>3432063
>but at times a FF will outperform an APS-C camera in the corners, especially with good lenses, but even then the lens has to be really good.
There's been a push this decade for corner sharp FF lenses. There's actually quite a bit to choose from now. Of course it's more $$$ than crop.

Capcode	All Only User Posts Only Verified Posts Only Moderator Posts Only Manager Posts Only Admin Posts Only Developer Posts Only Founder Posts
Show Posts	All Only With Images Only Without Images Only Spoiler Images Only Non-Spoiler Images
Deleted Posts	All Only Deleted Posts Only Non-Deleted Posts
Ghost Posts	All Only Ghost Posts Only Non-Ghost Posts
Post Type	All Only Sticky Threads Only Opening Posts Only Reply Posts
Results	All Grouped By Threads
Order	Latest Posts First Oldest Posts First

On these archives

Your latest searches

Resolution with FF lens on crop body