/p/ - Photography » Thread #4294803

[32 / 3 / ?]

65KiB, 800x601, Sensor_sizes_overlaid_inside.svg.png

APS-H

Anonymous Fri 15 Mar 2024 02:44:41 No.4294803 View View Reply Original Report

What do you guys think about APS-H? I just picked up a camera which uses it.
>about 50% more surface area than APS-C
>enables faster readout times than full frame
>larger photo-sites than a similar megapixel full frame
>better dynamic range than APS-C
>lots of APS-C lenses' image circle will cover an APS-H sensor
The only downside is needing to check the lenses yourself to ensure compatibility. It is truly the best of all worlds.

cinefag !CiNE/YT/e6

cinefag !CiNE/YT/e6 Fri 15 Mar 2024 02:51:58 No.4294806 Report

Quoted By: >>4294807

>>4294803
>larger photo-sites than a similar megapixel full frame
You mean than a similar megapixel APS-C?

Anonymous

Anonymous Fri 15 Mar 2024 02:58:59 No.4294807 Report

Quoted By:

>>4294806
Yes, you're correct. I got it backwards

Anonymous

Anonymous Fri 15 Mar 2024 04:06:43 No.4294822 Report

Quoted By: >>4294825

Unfortunately the last and best APS-H body was the 1D IV from 2009, it just can't compare to modern high res full frame bodies. Yeah they're pretty cheap and they're tough but that's about all they have going for them, so if you're not poor and shooting in really shitty conditions there's not much appeal.

Anonymous

Anonymous Fri 15 Mar 2024 04:14:04 No.4294824 Report

Quoted By:

>>4294803
Anything smaller than 4x5 is not worth fussing over. Unless you have the choice between 645 and 67.

Anonymous

Anonymous Fri 15 Mar 2024 04:14:07 No.4294825 Report

Quoted By: >>4294832 >>4294833

>>4294822
Actually Sigma used a very slightly smaller ASP-H sensor in the Quattro H in 2016. I actually think there could be an interesting use case for the Canon 1D mkii due to it's photo-sites being only 8.18 µm could be a good low light camera if it weren't for it's 1600 ISO cap. Maybe a good candidate for Astro?

cANON !!5Dz+C7v45HK

cANON !!5Dz+C7v45HK Fri 15 Mar 2024 04:29:51 No.4294832 Report

Quoted By: >>4294834 >>4294838

>>4294825
Photosite size is a spook, what matters is sensor size and then again only for base ISO. In reality, a good low light camera is one with a large equivalent aperture regardless of sensor size. APS-H's real advantage is extra reach compared to same MP FF.
>lots of APS-C lenses' image circle will cover an APS-H sensor
With a massive caveat though, in Canon's case they need to be EF mount rather than EF-S. That excludes most first-party APS-C lenses. Third party makers often used EF mount regardless of coverage so those can work.

Anonymous

Anonymous Fri 15 Mar 2024 04:32:10 No.4294833 Report

Quoted By:

>>4294825
>Actually Sigma used a very slightly smaller ASP-H sensor in the Quattro H in 2016
Technically yes, but Sigma bodies are pretty much irrelevant and it's also inferior to the Canon.
>the Canon 1D mkii due to it's photo-sites being only 8.18 µm could be a good low light camera
But it's also old as shit. The sensor is worse in every way compared to the IV and gets absolutely spanked by a recent full frame sensor.

Anonymous

Anonymous Fri 15 Mar 2024 04:34:58 No.4294834 Report

Quoted By: >>4294875

>>4294832
>a large equivalent aperture regardless of sensor size
Only depth of field scales with sensor size. The effect of exposure is exactly the same regardless of the sensor size.

>Canon's case they need to be EF mount rather than EF-S. That excludes most first-party APS-C lenses
That is true, I didn't think of that.

Anonymous

Anonymous Fri 15 Mar 2024 04:37:11 No.4294838 Report

Quoted By: >>4294839 >>4294875

>>4294832
>a larger photosite does NOT collect more photons and therefore less noise and more signal.
>but a larger sensor? oh yeah that has less noise because it gets more signal.
I fucking hate kasson and claff for putting this BS in idiots heads
>LOOK THE Z6 AND Z7 HAVE THE SAME SIGNAL TO NOISE RATIO!
>https://blog.kasson.com/nikon-z6-7/z6-and-z7-shadow-noise-at-nosebleed-isos/
The chroma noise is obviously way fucking worse.
>BUT IF YOU VIEW THEM AT THE SAME MAGNIFICATION LEVEL
>Color on the z7 is still way fucking worse
Same story for r6 vs r5. And yet DR CHART gearfags continue spreading this blatant fucking lie that anyone who used at least two cameras would already know is a blatant fucking lie.

Oh I know why. Because the "details" are still better through all the noise, details = sharpness and thats all these people care about. Small pixels = low ISO camera, sorry bitch.

Anonymous

Anonymous Fri 15 Mar 2024 04:38:23 No.4294839 Report

Quoted By:

>>4294838
based and true

big pixels are better in low light.
equivalence is not relevant outside of matching lines on a chart for nerd wank.

Anonymous

View Same Google ImgOps iqdb SauceNAO 1dvsa7r.png, 227KiB, 1189x2119

Anonymous Fri 15 Mar 2024 04:45:04 No.4294847 Report

Quoted By: >>4294850 >>4294854

Anonymous

Anonymous Fri 15 Mar 2024 04:52:55 No.4294850 Report

Quoted By:

>>4294847
Another thing to note with this, the 1D IV is 16mp, cropping the A7R to APS-H gets you like 24mp and even going down to APS-C is 18mp so there goes any reach advantage from the 1D

Anonymous

View Same Google ImgOps iqdb SauceNAO wojak-angry-soy-boy-holding-up-c (...).jpg, 158KiB, 1571x805

Anonymous Fri 15 Mar 2024 04:56:31 No.4294854 Report

Quoted By: >>4294859 >>4294875

>>4294847
This is you

Anonymous

Anonymous Fri 15 Mar 2024 05:06:24 No.4294859 Report

Quoted By: >>4294865

>>4294854
Are you arguing against the numbers? Are you going to telling me a modern FF sensor doesn't completely whoop a 15 year old APS-H sensor in every way?

Anonymous

Anonymous Fri 15 Mar 2024 05:13:08 No.4294865 Report

Quoted By:

>>4294859
>Are you arguing against the numbers?
no
>Are you going to telling me a modern FF sensor doesn't completely whoop a 15 year old APS-H sensor in every way?
no

All I am doing is simply pointing out what you are doing. It has nothing to do with whether or not you're correct.

cANON !!5Dz+C7v45HK

cANON !!5Dz+C7v45HK Fri 15 Mar 2024 05:17:35 No.4294875 Report

Quoted By: >>4294885

>>4294834
>Only depth of field scales with sensor size. The effect of exposure is exactly the same regardless of the sensor size.
No, SNR does as well. The game is counting photons, so if you collect the same you'll get similar results. The problem, and this is where FF gets an advantage, is well size in case you are filling them. But that only really applies to ISOs up to 200 in FF which are the ones MFT can't match. Basically with FF, because the same amount of photons is more spread out you can collect slightly more before your wells get filled. But again, that's only relevant at base ISO. When the MFT camera is blowing highlights and can no longer drop ISO because it's at the minimum value, the FF camera still has 2 more stops to drop. It's mostly edge cases but it does happen.
>>4294838
Anon, does it matter if you collect your photons across 1 or 4 photosites as long as you collect them all?
>Oh I know why. Because the "details" are still better through all the noise, details = sharpness and thats all these people care about. Small pixels = low ISO camera, sorry bitch.
a7RIII is proof that you're wrong, it outdoes many cameras with bigger pixels. And the a7SIII scamera is proof of the contrary, kek. 12MP, still gets mogged by higher resolution ones.
>>4294854
You got filtered by technical talk, lel. The chart is just a convenient representation of a measured quantity.

Anonymous

Anonymous Fri 15 Mar 2024 05:19:57 No.4294878 Report

Quoted By:

You need a full frame sensor for the best image quality

Anonymous

Anonymous Fri 15 Mar 2024 05:26:18 No.4294885 Report

Quoted By: >>4294895

>>4294875
>No, SNR does as well. The game is counting photons, so if you collect the same you'll get similar results
Except that the "extra photons" you're getting on the larger sensor is new information meaning it doesn't effect the exposure. Its like shining a flashlight on a postit note than swapping it out for a piece of printer paper, the larger paper will not be brighter. It will just capture more of the projection.

>You got filtered by technical talk, lel.
I didn't. I was just pointed out what he was doing and rhetorically asking if that's the kind of guy he wants to be.

cANON !!5Dz+C7v45HK

cANON !!5Dz+C7v45HK Fri 15 Mar 2024 05:38:50 No.4294895 Report

Quoted By: >>4294896

>>4294885
>Except that the "extra photons" you're getting on the larger sensor is new information meaning it doesn't effect the exposure. Its like shining a flashlight on a postit note than swapping it out for a piece of printer paper, the larger paper will not be brighter. It will just capture more of the projection.
Not when maintaining aperture and focal length equivalence. For ease of comparison's sake let's say you have a 42.5mm f/2 on MFT and you're at ISO 100 blowing highlights. The FF guy can shoot with his 85mm at f/4 and because f/4 is f/4 no matter what, he'll need ISO 400 to have an equivalent exposure to yours. But because he's at f/4 instead of f/2, his highlights aren't blowing as much. And even better, he can drop his ISO up to 2 stops. It's an edge case but it's an advantage of FF. It's not the most intuitive thing because both lux and lumen matter here.

Anonymous

Anonymous Fri 15 Mar 2024 05:49:23 No.4294896 Report

Quoted By: >>4294900

>>4294895
I'm pretty sure you're agreeing with me. But you can try this for yourself if you have a FF lens and body and another crop body, at a given aperture the exposures are the same. Because the aperture size doesn't change, it's a fixed size regardless of sensor size. So the effectively longer focal length of smaller sensors only changes the depth of field. It's physically the same amount of light being let in.

cANON !!5Dz+C7v45HK

cANON !!5Dz+C7v45HK Fri 15 Mar 2024 06:01:18 No.4294900 Report

Quoted By: >>4294901

>>4294896
No anon, the same f-number (relative aperture) doesn't lead to the same aperture size in lenses of equivalent lengths. That's why the Panaleica 42.5mm/1.2 is the same size of a 85mm/2.4.
The f-number is the divider in the formula for absolute aperture size.
42.5mm/1.2=35.41666666mm aperture (entrance pupil)
85mm/2.4=35.41666666mm aperture (entrance pupil)

Anonymous

Anonymous Fri 15 Mar 2024 06:06:54 No.4294901 Report

Quoted By: >>4294903

>>4294900
We're talking about different things, but we ultimately agree. I'm talking about moving a lens from one camera to another, and you're talking about equivalence, hence the confusion, but we fundamentally agree.

cANON !!5Dz+C7v45HK

cANON !!5Dz+C7v45HK Fri 15 Mar 2024 06:13:43 No.4294903 Report

Quoted By: >>4294904

>>4294901
I think I know what you mean now, which is what I use to increase reach. Basically the MFT-sized crop of a FF sensor will be virtually identical to the MFT image if using the same settings and the same lens across the two cameras, even noise-wise. But the denser sensor will get more reach. The denser sensor is usually the smaller one but it's not always the case (in fact it's what kept me from buying an Olympus E-5, it gave me less reach than a teleconverter on full frame).
It's also what some people on this board (fanatics of their camera of choice who feel the need to believe their camera is the greatest at everything) refuse to acknowledge when comparing the a7RIV to the EOS 90D: the latter has more reach and is superior for wildlife.

Anonymous

Anonymous Fri 15 Mar 2024 06:19:05 No.4294904 Report

Quoted By: >>4294905

>>4294903
What do you mean by a "denser senor"? Are you talking about pixel density? Because yes, if you have a 20mp MTF and a 30MP MFT-sized crop of a FF sensor you would obviously have more effective reach with the full frame camera

cANON !!5Dz+C7v45HK

cANON !!5Dz+C7v45HK Fri 15 Mar 2024 06:28:33 No.4294905 Report

Quoted By: >>4294907

>>4294904
That's exactly what I mean. It seems obvious but over the years people have been taught a lot of stuff that is simply incorrect, like crop factor being "magnification" so they become confused by this stuff.

Anonymous

Anonymous Fri 15 Mar 2024 06:30:47 No.4294907 Report

Quoted By: >>4294920

>>4294905
But that doesn't hold up with the a7riv and the 90d. The a7riv has a 61mp sensor. 61/1.5 = 40.667MP. The 90D is 32.5MP. the APS-C area of the Sony has more megapixels. Did you mean the a7riii?

cANON !!5Dz+C7v45HK

cANON !!5Dz+C7v45HK Fri 15 Mar 2024 07:04:53 No.4294920 Report

Quoted By: >>4294922

>>4294907
Canon is 1.6x, also you're dividing an area quantity by a linear one. 61MP/1.6/1.6 is 23.82, which gives you roughly the 80D (unsurprising given that they have almost the same pixel size).

Anonymous

Anonymous Fri 15 Mar 2024 07:17:39 No.4294922 Report

Quoted By: >>4294927 >>4296685

>>4294920
>you're dividing an area quantity by a linear one. 61MP/1.6/1.6
I'll have to think about this, I'm not sure this setup is correct. Because we're not trying to figure out the new area, we're just interested in the number of pixels that remains. If Canon's crop is 1.6. That means the sensor is 60% smaller or 40% of the area of FF. So you would only need to reduce the number of MP by 60%, not 60% twice. If you wanted to caluclate the dimensions of the sensor, you would reduce both the width and length by that amount.

Anonymous

Anonymous Fri 15 Mar 2024 07:49:37 No.4294927 Report

Quoted By:

>>4294922
That's how it works. MFT has a crop factor of 2 because the linear dimensions are approx half of FF (discounting aspect ratio differences). I.e. you could fit 2 MFT lengths on a FF length and ditto for height. With that in mind, if you overlay a MFT sensor on a FF sensor, you realize that you could fit 4 MFT sensors on a FF sensor.

Anonymous

Anonymous Wed 20 Mar 2024 19:45:07 No.4296685 Report

Quoted By:

>>4294922
>you would reduce both the width and length by that amount.
And the area (which governs the number of pixels) is length*width. That's where the "reduce by 60% twice" comes from. Once for length, and once for width.

Anonymous

Anonymous Wed 20 Mar 2024 20:07:50 No.4296694 Report

Quoted By:

I have an M43 camera exclusively for video, and I'm using a speedbooster so I can use lenses from my FF collection. My particular speedbooster cuts the crop ratio from 2x to about 1.28x, which is almost APS-H lens coverage (but it's still being recorded on fucking M43, lol).

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

APS-H