5D Mk II picture size cropped down to 50D size comparison?

I dont really know how to ask this but I remember seeing a post similiar to this idea.
Im still not sure if crop sensors give more reach such as a 100mm lense giving the reach of a 160mm lense, but assuming they do if you were to crop down a 5DII pic to the normal pic size of the 50D would it give the reach of the 50D?

SO would the 100mm lense on the 5dII be like a 160mm lense on a 50D if the 5dII pic is cropped down to 50D size?

Hopefully this makes sense.

No.

a 5Dii file cropped to 1.6 size is roughly 8mp, so 20D/30D size.
The 50D has roughly twice the pixels.

Thanks that was exactly what I was looking for.
So one more quick question how many mp on a ff camera would you need to equal the 50D size? Would around 40mp be correct?

A 1.6x camera has 39% of the area of a full frame camera.

So if you're cropping a FF image down to 1.6x size, you'll be left with 39% of the pixels you started with.

So if you start with 21.1 megapixels, you'll be left with 8.2 megapixels.

You would need 38.7 megapixels in a FF camera to crop down to 15.1.

Editorial comment: The extra pixel density of of the 50D compared to the 30D will rarely get you more detail in an image. There are too many other factors that limit image quality. I have a half-dozen 24 x 36 prints on my office wall taken with a 20D that have more detail than you can see at normal viewing distance.

No, but next to a 30D, yeah!

Thanks again!

Should i crop my H3DII-39mp to see how much it will equivalent to 1.6x 35mm DSLR size? What do you think it will be?

What is the sensor size of the H3DII?

Curtis N wrote in post #8418417
What is the sensor size of the H3DII?

I dunno, maybe 36.8x49.1mm for 39mp

Okay, I've been sort of working on this concept in my head and I'm coming up with some different conclusions than others on this issue.

The comparisons I will make will take into account the following specifications:

Camera 5D MkII
max pixel size 5616 x 3744
image sensor size 36 x 24 mm

Camera 50D
max pixel size 4752 x 3168
image sensor size 22.3 x 14.9 mm

Here is my thinking. If I put a 100 mm focal length lens on the 5D, I get an image that is 5616 x 3744 pixels. That image is comprised of most of the image circle created by the lens (this is the point of the full frame camera with an EF lens attached). If I were to crop out of this image the pixels that would be the equivalent of a 50D image with the same lens, then that image would be 3479 x 2324 pixels. Here is the math I used to arrive at that statement:

22.3 mm / 36 mm = 0.6194 (this is the ratio of the full frame sensor size to the APS-C sensor size in the horizontal dimension)
14.9 mm / 24 mm = 0.6208 (ratio of full frame sensor size to APS-C sensor size in the veritical dimension)

Now, multiple the pixel count of the full frame sensor by the appropriate ratio and you get

5616 * 0.6194 = 3479 pixels for the horizontal and
3744 * 0.6208 = 2324 pixels for the vertical dimension

This tells me that if I crop out of a 5D MkII image the same size frame that an APS-C sensor would see using the same 100 mm lens, this is the image that I would have left, 3479 x 2324 pixels.

Hypothetically, if I take the same lens and shoot an image of exactly the same scene (think of the lens as being mounted on a tripod and the 5D detached and the 50D put in it's place), I would have an image that was the equivalent of approximately 40% of the 5D's image, or the same field of view as the 3479 x 2324 pixel crop I made of the 5D MkII image. However, the image that I make with the 50D has 4752 x 3168 pixels. So, when I view both images at the same resolution, say 240 pixels per inch, the cropped image from the 5D MkII is 14.50 inches x 9.68 inches. The image from the 50D is 19.80 inches x 13.2 inches. And both of these images should show exactly the same scene!

Coming at this from another direction, If I take a shot with both a 5D MkII and a 50D with the same lens mounted on a tripod (like in the above example), the 5D MkII image will have a certain field of view and the 50D image will have a correspondingly smaller field of view. But since the approximately 40% field of view of the 50D has more pixels concentrated on that part of the image, when viewed at the same resolution, the 50D image will appear to be "closer" to the viewer. That part of the 5D MkII image that corresponds to the 50D image will have less detail just by virtue of the fact that more pixels were trained on that part of the image.

I'm sure that I have made some sort of error in my methodology here, so will someone please point that out? I "know" that a 100 mm lens is a 100 mm lens no matter what body you put on it, but something has to account for the apparent increase in lens focal length when taking the same image from the same location using a full frame sensor camera and an APS-C sized sensor camera.

Sorry for being so long winded.

•The math looks fine.
•Viewing images from the two cameras at 240 ppi means that the geometric enlargement is greater for the 50D image (50D goes from 210 pixels per mm to ~10 pixels per mm or 21x, vs 156 pixels per mm to ~10, or 16x)
*The angle of view of a given focal length is determined by the sensor dimensions, so change sensor and get different AOV from the same FL.
*This is often perceived as a change in the so-called "apparent" focal length because the speaker is used to a different sensor dimension.

MarKap77 wrote in post #8418656
Okay, I've been sort of working on this concept in my head and I'm coming up with some different conclusions than others on this issue.

The comparisons I will make will take into account the following specifications:

Camera 5D MkII
max pixel size 5616 x 3744
image sensor size 36 x 24 mm

Camera 50D
max pixel size 4752 x 3168
image sensor size 22.3 x 14.9 mm

Here is my thinking. If I put a 100 mm focal length lens on the 5D, I get an image that is 5616 x 3744 pixels. That image is comprised of most of the image circle created by the lens (this is the point of the full frame camera with an EF lens attached). If I were to crop out of this image the pixels that would be the equivalent of a 50D image with the same lens, then that image would be 3479 x 2324 pixels. Here is the math I used to arrive at that statement:

22.3 mm / 36 mm = 0.6194 (this is the ratio of the full frame sensor size to the APS-C sensor size in the horizontal dimension)
14.9 mm / 24 mm = 0.6208 (ratio of full frame sensor size to APS-C sensor size in the veritical dimension)

Now, multiple the pixel count of the full frame sensor by the appropriate ratio and you get

5616 * 0.6194 = 3479 pixels for the horizontal and
3744 * 0.6208 = 2324 pixels for the vertical dimension

This tells me that if I crop out of a 5D MkII image the same size frame that an APS-C sensor would see using the same 100 mm lens, this is the image that I would have left, 3479 x 2324 pixels.

Hypothetically, if I take the same lens and shoot an image of exactly the same scene (think of the lens as being mounted on a tripod and the 5D detached and the 50D put in it's place), I would have an image that was the equivalent of approximately 40% of the 5D's image, or the same field of view as the 3479 x 2324 pixel crop I made of the 5D MkII image. However, the image that I make with the 50D has 4752 x 3168 pixels. So, when I view both images at the same resolution, say 240 pixels per inch, the cropped image from the 5D MkII is 14.50 inches x 9.68 inches. The image from the 50D is 19.80 inches x 13.2 inches. And both of these images should show exactly the same scene!

Coming at this from another direction, If I take a shot with both a 5D MkII and a 50D with the same lens mounted on a tripod (like in the above example), the 5D MkII image will have a certain field of view and the 50D image will have a correspondingly smaller field of view. But since the approximately 40% field of view of the 50D has more pixels concentrated on that part of the image, when viewed at the same resolution, the 50D image will appear to be "closer" to the viewer. That part of the 5D MkII image that corresponds to the 50D image will have less detail just by virtue of the fact that more pixels were trained on that part of the image.

I'm sure that I have made some sort of error in my methodology here, so will someone please point that out? I "know" that a 100 mm lens is a 100 mm lens no matter what body you put on it, but something has to account for the apparent increase in lens focal length when taking the same image from the same location using a full frame sensor camera and an APS-C sized sensor camera.

Sorry for being so long winded.

The higher pixel density of the 50D will theoretically give you more detail compared to the same FOV cropped from a 5DII image if those pixels are of sufficient quality and the lens used can resolve that level of detail .You're assuming higher density equates to higher detail which is not necessarily true.

Say you had a lens that gave you a mosaic effect similar to the Photoshop filter and that mosaic had exactly as many squares as the 5DII. If you then put that lens on the 50D its higher density would give you no more detail because the lens was a limiting factor.

AJSJones wrote in post #8418800
•The math looks fine.
•Viewing images from the two cameras at 240 ppi means that the geometric enlargement is greater for the 50D image (50D goes from 210 pixels per mm to ~10 pixels per mm or 21x, vs 156 pixels per mm to ~10, or 16x)

Sorry, but this makes no sense to me. Can you elaborate on this?

Hmm, I think you have this wrong. The angle of view of any lens is a fixed value. When using a hypothetical 100 mm lens on a full frame sensor camera, the portion of the image cast by the lens that falls on the image sensor is a certain part of the image circle, that being a 3:2 rectangle where the corners are just shy of touching the edges of the image circle. When using an APS-C sized sensor, you are viewing a smaller portion of the image circle than that viewed by a full frame sensor, hence the field of view being "cropped".

It is perceived as an increase in focal length for the reasons I stated, because the image that is captured on the APS-C sensor has so many more pixels concentrated on that portion of the image, it "appears" as though the image was captured by a lens of longer focal length than was actually used.

Regards

MarKap77 wrote in post #8418935
The angle of view of any lens is a fixed value.

Incorrect.

The image circle created by a lens is always the same, regardless what camera it's mounted on. But the angle of view is limited by the film/sensor that captures the image, which is only part of that image circle.

I think you're overanalyzing this.

Over the years, Canon has made cameras ranging from 4MP to 15MP with essentially the same sensor size. As technology has advanced, pixel size has gotten smaller, pixel density higher and megapixels higher.

The amount of detail that can be captured by a camera is dependent on many factors. Pixel density is only one of them.