Understanding Digital terms using 35mm as your base: What is the equivalent of 35mm color slide film in digital? Where do those pixels come from?

What mega pixel camera can produce equivalent of a 35mm film? What mega pixel camera enlarges as well as a 35mm color film?

Let us work backwards using a Nikon coolscan LS-10 slide scanner that scans around 2700 dpi from a 35mm slide film measuring 24mm x 36mm (.94 x 1.42 inches). 

A 2700 dpi or ppi (pixels Per inch) produces 2700 X 1.42 = 3834 pixels from the width of a slide film (1.42 = 36mm) and it produces 2700 X .94 = 2538 pixels from its height. 

Now let’s multiply 3834 BY 2538 (3834 x 2538 = 9,730,692 OR approximately 10,000,000 which is 10 mega pixels. This means the entire surface of the 35mm film will produce about 10,000,000 pixels or it takes 10,000,000 pixels to reconstruct the image accurately in the digital world.

Since each COLOR pixel is represented by 3 primary colors (Red, Green, and Blue), then by multiplying 10,000,000 by 3 we get about 30,0000,000 characters or 30 Mega Bytes. This is the amount of storage needed to keep track of the color info for the image (each pixel has a 3 character overhead to remember its color/density attributes). Therefore it takes a 30 mega byte uncompressed TIFF (.tif) file to accommodate one scanned 35mm slide film @ 2700 dpi.
Practically, a slide film will generate a tiff file that is about 27 to 29 Mega Bytes in size. Since due to the corners and mount overlap a full frame is usually not scanned.

If you have a black and white negative (grayscale/No color), then the size of the file will only be 10,000,000 (10 mega bytes) since the density of each grayscale pixel is represented by only ONE informational byte or character.

Let’s assume that a typical 100ISO slide film (measuring 24mm X 36mm) has a resolution of about 107 lines/mm. 

Please note that this number could be more (up to 120) or less (say 80) for this type of film. A resolution of 100 to 110 lines/mm is typical. This number can be found in the technical specifications for a slide film from a manufacturer like Kodak. 

Let's draw 107 lines, alternating between black and white as shown below:

and 107 horizontal lines as illustrated below:

it will produce 107X107 distinct pixels (Picture Element: The smallest building block of any digital image) for each sq. mm. of the film.

This means the entire film produces (24X107) x (36X107) = 9.9 million Pixels.
Since it takes 3 colors (Red, Green, and Blue) to represent ONE color pixel, then, we need about 9.9 X 3 = 29.7 Mega bytes (characters) or about 30 Mega Bytes of storage when the file is created in tiff format.
This is to accommodate / keep track of the color information for these pixels (as was mentioned each color pixel has a 3 byte/character overhead) for a scanned 35mm color slide film at 2700 DPI.
This size is closely represented by the size of the scanned image saved in the form of uncompressed Tiff (.tif/Tagged Image File) format.

Please note that it takes about 21 floppies to accommodate this image or one 700MB CD to accommodate about 23 of these images. By the same reasoning, a 4.7GB DVD Rom can accommodate about 156 of these images.

A 6 mega pixel D Slr with an aspect ratio of 2/3 produces 3000 X 2000 = 6,000,000 pixel image. 

An 8 mega Pixel D-Slr camera with an aspect ratio of 2/3 will produce 3464 X 2310 = 8,000,000 pixel image.

A 35mm slide film with an aspect ratio of 2/3, scanned at 2700 dpi produces (1.42 X 2700 = 3834) by (o.94 X 2700 = 2538) or 3834 X 2538 = 9,730,692 or 10,000,000(approx)

What happens if one scans a 35mm slide film with the above specifications at 5400 dpi/ppi?

Not much. The scanner will enlarge the pixels (creates fake pixels) using a process called interpolation. It WILL NOT enhance the quality of the scan. A practical example of this is if you have 2700 dollars and the car you want to buy costs $5400 and you simply can double your wealth using a mirror in front of the original $2700! True, you will see $5400 but the image you see in the mirror is just an image and has no real value.