Print size question

[DynaSport]

Thanks guys. It’s not that I don’t appreciate the answers, but my communication skills must be lacking, or most don’t read, or both. I’m familiar with how to calculate largest size to print based on the height and width of a photo in megapixels. What I was trying to ask is that since full frame pixels are larger than aps-c megapixels and so on, if that size difference made a mathematical difference in the size of print that could be printed. Unfortunately I was short on sleep and had a nice relaxing bourbon or I wouldn’t have asked the question in the first place. I know that the pixel collects electronic data and passes it along to be combined with the electronic data of other pixels to form an electronic image. So, regardless of the physical size of the pixel, the size of the data it passes along is the same. That’s all. Thanks again.

 

[davidzvi]

Another reason I don't drink bourbon.

 

[DynaSport]

Another reason I don't drink bourbon.

I only had a tad left. But I decided to give it up for a while too. I bought Tennessee whiskey instead.

 

[bye]

Thanks guys. It’s not that I don’t appreciate the answers, but my communication skills must be lacking, or most don’t read, or both. I’m familiar with how to calculate largest size to print based on the height and width of a photo in megapixels. What I was trying to ask is that since full frame pixels are larger than aps-c megapixels and so on, if that size difference made a mathematical difference in the size of print that could be printed. Unfortunately I was short on sleep and had a nice relaxing bourbon or I wouldn’t have asked the question in the first place. I know that the pixel collects electronic data and passes it along to be combined with the electronic data of other pixels to form an electronic image. So, regardless of the physical size of the pixel, the size of the data it passes along is the same. That’s all. Thanks again.

Dynasport,

The short answer is yes. Given the same megapixels, a larger sensor can allow larger sizes to be printed compared to a smaller sensor of the same megapixel. I asked Bill Claff about the D3 vs D300, because I used to own them in the past and I noticed the full frame sensor will tolerate much larger prints with my D3 than with my D300 (despite both being 12MP). Basically, it has to do with SNR (Signal Noise Ratio) and also the medium you are looking at. With prints, Print Dynamic Range (PDR) scales up as your sensor size increases. Below is the explanation.

Subscribe to see EXIF info for this image (if available)

 

[jbinco]

Viewing distance needs to be taken into account when determining print size file resolution. For a whole variety of often arcane and likely debatable reasons, the "generally accepted" optimal viewing resolution is 227dpi at ~ 2feet (otherwise known as the length of the "average" human arm). Usually I go with 300dpi for anything I expect to be viewed from "arms length": 8x10 or smaller. Any print larger than ~8x10 is likely going to be viewed from further away and therefore requires less resolution to satisfy the "average" human eye.

Bottom line is that an 8x10 inch print at 300dpi = 2400x3000 points = 7.2 megapixels. Anything over 7.2 megapixels is being downsized and will not contribute to the final print output. Think that sounds silly? Well I was selling 40 inch prints from a 6 mp camera 17 years ago.

A 20 megapixel E-M1 II sensor with 5184 x 3888 pixels printed at 300 dpi would render at 17 x 13 inches. At 227 dpi you'd get a 23 x 17 print. At 150 dpi a 34 x 26 print.

Does each individual pixel matter? Yes. Insofar as each sensor pixel captures a certain depth of color (or gray scale shading). This matters if you are capturing raw images and printing 10 to 16 bit files through a RIP. If you are capturing JPEG or printing from JPEG you only have 8-bit files so the tonal range is not going to differ between a pixel captured in a medium format back or a m43 camera. If your entire workflow is greater than 8-bit (raw capture, 16-bit TIF or PS editing, 16-bit RIP) then each individual pixel can make a difference even if the tonal range of the printer is more like 8-bit. This is because the larger dataset of each pixel's tonal value can provide more subtle linearization in the print output. You can achieve the same thing (to a degree) in-camera or post-process when the raw image data is rendered to JPEG assuming that the JPEG suffers no further degradation at print rendering.