Useable Megapixels? - DPChallenge Forums

Nikon AF Zoom-Nikkor 18-135mm f/3.5- 5.6G ED DX

04/11/2009 12:39:44 AM · #1

An observation. While there are standard/expected conversions for megapixels to useable print sizes, such as 6 megapixels typically equals a 12 x 18 inch print, I've noticed my 40D doesn;t really give me 10 megapixel useable prints. In other words, while it gives me 10 mp RAW files, I can't print that file. I have to convert it to JPEG which leaves me typically with a 7mp file.

For all practical puposes I'm getting a 7mp file and its corresponding print sizes.

Is my understanding of this wrong. If so, how?

Its a little more than an academic concern for me as I'm wanting to print 4 prints at 24 x 36 or greater print sizes and I'm not sure I can really do that with 7mp jpegs.

Message edited by author 2009-04-11 00:40:58.

JeffryZ

Nikon D80

04/11/2009 12:48:03 AM · #2

I think you may be confusing megapixels which is the total count of the dots making up the image with megabytes which is the size of the file telling each of those dots what color and brightness to be. In Photoshop or whatever editing program you are using, try going to "resize image" and see what the dot dimentions of the photo are. Multiply the width by the height and that will tell you how many pixels you have. Unless you have cropped the image, it should still have the same total number of pixels.

Nikon AF Zoom-Nikkor 18-135mm f/3.5- 5.6G ED DX

04/11/2009 12:52:40 AM · #3

Yes and no. I should have referenced MBs not MPs. But, I still think the observation is correct if you switch the references to megapixels to megabytes or mps to mbs.

JeffryZ

Nikon D80

04/11/2009 12:57:17 AM · #4

A little information about print sizes:
//www.bythom.com/printsizes.htm

lovethelight

Canon EOS-5D Mark III

Canon EF 135mm f/2.0L USM

04/11/2009 01:02:12 AM · #5

Originally posted by Yandrosxx:

Yes and no. I should have referenced MBs not MPs. But, I still think the observation is correct if you switch the references to megapixels to megabytes or mps to mbs.

If you want nice really large prints I would try printing using TIFF files. Instead of converting raw to jpg convert it to TIFF and edit/print that. A lot of generic printing places might not take TIFFs but worthwhile printing places will.

Yo_Spiff

Olympus OM-D E-M5 Mark III

Olympus 12-100mm f/4.0 M.Zuiko ED IS PRO

04/11/2009 01:02:39 AM · #6

There is not a direct correlation between Megabytes and Megapixels. While they do increase in size together, you cannot determine what one is going to be based on the other. A highly compressed JPEG from your 10 Megapixel camera, still has 10 megapixels in it, as does the RAW file. What the RAW file has is way more information on the colors within the image. The JPEG has thrown out tonal data that is (in theory) not seen in order to save space.

After reading your original post and your second response, I'm not quite sure what the question really is. Are you talking file sizes, or the number of dots that make up the image?

As previously mentioned, TIF files do not have that lossy compression, though if you have an 8 bit TIFF, it will toss out some tonal data. Graphic professionals prefer TIF.

Message edited by author 2009-04-11 01:06:06.

04/11/2009 09:38:57 AM · #7

That's what I got confused about. I figured it out (I think) after I posted this. I confused 10.1 megapixel resoltuion with 10 megabytes in data because the resulting RAW file I get is typically 10 mb file. I equated the two in my mind.

In truth 10.1 megapixels deals with the resulting resolution of the picture typically measured in vertical and horizontal pixel dimensions. The more megapixels the bigger the physical pixel dimensions of the resulting image. Coupling those dimensions with DPI and that determines the size (essentially) of the maximum useable print size?

Correct?

In short a 6 megapixel camera gives you smaller vertical and horizontal pixel dimensions than a 10.1 megapixel camera. Thus, it really doesn't matter what the mb size of the resulting RAW or JPEG file is because that has nothing to do with resulting pixel dimensions of the image. Obviously smaller pixel dimensions is going to use result in smaller mbs such as when you crop a photo, but that's not an issue concerning determining the resolution of the camera and its maximum useable print sizes.

If that's correct then I think I understand why chasing pixels at some point won't matter because unless you want to make billboards at some point you really don't need larger and larger megapixel cameras????

Message edited by author 2009-04-11 09:43:10.

Nikon AF-S NIKKOR 24-70mm f/2.8G ED

04/11/2009 09:51:48 AM · #8

I do from time to time convert to tif, but usually avoid it because of the resulting file size and because the local print shops can't print from it. I'm involved in an industry that puts me in contact with larger commercial printers from to time but I typically only use them to print plans. Guess, I need to have a conversation about printing large prints for me.

Covert_Oddity

Nikon D700

04/11/2009 10:09:25 AM · #9

Originally posted by Yandrosxx:

If that's correct then I think I understand why chasing pixels at some point won't matter because unless you want to make billboards at some point you really don't need larger and larger megapixel cameras????

There is a point where this is true, but those extra megapixels come in handy when you want to crop an image and still achieve a print quality finished shot. When I was using my 6MP D70 I could hardly crop at all if I wanted to keep a print quality image, but with my 12MP D700 I can crop quite a small area from the original shot and still achieve a printable finished shot.

There are other considerations to make when doing this though, if your focus isn't absolutely sharp cropping shows up the softness.

Canon RF 24-70mm f/2.8 L IS

04/11/2009 10:38:15 AM · #10

If your intent is to print very large, you should also consider typical viewing distance. Anyone looking at a billboard, would typically step back far enough so he can see the whole bloody thing. Sharpness, obviously, is less of an issue from six feet away than it would be up close.

An aside: whenever I want to print larger than 16x20, I try to start with a minimum file size of 45MB. My print-ready, edited files, usually, are between 45MB and 74MB.

kirbic

Canon EOS R5

04/11/2009 10:52:23 AM · #11

Originally posted by zeuszen:

...An aside: whenever I want to print larger than 16x20, I try to start with a minimum file size of 45MB. My print-ready, edited files, usually, are between 45MB and 74MB.

I take it you're going to print with TIFF files, then? For JPEG, the file size is *so*( dependent on image content, noise levels and compression settings that it's impossible to predict the output file size just based on required print dimension

PGerst

Canon EOS-20D

04/11/2009 10:52:34 AM · #12

Here is a chart from West Coast Imaging:
Pixel Chart

And if you click here and then click on MegaPixel to Print Size Chart, you'll see the above image, plus some additional information.

04/11/2009 10:58:24 AM · #13

Since I tend to crop a lot, I suspect I'm better offer looking at pixel dimensions to determine what the maximum useable print of the image is likely to be.

Canon RF 24-70mm f/2.8 L IS

04/11/2009 11:05:27 AM · #14

Next question, is the maximum DPI limited by the the total megapixels of the image? So the more megapixels you have the greater number of dots per inch that can be used in the final print. That's what I'm gathering from the links psoted here.

Seems to me that's the real limitation on useable print sizes that occurs as a result of limitations on megapixels?

Message edited by author 2009-04-11 11:06:41.

kirbic

Canon EOS R5

04/11/2009 11:19:26 AM · #15

Originally posted by Yandrosxx:

Yes, that's true. Take the image width in pixels (3888px for the 40D) and divide by the print dimension. Example:

You're planning to make an 18x24 inch print from an uncropped 40D image. 3888px divided by 24 inches gives 162 px/inch (ppi). If the image you are starting with is sharp, that's enough for a good quality print (when viewed from normal viewing distances).

Nikon AF-S Nikkor 28-300mm f3.5-5.6G ED VR

04/11/2009 11:21:46 AM · #16

And truthfully, considering that more megapixels equals greater flexibility toward cropping an image that you can still make a decent print of dispels the notion that more megapixels don't really matter.

I usually tend to get the image I want when I shoot it, but I always crop a little and I sometimes crop significantly. That's fine for web images, but not for prints and I'm starting to print many of those images. I just discovered one of my favorite images I've ever created was cropped significantly from the original and it was originally shot on a 6 mp camera.

In other words, after this discussion I really doubt I can print it to 24 x 36 + print. Maybe with upsizing, but the problem highlights the point.

Greater megapixels leads to greater flexibility in some applications. Might not matter to a novice or amateur, but I can certainly see why it matters to pros.

Message edited by author 2009-04-11 11:22:16.

NikonJeb

Nikon D850

04/11/2009 11:55:20 AM · #17

And the MP count doesn't mean doodly-squat if the image is a bad image.....OOF, blown out, underexposed.....

I have an image that came from my 6MP D70s that looks absolutely fabulous at 6 feet by 11 feet.

My printer couldn't believe it, nor I for that matter, and I'd have to say I believe it to be a fluke.

But I also have 8x10s I'm unhappy with because of an error, or seven, on my part that just made them crappy images.

I am interested in a newer, more MP camera because I have been getting requests for larger images, but I'm interested moreso in the better quality that the later Nikons have in their higher ISOs more than the more MP count.

The higher MP is just a bonus.....8>)

As an aside, I think it's funny when some people see my images and say things like "Wow! You must have like a jillion mega-pixel camera!!!!": and then watch their jaw drop when I say "Nope! Just a 6!".

ETA: I also crop quite a few of mine pretty hard, and I print mostly 8x10; I'm very happy with the majority of what I get.

Message edited by author 2009-04-11 11:57:18.

04/11/2009 12:12:05 PM · #18

Originally posted by kirbic:

Originally posted by zeuszen:
...An aside: whenever I want to print larger than 16x20, I try to start with a minimum file size of 45MB. My print-ready, edited files, usually, are between 45MB and 74MB.

I take it you're going to print with TIFF files, then? For JPEG, the file size is *so*( dependent on image content, noise levels and compression settings that it's impossible to predict the output file size just based on required print dimension

Yes, they're 8-bit TIFFs. I'm contemplating 16-bit, but most of those I'd have to re-convert from 8-bit to 16. I'm not sure, if there would be much to gain, unless they stayed in 16-bit mode throughout the process... Any thoughts, Fritz?

Canon EF 17-40mm f/4.0L USM

04/11/2009 12:16:08 PM · #19

Originally posted by Yandrosxx:

... after this discussion I really doubt I can print it to 24 x 36 + print. Maybe with upsizing, but the problem highlights the point...

If you have to upsize, do it in increments of 10%, i.e. 110%, then 120% and so on.

soup

Canon EOS-30D

04/11/2009 12:43:09 PM · #20

you've got people requesting prints larger than 6' x 11' ?

WOW !!! - what do you charge for one of those ?

:)

Originally posted by NikonJeb:

I have an image that came from my 6MP D70s that looks absolutely fabulous at 6 feet by 11 feet.

My printer couldn't believe it, nor I for that matter, and I'd have to say I believe it to be a fluke.

But I also have 8x10s I'm unhappy with because of an error, or seven, on my part that just made them crappy images.

I am interested in a newer, more MP camera because I have been getting requests for larger images, but I'm interested moreso in the better quality that the later Nikons have in their higher ISOs more than the more MP count.

The higher MP is just a bonus.....8>)

Message edited by author 2009-04-11 12:43:27.

kirbic

Canon EOS R5

Canon RF 24-70mm f/2.8 L IS

04/11/2009 12:45:26 PM · #21

Originally posted by zeuszen:

Even with a complete 16-bit workflow, almost nothing to gain, methinks. The data streams sent to most printing devices are 8-bits/channel anyhow, and it's better that you control the conversion to an 8-bit space. 16-bit entails double the file size, a huge penalty even in these days of cheap storage. For images already in 8-bit, absolutely nothing to gain.
Now that said, I'd never throw away the high-bit "originals" of my images, I just don't see the need to go to print with high-bit images.

soup

Canon EOS-30D

Canon EF 17-40mm f/4.0L USM

04/11/2009 12:52:58 PM · #22

ONE thing to gain with a 16bit workflow is - as you said - a 16bit original file - that has the potential to produce a better print output than its 8bit sibling - if and when - the printers start to accept a 16bit file.

and with a 32bit workflow starting to come about - methinks that day isn't too far into the future.

Originally posted by kirbic:

Even with a complete 16-bit workflow, almost nothing to gain, methinks. The data streams sent to most printing devices are 8-bits/channel anyhow, and it's better that you control the conversion to an 8-bit space. 16-bit entails double the file size, a huge penalty even in these days of cheap storage. For images already in 8-bit, absolutely nothing to gain.
Now that said, I'd never throw away the high-bit "originals" of my images, I just don't see the need to go to print with high-bit images.

Message edited by author 2009-04-11 12:53:05.

Yo_Spiff

Olympus OM-D E-M5 Mark III

Olympus 12-100mm f/4.0 M.Zuiko ED IS PRO

04/11/2009 02:04:40 PM · #23

DPI is a point of regular confusion. Printers and graphics people regularly say they need a 300 dpi file, but often fail to mention the needed print size. It came up the other day in this thread.

DPI is not the size of the image, but rather a measure of pixel density. It really does not make a difference until you are talking about printing or scanning.

Message edited by author 2009-04-11 15:15:42.