In the workflow settings you can change the PPI of the image opened in Photoshop. BTW, the 200 PPI has nothing to do with MP. Your camera captures an image that is a certain pixel dimension by a certain pixel dimension but with no specific PPI. For print output you probably want 300 PPI. You can resize without resampling in Photoshop. But if you need web output, you'll want to resample and have the final resolution be 72 PPI.
What I guess I am trying to figure out, I can change the 21 mp camera into a 50 mp in ACRaw,, it still looks very nice, pixels are finer,, I am just wondering if the extra pixels are added, if they "borrow/copy" the closest information on color, depth, to make it a 50mp file? if you zoom down to 100 percent or more, where you see the actual pixels form, and then just back out where they come together,, it looks great.
Then also, with importing the files from my camera, a 21 mp into my computer, at resolution 72 in raw (the resolution i think when i looked at the raw file), I get a print that is double in size than importing the file in at 2-300 dpi then i can raise the resolution in photoshop I still have the same file of a 21 mp, resolution 266, and matin the file of the picture that is double the size than importing it at the native 72 ppi
1 person found this helpful
So you’ve discovered upsizing. And the fact that some software uses the DPI of a file to also upsize. The native pixels of the camera have not changed. Many different photo processing programs allow resizing. What is the amazing thing about ACR doing it in your mind, or what part of what’s happening don’t you understand? You’re not somehow getting a photo taken from a higher resolution sensor so there’s no any more detail in it than the original photo.
what is the native resolution of a 21 or a 36 mp camera's sensor?
1 person found this helpful
You can look at the camera specs to find the exact pixel dimensions but some algebra on two equations that define what megapixels and aspect ratio are gives a good sense of what size numbers we're dealing with:
W x H = megapixels (megapixels approximately equals million pixels squared)
W / H = aspect ratio -> W = H x aspect ratio << solving for W
(H x aspect ratio) x H = megapixels << substituting the second equation for W into the first equation in place of W
H x H x aspect ratio = megapixels << rearranging
H x H = megapixels / aspect ratio << dividing both sides by aspect ratio
H = sqrt ( megapixels / aspect ratio ) << sqrt of both sides
W = H x aspect ratio << 2nd equation at the beginning
Substituting your specific numbers and assuming a 3:2 = 1.5 aspect ratio:
H = sqrt (21-million / 1.5) = 3741.7
W = 3741,7 x 1.5 = 5612.5
H = sqrt(36-million / 1.5) = 4899.0
W = 4899.0 x 1.5 = 7348.5
These width and height numbers are approximate because the megapixel numbers are only approximate to begin with, being rounded to only 2 significant digits, so 5 digits are unwarranted, and the "mega" in megapixels is actually 1024x1024 not 1000x1000 = 1-million, but you can get a sense of what native pixel dimensions of a sensor would result in a particular megapixel value.
What I am trying to figure out, and all can try this on your PC's, I have PSCS6, Adobe raw version 9.1.1, i can open my 21mp file at 96 or 300 resolution, and change it to a 50 mp image at 96 or 300+resolution or dpi,, the sensor is ppi, I am just thinking if the camera raw is mmm, adding more pixels are in the image by copying/duplicating to the closet color, they both look great when you go down even to the pixel or far away,,, or if it is my IMac 27 in monitor allowing me to see the quality,, can't print it out right now(printer is down)
Upsizing from 21MP to 50MP is resizing up by a factor of sqrt(50) / sqrt(21) = 1.543 (153% larger) which probably yields an ok result since it is less than doubling the size in each dimension.
If you resize in Photoshop instead of ACR then you have a choice of what resizing algorithm you want to use:
Your original post asked why not always upsize to a pseudo-50MP camera size because it's bigger and still looks great.
So while it may be amazing that you can upsize an image to 8660 x 5773 and it still looks great and you chose those dimensions because you like the number 50 megapixels because it's more than twice a large as your camera sensor's native MP of 21, that's really not the way to go about resizing.
Generally, you want to resize to whatever the pixel dimensions of your display or output medium, or not resize at all if something else will be doing it after you pass the image on to the next step in the process.
For a monitor you probably already know the pixel dimensions and can just resize to that: 1920 x 1080 or whatever dimensions a higher-res 4K or 5K monitor might be.
For printed output you should have some sense of how many inches something will be and how many DPI/PPI the printer has, and you can just multiple those two -- inches and pixels-per-inch -- to get the pixel dimensions to resize to. Or use the shortcut in the resize box where you can specify the PPI and Inches separately.
Why should things be resized to exactly the size of the output medium, well, because resizing necessarily adds some smoothness to the image and you may want to counteract this by some sharpening, and adding a touch of sharpening after resizing will keep the sharpening from being overly noticeable in a bad way. Sharpening should make things look better, without adding obvious artifacts to the image.
what i am trying to do is think outside the pixel… some customers want supersize prints for their whole wall, i'm just experimenting with different ways to see which might get the best results for the tools you have to work with.
i really do appreciate all this input.. always learning