When I open an image it defaults to opening at 72dpi. I need to work at 300dpi most of the time so I would like to set the default to 300dpi if possible
Give your photos a professional finish with sharpening in Photoshop. Learn to enhance details, create contrast, and prepare your images for print, web, and social media.
Resolution is irrelevant. Picture dimension is not. You can change to 300 ppi for printing through Image>Image Size. Adjust with resampling turned off and you will be fine
Pixels Per Inch. DPI (dots per inch) relates to the way a printer lays down the ink.
300 PPI is generally accepted as the norm for images destined for print. This is certainly what most photo labs ask for in terms of resolution when you send them your images.
In reality, an inkjet printer will be quite happy if you feed it 240 PPI, assuming you are printing at 1440 DPI (divide 1440 by the number of inks [six] equals 240)
how do you turn off resampling??
In the IMAGE>IMAGE_SIZE dialogue, UNCHECK the "Resample" box at the bottom. This way you can change the "resolution" without messing with the pixels. The "PPI" of a file is only a tag which tells the printer how many pixels to fit into each inch.
It also makes it easier for you to assess the final size when you come to resize the image for printing. Always work at the generic size (as it came off the camera). Always save your "Master File" (after editing) at the generic size. When you resize for printing/web use etc, always save a new version, sharpened at the finished size. If you need a different size, go back to the "Master File" and resize from there.
It opens your photos in 72 DPI because that’s what your camera shoots them in, regardless of how many megapixels your camera is. My 6MP Canon PowerShot S3 IS shoots 180 DPI photos, and they open at 180 DPI in Photoshop.
I wish the digital camera gods had not blessed the world with the muddled concepts of dpi, ppi, resolution, mega pixels, picture size, and the like.
But I do confess — and no disrespect intended — that Tim’s posting has given some of these terms another slant that can only contribute to more confusion. But it did make me smile.
You cannot equally compare two images unless they are at the same resolution. To compare your 6MP and 8MP images, turn off resampling in the Image Size dialog and set both images to the same resolution. Doing so will show what the math also tells us, that the 8MP will yield a larger printed size than the 6MP. If you print them out and compare them side by side, and still find the 6MP image appears sharper, then that is likely due to the optics of your camera. If you instead resized the images so the 8MP image printed at the same size as the 6MP image, with resampling still disabled, then that would yield a higher resolution file with the 8MP image because there is more data to pack into a smaller space. Likewise, the 8MP image theoretically should yield a finer quality print since more data is being used to produce the same size of print than the 6MP image offers.
Tim. No camera "shoots" at any particular "resolution". It records an image at xxxx pixels wide by xxxx pixels high.
If the file info states 72 ppi, 180ppi, or whatever, that is just a "tag", a label if you like, which is added to the file’s metadata.
You don’t have to "upsample" to 300 ppi. Simply changing the "resolution" setting without resampling (in Photoshop’s Image_Size dialogue for instance) will indicate the size in inches that the image would print at its generic size.
For example, a 3000 x 2400 pixel camera file that declares itself to be 41.6 x 33.3 inches at 72 ppi, would in fact print at 10×8 inches at 300 ppi without resampling the image.
To sum up: During your image editing workflow, it matters not a jot what "resolution" is stated for the image. Only when you come to resize a final version of the image for print or for the web, do you need to set the appropriate ppi along with the required dimensions.
Every digital camera has a processor that interprets the light entering the sensor from the lens. That processed data is then written to the memory card in the camera’s native resolution (typically 72 dpi) regardless of what size or quality (compression) settings have been chosen by the user. This is the dpi that Photoshop reports when it opens the photo — it’s not just some esoteric tag or label.
Prove it to yourself. Take a 6MP photo and a VGA-size photo and look at them in Photoshop. They both come out of the camera at the same dpi. Why? Because the camera only has one resolution it can save files in. Sure, you can change the size and the quality on the camera, but the resolution — the dpi — remains the same.
Digital cameras are not like scanners where you can choose what resolution to scan something at. Constrained by the number of holes that can be drilled in a CCD, they can only make the sensor larger allowing people to take bigger, not necessarily better, photos. (Sure, bigger photos generally make better prints, but that’s a separate topic.)
Canon’s processing is akin to having a CCD with the holes closer together, so the result is smaller-sized photos that are sharper than a typical digital camera is capable of. So whereas a typical 6MP photo is basically 39 x 29 inches @ 72 dpi, the Canon’s 6MP photo is 15 x 11 inches @ 180 dpi. Both have the same pixel dimension of 2816 x 2112.
I have encountered a number of people who think that they can shoot smaller-sized photos with their high-MP camera (in order to save memory card space) and still somehow benefit from the higher megapixels. What they don’t realize is that by choosing a smaller size, the camera is simply switching off a band of pixels around the outer edge of the sensor leaving fewer effective pixels in the center and, hence, producing a smaller-size photo. That photo, however, still has the same number of pixels per inch as a larger photo does.
My suggestion to upsample the images to 300 dpi for production work is a matter of professional taste. I wouldn’t place any graphics in InDesign that weren’t at least 300 dpi, but that’s me and others may have a different take on that.
Umm, Chris, your division is correct, but any implication that resolution equals that quotient is not.
The historical habitual incorrect use of the term dpi has been the cause of much confusion.
Let us take, as an example, one of the printers sitting in front of me now. It’s "Photo Quality setting is 1440 dpi which means it is capable of squirting 1440 dots of ink per inch whilst it attempting to recreate on paper the image contained within a file made up of (say) 300 pixels per inch.
In practice, whatever number of pixels per inch you feed into the printer, the driver software will interpolate that data into a quotient divisible by the number of ink jets. If any given "dot" can be represented by varying amounts of the six inks, then the true "resolution" of this particular printer is 240 dots per inch, each dot being made from (up to) six smaller dots of ink.
So in theory my premise is correct. Feed a 300 ppi file into the printer and it will effectively be interpolated down to 240 ppi to produce those 1440 dots per inch.
Therefore, theoretically, you would achieve no lesser quality by feeding in a file which is already at 240 ppi.
Theoretically (that word again) the quality should actually be better from a 240 ppi file because no further interpolation needs to take place.
Theoretically – because I defy anyone to tell the difference with the human eye.
Let’s not also confuse the issue with "holes being drilled in CCDs.
Let’s not confuse the issue with "VGA-size photos" (Video Graphics Array/Adaptor??)
Let’s not confuse the issue with camera’s having a "native resolution (typically 72 dpi)" ?!!
Let’s not confuse the issue with "cameras only having one resolution it can save files in."
Let’s also not confuse the issue by implying that you have to "upsample" a file to 300 (dpi [sic]) for production work.
Upsampling a file increases the number of pixels by interpolation.
You could simply change the resolution to (say) 300 ppi without resampling.
Only if the image is not physically the correct size do we have to upsample (or downsample).
True, we often do both operations at the same time via the Image_Size dialogue by setting the new size and resolution with the "resample" box checked. But "resampling" and "changing the resolution" are mutually independent.
The truth of the matter is that dpi and ppi tend to be proprietary linked rather than absolutes.
Whatever the program does it is probably better to respect that methodology as a preferred option taken by that program’s development team.
Some are screen oriented, some are print oriented, some are both print and screen oriented so may indeed need to define things as that program will handle stuff.
Wow what a response and thank you all so much for your kind help, I afraid it looks like if fell into the marketing trap of the digital camera manufactures, So the cameras should actually be sold as what size the photos it produces when printed at 300ppi and NOT now many pixels it has.
So the cameras should actually be sold as what size the photos it produces when printed at 300ppi
That’s potentially a good idea – except it is never likely to happen because it would give the impression that a 12MP camera was only capable of producing prints up to 14×9" or so. In reality we regularly produce 30×20" prints from this size, though careful processing at our studio.
The important thing to remember is that a digital image is measured in pixels. Nothing else. Just pixels.
What you do with those pixels is dictated by the final use to which the image will be put.
The final irony, of course, is that for really large prints, you can get away with a much lower ppi file size and a lower dpi print resolution.
I’m talking about those giant advertising banners which look great from across the street but become a mess of blurry dots when viewed close up!
A printer dot is not a cluster of multi-colored smaller dots. a dot is a dot is a dot no matter what its color. Pixels are composites of dots, but dots are not composited. This seems irreconcilable with your wording, quote "…any given "dot" can be represented by varying amounts of the six inks,…each dot being made from (up to) six smaller dots of ink."
George: For dot’s sake… We’re trying to simplify things here! 8-|
Yes, a dot is a dot is a dot. But can YOU perceive 1440 dots in an inch?
A group a six different coloured tiny dots will visually combine to produce a "dot" (for the want of a better word) of one particular colour. And each of those dots visually combine to produce an impression of continuous tone.
I’m repeating myself, but printer dots of different color are not clustered contiguously to form a larger dot. That’s just not so.
The only spatial requirements for an individual C,M,or Y dot is that it lie within the pixel boundary and that it not overlap a dot of the same color.
A Cyan dot absorbs red and is transparent to green and blue which pass on thru it and, if no other ink dot lies beneath, get reflected by the white substrate to pass back through the same dot a second time and out to the eye which then sees green and blue (called Cyan).
If, for example, a Magenta dot lies beneath the Cyan dot, the Magenta dot still performs its function of removing green—to the extent that the Cyan dot on top of it is really transparent in green. (It isn’t totally transparent, so some loss is incurred).
Now here is the crucial point about the alignment or non-alignment or partial alignment of the Cyan and Magenta dots (which we’re using as examples where we could, as well, have used yellow/cyan or yellow/magenta). To the first order of magnitude, the alignment does not matter!
Each dot does its job of removing one color color—R,G,or B—from the pixel area regardless of relative positions. That’s because it’s happening on a scale smaller than the eye can discern.
The eye only knows that there has been some reduction in red (for cyan ink) or green (for magenta ink) for that pixel. So the printer is relieved of any requirement to "register" the different colored CMY dots—except to keep them within the pixel area. And, thus, you don’t see multi-colored dot clusters.
Can I perceive 1440 dots per inch of a single color repeated for multiple colors? Yes. Remember, the amount of ink per dot is only 4 PICOliters. Do the math on dot dimensions for such small globs, even allowing for dot spread.
I think the idea that printer dots are clustered comes from viewing CRT screens where the unditherd additive RGB components ARE clustered. Dithered, subtractive printer dots are a different ballgame.
At the risk of repeating myself, George, you are over-complicating the issue in relation to the Original Post.
The thread you started about combining RGB and CMYK inks is the place to discuss how the ink dots are deposited, and whether they are clustered, dithered, or overlapped in an attempt to convince the viewer they are seeing a continuous tone.
Different technologies are employed by the various manufacturers to produce the definition and impression of continuous tone that we have come to accept as "photographic quality".
What we are discussing here is how the print engine interprets the image pixels before passing that information to the print heads.
The fact remains that each pixel of an image needs to be represented on a finished print by one "dot" – regardless of how many 4pl sized (let’s call them) micro dots go to make up that dot and how they are deposited onto the paper.
In the case of the printer cited in my example, it has six inks producing 1440 "micro" dots, representing 240 "dots".
It follows that whatever the ppi resolution of the file, the print engine will interpolate that information into 240 dots per inch and that each dot will then be represented by an appropriate signal being sent to each of the six ink heads.
If you remember, my original proposition back in post #3 was that no loss in quality should be evident when printing a 240 ppi image rather than 300 ppi.
In fact, the closer the image data matches the requirements of the printing engine, the better the image will be rendered.
Yes, this and the thread I initiated are criss-crossed. One should not feel constrained, however, to avoid twists and turns in a given thread simply because they are addressed elsewhere.
Following your logic, the Canon iPK5000, having 12 inks and 1200dpi, would provide only 100ppi. Your rule-of-thumb is simply invalid.
George. As you well know, the Canon IPK5000 is a completely different animal.
In addition to the usual CMY (plus "photo cyan" and photo magenta), this printer (uniquely) also employs Red, Green and Blue ink. It also has two shades of grey and two of black.
Clearly, not all these inks will be used at the same time, therefore to imply that my analogy would result in 100ppi being the ideal is complete nonsense.
Regardless of the number of inks, be it four, six, or more, every printer will have a an ideal file resolution. If the manufacturers were to publish it, we could prepare our file to that resolution rather than the "regulation" 300ppi.
But when all is said and done, George, you and I batting this back and forth is a complete waste of bandwidth because the naked eye would not be able to tell the difference between a 1440 dpi print of a 300 ppi file and one produced from a 240 ppi file of the same image. 😉
As far as I’m concerned, this thread has been about the optimum file resolution for any given printer, but it appears we have both misunderstood one another and the whole thing’s gone off on a tangent.
So how about we agree to disagree, rather than further water down any benefit others might glean from attempting to follow the original theme.
Although Chris has opted out, others may appreciate this retraction. Adding more ink colors DOES affect resolution after all—but in the opposite sense—INCREASING rather than decreasing resolution. There is a trade-off between resolution and color depth (tones per channel). Additional ink shades relieve some of the tonal burden on dithering, allowing the same number of tones to be achieved with a smaller pixel (higher res) or more tones for the same-sized pixel.
Actually with FM Screening the ‘ideal file resolution’ is determined by the individual dot size from the print head. However, printer manufacturer’s describe this not by dimension, but by volume (pico-litres) because the dimension of the dot size is highly dependent upon the media. There is far more dot-gain on matt paper than photo glossy for example.
Thus it follows that the ‘ideal file resolution’ is not only dependent upon the printer model, but also upon the media being printed upon. If talking about ‘ideal’ the image contents also have some bearing on the choice.
Useful insight Len, and as printer technologies progress apace, there will I’m sure be many more facets to the subject of resolution/print quality.
It would actually be quite useful if printer manufacturers openly stated the ideal minimum recommended image file resolution for a given dpi, if only to save an image being interpolated twice – once by us in Photoshop and again by the printer driver.
In fact, if it weren’t for the need to sharpen an image at its final size, we could probably simply feed in the files at their generic resolution and let the print engine do the interpolation anyway!
* For any newcomer attempting to disseminate the information in this thread, I’d just like to clarify something in George’s (otherwise useful) post #32:
When he talks about "the number of tones achieved with a smaller…[snip] …same size pixel" he is actually referring to the dots which make up the printed image. The ppi / dpi issue was queried by the OP back in post #2.
Just to recap: Pixels are picture elements in a digital file, which the printer driver then translates into information to control the way the dots are laid down by the ink nozzles.
Sadly dpi has almost become a generic term and its misuse (even by software vendors, who should know better!) only leads to confusion and misconception.
I couldn’t help smiling to myself as I evesdroppped in a camera shop recently, when I overheard a salesman telling his customer that "this lens had a higher dpi than that one"! 🙂
Actually, Chris, the pixel I refer to (and it’s an ideal for descriptive purposes, not literal) is a matrix of possible dot positions. Like, say, an 8x 8 chess board with 64 squares. The positions may or may not be occupied. The white space remaining when a square is unoccupied by an ink dot plays a role—it dilutes the pixel’s color to a lower tone.
The deployment of dots in the matrix is determined by the image-specified color value. The higher the value, the more squares are filled. The larger the matrix (of filled and unfilled squares), the more tones available, but the lower the resolution (more squares take more space). So, the PIXEL is this ENTIRE ARRAY of filled and unfilled squares. The unfilled areas are as important as the filled in producing the specified tone. And, BTW, the DPI is the number of matrix squares per inch—whether occupied by dots or not.
Again, as I had once been advised by a printing technology expert consulted by Epson, this model is not literally constructed by the printer, but it serves perfectly well in visualizing the process.
Give your photos a professional finish with sharpening in Photoshop. Learn to enhance details, create contrast, and prepare your images for print, web, and social media.
Related Discussion Topics
Nice and short text about related topics in discussion sections
