Why change dpi in Photoshop?

DPI really for offset press printing applications.
For ALL printing.
Yes you are correct. I should have been more specific. I realize it is printed dots. We work in PPI when it comes to the printing we do. Anyway I do because that is what I see in PS.
PS only understands and works with pixels, not dots.
I never worry about DPI settings.
Read the Schewe piece, maybe you should. Again, IF you're printing AND the document's pixel density is less than what will transfer to dots, something has to interpolate the data up (and sometimes down). Not all interpolations are equal.
I don't do my own printing anymore so there is one disadvantage. It has been a while. I have your book which was helpful but I could never keep to you on this stuff.
Forgot to say. Anything that goes to my local printer is exported using LR. I don't worry about too much these days.
Doesn't matter if it's exported, how it's exported can play a role in the resulting print. Again, the Schewe PDF shows examples of this.
I'll check it out. Thanks
--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net
 
DPI really for offset press printing applications.
For ALL printing.
Yes you are correct. I should have been more specific. I realize it is printed dots. We work in PPI when it comes to the printing we do. Anyway I do because that is what I see in PS.
PS only understands and works with pixels, not dots.
I never worry about DPI settings.
Read the Schewe piece, maybe you should. Again, IF you're printing AND the document's pixel density is less than what will transfer to dots, something has to interpolate the data up (and sometimes down). Not all interpolations are equal.
I don't do my own printing anymore so there is one disadvantage.
But whoever handles your data may NOT be ideally resampling to the output device.
It is possible. It is not a sideline photo lab in a grocery store. I send send with PPI they request.
You could ask!

They might very well simply send the data as is to the front line and that data may not be interpolated to the necessary resolution ideally.

Do they demand you also send them sRGB? If so, the idea that they are providing the best print path because they are not a grocery store lab proves they are not handling the data optimally!
 
DPI really for offset press printing applications.
For ALL printing.
Yes you are correct. I should have been more specific. I realize it is printed dots. We work in PPI when it comes to the printing we do. Anyway I do because that is what I see in PS.
PS only understands and works with pixels, not dots.
I never worry about DPI settings.
Read the Schewe piece, maybe you should. Again, IF you're printing AND the document's pixel density is less than what will transfer to dots, something has to interpolate the data up (and sometimes down). Not all interpolations are equal.
I don't do my own printing anymore so there is one disadvantage.
But whoever handles your data may NOT be ideally resampling to the output device.
It is possible. It is not a sideline photo lab in a grocery store. I send send with PPI they request.
You could ask!

They might very well simply send the data as is to the front line and that data may not be interpolated to the necessary resolution ideally.

Do they demand you also send them sRGB? If so, the idea that they are providing the best print path because they are not a grocery store lab proves they are not handling the data optimally!
No they don't demand and I could send on other formats bit it is always Jpeg.
--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net
 
DPI really for offset press printing applications.
For ALL printing.
Yes you are correct. I should have been more specific. I realize it is printed dots. We work in PPI when it comes to the printing we do. Anyway I do because that is what I see in PS.
PS only understands and works with pixels, not dots.
I never worry about DPI settings.
Read the Schewe piece, maybe you should. Again, IF you're printing AND the document's pixel density is less than what will transfer to dots, something has to interpolate the data up (and sometimes down). Not all interpolations are equal.
I don't do my own printing anymore so there is one disadvantage.
But whoever handles your data may NOT be ideally resampling to the output device.
It has been a while. I have your book which was helpful but I could never keep to you on this stuff.
Well my book doesn't cover any of this kind of stuff, Jeff's does:

https://www.amazon.com/Digital-Print-Preparing-Lightroom-Photoshop/dp/0321908457
I have learned valuable bits and pieces from Jeff's blogs, etc. The last key one for me was how the LR detail slider in the detail palette works between halo suppression and deconvolution. Interesting stuff.
--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net
 
Last edited:
DPI really for offset press printing applications.
For ALL printing.
Yes you are correct. I should have been more specific. I realize it is printed dots. We work in PPI when it comes to the printing we do. Anyway I do because that is what I see in PS.
PS only understands and works with pixels, not dots.
I never worry about DPI settings.
Read the Schewe piece, maybe you should. Again, IF you're printing AND the document's pixel density is less than what will transfer to dots, something has to interpolate the data up (and sometimes down). Not all interpolations are equal.
I don't do my own printing anymore so there is one disadvantage.
But whoever handles your data may NOT be ideally resampling to the output device.
It is possible. It is not a sideline photo lab in a grocery store. I send send with PPI they request.
You could ask!

They might very well simply send the data as is to the front line and that data may not be interpolated to the necessary resolution ideally.

Do they demand you also send them sRGB? If so, the idea that they are providing the best print path because they are not a grocery store lab proves they are not handling the data optimally!
No they don't demand and I could send on other formats bit it is always Jpeg.
So they would accept a document in ProPhoto RGB?
 
DPI really for offset press printing applications.
For ALL printing.
Yes you are correct. I should have been more specific. I realize it is printed dots. We work in PPI when it comes to the printing we do. Anyway I do because that is what I see in PS.
PS only understands and works with pixels, not dots.
I never worry about DPI settings.
Read the Schewe piece, maybe you should. Again, IF you're printing AND the document's pixel density is less than what will transfer to dots, something has to interpolate the data up (and sometimes down). Not all interpolations are equal.
I don't do my own printing anymore so there is one disadvantage.
But whoever handles your data may NOT be ideally resampling to the output device.
It is possible. It is not a sideline photo lab in a grocery store. I send send with PPI they request.
You could ask!

They might very well simply send the data as is to the front line and that data may not be interpolated to the necessary resolution ideally.

Do they demand you also send them sRGB? If so, the idea that they are providing the best print path because they are not a grocery store lab proves they are not handling the data optimally!
No they don't demand and I could send on other formats bit it is always Jpeg.
So they would accept a document in ProPhoto RGB?
Yes they would. Long ago in the forums you suggested to always work in aRGB then convert to profile which I did and never looked back. These days ACR is set to ProPhoto to match what LR works in.
--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net
 
DPI really for offset press printing applications.
For ALL printing.
Yes you are correct. I should have been more specific. I realize it is printed dots. We work in PPI when it comes to the printing we do. Anyway I do because that is what I see in PS.
PS only understands and works with pixels, not dots.
I never worry about DPI settings.
Read the Schewe piece, maybe you should. Again, IF you're printing AND the document's pixel density is less than what will transfer to dots, something has to interpolate the data up (and sometimes down). Not all interpolations are equal.
I don't do my own printing anymore so there is one disadvantage.
But whoever handles your data may NOT be ideally resampling to the output device.
It is possible. It is not a sideline photo lab in a grocery store. I send send with PPI they request.
You could ask!

They might very well simply send the data as is to the front line and that data may not be interpolated to the necessary resolution ideally.

Do they demand you also send them sRGB? If so, the idea that they are providing the best print path because they are not a grocery store lab proves they are not handling the data optimally!
No they don't demand and I could send on other formats bit it is always Jpeg.
So they would accept a document in ProPhoto RGB?
Yes they would.
Cool; do they take files from anyone remotely or just a local lab? There are folks in these parts looking for labs that do not demand an sRGB workflow.
 
DPI really for offset press printing applications.
For ALL printing.
Yes you are correct. I should have been more specific. I realize it is printed dots. We work in PPI when it comes to the printing we do. Anyway I do because that is what I see in PS.
PS only understands and works with pixels, not dots.
I never worry about DPI settings.
Read the Schewe piece, maybe you should. Again, IF you're printing AND the document's pixel density is less than what will transfer to dots, something has to interpolate the data up (and sometimes down). Not all interpolations are equal.
I don't do my own printing anymore so there is one disadvantage.
But whoever handles your data may NOT be ideally resampling to the output device.
It is possible. It is not a sideline photo lab in a grocery store. I send send with PPI they request.
You could ask!

They might very well simply send the data as is to the front line and that data may not be interpolated to the necessary resolution ideally.

Do they demand you also send them sRGB? If so, the idea that they are providing the best print path because they are not a grocery store lab proves they are not handling the data optimally!
No they don't demand and I could send on other formats bit it is always Jpeg.
So they would accept a document in ProPhoto RGB?
Yes they would.
Cool; do they take files from anyone remotely or just a local lab? There are folks in these parts looking for labs that do not demand an sRGB workflow.
I just called them. I have not used them in over year and a half. Gave up event shooting after my knees were replaced last spring. They moved to a new location in the city. They will accept it but render it to RGB. Maybe something changed. I knew they accepted it but I never looked into it as always submitted sRBG.I can PM you a link if you like.
--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net
 
DPI really for offset press printing applications.
For ALL printing.
Yes you are correct. I should have been more specific. I realize it is printed dots. We work in PPI when it comes to the printing we do. Anyway I do because that is what I see in PS.
PS only understands and works with pixels, not dots.
I never worry about DPI settings.
Read the Schewe piece, maybe you should. Again, IF you're printing AND the document's pixel density is less than what will transfer to dots, something has to interpolate the data up (and sometimes down). Not all interpolations are equal.
I don't do my own printing anymore so there is one disadvantage.
But whoever handles your data may NOT be ideally resampling to the output device.
It is possible. It is not a sideline photo lab in a grocery store. I send send with PPI they request.
You could ask!

They might very well simply send the data as is to the front line and that data may not be interpolated to the necessary resolution ideally.

Do they demand you also send them sRGB? If so, the idea that they are providing the best print path because they are not a grocery store lab proves they are not handling the data optimally!
No they don't demand and I could send on other formats bit it is always Jpeg.
So they would accept a document in ProPhoto RGB?
Yes they would.
Cool; do they take files from anyone remotely or just a local lab? There are folks in these parts looking for labs that do not demand an sRGB workflow.
I just called them. I have not used them in over year and a half. Gave up event shooting after my knees were replaced last spring. They moved to a new location in the city. They will accept it but render it to RGB.
sRGB?
Maybe something changed.
Yeah, if they manually convert to sRGB, not so good.
I knew they accepted it but I never looked into it as always submitted sRBG.I can PM you a link if you like.
Not necessary really. I'm just always on the look out for labs that implement a full color managed path. This lab doesn't sound like it would fit the bill.
 
There is no technical advantage of setting one or another PPI so far as editing a digital image's content. IMO it only makes sense to think about PPI at all, once you have got a specific print size or other concrete usage in mind; and only then in respect of a particular version of your image which is dedicated toward that particular output.

There are some practical differences from PPI so far as placing a digital image, in certain software. For example, into page layout software such as InDesign its initial sizing may follow the PPI that is present.

But there's more: real world case - after glibly telling a print manager that it was the same thing whether I sent in my image set to 300ppi or to any other number, so long as it contained sufficient pixels for the usage size, she patiently explained how in practice it was NOT found to be the same thing at their end.

IOW theory trumps data, but only in theory ;) .

For example - some layout software creates a screen-res preview internally, for best performance and also so as to still manipulate images that are externally linked, and may be at times unavailable. This software works on the basis that images will have always been SIZED in an intended way - as to their placement in a layout - and that their PPI and therefore notional print size is therefore significant. It makes a screen preview resampled to 72ppi on that presumption.

Consequence: say you provide an image which is 4000px wide and 400ppi - the software concludes this is intended to print 10" wide, generates a 72ppi screen preview representation suitable for a 10" sized image (approx. 0.5 megapixels), and embeds that inside the layout file.
OK I read this again. So you are saying when I view at this size I will see a difference between my default of 300 ppi and 72?
I am saying that if you view the image in an image viewer program, or an image editing program, or on a website, or using your computer's file browser, etc - there will be no difference whatever. That's provided the number of pixels is still the same.

If you have changed the number of pixels by "resampling" when you switched from one PPI to another - then this will be a real and visible difference. It's a completely reconstituted digital image now, with a different amount of detail within it.

Analogy: say a gallon of goop costs $2. Two different ways I could sell you $1 of goop instead: I could give you just half a gallon, or else I could halve the price for a full gallon. Resampling the image pixels to a smaller number of pixels, is like reducing the amount of goop. Changing the PPI without resampling, is like reducing what the price is, for the same amount of goop.

And it's the number of pixels and the way you save it, that cause a JPG or a TIFF etc, to be a certain file size. PPI has got nothing at all to do with that, per se.

But there are some OTHER programs that DO pay attention to PPI. If you put a photo into Microsoft Word and this photo "thinks" it is 1 inch x 2 inches big (because of its PPI) then you will get it scaled to 1 inch x 2 inches big on the page by default.

If you put the same photo onto the same page but the PPI is different, then the default scaling at which it is placed, will correspondingly differ by default. But if you artificially then make both of these fill the page, or else both be four inches wide, or whatever - they will then look the same as each other (because they are the same as each other in all important respects). They are like two buckets both containing the same quantity of goop, even though one bucket's goop was "priced" differently than the other. If I offer to sell either bucket to you for $5, it is the exact same deal so far as what you then get.
Long ago at Luminous Landscape there was a thread about LR where someone had said 72 ppi as well. I have only ever found one reference to this and that was on that forum. I looked quire a bit and if it had been this important I should have found more references.

The LR web module does not even offer a choice so I'm not sure what it is set for or if they even bother with a ppi setting somewhere in the background.
The LR web module concentrates on pixel dimensions because that is what is important in on screen viewing. PPI is chiefly of interest as part of scaling images onto a page layout, for hardcopy output of some sort.
If you instead provided 4000px wide and 4 ppi, the software then generates a screen res preview to suit a 1000" sized image. That's a screen preview containing approx. 500 megapixels. Needlessly embedded inside the layout file, and gluing up the operations of the program.
 
When I'm sending images out I do this out of sheer paranoia.
If my images have the "correct" size then it's less likely that some outside workflow, that I don't control, will mess up.
This doesn't mean I always use 300 ppi; it means I use a combination of pixels and ppi that correspond to my desired output size.
 
But just a question. have you tried editing at native resolution and then re-sample? And then try it the other way around. Re-sample and then edit. Which gave you better results?
I haven't tried it, but in my workflow I could never resample at the start because editing of my serious photos is print-size independent. I produce metal prints, ranging from 5x7 to 24x36, depending on what the buyer wants.

I store a single file per photo of my portfolio at Adoramapix, and then just order whatever I need. (They're great, by the way.)

If I'm ordering a small print and it turns out to be 900 ppi, that's they're problem, and they're perfectly capable of handling it. They know how to down-sample as well as I do. Better, probably, as they no doubt use a high-end RIP.
 
Well here's one. You KNOW you need a much larger image file than the native capture of raw. At least in LR/ACR, it's interpolation engine is very good and after all raw parametric edits but before rendering to further edit in Photoshop, you would want to have these products interpolate from the raw data.
 
IT'S PIXELS PER INCH (PPI) NOT DOTS PER INCH (DPI)
Probably shouldn't get too picky about this... I probably was speaking loosely, and he might have said ppi. But the issue I was addressing wasn't about correct terminology. It was about his basic misunderstanding of the (non-)role of resolution in the workflow.

--Marc
 
DPI really for offset press printing applications.
For ALL printing.
DPP still calls it it DPI but it is really PPI or pixels per inch which is for the printing we do. Adobe calls it resolution which is PPI.
Images have pixels. It's PPI. Displays have pixels, it's PPI. Printers produce dots; it's DPI.
When I took PS course about 10 years ago they told me 72 PPI for the screen as well.
Rubbish and an old urban legend that just never dies.
That is incorrect. PPI for is printing only it has zero effect on the screen resolution.
Well the incorrect part is more about the number than the letters after.

IF you want to know the exact output resolution of your display, measure the width of your display and divide that by the number of pixels its displaying.

For example, on my NEC 3090, the width is 25.25 inches. Its resolution is 2560x1990. 2560/25.25=101.4 PPI.

On my NEC PA272W, the width is 23.5 inches. Its resolution is 2560x1440. 2560/23.5=109PPI.
 
DPI really for offset press printing applications.
For ALL printing.
DPP still calls it it DPI but it is really PPI or pixels per inch which is for the printing we do. Adobe calls it resolution which is PPI.
Images have pixels. It's PPI. Displays have pixels, it's PPI. Printers produce dots; it's DPI.
When I took PS course about 10 years ago they told me 72 PPI for the screen as well.
Rubbish and an old urban legend that just never dies.
That is incorrect. PPI for is printing only it has zero effect on the screen resolution.
Well the incorrect part is more about the number than the letters after.

IF you want to know the exact output resolution of your display, measure the width of your display and divide that by the number of pixels its displaying.

For example, on my NEC 3090, the width is 25.25 inches. Its resolution is 2560x1990. 2560/25.25=101.4 PPI.

On my NEC PA272W, the width is 23.5 inches. Its resolution is 2560x1440. 2560/23.5=109PPI.
 
DPI really for offset press printing applications.
For ALL printing.
DPP still calls it it DPI but it is really PPI or pixels per inch which is for the printing we do. Adobe calls it resolution which is PPI.
Images have pixels. It's PPI. Displays have pixels, it's PPI. Printers produce dots; it's DPI.
When I took PS course about 10 years ago they told me 72 PPI for the screen as well.
Rubbish and an old urban legend that just never dies.
That is incorrect. PPI for is printing only it has zero effect on the screen resolution.
Well the incorrect part is more about the number than the letters after.

IF you want to know the exact output resolution of your display, measure the width of your display and divide that by the number of pixels its displaying.

For example, on my NEC 3090, the width is 25.25 inches. Its resolution is 2560x1990. 2560/25.25=101.4 PPI.

On my NEC PA272W, the width is 23.5 inches. Its resolution is 2560x1440. 2560/23.5=109PPI.
 
DPI really for offset press printing applications.
For ALL printing.
DPP still calls it it DPI but it is really PPI or pixels per inch which is for the printing we do. Adobe calls it resolution which is PPI.
Images have pixels. It's PPI. Displays have pixels, it's PPI. Printers produce dots; it's DPI.
When I took PS course about 10 years ago they told me 72 PPI for the screen as well.
Rubbish and an old urban legend that just never dies.
That is incorrect. PPI for is printing only it has zero effect on the screen resolution.
Well the incorrect part is more about the number than the letters after.

IF you want to know the exact output resolution of your display, measure the width of your display and divide that by the number of pixels its displaying.

For example, on my NEC 3090, the width is 25.25 inches. Its resolution is 2560x1990. 2560/25.25=101.4 PPI.

On my NEC PA272W, the width is 23.5 inches. Its resolution is 2560x1440. 2560/23.5=109PPI.

--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net
Better yet look up the pixel pitch specifications for your monitor(s) and get the PPI value that way.
Much better if you don’t own a ruler and the math is too difficult! 😀
Or, if as is normal, the edges of the screen is covered by the bezel so you can't get an accurate measurement of the screen size.
No! You only measure image area. Just as accurate. That IS the screen size!

Are you serious that you believe there are image pixels behind the bezel? It isn't close to April 1st :-D

You do realize that your camera sensor has pixels that do not capture image data you'll ever see or get (used for white balance and other work) yet when those seriously speak of image resolution of their capture, they only count the sensor pixels used to produce an image.

You do realize that if you're trying to figure out the resolution for a print, you don't take the white border into account; just the image pixels.

You think your display is different due to a bezel?

--
Andrew Rodney
Author: Color Management for Photographers
The Digital Dog
http://www.digitaldog.net
 
Last edited:

Keyboard shortcuts

Back
Top