Skip to main content
A
Augusto24716166az1l
Participating Frequently
June 19, 2022
Question

Resizing using free transform without losing quality

  • June 19, 2022
  • 1 reply
  • 4761 views

Hello,

 

I am currently working on an image collage in Photoshop that I'm hoping to print with the best possible quality.

 

I created a new document with the dimensions I want, 300dpi, ... I then started creating the collage using this tutorial which is pretty straight forward, I think:

 

https://www.youtube.com/watch?v=DAwRrl9tXKE

 

My question is, when I'm doing the positioning and resizing of each image into my working file, am I keeping the best possible quality for each image doing doing it as explained on the tutorial?

 

Before I drag any of the images into my working file I ensure I have the correct color profile (Adobe RGB 1998) and I make sure I chage the image size to 300dpi (as some of my old scans had different DPIs) to match the destination, and I leave the 'resample' box unticked. I also make sure that the images I'm using will be big enough for the propuse of this collage so I only need to downsize them and not the other way around. 

 

I then resize them to the desired size/dimensions and create a clipping mask to remove the pixels I don't need as explained on the tutorial.

 

So far it looks ok but I want to ensure I am keeping the best possible quality for each image and not losing quality when risizing. Am I doing this correctly? I want the image to look as crisp and clear when printed.

 

Should I be cropping and resising the images before dragging them into the collage?

 

Do I need to resample or do I leave the box unticked when changing to 300dpi as I only want to reduce and not increase the size?

 

Do I need to convert the layers into smart objects?

 

Still learning some of the Photoshop basic tools.

 

Thank you!

 

 

    1 reply

    D Fosse
    Community Expert
    D FosseCommunity Expert
    Community Expert
    June 19, 2022

    The best advice I can give is to disregard ppi, it's actually just confusing the issue.

     

    This all goes by pixels. The pixels in the pasted image align to the base pixel grid, 1:1. If you bring in a 2000 x 3000 pixel image, it will occupy 2000 x 3000 pixels in the base document. And that's all there is to it.

     

    People tend to think ppi is a very important property of the file, but it isn't. It's just a way to translate pixels into a physical print size. Pixels per inch! Say it out loud. It means exactly what it says. It's a measure of pixel density on paper.

     

    A random ppi number can be assigned to any file without altering the file in the slightest. It will just print at different sizes.

     

    So. Resizing after you bring it in is as good as anything else - provided you're not going to readjust the size later. If you are, work with Smart Objects. This embeds a full size copy in the file, so that repeated transforms will not inflict further damage. You always go back to the full copy. Of course, the final resize is as destructive as any other method, but you only get it once.

     

    Look at your resampling algorithms. Bicubic Auto and Bicubic Sharper tend to (IMO) oversharpen and introduce a lot of artifacts. I prefer Bicubic Smoother, which doesn't sharpen, and then apply a bit of sharpening after the resize. Be careful and don't overdo it.

      A
      Augusto24716166az1lAuthor
      Participating Frequently
      June 19, 2022

      Thank you for your answer.

       

      When I mentioned I resize the images first I meant that I change the resolution to 300dpi as I thought I need to use the same DPI I'll be printing the file with. So I don't touch the pixel dimensions, just change resolution to 300dpi and leave 'resample' unticked. But I guess by reading your message that if my working file is 300pi and I drag an image that it's let's say 72dpi the image will adjust itself and appear with the size it'll print at 300dpi?

       

      I still find find myself getting confused with the whole ppi vs dpi but when it comes to printing I only pay attention to the dpi/resolution.

       

      Also, since you mentioned it on your message. If i convert the image/layers into Smart Objects, do I need to do it before I do any adjusting/resizing?

       

      Thanks again for your help!

      D Fosse
      Community Expert
      D FosseCommunity Expert
      Community Expert
      June 19, 2022

       

      But I guess by reading your message that if my working file is 300pi and I drag an image that it's let's say 72dpi the image will adjust itself and appear with the size it'll print at 300dpi?

       

      Yes, that is precisely correct.

       

       

      I still find find myself getting confused with the whole ppi vs dpi but when it comes to printing I only pay attention to the dpi/resolution.

       

      You need to make sure you have enough pixels to print it at the required size at the required pixel density (ppi). That's all you need to worry about.

       

      This is a perfectly straightforward equation. You have, say, 1000 pixels long side, and you need to print this at a standard book print resolution of 300 ppi. Just do the math! Pixels per inch, remember? 1000/300=3.33. So that's your answer: it will be 3.33 inches wide.

       

      If you print this any larger, there are fewer pixels to go around per inch. So the ppi number goes down. And vice versa.

       

      Of course, you don't have to actually do the math. Photoshop's Image Size will do the math for you.

       

      Yes, you convert to smart object before you do any resizing/scaling.

        Terms & ConditionsAccessibility statement
        Using the community Terms of use Privacy Cookie preferences Do not sell or share my personal information ad choicesAdChoices