Resolution Factor

300 dpi is the common number for photos but that could change depending on the type of printing you are doing.
 
As a rule of thumb 300 dpi at the size you are going to reproduce will work, but we have found that 1.5 x the screen ruling is the "minimum" for good reproduction, ie: for a 150 line screen, 1.5x150=225 so 225 would be the number there. The most important thing to remember is you cannot take a 300 dpi photo/file and enlarge it to 300% - that would reduce its effective dpi to 100 dpi.

It can get confusing
 
At what resolution should I save my photos and graphics so that I get best print?

This is a link that explains the issue in detail with images (start at the bottom of the web page): Quality In Print: lpi/dpi

The very short version:
• Your image is made of pixels.
• Your image will be broken down into halftone dots in order to be printed.
• If there are more halftone dots per inch than there are pixels then the halftone will resolve the pixels and they'll be reproduced and look like artifacts.
• If there are more pixels per inch than there are halftone dots then the halftone will not resolve the individual pixels and the printed image will look continuous tone.
• Having more pixels per inch than are necessary just increases file size and processing time.
• The rule is that the dpi of your image at 100% of final size should be a multiple of the lpi of the halftone screen that will be used to print it.
• You can use any dpi from a minimum of 1x to a maximum of 2x the lpi of the halftone screen. (e.g. if I'm doing an ad for a magazine the halftone screen will be 133 lpi. Therefore the image resolution in dpi should be between 133 dpi and 266 dpi. Any less and I'll see pixels in the print - anymore and I'm wasting space and time.

BTW, it's always best to have your images at the highest resolution you can - then down sample copies for the final dpi you need for any particular application.

best, gordon p
 
BTW, it's always best to have your images at the highest resolution you can - then down sample copies for the final dpi you need for any particular application.

Gordon, if we are talking about the high res "master" and down sampling are you following the traditional logic of 100% size 2xLPI that you mentioned above? I ask because it is common (at least in my experience) that users take their photos, make one all nice and pretty and perfect. Then they will place it and scale it to the desired size in the layout application. You may have the one link, but used several times in the layout. And they will rely on say InDesign or Distiller to down sample to the desired resolution. What do you think of that method? I think I know your answer (pretty confident I do), but I think the OP might want to understand the differences.
 
Of course I don't follow the traditional logic of 100% size 2xLPI - but it's a good and practical guide to follow.

I prefer to down sample images myself in PShop so that I can choose which method of down sampling is best suited to the project and therefore gives me control over the final quality/appearance of the smaller image and also allows me to see if any artifacts such as moiré have been introduced by the resampling process. Automatic down sampling by the application, e.g. InDesign or Distiller or the RIP will also tend to "soften" the image - killing detail - as a result of the down sampling.

Attached is a sample of how several PShop methods affect detail when down sampling to give you a rough idea of what happens. Best to download and take a closer look in PShop to see the differences.

Maybe Matt has some other ideas?

best, gordon p
 

Attachments

  • Truck.jpg
    Truck.jpg
    115.2 KB · Views: 209
No, no other ideas. I agree with you.

In a former life what we would do is scan our original, color correct/retouch that. Then we would create "optimized" versions for various common sizes. We'd pick the one that fit most closely the desired final usage size. It was a bear but it worked well. We weren't doing anything that was hyper critical or with fabrics or subjects that had a lot of "natural" patterns. But a lot of times now PDF exporting/generation, preflight processes and RIP's will automatically re-sample images. I think Dan Magullis in one of his books showed some examples of what happens to the image when this happens.

It's just not as convenient to "tailor" an image to a size when you can just place it, resize it and export it at the desired resolution. Too much work the other way(s)...
 
It's just not as convenient to "tailor" an image to a size when you can just place it, resize it and export it at the desired resolution. Too much work the other way(s)...

Yup. And most people wouldn't recognize any degradation in their images, anyway, unless an image totally craps out because of the resampling. Luckily there's always time to reprint the job at the printer's expense.

best, gordon p
 
We tend to give a range of 300 - 450 dpi (when placed at 100% size) for a stock answer. We print with a 20 micron Staccato dot so we can use the extra info if it's there. In our testing more than 450 dpi wasn't visibly much better and increased file size by too much. Most of the time we're lucky to see 300 dpi, though.

Another factor to consider, is that the image content can make a big difference in the resolution required for good reproduction. A softer focus shot -- like a bunch of clouds -- is going to be more forgiving of a lower resolution than one that has lots of fine detail -- like a pine tree.

In a slight tangent... One thing that has changed in the last few years is that the majority of images are captured with a digital camera instead of being scanned from a print or film. A good rule of thumb, if you're the one doing the shooting, is to set your camera to capture the highest resolution and the lowest amount of JPG compression. That will give you a little leeway for cropping, etc and the image can be down-sampled later if necessary. (Low light shooting might be a good exception to the max res rule, but I don't want to go too far off the main topic.)

I've seen lots of images that would be otherwise acceptable ruined by agressive JPG compression settings. JPG artifacts are also much more pronounced with low-res images so no matter what keep that set to the best quality available.

Shawn
 
If you'll note, the OP said "photos and graphics".

Graphics is where I have a bone to pick with folks. At the lower end, there are far too many folks using Photoshop for page layout and composition duties. Last week I criticized a poster on this board for using Photoshop in that manner, and we routinely are forced to take 'desperate measures' to get 1" X 1" .JPG logos saved at 300 PPI to work in high quality reproduction.

Plainly put, you need higher than 2 x LPI for raster images that have type or other kinds of hard edges in them. I would like to see a minimum of 5 x LPI for raster images that have hard edges in them. I would really prefer to see 8 x LPI in this situation, but it's rare I can get 2 x LPI.
 
It happened again. This morning I was sent a small container (packaging) job that was completely created in Photoshop. It came to me as an RGB image 150 PPI at 100% size. Lots of hard lines and small type.

I now have to send a reply email explaining why I can't use this file.

This never gets old. (/sarcasm)
 
Hi geozinger,

It's a slightly different discussion, but I feel your pain. I just had to process a 32 pg catalog done in Photoshop 7 (to save money...) and once had to deal with an entire 100+ page book done in it.

I usually tell people that using Photoshop as a layout tool is like trying to use a screwdriver to pound in a nail. It can be done, but it's no fun for you, or the screwdriver -- and more often than not someone gets hurt before you're finished.

Shawn

It happened again. This morning I was sent a small container (packaging) job that was completely created in Photoshop. It came to me as an RGB image 150 PPI at 100% size. Lots of hard lines and small type.

I now have to send a reply email explaining why I can't use this file.

This never gets old. (/sarcasm)

----

If you'll note, the OP said "photos and graphics".

Graphics is where I have a bone to pick with folks. At the lower end, there are far too many folks using Photoshop for page layout and composition duties. Last week I criticized a poster on this board for using Photoshop in that manner, and we routinely are forced to take 'desperate measures' to get 1" X 1" .JPG logos saved at 300 PPI to work in high quality reproduction.

Plainly put, you need higher than 2 x LPI for raster images that have type or other kinds of hard edges in them. I would like to see a minimum of 5 x LPI for raster images that have hard edges in them. I would really prefer to see 8 x LPI in this situation, but it's rare I can get 2 x LPI.
 

PressWise

A 30-day Fix for Managed Chaos

As any print professional knows, printing can be managed chaos. Software that solves multiple problems and provides measurable and monetizable value has a direct impact on the bottom-line.

“We reduced order entry costs by about 40%.” Significant savings in a shop that turns about 500 jobs a month.


Learn how…….

   
Back
Top