Opinions on Lowest Usable PPI?

[mephetic436]

I am trying to make the largest print I can from a 3072x2048pixel jpeg file, that I expect to be viewed from a few feet away.

So at the range of 2-4ft viewing distance what is the lowest ppi I can use that will not degrade the quality of the picture?

100ppi for a 20x30? 130ppi for a A2? Lower?

I understand this is a very subjective question about what "acceptable, but I've never made enlargements so any experienced advice would help.

[eddarr]

Do you have a Sam's club or Costco nearby? 20x30's only cost about $12 to find out for sure.

Just resize the photo to 20x30 at 300ppi in PS.

[mephetic436]

Just resize the photo to 20x30 at 300ppi in PS.

Dont u mean dpi?

[Damo77]

Dont u mean dpi?

No, DPI (dots per inch) specifically refers to printing. PPI (pixels per inch) is the correct terminology for digital photos.

The two are not the same - for example, for offset printing (eg business cards) you would need to supply a 300ppi file, but it's printed at 150dpi.

[Damo77]

Just resize the photo to 20x30 at 300ppi in PS.

I agree. Then sharpen a little if necessary.

[macropod]

So at the range of 2-4ft viewing distance what is the lowest ppi I can use that will not degrade the quality of the picture?A 2-4ft viewing distance is a huge range to accomodate; the print resolution you need at 2ft is double what you need at 4ft.

About the largest you can go with your camera for a really sharp print is about a:
. 13.5*9in print @ 255ppi for viewing at 18in,
. 18*12in print @ 191ppi for viewing at 24in,
. 27.5*18.33in print @ 127ppi for viewing at 36in, and
. 36.75*24.5in print @ 96ppi for viewing at 48in.
Now, if you want the same print to appear sharp at all those distances, you'll need to use the ppi required for the shortest distance. Unless you're into resampling to produce a larger print (thus masking the lack of pixels in the original image), you're not going to be able to produce a really sharp print larger than the dimensions corresponding with the shortest viewing distance.

Of course, you may not need your image to be so sharp, in which case you could increase the image dimensions by up to 40% on each side (which would reduce the potential ppi figures by 30%). For a calculator that'll help you work these things out, see my posts in:http://photography-on-the.net/forum/showthread.php?t=408465

Cheers

[Bobster]

for example, for offset printing (eg business cards) you would need to supply a 300ppi file, but it's printed at 150dpi.
actually its printed @ 1200-2400 DPI, but 300DPI photos will be printed @ 150-200LPI ;)

[Damo77]

300DPI

?? Come again?

Yes, well, I didn't go into LPI, but for all intents and purposes it's the same thing. And yes, in the offset printing process an image is printed twice - once onto the plate, then onto the paper.

[Bobster]

offset print isn't limited to 300DPI it actually prints text and line art @ 1200-2400DPI depending on the quality of the imagesetter, and what is actually being printed.

saying that business cards are printed @ 150DPI is false, they're probably output at something like 1270DPI @ 150LPI

and DPI and LPI aren't the same thing ;)

[Damo77]

Yeesh! Do you think it matters?

The question was regarding the difference between DPI and PPI, which I felt was relevant enough to photographers to provide clarification. The average photographer does NOT need to know about imagesetters, and the resolution of text on a business card certainly doesn't keep them awake at night, I'm sure.

[mephetic436]

Thats great fellas, but in regards to my question the 2 answers I've received so far are:

Just resize the photo to 20x30 at 300ppi in PS.
and

About the largest you can go with your camera for a really sharp print is about a:
. 13.5*9in print @ 255ppi for viewing at 18in,
. 18*12in print @ 191ppi for viewing at 24in,
. 27.5*18.33in print @ 127ppi for viewing at 36in, and
. 36.75*24.5in print @ 96ppi for viewing at 48in.
Now, if you want the same print to appear sharp at all those distances, you'll need to use the ppi required for the shortest distance. Unless you're into resampling to produce a larger print (thus masking the lack of pixels in the original image), you're not going to be able to produce a really sharp print larger than the dimensions corresponding with the shortest viewing distance.

Of course, you may not need your image to be so sharp, in which case you could increase the image dimensions by up to 40% on each side (which would reduce the potential ppi figures by 30%).


Is that generally accepted?

[SilverHCIC]

When you resize your image to 20x30 @ 300dpi in PS, make sure you select "bi-polar sharpen" as your resize mode. ... I have printed a number of shots taken from a Canon A460 P&S at 24x36 which came out just fine after resizing @ bi-polar sharpen mode.

[SilverHCIC]

BTW - your file size will be HUGE, so if you are uploading to Shutterfly, Winkflash, etc. to order the print it will take a while to load.

[Damo77]

When you resize your image to 20x30 @ 300dpi in PS, make sure you select "bi-polar sharpen" as your resize mode. ... I have printed a number of shots taken from a Canon A460 P&S at 24x36 which came out just fine after resizing @ bi-polar sharpen mode.

LOL ... it's "Bicubic" not "Bipolar".

And I believe it's advised to use Bicubic Smoother when upsizing, and Bicubic Sharper when downsizing.

[eddarr]

LOL ... it's "Bicubic" not "Bipolar".

And I believe it's advised to use Bicubic Smoother when upsizing, and Bicubic Sharper when downsizing.

Dude, your bipolar:lol:

I have not personally tested this. But there are a few threads on here showing the results of using the different algorithms and doing it all at once and in 10% steps. When viewed on a monitor, I could not see any difference between them.

[SilverHCIC]

LOL ... it's "Bicubic" not "Bipolar".


^^^^^ That's funny !!! I didn't even notice I had typed that!! But you're right, it's bicubic-sharper. But anyway, I've gotten decent results even from my A460 P&S.

[tim]

Prints are cheap, try it out. I have great 50" prints from a 5D, and 40" prints from a 30D. There's a printing FAQ in my sig.

[macropod]

Is that generally accepted?Probably not, but I have no idea for sure. There are many who doggedly stick to a fixed ppi, like 300ppi, regardless of the print size and viewing distance. All that achieves is creating unnecessarily large print files.

It's simple enough to work out what you can do with a bit of basic knowledge, though. In essence, for a given print size you can simply divide the number of pixels in a given dimension by the size of the print in the same dimension to work out the required ppi. That bit's pretty simple. To work out the minimum acceptable viewing distance for that ppi requires a decision on how sharp the print needs to look. The figures I've given are based on a visual resolution limit of 1.5 arc-minutes. That equates to a 300ppi/dpi print viewed from 15.3 inches. Everything else is scaled from there and rounded off to the nearest ppi and inch of viewing distance. The alternate figures I suggested are based on a visual resolution limit of 2 arc-minutes, which is what you'd see in a 200ppi/dpi print viewed from 17.2 inches.

Cheers

[Damo77]

There are many who doggedly stick to a fixed ppi, like 300ppi, regardless of the print size and viewing distance.

Yes, I guess I am pretty anal about it, now that you mention it.

The reason I stick to a standard resolution (250ppi at our lab) is because of sharpening. I know exactly what USM radius I need for premium sharpness for various print sizes. To achieve a standard USM radius, obviously I need rigid resolution.

My choice of Radius is dependent on print size (and, by association, viewing distance), so in a way I guess it's related to this topic.

[macropod]

Yes, I guess I am pretty anal about it, now that you mention it.

The reason I stick to a standard resolution (250ppi at our lab) is because of sharpening. I know exactly what USM radius I need for premium sharpness for various print sizes. To achieve a standard USM radius, obviously I need rigid resolution.

My choice of Radius is dependent on print size (and, by association, viewing distance), so in a way I guess it's related to this topic.While I understand (and agree with) the need to take viewing distance into account when deciding on the USM radius, I can't see any necessary connection between that and the choice of a fixed resolution. Since either method means calculating the required radius, and there's a limit to how much detail can be seen from a given distance, what do you gain by creating the larger print files that applying a fixed resolution to large prints requires? IMHO, all you're really achieving is creating work for yourself and overly large print files.

If I'm wrong about this, perhaps you could enlighten me?

Cheers

[Damo77]

Well, I've rarely printed bigger than 12x18, so file size isn't a problem in my situation.

If I was printing bigger than that, and file size became an issue, I'd be perfectly happy to drop the res down to 150, or 100, or whatever. The point is, I'd establish a standard for myself, and stick to it.

Where I see a problem, is in the weird resolutions you get when you resize without resampling. If I want to use my preferred USM radius at a given print size, but one file is 381.76ppi, and another one is 268.91ppi (because of a tighter crop), then I'm reaching for a calculator every time to work out the correct radius. Now that's inefficient!

[macropod]

Where I see a problem, is in the weird resolutions you get when you resize without resampling. If I want to use my preferred USM radius at a given print size, but one file is 381.76ppi, and another one is 268.91ppi (because of a tighter crop), then I'm reaching for a calculator every time to work out the correct radius. Now that's inefficient!But surely you're having to do that anyway given that the normal viewing distances will likely be different?

Cheers

[Damo77]

No, not at all.

I know exactly what USM radius I need for premium sharpness for various print sizes.

[macropod]

No, not at all.

Quote:
Originally Posted by Damo77 http://photography-on-the.net/forum/images/buttons/viewpost.gif (http://photography-on-the.net/forum/showthread.php?p=5241216#post5241216)
I know exactly what USM radius I need for premium sharpness for various print sizes.
So how is that any different to knowing what you'd need with just the one (unsharpened) file that you could either (a) sharpen before printing then discard the changes, or (b) make multiple copies of without changing the resolution but with different amounts of (remembered) USM radii applied? I just can't see how making larger copies of the file gains anything for what you've described.

Cheers

[Damo77]

This has gone on so long I can't remember what we were talking about.

Look, I'm not proposing anything revolutionary here. My workflow goes thus:

1. The file from the camera - either Jpeg or Raw, whatever I'm given - is archive unchanged.

2. I save a Working file to which I do all my edits (directly to the image, I see very little need for adjustment layers - but that's an argument for another thread). I don't resize this Working file, and rarely crop it. Every pixel is sacred. Of course, I pay no attention to resolution at this stage. As I've said frequently, image resolution is arbitrary and irrelevant until output.

Oh, and I definitely don't sharpen the Working file.

3. I duplicate the Working file to create an Output file. This is the one I resample to the required size and res, then sharpen to suit (as previously discussed), and save and send to the lab.