res setting for web

[Shadowman]

downsized to 72pixels.



Image on the right is saved at resolution 300pixels/inch res.

[Stagecoach]

Looks OK to me John, but then would really need to do a side by side comparison.

[george013]

downsized to 72pixels.

What do you mean? The picture is still 6016x3357 in the exif.

George

[Shadowman]

What do you mean? The picture is still 6016x3357 in the exif.

George

George,

When I download dimensions show as displayed here.

What type of reader are you using?

[Stagecoach]

John, the second image (300ppi) you posted seems to have changed in content, the birds got thinner

[george013]

The resolution and the wanted output size are related to each other. I think you did choose a metric dimension and a resolution. The output device is calculating a new file size.
When I look in the exif I get a filesize of 6016x3357, the size of the file. When I download it I get 1444x806, the size of the output.

George

[Shadowman]

John, the second image (300ppi) you posted seems to have changed in content, the birds got thinner

Hi Grahame,

Yes, I noticed that when resizing, I had to decrease the pixels a bit (1000x1792 on left) and 806 x 1444) to get the file size below what I feel is TinyPics file size constraints. I noticed that any file above 5MB usually takes longer to upload if it uploads at all.

[Shadowman]

The resolution and the wanted output size are related to each other. I think you did choose a metric dimension and a resolution. The output device is calculating a new file size.
When I look in the exif I get a filesize of 6016x3357, the size of the file. When I download it I get 1444x806, the size of the output.

George

George,

What app do you use for viewing EXIF?

[xpatUSA]

George,

What app do you use for viewing EXIF?

Are we actually getting confused between the ppi which is only for printing and the image appearance on-screen?

The posted images are a bit different in pixel sizes as already stated. The file sizes (Kb) are very different because the JPEG sub-sampling is 4:4:4 (no chroma sub-sampling) for the 72ppi (oddly enough) but 4:2:0 (both horizontal and vertical sub-sampling) for the 300ppi. This from Harvey's ExifToolGUI and confirmed by JPEGsnoop. Seems backasswards to me.

Neither app seems to mention an (original?) size of 6016x3357px - perhaps it's buried in the XMP meta-data (added by Elements 14) which is far too long and boring to read.

Any reason for the embedded ICC profile being ProPhoto RGB, John, other than "Adobe strikes again"?

My memory of Elements (6) is that it often selected what Adobe thinks is best for you - as opposed to what you actually wanted, LOL.

[george013]

George,

What app do you use for viewing EXIF?

An add-on in FF, Exif Viewer. How big was the file when you uploaded it? I mean in pixels.

What is your purpose to play with the print resolution? Find yourself a filesize that suits you. I mostly use something as longest side 1600, or something like that.

George

[Shadowman]

Are we actually getting confused between the ppi which is only for printing and the image appearance on-screen?

The posted images are a bit different in pixel sizes as already stated. The file sizes (Kb) are very different because the JPEG sub-sampling is 4:4:4 (no chroma sub-sampling) for the 72ppi (oddly enough) but 4:2:0 (both horizontal and vertical sub-sampling) for the 300ppi. This from Harvey's ExifToolGUI and confirmed by JPEGsnoop. Seems backasswards to me.

Neither app seems to mention an (original?) size of 6016x3357px - perhaps it's buried in the XMP meta-data (added by Elements 14) which is far too long and boring to read.

Any reason for the embedded ICC profile being ProPhoto RGB, John, other than "Adobe strikes again"?

My memory of Elements (6) is that it often selected what Adobe thinks is best for you - as opposed to what you actually wanted, LOL.

Using the Image Size Dialog box, it was suggested in Kelby's text that resolution of 72 pixels/inch was sufficient for posting online. Regarding the ICC profile, I normally use ProPhoto RGB for print work, didn't alter for the experiment.

[Shadowman]

An add-on in FF, Exif Viewer. How big was the file when you uploaded it? I mean in pixels.

What is your purpose to play with the print resolution? Find yourself a filesize that suits you. I mostly use something as longest side 1600, or something like that.

George

George,

See post #11 for reason, I noticed that file size normally uploaded were at maximum document size, figured why do so if its not necessary for simple viewing.

[xpatUSA]

Using the Image Size Dialog box, it was suggested in Kelby's text that resolution of 72 pixels/inch was sufficient for posting online.

Then Kelby is either wrong or has been misunderstood, sorry John. The box "Document Size" sets parameters for printing only and definitely not for viewing by a browser on-line.

[Shadowman]

Then Kelby is either wrong or has been misunderstood, sorry John. The box "Document Size" sets parameters for printing only and definitely not for viewing by a browser on-line.

Word for word Kelby writes, "A resolution of 72 ppi is considered "low resolution and is ideal for photos that will only be viewed onscreen (such as Web graphics, slide shows, etc.). This res is too low to get high-quality results from a color inkjet printer, color laser printer, or for use on a printing press." p. 119, the photoshop elements 8 book for digital photographers. I decided to give it a try.

[xpatUSA]

Word for word Kelby writes, "A resolution of 72 ppi is considered "low resolution and is ideal for photos that will only be viewed onscreen (such as Web graphics, slide shows, etc.). This res is too low to get high-quality results from a color inkjet printer, color laser printer, or for use on a printing press." p. 119, the photoshop elements 8 book for digital photographers. I decided to give it a try.

His statement is wrong or misleading and has been misunderstood.

Please regard these two images:





Both 1320x880px, 496 Kb

One is set to 72ppi, the other 600ppi. Flip them back and forth in the LightBox.

If we take Kelby literally, there should be some sort of huge visible difference but there is none whatsoever.

72ppi is not "ideal for photos that will only be viewed onscreen" - 72ppi, or any other ppi for that matter, is meaningless for on-screen viewing.

Kelby is wrong, as proven by the above two images.

P.S. any Sigma output that I post here will be and always has been their default of 180ppi and I have never changed that until today.

[Stagecoach]

Oh no he's not

This sentence of his statement .....................

"A resolution of 72 ppi is considered "low resolution and is ideal for photos that will only be viewed onscreen (such as Web graphics, slide shows, etc.)..

is in relation to the'default' setting that is in the box which is 72ppi and he then goes on to say this is no good for printing and explains how and why you change this to get the print size you want from your 6000 x 4000 (forgot the figures and have put the book away) px file that came out of the camera.

[xpatUSA]

Oh no he's not

This sentence of his statement .....................

"A resolution of 72 ppi is considered "low resolution and is ideal for photos that will only be viewed onscreen"

is in relation to the'default' setting that is in the box which is 72ppi and he then goes on to say this is no good for printing and explains how and why you change this to get the print size you want from your 6000 x 4000 (forgot the figures and have put the book away) px file that came out of the camera.

Oh yes he is

Please explain by formula or by example the relationship between "photos that will only be viewed on-screen" and the desired print resolution. [none?]

Please explain what effect the EXIF IFD0 tags 'XResolution' and 'YResolution' have on an on-screen image. [none?]

Please explain why there is a further 'ResolutionUnit' and what the value of that has to do with an on-screen image. [none?]

These tags have nothing to do with on-screen viewing - that is my only point. Mentioning them in the context of on-screen viewing is pure obfuscation.

I must be going insane because nobody here is understanding my point.

glurk . . .

[Stagecoach]

Oh yes he is

Please explain by formula or by example the relationship between "photos that will only be viewed on-screen" and the desired print resolution. [none?]

Please explain what effect the EXIF IFD0 tags 'XResolution' and 'YResolution' have on an on-screen image. [none?]

Please explain why there is a further 'ResolutionUnit' and what the value of that has to do with an on-screen image. [none?]

These tags have nothing to do with on-screen viewing - that is my only point. Mentioning them in the context of on-screen viewing is pure obfuscation.

I must be going insane because nobody here is understanding my point.

glurk . . .

Ted,

If you wish to diagnose what has been written in that section in the book and also to bring in formulae, Xresolutions and tags you are welcome, some of us simply aren't interested.

But for me, and I have the book, what he is saying makes sense and I would not spend my time trying to prove that his terminology may be used differently to what I think it should be.

I do not see any indication or suggestion in what he has written in that section to suggest he is "literally" meaning what you have posted with those two pics, but you do?

[george013]

I think you're forgetting the "resample image" check. Without that marked it's an order to the printer to change the image size so it looks like the printer is using 72dpi when printing at full size. When resample image is checked, the image size is changed on the forehand. That's what I learned here a while ago. I can't see any difference with resizing.
I can't read the reason from post #1. But what I think to understand now the limitation of the filesize by Thinypic is the reason and how to include an as big as possible image in that file.
When this is the problem a lower compression might help, or don't include all the metadata.

George

[xpatUSA]

Ted,

If you wish to diagnose what has been written in that section in the book and also to bring in formulae, Xresolutions and tags you are welcome, some of us simply aren't interested.

But for me, and I have the book, what he is saying makes sense and I would not spend my time trying to prove that his terminology may be used differently to what I think it should be.

I do not see any indication or suggestion in what he has written in that section to suggest he is "literally" meaning what you have posted with those two pics, but you do?

Grahame and John, I fold.

By all means, decline to answer my questions, that is your privilege.

I can find no way to make my point (that the ppi setting has no effect on an image posted for viewing on the web) any clearer. Here's another question to ignore:

Is that point correct or not?

I simply do not understand why a print resolution of 72 ppi is any more or any less "ideal" than any other resolution for photos that will only be viewed on-screen.

Sorry.