What's the DIFF?

[Andrea B.]

TEST:  Can you detect any differences between these two images?

One is a compressed PNG and one is a high-quality JPG.

Both were saved with width 1800 px.

 

Please view these images within this post and don't click them up so far that the file extensions can be seen.

 

IMAGE A

 

 

IMAGE B

[Andrea B.]

TEST:  Can you detect any differences between these two images?

One is a medium-quality JPG with width 1800 px. 

The other is a medium-quality JPG with width 1093 px.

(Actually, I requested a resize to 1093 px, but I got a resize to 1092 px.)

 

Please view these images within this post and don't click them into the lightbox where the sizes will be obvious.

 

IMAGE C

 

IMAGE D

 

[Andrea B.]

On my Macbook Retina and in Safari browser which is fully expanded with no sidebars.

• I can't see differences between Image A and Image B.
• I think that Image D looks more detailed than Image C.

 

If I downsize the browser window, then I can't see differences between C and D. Interesting.......

[Andrea B.]

TEST:  Can you detect any differences between these three images?

One is a compressed PNG, one is a high-quality JPG saved in Photo Mechanic, one is a high-quality JPG saved in Photoshop Elements.

All were saved with width 1000 px.

 

IMAGE E

 

 

IMAGE F

 

 

IMAGE G

[Andrea B.]

I can't see any obvious differences between Images E, F and G.

 

*********************

 

I would like to repeat some of these tests with a close-up image to see what happens. Must go find a good candidate image to play with.

[Nate]

For A and B no visual difference viewed on 32in 2560x1440 monitor until downloading images and magnified 814% and in the dark areas in the sky where I see image B has better pixel smoothness(If that's a thing)

 

 

Out of images C and D,  D looks a little sharper

 

For E,F and G. No visual un zoomed differences until magnified 1442%, then F is the clear winner. Same part of the sky for all 3

 

 

 

[Fandyus]

Image B seems a tiny bit sharper than A but negligible.

[photoni]

I downloaded the four files
A (1800x1012 pixels) is equal to D and very similar to B
A.Jpeg weighs 1500KB
B.Png weighs 6600KB
C.Jpeg (1092x614 - weighs 564KB) visually it is the same as the other three, enlarged it has less details
.
Since PNG uses a different compression method than Jpeg (which works in LAB and increases the micro-contrast) everything seems logical to me

.

E.Jpeg = 478KB
F.PNG = 2000KB
G.Jpeg = 475

I opened the images as Smart Objects with photoshop all at 300 DPI...they are the same

.
Where do you want to go?
If the Brouser doesn't keep the original but creates a new Jpeg or PNG file with different compression (around 8/12 of Photoshop) the problem remains the same.

.
OT
PRO-licensed Flickr preserves the original Jpeg or PNG or TIFF file
(like a 11648 x 8736 from the GFX100)
then they produce 4-5-6-7 scaled files for display.
If it is not granted, only the author can see the original file, the others can at most see a 2000x2000 pixel (if I remember correctly)

[lukaszgryglicki]

There should just be an option to store unmodified files and that would solve all issues.

 

[Doppler9000]

I thought B was more detailed than A, and D was more detailed than C, based on the images as they were presented.

[Andrea B.]

Thanks to all for the comments !!  I've been working to determine what are the optimum dimensions for viewing a photo on UVP. But there is a very wide range of monitors and browsers in use by viewers. 

 

The only solid facts I have so far are these:

 

If you resize your photo to a maximum width of 1800 pixels and save it as a losslessly compressed PNG, then the forum software will not alter the file. The photo will look very good both on the page and in the lightbox under the assumption that the viewer is using a fully expanded browser (no sidebars) on a PC monitor or a large laptop monitor. The photo will be dynamically resized to about 1100 pixels for display within a post. When clicked up for display within the lightbox, it will display at 1800 pixels for most monitors and large laptops. 

 

If you resize your photo and save it as a JPG, then there is a minor quality hit on the JPG quality by the forum software but no change to the dimensions**. I have not yet experimented with extra sharpening as a way to overcome the quality hit. Extra sharpening is typically done before printing to overcome the "spread" of the printer inks, so I'm thinking that extra sharpening might be useful for posted JPGs.

 

This topic was a follow-up to my lengthy test topic here:  TESTING image sizes again. TIF upload for you to play with in Post #12.

 

 

 

**No change to the dimensions in the stored file. When displayed, a photo is dynamically resized. If your monitor/browser combo does not support the original file dimensions, then you will see a dynamically reduced size within  a topic or in the lightbox.

 

TEST THIS for yourself!

Upload a 600 pixel width photo. On a PC monitor or laptop monitor, fully expand your browser and close all sidebars. You will see the full 600 pixel width, unresized.

Now test upload a 1600 pixel width photo. It should display unresized in the lightbox but will be dynamically resized within the post.

 

You can test sizing uploads over in the test topic linked above.