Smugmug experiment

[rpcrowe]

I wanted to see if the image size that I upload to smugmug displays any different when I post it on CiC

I uploaded this to smugmug at 300 ppi giving it a size of 3458x3458



I uploaded this to smugmug at 72 ppi giving it a size of 800x800 pixels. I reduced the size using Perfect Resize

[Geoff F]

Second image appears slightly sharper to me.

[DanK]

I also think the second is slightly sharper.

It's not clear to me what exactly you are comparing. How did you post from smugmug? Did you use a link? If so, the size you chose in creating the link would determine how much of the downsizing was done by Smugmug and how much was done by this site.

As a practical matter, I don't pay any attention to this when posting here. I first post full-size images on Smugmug. I then past a link here using the "XL" size, which doesn't require much further downsizing to display here. Works fine.

[Dave Humphries]

I'm not surprised the second looks sharper.

First off, the PPI figure is irrelevant to actual displayed image - the image is the size it is; either 3458x3458 or 800 x 800. So forget 300 vs 72 ppi, set image size in pixels for web display, not: inches, centimetres, etc.

It is actually our browsers that do the downsizing to 700px wide (to fit the forum page width), but under instruction from code on the web page.

If you click the image, you'll see it larger in Lytebox (size now depends upon your screen size, browser banners, tool bars, etc.), if you click the square 'enlarge' icon in LyteBox, or hit the "F" key, it will expand until it is 1:1 pixel mapped to your screen and you'll need to scroll to see all the big one. Because the browser is downsizing so much, that's why the first looks softer when viewed at 700px in the thread or slightly larger in LyteBox filling (but not overflowing) the screen.

Ideally, it is best to follow my advice in the image posting thread, it was the result of much research a few years back.
Failing that, Dan's approach is close enough - although personally, I remain sceptical of web sites alternate size options, since they are producing images to reduce their bandwidth usage, not display optimally. That said, I haven't tested on SmugMug, so perhaps I am being unfair to them.

I prefer to downsize and sharpen under my control, uploading a size that will fit most people's screens at 1:1 in LyteBox at a jpg quality of 75% (or 9 of 12) - anything more is not visible, but dramatically increases file size and download time for all members, especially those on slower connections. I then link the 'original' size, I don't use (in my case) PBase's sized options.

[Mike Buckley]

When the two images are displayed at their largest size in the Lytebox, they are equally sharp. I assume that's because Richard sharpened the two images appropriate to their respective sizes. That being the case, it's understandable that the large image looks a little soft when being displayed in the thread as opposed to the Lytebox; it hasn't been sharpened after the downsizing has taken place.

Sort of the short version of what Dave explained.

[rpcrowe]

I realized that when I reduced the size of the 800 pixel image, I added sharpening through Perfect Resize. I did not add any output sharpening for the larger image.

Additionally, looking back on Perfect Resize, I realized that the output sharpening was in the high detail mode. I usually select Portrait mode when working with dogs. The other choices I have are General Purpose, Low Res JPG, Landscape and custom.

I think that the high detail mode might have been the reason that the second image seemed sharper.