As I understand it, the 720 / 600 would be for images with native “on the paper” resolutions that are above 360 / 300 ppi. If the native is below those numbers, have LR upsample (user controllable) only to 360 / 300. Unless there’s new thinking on this?
The general idea is that interpolating, beyond the native file resolution for a given output size, will not add real image detail. While that is generally true, it is IMHO only part of the equation.
1. There are algorithms that do add additional detail beyond what the image offered natively. "
Photozoom Pro" by Benvista, and OnOne's "
Resize 2018", both add additional resolution detail, so even above 300/360 PPI.
2. Having more pixels, allows improving the sharpening quality (because there are more pixels to work with at final output size and render edge detail with more precision).
3. There are a number of applications that allow improving micro-contrast at the output size, thus counteracting losses due to ink diffusion.
4. There are applications like
Qimage that use their proprietary interpolation and Smart Sharpening algorithms, and resample to whatever the printer driver expects
Whether there is enough of an improvement over the relatively decent Lightroom upsampling to warrant the additional steps and time, depends on the image and on potential printer buffer limitations.
Cheers,
Bart