Output Sharpening during Export in Lightroom Classic

Hello,

I have posted this in follow up to the Zoom Event last week about post-processing in Photoshop.

I use Lightroom Classic for exporting my images and would appreciate any feedback please in regard to output sharpening during the export process.

If I am starting with a RAW image that is between 4000px and 6000px on the long edge and I want to export to a JPEG that is 1800px on the long edge, how should I approach output sharpening?

The options are to: not sharpen at all, or sharpen for screen, matte paper or glossy paper, by an amount of low, standard or high.

I may have already sharpened the image during post-production, but should I be applying shapening again as part of significantly reducing the length of the long edge?
And if I should be doing that, should the amount that I sharpen by (be it low, standard or high) depend upon the proportion that the exported long edge is compared to the long edge of the original?
Or should I open the image in Photoshop and export from there, where I can choose a more tangible amount?

Any advice greatly appreciated. Thank-you.

Best regards,
Craig

Hi Craig,

If using Lightroom for RAW conversion and initial processing, the advice I follow for output sharpening is to transfer the image from Lightroom to Photoshop (by right-clicking on the image and selecting the 'Edit in' command), then downsizing the image in Photoshop, then applying the final sharpening (typically using Smart Sharpen) only after the downsizing. This method gives a lot more control over the sharpening process. The parameters to use for downsizing and final sharpening are discussed in Ian Wilson's "Notes on Using DPP4 and PSE14, Part 2: PSE14", available in the Resources section on the BLP website, click here . While Ian prefers DPP4 for RAW conversion (and it is worth also reading his Part 1 to see why), his PSE14 notes are applicable to both versions of Photoshop.

To save as a JPEG from my version of Photoshop requires a couple of extra steps. Firstly, converting the colour space to sRGB using Edit->Convert to Profile, then using File->Save a Copy to get the option of saving as a JPEG file.

Cheers, David

Thanks David for your feedback. I will take a look at those resources you mention too.

Best regards,
Craig

Hi Craig

It's a long time since I used Lightroom. However, I think the different options for output sharpening will give you pre-set sharpening levels intended to optimise sharpening for the medium with which you wish to display the output. I think you would be best off using sharpening for screen as this is how the files will be viewed on the Birdlife website.

I don't think there is any hard and fast rule for the amount of sharpening, and it will depend on the file. When I downsize for BLP I find that sometimes the files work best with a lower level of sharpening and sometimes with a higher level. I think you just need to export with different settings and see which you prefer. If the file gets a 'hard' grainy look to it, and there are noticeable bright halos around precise edges (generally first evident around sharply focused areas which have abrupt edges (eg the bill and legs, particularly if viewed against the sky) then you have overdone it. I am not aware that the low/standard/high values have any direct link to the length of the sides of the image - in the sense that for a particular range of pixel numbers, high is better than standard for example.

You can keep sharpening as you normally would during your post processing. When it comes to exporting files downsized, you need to apply additional output sharpening in most cases to get an optimal output. This is because the downsizing process normally reduces sharpening so the resultant files look softer. Don't ask me to explain the technical process involved, but part of it may be that when you work on the file at full size, you can apply a relatively fine level of sharpening (across only a few pixels). However, this fine sharpening becomes less visible when you downsize the file (which involves combining/averaging the values of adjacent pixels in the file).

It is often worth doing online searches for specific advice on Lightroom. There are vast resources on the web and youtube on just about every aspect of Lightroom. Of course these are of variable quality, but Adobe has lots of material and tutorials even if you ignore the youtube 'experts'. I learnt a lot about the more exotic aspects of LR by trying lots of these resources.

Regards
Simon

Craig, I can't comment specifically on the export sharpen option in Lightroom but can add two points that might help:
1. In terms of amount of sharpening, it's critically important to view your final downsized image at 100% (meaning one pixel on your screen is exactly one pixel in the image). If you don't do this and display at a different zoom level, the zoom function itself will introduce blur and artefacts (see point 2 to understand why). Also, normally an image will display at 100% when loaded from our web page, so that's what most users are going to see when they view your downsized photo.

2. Blur is introduced during downsizing because you are asking the software to take, for example, a block of 3600 x 1800 pixels in the original image and turn them into a block of 1800 x 900 pixels. So for every pixel in the final image, it will have to be derived from 4 pixels in the original image : 2 horizontal and two vertical pixels will combine to one final pixel. If there is a sharp tonal boundary such as a feather filament in two of the starting pixels such that two pixels are pure black and two are pure white, the simplest way is to average the content of the pixels to get a single pixel from them. This will result in one grey pixel in the final image. This essentially destroys the sharp tonal boundary in the original image pixels. Imagine now if there were alternating black and white pixels in the original image. When this was downsized, you'd end up with a blob of grey pixels with no noticeable tonal difference between them = no detail. People who create image manipulation software know about this and try to create processes that reduce this effect. I personally don't know anything about the detail of these or the mathematics behind them but the reality is that blur will still be introduced to some extent. So sharpening a downsized image is critical to try to remove some of the blur.

Thanks Simon and Glenn for your thought provoking advice - certainly will take it on board.

Best regards,
Craig