You are at netAirspace : Forum : Spotting and Photography Forums : Photo Editing & Feedback

When it comes to sharpening images after processing them in ACR, is it best to do it before cropping and resizing or beforehand? Historically I always used to do it on the image once cropped to a usable size, but I think this may have had as much to do with my PC performance than the end results.

With performance no longer an issue, I'm using various methods on the full size image without seeing any significant changes to the image and so am wondering if perhaps I need to be a bit heavier on the sharpening due to the additional pixels.

My steps are usually:

• Level and crop
• Sharpen, using Smart Sharpen (40%, 0.4px)
• Downsize, using "bicubic, sharper"
• Sharpen again, using the High Pass filter (0.4px, hard or soft light blending option, depending on the photo)

I would recommend that workflow only for RAW shooters, though.

When shooting JPEG, the camera will have added some sharpening already, depending on the settings. I would then leave the first sharpening (before downsizing) away.

Ah, that could also be a discrepancy between other workflows: I tend to crop to a specific size (usually 1400x933) to be reduced only if required. I think I'll try cropping at full size as well to see if that has any effect. Thanks for that, Stefan.

mhodgson wrote:Ah, that could also be a discrepancy between other workflows: I tend to crop to a specific size (usually 1400x933) to be reduced only if required.

Sorry, not sure what you mean by this.

Anyway, my workflow goes:

RAW Editing:
1. Rough adjustment of exposure/levels/contrast and white balance.
2. General NR if needed.
3. Sharpening is usually left at whatever I've set the default to (currently 3 in Canon DPP; usually use radius 1.3 / amount 70 in ACR).
4. Export to JPEG.

JPEG Editing:
5. Level and crop.
6. Fine adjustment of exposure/levels/contrast and color balance.
7. Resize in one step (standard bicubic).
8. Sharpen (Unsharp Mask, typically radius 0.4, amount 60-90, threshold 2). I sharpen on a duplicate layer then erase OS areas, then flatten the layers.
9. Any local NR if necessary.

I've been sharpening most of my stuff through the noise reduction interface lately and then using a .2 high pass.

vikkyvik wrote:

mhodgson wrote:Ah, that could also be a discrepancy between other workflows: I tend to crop to a specific size (usually 1400x933) to be reduced only if required.

Sorry, not sure what you mean by this.

Basically, I'd crop and resize at the same time using the functions of the crop tool - so rather than simply retaining the image proportions I'd crop it straight to the dimensions of 1400x933. I've discovered today that keeping it as two steps seems to produce much better results!

mhodgson wrote:Basically, I'd crop and resize at the same time using the functions of the crop tool - so rather than simply retaining the image proportions I'd crop it straight to the dimensions of 1400x933.

Oh. I didn't even know you could do that!

I do my sharpening as the final step before saving the file.

1. Level & Crop if needed.
2. Add a bit of saturation.
3. Resize.
4. Sharpen.
5. Save file.

I do:

1. Contrast Mask or something similar
2. Curves
3. Levels.
4. Saturation.
5. Specific color correction.
6. Level & crop
7. Resize
8. Sharpen (usually smart sharpen)

I need to learn how to use NR now that I'm uploading old slides.