civilized ku # 597-99 ~ printing

Former clubhouse, renovated in the Adirondack Rustic Tradition ~ Whiteface GC - Lake Placid, NY - in the Adirondack PARK • click to embiggen

Whiteface GC - 17th hole / rain shelter ~ Lake Placid, NY - in the Adirondack PARK • click to embiggenOn yesterday's entry, re: printed pictures/photo magazines, Dennis Allhouse (no link provided) commented/inquired:

I gather you print larger than 'native' size ie you've made references to sizes like 22 inch square images. What is your technique? Here I supposing that you must be uprezzing by one means or another, so I'm curious as to your approach.

My response: My first approach is to ignore all the gearhead, techno-fan, pixel-peeping stuff, re: inkjet printing, that is spread like a plague all over the internet.

Instead, my approach has been to make prints using various techniques (file sizing and sharpening) and then ... gasp ... look at them as a "casual" - read as, non picture maker - viewer would look at them. That is to say, look at them from a "proper" viewing distance that allows for taking in the whole of the picture (even though one's eye will inevitably move about the print's surface).

Using that approach, I have arrived at an approach that yields very very nice 24×24 inch prints - 22×22 inch centered image on 24×24 inch paper. The resulting prints/images appear to be quite sharp and artifact free. That's because they are quite sharp and artifact free. It is worth noting that, if I had an Epson 9800 (44' wide-format printer) instead of an Epson 7800 (24' wide-format printer), there is no doubt in my mind that much larger prints could be made using the same approach.

In short, my approach is as follows:

1) apply "normal" sharpening during RAW conversion/processing (in RAW Developer).

2) apply "normal" sharpening* as the very last step in PhotoShop processing:

*for me, normal sharpening in PS = convert image to LAB and, on the Lightness channel, apply Unsharp Mask - Amount 500, Radius 0.2, Threshold 0. Converting to LAB is extremely important - something that is rarely mentioned in all the gearhead blather about sharpening techniques. In LAB, the Lightness channel contains only the tonal/contrast image info, not any of the color info. Consequently, one can apply much higher levels of sharpening without any of the "halo" that results from applying the same level in RGB.

3) Using the Image Size function in PS, I turn on Resample Image and select Bicubic, set the Resolution to 180 pixels/inch, and the Width/Height to 22'. My file is uprezzed from 56.8Mb (native) to 89.7Mb.

4) convert to LAB.

5) apply the same "normal" sharpening to the Lightness channel.

6) convert back to RGB and print.

7) view the print and make any sharpening adjustments (more or less), if needed (rarely ever needed). IMPORTANT - first, revert to step # 4 in the History window, which will undo the previously applied sharpening, and then reapply new levels.

All of that said and done, I would guess that a gearhead, techno-geek, fanboy/girl might be able to apply some other more involved and convoluted uprezzing approach that might produce a print that is sharper when viewed from a pixel-peeping distance. However, that is not at all important to me and what I am doing with my picture making endeavors.

Not to mention the fact that it is also not very important to those who view my prints since they are usually forewarned that, if I catch them pixel-peeping one of my pictures, I will kneecap them with an aluminum baseball bat and then drag their crumpled body back to a "proper" viewing distance.

Comments

[Andre]

Mark,

Thanks for the info. I had often wondered about your workflow for getting 12mp files to print to 22".

Refreshing in these days of over sharpened images.

[Sven W]

Mark, one reason that converting to LAB is "rarely mentioned in all the gearhead blather" is that it has fallen from favour.

Authors like Martin Evening (who I believe derives his ideas from Bruce Fraser) use a blend mode of Luminance (assuming PS layers functionality). Apparently it is less destructive and faster than converting to/from LAB.

Further, by sharpening a copy of the image (in a layer with blend mode of Luminosity) it is possible to visually vary the "level of sharpness" by varying the opacity of this layer.

I'm no expert in these matters, so before taking me to task, pls refer to the relevant books from the above two authors.

Note - I agree with your comments re the value of producing prints and normal viewing distances. I have a small number of 21" x 28" prints from a 10Mp point-and-shoot camera hanging on the wall. Visitors like the images and are surprised when I tell them what camera I use.