Raw & Post Processing, Printing > Printing: Printers, Papers and Inks

Standard print and mat sizes

<< < (3/3)

sojournerphoto:

--- Quote ---I'm in a similar position, so this thread is interesting.  I, too, recently got my 3800, and so far have been printing 8x10's and 11x14's for mounting in standard pre-cut mats.  I've been using 11x14 mats (8x10) opening, and 16x10 mats (11x14 opening).

However because I have to crop my images severely to fit those dimensions, I'm noticing a considerable lack of sharpness in my prints.  I'm only working with 6MP images, so I suspect the cropping is a real issue.  I'm using Photoshop CS3 for cropping, using the methods it recommends based on the direction of my cropping.

On occasion I've seen prints matted, where the image size is smaller than the actual opening in the mat.  For example, I'm experimenting with an uncropped image which prints (at 360dpi for the 3800) at 5.5 x 8.5.  This leaves a border around the image inside the mat opening.  At this uncropped size, the image appears much sharper than the cropped ones.

Is this practice considered "acceptable".  Is there room for "artistic freedom" with this, or are there established guidelines or "best practices" on this sort of thing?
[a href=\"index.php?act=findpost&pid=153645\"][{POST_SNAPBACK}][/a]
--- End quote ---

Cropping shouldn't directly lead to unsharp images. Upressing may lead to a noticeable loss of detail - look at one of Michael's Madagascar images on the road - so you may be are doing more than you think when cropping.

Perhaps you could run through your typical process.

By way of info I print according to what I think the image needs and offer sizes that I think are appropriate. Sometimes that means a crop, other time not. I either get prints matted by a trusted local frameshop or do anything up to 40inches per side myself (as an IPF5000 used that really means up to a mat of 40 by 22 or 24.

I'm also starting making my own frames for smaller work.

Mike

bheiser:

--- Quote ---Cropping shouldn't directly lead to unsharp images. Upressing may lead to a noticeable loss of detail - look at one of Michael's Madagascar images on the road - so you may be are doing more than you think when cropping.

Perhaps you could run through your typical process.

...

sojournerphoto:
Hi

I'm sure that workflows do vary quite a bit, but here are a few thoughts:) Also Camera to Print is useful I think - Lightroom and ACR are very similar and have the same engine inside.

1. Don't worry about dpi at this stage of the workflow. Do learn to use ACR/LR/other as well as you can to get the best overall image into photoshop. I'm still learning loads and recently have been amazed at how far you can pull an image in LR and still get ded=cent prints.

2. I tend to do levels, curves, colour work local contrast enhancement (usm with radius 30 to 60 and amount 10 to 30) and creative sharpwning/blurring in CS3, using masking, blending options to protect highlights and blacks and luminosity blend mode for curve/levels adjustments on rgb.

3. yes, save a master file at original resolution.

5. yes

6 and 7. I wouldn't do this at this stage.

10. I wouldn't fix the resolution at this stage. If you want to fix the aspect ratio just leave the resolution to fall where it falls. In your workflow you will actually be upressing the file at this stage if you crop in both directions.

11. use image size to check your output resolution (no resizing yet) as this will guide your output sharpening. For output sharpening you could use usm, smart sharpen or a plugin like photokit. The radius and amount required will depend on resolution, media, subject and print size! By way of illustration yesterday I printed a set of wedding album pages at image sizes from 8 by 5 up to 12 by 8 on innova smooth rag paper. Native resolution varied from about 250ppi to 550ppi and I used usm with radius ranging from 1.9 down to 1.1 and amount from 100% up to 175%. I also sharpened on a layer and masked some areas:) Basically I find matte rag paper needs more sharpening than glossies and more detail means lower radius. resize without interpolation if necessary.

12. It's not clear if you are converting to the output profile or not here. If not this would explain why prints aren't working. If the printer doesn't manage the colours then you need to either convert to the paper profile in CS3 or tell the driver to do it. Also, what is your working profile in CS3?

Finally, a couple of other thoughts:

- the print driver will interpolate your data for you without the need yo do it in CS3. My IPF5000 allows me to use bicubic so I usually just let it, but you would need to do a side by side test of the driver against CS3 resizing to decide which worked best for you.

- Softproofing is really useful when you get the colour management sorted out. Particularly on matte papers it makes a huge difference to output quality.

Hope there is something of help in there:)

Mike



--- Quote ---Hi Mike,
Thanks for the suggestion.  I think you're right - my workflow is just what I've cobbled together from taking bits and pieces of what I've read on various sites.  I'm not convinced it's the "right" workflow.  I'm probably doing something to clobber the images.

I've managed to get them to look "ok" but definitely nowhere near how I want them to look.

What I've put together so far is something like this:

1.  load the raw file [question here of what dpi I should use reading the file.  Default from my D70 in ACR is 72.  Default for the D300 is 240.  I need to end up with 360 for printing on the Epson 3800.]

2.  Some combination of edits, usually Levels, Curves, etc in PS CS3.  My goal (which I rarely achieve so far) is to (as they say) "make the image really pop".  So I increase saturation (but only by a tiny bit from the D300, more from the D70), and use Curves to increase contrast.

3.  I typically want to end up with an "edited" file that I can then later use for various print sizes, say 11x14 and 8x10.  So at this point I save the edited file in a "edited-not-sized" directory.

4.  Then for each size print I want I do something like this:

5.  Load the edited-not-sized file into PS CS3.

6.  [this is a step I have just refined based on a tutorial on http://www.lynda.com.  In Image Size dialog, turn resample off, then change PPI to 360, and OK.

7.  Then in Image Size dialog again, turn re-sample on, and set the right dimension for the shortest side of the image (e.g. "8").  That results in something that's 8" wide, but longer than 10".  For this step I use Bicubic Smoother (best for enlargement) because the image size I want is usually greater than what's displayed in the Image Size dialog box.

10.  Then go to the crop tool, and set dimensions of 8x10, set Resolution to 360, and crop.  My PS default Image Interpolation setting is set for Bicubic Sharper (best for reduction) since at this point I'm actually reducing.

11.  Then use Unsharp Mask, usually around 100%, radius 3 [honestly I'm not really sure what's right here, but these are values I picked up somewhere in a tutorial].  I've also used radius 1, and %'s varying from about 50 to 120].

12.  Then I save & print the file.  In the print dialog I use 2880x1440dpi, no color adjustment, Photoshop manages colors, the profile for the paper I'm using, and Rendering Intent: Perceptual, and uncheck Black Point Compensation.

I've hesitated to post this here becuase I suspect there probably as many different workflows as there are photographers (and because I am not sure this flow even makes sense .
[a href=\"index.php?act=findpost&pid=156263\"][{POST_SNAPBACK}][/a]
--- End quote ---

Navigation

[0] Message Index

[*] Previous page