scanned pt/pd prints grainy?

[alecmoody]

I scanned some of my 5x7 pt/pd prints and they look grainy (the originals do not). I think this is just the paper texture but I was wondering if anyone has any other ideas? The prints themselves are printed with all mostly A (1 sometimes two drops of B) on bergger cot-320. also, they are 6pd 2pt so they are in fact smooth.
Scanning at a high resolution on an epson v700 as a 16bit color tiff.

examples:

 

[MattKing]

alec:

First, your work - wow that's fascinating/great!

Secondly, I would guess that you are having aliasing problems with respect to the texture in the paper and the scanning process.

I don't know for sure, but that is what I suspect.

Matt

 

[David A. Goldfarb]

This is one of the problems I have with print scans in general--surface artifacts, even with many silver gelatin papers. One solution for presenting traditional prints in the APUG galleries or on the web in general is to photograph them with a digital camera, and use the lighting to control the appearance of texture or cross polarize to reduce it as much as possible.

 

[Photo Engineer]

David and I have presented data on this in the past. Paper fibers tend to make a grainy presentation of a picture. Use of a subbing layer or a baryta paper will reduce this problem.

PE

 

[alecmoody]

I will try photographing them. I haven't ever shot copy of something as small as 5x7 but hopefully I wont lose much in the way of tonality.

 

[Ian Leake]

In my experience, scanning Pt/Pd throws up all sorts of problems. Paper grain, dust, things that are imperceptible on the print being magnified, things that are obvious on the print being lost, and other irritants (depending on paper and a host of other factors). Most importantly, though, the tones get violently distorted. Pt/Pd is sensitive to the kind of light it's under and scanner lights are about as bad as you can get. (If you've never done this before then try looking at a print under halogen, tungsten, or fluorescent lights and then taking it straight into the daylight - you should see significant changes in the tones.)

I've found the best way to reproduce Pt/Pd prints for online use is to use a copy stand and a digital camera, ideally under plentiful natural light (be aware of the angle of the light as this may exaggerate paper texture). You can use a blower brush to get rid of any dust on the print. Usually all I need to do in Photoshop is to tweak the curves and then desaturate by 20-60% depending on the type of light used.

But even this process is flawed. Unfortunately the bottom line is, I believe, that a digital copy simply cannot show a Pt/Pd print in it's full magnificence.

One thing to think about though... Digital repro can highlight underlying problems with a print which you haven't noticed, so you may want to consider this before blaming your scanner. For example, if you have used too little solution then you can sometimes get a grainy effect...

 

[alecmoody]

Okay, the files have been replaced with versions i shot as copy today. I think the tonality and especially the color are more accurate to my prints now. The noise is slightly less and I think that when you take a step back and look at it closer to original size (5x7) its not so strong.

Its too bad the paper texture comes across as noise because that textured sheen on the real print is part of what makes them look so appealing.

 

[Ian Leake]

These two look much better, Alec (although obviously I can't compare them with the originals).

Paper texture is always going to be a problem. I print on Buxton which has much more texture than COT320. The paper texture is fundamental to the appeal of the print, but it looks bad on the copy. Perhaps there's something clever that can be done in Photoshop to reduce it, but for the moment I've learned to live with it. After all, it's the print that counts not the copy

 

[Ian Grant]

Have to agree that copying with a Digital camera is possibly the best option, but studio flash is a far better & more controllable light source than daylight. I recently copied virtually all the images I've printed since 1986 that way, some still framed and it works well.

Scanning works exceptionally well with glossy Resin Coated prints, but the quality drops as soon as you try scanning Fibre based images and gets worse as you scan hand-coated papers.

Ian

 

[PVia]

I've found that scanning at about 600 dpi, 16bit RGB works pretty well for my pt/pd prints on Fabriano EW paper. Obviously, areas of flat color (skies, etc) will show the most paper texture.

Here's one of the latest that I'm generally happy with: http://www.flickr.com/photos/viapiano/3553958781/sizes/o/in/photostream/

 

[Ian Leake]

Have to agree that copying with a Digital camera is possibly the best option, but studio flash is a far better & more controllable light source than daylight.

Studio flash is certainly much more consistent and controllable than natural light, but artificial light also warms up the print considerably which then requires more extreme Photoshop tweaks to get back to a 'close-to-print' copy. This, of course, assumes that you consider the print in daylight to be the true warmth as opposed to the print under tungsten (I do but that's just an opinion).

One thing I forgot to add in my earlier post, is that it's vital to shoot in raw mode on your camera. I use a QPcard to enable me to white balance the picture in ACR before doing anything in Photoshop. I've seen some really strange effects when the camera is doing its own white balancing. If you've only got a jpeg then you're stuck with whatever the camera thinks is best...

 

[Ian Grant]

Studio flash is quite different in colour temperature to artificial light, it should give very accurate colour balance if the camera is set correctly.

Ian

 

[pateeid]

David. V700 comes with silverfast programe, isn't it?
u should scan with RGB and turn it to grayscale and working on that. i think it's films problem.

 

[keithwms]

Scanned paper grain is a thorn in my side. I very much like matte papers and it's always tough for me to get good flatbed scans of prints. I have tried re-photographing prints digitally and can't say I like the results much more- the subtle colour matching is pretty hard. I did experiment with scanning wet prints and that can work but... what a mess!

Guess I need to set up a proper copy stand.

 

[archphoto]

There might be an easier way.

Sorry, but I downloaded the first pic and ran it through Noise Ninja and it realy cleared up.
Noise Ninja can be used in PhotoShop, but you have to pay for it, about $ 50,- if I remember correctly.

I can send you the file by email if you want to.

Peter

 

[brianmichel]

The texture of the paper will most certainly factor in this looking grainy, using dICE will only DESTROY the image you're trying to make as it does not handle b&w well at all (makes windows disappear and such).

Lovely work though!