billkantor

Hi Thomas,

Thanks for both the offer and the research on this issue. I just sent you my email address directly so you

can send me the file. Until I get a new target, an old reference file is the best I can do.

Thanks,

Bill

Indeed that works Michael. (Much better than right clicking and selecting "Save Linked File to

'Downloads'" which was what I had been trying.) Thanks!

It still would be nice to get the file from the source so I know it has not been modified.

So does anyone know if X-Rite no longer posts reference files?

Anyone know how to get an IT8 reference file for a Monaco/X-Rite target?

I have an X-Rite IT8 target (it came with an Epson scanner I got in 2005) that I am trying to use for calibration (of my

drum scanner—a ScanMate 5000, using ColorQuartet 5.3.1) and can't find the applicable reference file. According to

the X-Rite website, I need the reference file MONT45.2005.05.txt but I can't find it from their site.

Searching the web, I found an html version of the file here:

ftp://ignotus.ru/distrib/Graphics/Color_profiler/MonacoPROFILER%20v483/Installer/IT8%20Reference%20Files/MONT4

5.2005.05.txt

But I can't download the .txt file, just the html.

I found a posting here http://forum.silverfast.com/post30238.html

leading me to believe that X-Rite only provides reference files in a proprietary format.

Can anyone tell me:

(1) if I can get the appropriate reference file somewhere,

(2) how to get the file from the above site, or

(3) is this X-Rite's way of forcing you to buy another target?

Thanks,

Bill

Hi John,

If the problem is not RFI/EMI then why would it only show in this image? A friend speculated that there

was some torque applied to the lens/camera that may have temporarily deformed the sensor and

masks. Who knows?

Regrettably, no RAW file is available for a re-conversion.

Wow! I very much appreciate your technique to remove the pattern. I had already solved the color

components of the pattern using a much more destructive approach. Here's what I did (I like your

technique much better). I started by converting to LAB space and blurring the a and b (color) channels

(using different radii for the background and the lion). Blurring the color does not degrade the image

much, as most of the detail the eye perceives is carried in the L channel. This left me with visible

luminosity patterns in the background (the lion's fur probably still had some but they were not

noticeable), I then blurred the luminosity channel with a mask (excluding the lion) to get rid of the

remaining luminosity patterns in the background. This technique obviously only works because the area

where the pattern was most prominent was already blurred.

Thanks so much again for the new technique.

Bill

Yes, it appears in the print.

FWIW, looking at the other RFI samples referenced here, I would say that my pattern is slightly less regular, way less linear and more squiggly. I suppose that could be caused by the different types of RF modulation but I seem to be the only one showing this.

Bill

Well the moire is on the feathers. This is across the whole image. So what would cause that? RFI seems

most plausible of the explanations so far.

Parv; Reproducing may be possible but not without challenges. Good idea. If it were my personal image

and camera, I'd be all over it. But it was done by my friend's son on their camera. I view this as a curiosity

now so I can avoid this in the future.

Rob; I'll check out the RF thread. Thanks.

Randall; the pattern is visible over the entire frame. Both the detail and the detail of detail are 100% crops.

Bill

Second post is blank. Third post is what was supposed to be attached to the second one.

Beaches are beautiful at night indeed.

Cheers,

Bill

Can anyone tell me what the cause of this was. I am seeing funny squiggles in one image that a friend of mine shot on

an African safari and just asked me to print. Looks to me like a moire pattern but I can't explain what exactly would

cause it. (I asked if this might have been shot through a screen and was told no.) The pattern is most apparent in the

blurred background but in fact it continues throughout the entire image--it is just less prominent in the complex texture

areas of the lion's fur. The problem only shows on this image. Others look fine. Only possible difference I see is that

this image was underexposed by a stop or two. Image was shot with a Canon 20D and saved in jpg format. While there

is a bit of luminance noise, it shows most prominently in the A channel when converted to the LAB space.

Detail views follow. Photo is reproduced with permission.

Thanks,

Bill<div></div>

Dan,

"# of listings is up year over year."

What is your source on this?

FWIW, I heard about two years ago that LF was the only film segment that was growing. I agree with

Michaels assessment. Nothing beats a LF experience. My 21 Mpixel Canon 5DM2 is nice but

disappointing by comparison. Still I think that the time will come for a MF digital back for me.

Bill

Thanks Sheldon.

Blurring color channels to eliminate color noise may be an advanced technique but it is very effective. I can't be the only

one who wants to do this. It's a natural that it should work with a smart filter. It would be nice if this gets supported in a

future version. I am sure there are other reasons that this would be a useful feature.

Cheers,

Bill

Wouldn't that be a destructive process? I am looking for a non-destructive process that does not double my file size.

And I am not sure why I would need a smart filter for that... why not just run the filter on the channel in the first layer?

Bill

Anyone know if it is possible to apply a smart filter to a single channel? I want to do this so I can blur the color (a and b) channels of an

LAB-space image to eliminate color noise. This is easily done with standard filters on a separate layer. Just select the layer and apply the

filter. When you try this with a smart filter layer it applies the filter to all the layers.

I am running CS3 but will upgrade if this is covered in 4 or 5. Heck, I'll upgrade anyway... just want to know if this can be done.

Best,

Bill

Seems like you are getting the hang of it Tim. Can always pump the saturation a bit but I prefer to convert to LAB space

and enhance color definition with the a and b channels.

Enjoy,

Bill

Tim,

The HDR effect is what happens when the technique is misapplied. Try it on a contrasty image rather than a low contrast

one. I've never found the technique useful on a low contrast image. Also, if the opacity is high and blur radius is high

then you get the cartoonish HDR look. Try lowering both. Finally You can get funny halos around dark to light area

transitions. To avoid the bright Halos which contribute tot eh cartoonish HDR look when you use a large radius, try (after

blurring the contrast mask) Edit/Fade then set blend mode to darken and click ok. The halos will go away. You may want

to then do another blur at a very small radius to ease the transition.

As for how it is better than using ACR/Lightroom Clarity/Fill sliders, I can't comment because I use neither. On the other

hand, I find this technique is similar to the Shadow/Highlight tool but I can control things after the fact by changing the

opacity. Also if you make the Contrast Mask layer a smart filter you can change the blur radius.

Your call on what you prefer to use.

Patrick,

Thanks for the tips.

Bill

Wow! Who would of thought...

It works. There must be something about the gaussian blur that changes the color of the mask.

Thanks for the help. Probably saved me a few bucks in disk storage and hours of time waiting to save/open some super

large files.

Cheers.

For those of you unfamiliar with the contrast mask concept, see here for a description of the technique.

p://www.outbackphoto.com/workshop/photoshop_corner/essay_06/essay.html

The technique employs a mask (created from the original image--typically a desaturated, inverted, and blurred version of the original image

or one of the original image channels) applied with the Overlay or Soft Light blend modes in Photoshop to control local contrast. It's very

effective, with one drawback--the Overlay and Soft Light blend modes can change color. One workaround I found to this is to

duplicate the background layer on top of the contrast mask and set the blend mode of the duplicate background layer to Color which undoes the color shifts caused by the underlying Overlay or Soft Light blend mode--at the expense of doubling the file size.

My question: Does anyone know how to apply a contrast mask and have it only effect the luminosity without doubling the file size? Ideally

this should be an adjustment layer and not introduce destructive edits.

Thanks,

Bill

Thanks for the insight Kelly and for pointing out some of the unanticipated possible variations. I think this is the point. If

you have a paper with OBAs in them then you have another variable that adds to the possibility of getting inconsistent

results. The "old paper" approach seems akin to avoiding papers with OBAs.

Bill

Thanks Colin.

Yes I have seen this too. But it is difficult to go to each manufacturer and look up each paper. Sure would be easier if there were a central source--ideally a supplier whose website had the ability to filter papers based on whether they include OBAs.

Anyone know of a central source where one can determine if a particular inkjet paper has optical brightener additives (OBAs) in them? It

would be nice to know before purchasing and without having to test.

Thanks,

Bill

Hi John,

Thanks for the suggestions. I will try this and see how it works.

Kelly,

I like the resourcefulness of your suggestion. Similar to synthesising a GND by dodging in front of the lens during the exposure. It works by the way.

A "Gaussian hen peck." Could have multiple interpretations. I'll experiment but I'm just thinking that anything I create that way would rub off easily.

Happy New Year,

Bill