The Info Palette

Diagnosing Facebook colour problems

2013-05-25T18:02:00.003+10:00

I often notice people complaining that Facebook is changing the colour of their images. I can assure you this is not true - in fact, Facebook handles colour better than quite a few other sites.

If your Facebook photos don't look the same as when you edited them, something must be wrong somewhere.

So I've designed a page to help you find the problem.

Troubleshooter: When your Facebook photos don't match Photoshop

It's as comprehensive as I can possibly make it. If it doesn't resolve your issue, you must have a VERY obscure issue :)

Using a Photo Filter layer to fix casts

2013-02-15T09:15:00.000+10:00

There are lots of ways to fix colour casts using Photoshop. Filters are only one. If you have another method which works for you (eg Curves, Color Balance, Selective Color), this tutorial doesn't intend to sway your opinion. Stick with what you know.

But this works for me, so it's time I shared it.

At the outset, I must remind you NOT to attempt to fix casts by adjusting white balance. Raw edits such as white balance are for the whole photo ... the "greater good", if you like. Try to turn a blind eye to areas of cast while you're choosing the correct white balance for the overall photo. (I know, I know, it can be hard.) The truth is, sometimes choosing the correct white balance will make casts look even worse, and this will scare you. Stay strong, friend. Photoshop is the place to worry about casts, not your raw program.

This tutorial is demonstrated in Photoshop CS2, but the method works just the same in all recent versions of Photoshop and Elements. It assumes a basic knowledge of adjustment layers and masks, which are the beating heart of Photoshop - if you can not easily follow these instructions, I urge you to consider my little Layers and Masks Class.

The other thing you need is familiarity with "colour pairs". Every colour has its opposite colour on the colour wheel:

Red opposes cyan, green opposes magenta, and blue opposes yellow. (More colloquially, but no less important, "warm" opposes "cool", as we'll see soon.) So, a cast caused by green grass or trees would be fixed with magenta. A cast coming from a red umbrella would be fixed with cyan. And so on.

Many thanks to Barbara from Florida for permission to use this photo. As you can see, it's a lovely warm photo, embracing that afternoon light, but the lady's makeup makes her look too warm.

So I begin by adding a Photo Filter adjustment layer:

The Photo Filter interface is very simple - it hasn't really varied for many versions:

The first thing I do is move the Density slider up to 50%. I do this almost every time, no matter whether I'm dealing with a strong cast or a minor one:

(I almost always leave the "Preserve Luminosity" box checked. I only uncheck it in situations where I want to fix the colour cast and darken the area at the same time. This is very rare.)

Now it's time for me to choose the best colour. It seems to me that her face is too warm ... or is it too orange? Or even yellow? If it's possible to instinctively know which colour to choose in every situation, I haven't reached that skill level yet; so don't feel bad if you don't always know either. Just do what I do - try a few different options.

There are three different cooling filters, so I try all of those. I also try the Cyan filter, and the Blue filter. Eventually I decide that "Cooling Filter (82)" is the best option:

Of course, while I'm playing with the filters, I'm concentrating solely on her face, and ignoring the rest of the photo. Why? Because the rest of the photo is going CRAZY!! The sky is insane, the little girl looks like a smurf, and so on. None of that matters right now - all that matters is choosing the best colour for the problem area.

Then I invert the layer mask:

Then choose my Brush Tool and set it to 10% opacity:

(Sometimes it's necessary to go down to 5%, or even 3%. But mostly, 10% works well.)

Then I gradually paint on her face to add the cooling where needed, until she looks good:

Here's my finished result (before and after):

Needless to say, if you have trouble following any of these steps, or if you need help choosing the best colour filter for your particular photo, visit me at Ask Damien.

This tutorial is dedicated to my father, who lost his fight with cancer yesterday. He never really understood what I do, but he was proud of me for doing it well. I love you, Dad.

Resizing photos in templates (PS & PSE)

2013-02-03T08:03:00.000+10:00

If you've ever had trouble resizing a photo in a template without stretching it out of shape and making it look all wonky; don't worry, you're not alone. It's easy to get confused, because for some reason Photoshop and Elements have opposite default settings. So if you have Elements and have been trying to follow Photoshop instructions, or vice versa, it might have gone badly for you.

So I've created two videos - one for Elements, and one for Photoshop. I hope they help clarify things.

These videos specifically discuss the resizing step of the template workflow. They don't explain setting up the clipping masks, or placing the images, or anything like that. If you need assistance with those broader issues, I've written tutorials here and here (plus some more here, here and here).

Here's the Elements video:

Here's the Photoshop video:

Using Photoshop's Gradient layer

2012-10-30T19:46:00.000+10:00

This one is for Julie, from my Facebook group. Julie wanted to remove the distracting bright orange toy from this photo, and was having trouble cloning it out.

Now, let's be clear - cloning definitely is the "correct" way to do this. It could be done with some patience, using this method. But when I asked Julie if she'd be happy with the "lazy solution" (my area of expertise!!!) she said yes.

So, here we go, Julie. I did this in Photoshop CS2, but it'll work in all versions of Photoshop and Elements as far back as I can remember. This method uses the Gradient layer, which I've discussed a little in other tutes, particularly regarding skies.

First, I used the Lasso Tool to roughly select the background:

Then I added a Gradient layer. (Don't confuse it with a Gradient Map layer.) By default, it gave me a black-to-white gradient. Immediately, I changed the angle to 90 degrees, so it ran vertically:

But black-to-white was no good, of course. My goal was a three-colour gradient to roughly mimic the three main shades of brown that currently existed in the background. So I clicked on the little arrow at the top of the dialog, and chose a three-colour gradient preset. A garish blue/red/yellow one:

Next, I clicked on that gradient ...

... to get into the "engine room" of the layer. You can see the gradient illustrated there, with the three "stops" of colour indicated - blue at the start, red in the middle, and yellow at the end:

(One annoying thing about this method is that you get two or three little windows piled on top of each other. Not a problem if you have a big screen, but annoying on my laptop!)

Time to swap those loud colours for the three shades of brown I needed. I decided to start with the brown at the top, so I clicked on the little yellow stop (the little square), followed by the Color swatch (the rectangle):

When the Color Picker window opened, all I had to do was click on the brown above the hat, to choose it. Then pressed OK:

In a flash, I'd done the same with the other two stops ... click the square, click the rectangle, click on the desired colour in the photo, OK. Bam! Done:

Well, not quite done. The colours were right, but their position wasn't. For example, the light brown colour in the middle of the gradient was ... well, in the middle of the gradient. But in the photo, that light brown was a bit below middle. No problem, all I had to do was slide those stops around a bit until the colours were where I wanted them:

What was the point of that? Well, the closer the new colours are to the old colours, the easier the next step is ... the masking step.

So, once I was happy with the gradient, I pressed OK a couple of times to get out of all those little windows, back to my photo. Then all that remained was to zoom in, choose a small white brush, and mask the gradient properly around the subject:

Voila! Job done. A bit cheesy? Yes? Dead easy? Also yes. Gotta love that.

One more thing. Any time you add a gradient, you should also add a bit of noise. Read this tutorial about making a noise layer.

Have fun, Julie!

Resetting tools

2012-09-29T20:21:00.001+10:00

Sometimes, just occasionally, one of your tools will start misbehaving in Photoshop. It might be the Text Tool which is putting enormous spacing between letters and you don't know why; or it might be the Crop Tool which keeps cropping to a fixed shape even though you've got no values in the Options Bar; or it might be the Clone Stamp Tool which won't sample the right colours ... or whatever. You know, any one of those annoying little glitches that can happen with any software, any time.

If you're lucky, the problem can be fixed by simply resetting the tool. You do this via the little "Tool Preset Picker" at the left end of the Options Bar, then going to the submenu, and choosing "Reset Tool":

This won't solve every Photoshop problem you ever have, but it's an important little troubleshooting device to have up your sleeve.

What is "clean processing"?

2012-09-04T10:58:00.000+10:00

"Clean processing" is a term you'll see frequently on photographic forums, and especially in my Facebook group. To a degree, everyone has their own interpretation of the concept, but maybe it's new to you.

In a nutshell, a clean-processed image is one to which it's not obvious that any editing has been performed (even if it has). The kind of image that makes somebody say "Wow, you're a good photographer!", rather than "Gee, you're good at Photoshop!".

A clean-processed photo has no particular effect or style applied, and as such, it should sit as comfortably in your portfolio in five or ten years' time as it does today. However, it does provide the basis of all other editing styles. Consistent clean processing means consistent and predictable results from your artistic editing, especially if you use actions or presets.

Of course, even within the narrow description of "clean", there is room for personal interpretation. You may prefer your images warmer or cooler, or brighter or darker, or richer or paler, than the next person. But in general terms, a "clean" image is one where the whites are white, the blacks are black, and the grays are gray.

Is it quick?

Maybe, maybe not. It depends on how good the original photo was. Clean processing might take a few seconds, or a whole hour. The point is, nobody can tell.

In a way, clean processing is like housework. When your mother-in-law comes to visit, and finds your house tidy and spotless, she doesn't know if you've done a five minute spruce-up, or hours and hours of hard labour, to get it looking that way. (In my case, invariably the latter!) She just sees a clean house.

If you take a look at my portfolio, you'll see examples of clean processing that took an uncommonly long time, due to difficult lighting, or equipment malfunction, or whatever problems can arise from time to time. But for most photographers, most of the time, clean processing should be quite fast.

Is it easy?

Again, it can be. It should be. And it usually is. Those times when your lighting and your camera settings were perfect, the processing is a piece of cake, once you know how.

But when the photo wasn't perfect, clean processing gets hard. Not just hard - the hardest. There is no greater challenge in post-processing than making a ballsed-up photo look (a) not ballsed-up, and (b) not edited at all.

These are the skills that I have devoted my career to perfecting. So far so good, but still a way to go.

Postscript

There's one more thing I need to mention ...

Clean processing is the only type of editing that can be constructively critiqued. It's the only time you can legitimately say "Your such-and-such is incorrect" or "It needs less such-and-such", etc. In short, it's the only type of processing to which rules apply.

The moment you apply any kind of artistic effect to your photo, the only "critique" that can be offered is an expression of personal opinion - "I like it" or "I don't like it". People can still legitimately offer feedback about the photographic aspects of the image (composition etc), but not the editing.

This is why I have no patience for people posting artistic editing for CC in the AD group. There's no point to it. Your hazy crap (or whatever the fad of the week is) will get no gushing praise from me. I operate on a "guilty until proven innocent" basis - unless you show me a clean-processed version of your photo, I will assume that you've added the effect to disguise your shortcomings as a photographer, or editor, or both.

I want you to prove me wrong.

ColorMunki Create - thumbs down

2012-08-24T13:48:00.001+10:00

A recent addition to the monitor calibration market is the Pantone ColorMunki Create.

In reality it's an X-Rite device, like all the other ColorMunkis, but it's badged by Pantone and sold as part of a "professional colour creation solution" package, which apparently enables you to choose sets of colours, build colour palettes ... blah, blah, blah, I don't know. I didn't look into that stuff.

I bought it because I wanted to see if it's a viable option for people seeking a low-cost monitor calibrator. And I'm sorry to say it isn't. It gives even fewer calibration controls than the Spyder Express, and that's really saying something, because the Express gives you very little control.

So, my advice remains the same as it has for a while now. The bare minimum device for anyone serious about their monitor calibration is the X-Rite ColorMunki Display; and anyone who is more than average serious about it should go for the X-Rite i1Display Pro.

However, having said all that, I will (when I get a chance) be writing a tutorial for the use of the ColorMunki Create, for anyone who has one. It'll be included with the other calibration instructions on my website.

As always, if you have any questions about monitor calibration, or anything else, please visit me at Ask Damien.

Note: This device is also marketed as the ColorMunki Smile. Again, it'll give you no reason to smile at all.

Case study: Bright red clothing

2012-06-11T21:35:00.001+10:00

Thanks to Chasity from Ohio for allowing me to use this gorgeous photo in this post.

This article relates to my previous Strategies for managing out-of-gamut clothing post, which I suggest you read first if you haven't done so already.

Red clothing is a hassle, isn't it. Well, any vivid-coloured clothing can be, but red seems to plague us most often. Check out this photo. That red outfit is as cute as all get-out, and so wonderfully bright! It makes you swoon (although it also burns a cyan imprint on your retina if you look at it for too long!!!).

But there are two problems. Inter-related problems, but different.

The first is that there's channel clipping. Not just at the bright end of the red channel, but also at the dark end of the green channel. See the two severe spikes at the ends of the histogram?

Here is a view of all three channels. Both the red and green channels are clipped:

"So what?" you might say. "So it's clipped - big deal. I still love it."

Well, I love it too. I don't mind a bit of clipping in my images, if the circumstances call for it.

But here's the second problem - print gamut. Your lab (almost certainly) can't print reds as bright as your screen can show them, so the colour dies another death when it hits paper.

Here, I've soft-proofed the photo using WHCC's lustre profile. This is how it would print. You can see that the red is duller, and somewhat darker. Most significantly, there's no shape to it. It's pretty much just a blob of flat red:

Read more about soft-proofing here, if you need to. In that article, I wrote about ... "the perpetual trade-off when dealing with out-of-gamut areas ... you can have the brightest possible colour, with no detail, or duller colour, with correct detail. Your choice!".

So, we come to a fork in the road. To our left is the "Meh, I don't care if it's flat and shapeless; I've got to get this job done in a hurry, and I'm not charging much for it" route. On occasions when you want to take that easy path, don't bother reading any further. Just send it off to the lab, and get on with your life.

To our right is the "Let's do our very best to get a good print" road. This route involves massaging the out-of-gamut colours into the print gamut, so that they'll be identical between screen and print. And yes, it'll mean duller colours. But they'll be duller colours with visible detail, which is important to a nerd like me.

So, here's what I do. Take a look at the one on the right:

I've desaturated the red clothing until the colours are all within gamut. That is, when I toggle my soft-proof profile on and off, there's no change (well, not enough to worry about, anyway).

I hope you can see on your screen what I can see on mine. I see a duller red, but I see the folds and creases and undulations of the fabric, that aren't visible on the printed (middle) one.

See what I mean about making a choice? Because I'm a raving nerd, I prefer the desaturated one. I value shape and detail over in-your-face brightness. However, your opinion may well differ from mine, and that's absolutely fine. This is one of those little tiny aspects that form part of your style. If your style is rich colour, then go with it.

By this point in the article, I expect only my fellow nerds are still reading. Solidarity, comrades!

So are we, the nerdiest of the nerds, happy with the outcome of this image? Well, actually, no. When we look closely at the detail, it's still ... "mushy", isn't it?

Nerds don't just want a bit of detail. We want all the detail, dammit! And the above close-up doesn't satisfy us. So, what's the problem? Well, it comes all the way back to the clipping. Channel clipping, by its very nature, means that detail is lost.

Let's look at those channels again:

The red channel has got maybe half of its detail, and the green channel has none at all. Only the blue channel is showing proper detail. So really, when you add it up, the fabric really only has about 50% of the visible detail that it should have. It doesn't surprise us, then, that when we desaturate into print gamut, the results are only about half as good as we'd like.

What does it all mean, Oh Nerdy One? Well, it means this: if we want the best possible detail in our prints, we have to avoid channel clipping first and foremost. By "first and foremost", I mean when processing our raw file.

Take a look at this one. The one on the left is the original clipped file, for reference. The one on the right has been desaturated to within print gamut from a non-clipped version of the raw file:

Look at that beautiful result! Look at those lovely folds and creases and undulations and detail in the red. And all printable!

Wanna see a close-up of that? Here's a comparison between the previous attempt and the no-clipping version:

Look at that gorgeous detail. Perfectly visible, perfectly reproducible. Nerdvana!

How did I achieve it? Well, take another look at the Strategies article. Options 2, 3 and 4 are all perfectly legitimate ways to approach this situation.

As it happens, I used Option 4 for this demonstration (it was the easiest way to capture all the screenshots you've seen). In my normal editing, I use Option 3 almost every time, because I'm old-school. Folk who are more modern and hip than me might prefer Option 2.

Why have I written this article?

Well, I've seen this issue discussed a few times over recent days, and have been alarmed by two recurrent misunderstandings:

"I just add a Hue/Saturation layer and desaturate it a bit - that fixes the clipping". No, as we've seen, it doesn't. Well, it kinda does, but not very well at all.

"I just add a Levels layer and pull in the output sliders a bit - that fixes the clipping". Nope, sorry. It's an even less successful "solution" than the desaturating. It does absolutely nothing to recover detail - it just makes things flatter.

Clipping can't be fixed after it's happened. You have to go back to wherever it occurred (usually back to the raw file) and fix it at the source. Like in a time-travel movie where the hero goes back in time to stop something bad from happening before it actually happens ... if you know what I mean.

Elements doesn't have soft-proofing. What can Elements peeps do?

I wish I had a great answer to that, but I don't. All I can tell you is that you'll get a "feel" for how bright is too bright to print, after a while. How do you develop that instinct? Only by getting prints done, I'm afraid. It'll cost you a few bucks in test prints and reprints, until you get in the zone.

Free texture files - old leather

2012-05-19T20:16:00.000+10:00

When you keep your pommel horse in your back yard ...

... its leather inevitably starts to deteriorate. I was looking at it today (thinking about re-covering with all-weather canvas) and I realised that the old leather is as charismatic as anything!

Well, I couldn't pass up such a great opportunity to add to my texture library, so I borrowed my wife's camera, with the macro lens, and grabbed a bunch of photos. Needless to say, my complete lack of photographic skill resulted in fewer than one in five of the photos being usable, but still, I ended up with eight keepers.

I'm not sure if or when I'll ever use them; but hey, it can't hurt to have them on file, right? And since I know some of my readers enjoy adding textures to their photos sometimes, I thought I'd share them for you.

Download link (please let me know if it doesn't work, and I'll try hosting it somewhere else).

The files were captured with a 21 megapixel 5D Mark II (although some of them have been cropped a bit). Saved as high quality Jpegs, and zipped up. If you'd like them, you're welcome to them.

Here's what they look like:

Create your own wall guides

2012-05-07T09:33:00.000+10:00

Thanks to Crystal for asking this great question.

Recently I wrote a tutorial about preparing an image for canvas printing. But before you prepare it, you've got to choose it, and its size. The latter is really important - it's easy to think that a print size (eg 20x30") is really big, but in fact it's very small once it's on your wall.

So, the best way to choose a photo and an appropriate size is to see it on your wall first. And it's very easy to do. First, take a photo of the wall on which you'd like to hang the print. Here's an example wall:

To make it simple for yourself, don't photograph the wall at an angle. Stand perpendicular to it to take the shot. This tutorial can be adapted (sort of) for an angled photo, but it won't be as easy, or as accurate.

Next, get out your measuring tape and measure something on the wall. In this example, I know that the brown cabinet is 79 inches wide. In other cases, you can measure the wall itself (if its corners are visible in the shot), or a windowsill, etc. You need a reference width to go by.

(Make sure you measure something on or very close to the wall. There's no point measuring a table in the middle of the room, for example.)

Choose your Crop Tool, and enter the width of the object you measured (79 inches in this example). Make sure there are no values in the Height or Resolution fields:

Crop the image from one end of the object to the other, as precisely as you can. In the case of furniture, like this, make sure you use the corners closest to the wall, not closest to the camera:

Here's the cropped portion I end up with:

Now, open your Image Size dialog:

You don't need to do anything at all in this dialog. Just write down (or select and copy) the Resolution value. In this example, the resolution is 26.203:

Then press Cancel to exit out of the Image Size dialog:

Then return to the original full-sized image. You can do that by going back in the History Panel, or just by Reverting as I've done here (or even by closing without saving, then opening again):

Once you've got your original photo back, go to the Image Size dialog again:

Make sure "Resample Image" is turned off, then enter the Resolution you noted earlier. When you do this, the Width and Height fields will change of their own accord, which is exactly what you want:

Great! Now your image is "to scale", so to speak. Time to add the clipping mask for the canvas.

Choose your rectangular Marquee Tool, and choose "Fixed Size" as the style. Then enter some dimensions, eg for a 20x30 print:

Then simply click (just click, not click and drag) somewhere in the image to make the marquee. Then drag it around to put it where you want it:

Add a Solid Color layer, and there's your clipping mask, ready to go!

In this example, you can see I've got a few different print sizes on different layers, so I could try out a few different photos and sizes, to help me choose a wonderful print for the wall:

(If you don't know how to use clipping masks, please visit this tutorial.)

As always, please visit me at Ask Damien if you have any questions or problems with this tutorial.

Cloning in the workflow

2012-04-14T20:56:00.000+10:00

All editing can loosely be divided into two categories. In one category are colour and tonal adjustments, using Curves or Levels or Hue/Saturation or Photo Filters or whatever. These edits change the appearance of the pixels in the image, and are done with adjustment layers and masks.

In the other category are pixel edits. These edits actually change the pixels themselves (by moving or copying or blurring or hiding them), and can't be done using adjustment layers - they must be performed directly to the pixels.

In this "pixel editing" category, we find:

Noise reduction
Liquifying
Skin smoothing
Cloning
Healing

Noise reduction should always be done in Raw, so is almost a moot point. Liquifying is a special case, and I'll write a dedicated article about it soon. Skin smoothing, cloning, and healing are the most common pixel edits. This is not a tutorial about those specific methods ... I just want to discuss where they fit in the workflow.

Often I've seen discussions on forums about when to do pixel editing. Questions like: "Do you do skin smoothing at the beginning or end of the workflow?" or "Should I clone something out as soon as I open the image, or wait until I've done other editing?". I'm here to tell you it doesn't matter. You can do it first, last, or in-between. As long as you have a good layer-based workflow, the timing is of no consequence. You can do it whenever you feel like it.

What's important is where you do it.

Whenever you're ready to do some pixel editing, you must make a new layer immediately above the Background layer, and do it on that. Usually, the new layer is a duplicate of the background layer, which you create by selecting the Background layer, then pressing Ctrl/Cmd J:

But sometimes, it might be a plain blank layer:

Blank layers are suitable for cloning, and some of the healing tools. They don't work with the Patch Tool, though, nor for skin smoothing or noise reduction.

Duplicate layers are suitable for everything, but create a bigger file size. I don't usually care about the file size (since I have a very powerful computer), and it suits me to use a duplicate layer in most cases, so that my tiny brain doesn't have to remember which tools will work, and which ones won't, on a blank layer.

Whether you use a blank layer or a duplicated one, the really important thing is that it's at the bottom of the layer stack, right above the original image, and below all of the adjustment layers. This is the critical part. All adjustment layers must be above all pixel layers in a good layer-based workflow.

So can you see why I say it doesn't matter when you decide to do some pixel editing? At any time during your workflow, you can simply return to the Background layer, make the new layer, and do whatever you need to do. If you happen to mess up the editing somehow (which we all do sometimes!) then delete it, make another layer, and try again.

Of course, after the pixel editing, always remember to return to the topmost layer in the stack before creating any new adjustment layers. You can get some unexpected and wacky results if you add new adjustment layers beneath old ones.

Some related articles:
The non-destructive workflow
The role of masks when cloning
Fixing skin blemishes

Opacity shortcuts

2012-03-22T22:16:00.001+10:00

There are about a bazillion keyboard shortcuts in Photoshop and Elements, and you can't learn them all at once.

At first, focus your attention on the most important ones in the toolbar - C for the Crop Tool, B for the Brush Tool, V for the Move Tool, etc. Those are the ones that you'll use frequently, and therefore they'll save you the most time.

Once you've mastered those, you should aim to gradually add more shortcuts to your subconscious arsenal. Every time you can press a key instead of moving your mouse to a tool/menu/slider, you've saved yourself a second or two. And even though a second or two doesn't seem like much, a lot of little time savings can seriously add up to a faster workflow.

I want to tell you about the keyboard shortcuts that have saved me the most time - brush opacity.

Changing the opacity of your brush is a tedious thing to do with the mouse. It's not just one click, it's two or three. You have to reach up to the toolbar, click for the slider to appear, then slide it; or click to highlight the numeric value, then type a new one in. Seconds wasted!

Embrace the shortcuts, my friends. They're staggeringly simple - just press 1 for 10%, 2 for 20%, 3 for 30% etc. That's all! Just a single keystroke will change the brush opacity, and you barely have to pause painting to do it.

Most importantly, press 0 to put it back to 100%.

If you want a more precise value, type two digits. Type 25 for 25%, or 03 for 3%, etc.

Try it and see! It's great. It also works for the Clone Tool, and other brush-related tools.

Word of warning, though. If you happen to have a non-brush-related tool selected (such as the Move Tool, or the Crop Tool, or whatever), and you press a number, it'll change the opacity of the layer instead. Watch out for that little trap - it's caught me many times!

Dealing with "hot" areas of skin

2012-03-20T19:55:00.000+10:00

This is a very simple tutorial that I've been meaning to write for ages.

Many thanks to Nicole of Essence of the Soul Photography for allowing me to use this beautiful photo:

So gorgeous! But it has a little problem that seems to pop up in quite a lot of photos ... it has some areas of very "hot" skin - that is, the skin is glowing quite bright red in some parts:

This problem almost always occurs in shadowy areas of skin. Sometimes it's caused by a red cast - perhaps the subject is wearing a red shirt, or standing beside a red car, or whatever.

But more often it happens in post-processing. It can happen for a number of reasons, I guess, but I think the main culprit is Curves. If I took a poll, I bet I'd find that this problem happens to Curves users more than Levels users. (Read about the difference)

Anyway, no matter how it happens, it's very easy to fix, in either Photoshop or Elements.

Start by adding a Hue/Saturation adjustment layer:

Choose "Reds", and drop the Saturation slider to -50:

This will make your subject look awfully ill for a moment, but don't worry about that ...

... because you immediately invert the mask to hide the adjustment:

Choose the Brush Tool, and lower its opacity to no higher than 10%. Also make sure that white is your foreground swatch colour:

Then paint gently over the hot areas, until they are desaturated just enough:

The difference is very subtle, but it's important all the same.

As a guide, here's a black and white view of the mask I used for this edit:

Variation

The -50% saturation setting will work quite well for most photos. But sometimes you'll find that the desaturated area is either too light or too dark compared to the rest of the skin. If this happens, just go back to the adjustment layer, and move the Lightness slider up or down as needed.

Since the Lightness slider has a side-effect of reducing saturation by itself, watch out that the area doesn't become too undersaturated. Generally, I compensate one with the other - if I move the Lightness slider by 10% either way, for example, then I'll increase the Saturation slider by about 10% (to -40) to make up for it.

Your mileage may vary, of course, so play around and see what settings work best for you.

Jpegs in ACR (Elements)

2012-03-02T12:54:00.000+10:00

I don't know why it's taken me so long to write this simple post!

Photoshop Elements has been able to open Jpegs in Camera Raw for quite a few versions now (since V6, maybe?). It's very easy to do, as I'll show you.

First, let me reiterate - I can take my cat to a zoo, but that doesn't make it a tiger! Likewise, you can open Jpegs in a Raw program, but that doesn't make them Raw files! Editing a Jpeg in ACR is fun, but it's still a Jpeg. If you want real control and flexibility when editing, shoot Raw. [/rant]

First, go to File>Open As:

At the bottom, change the format to "Camera Raw":

Choose your image, and press "Open":

Voila! There it is in ACR:

Do your editing, then press "Open Image" to bring it into Elements for further work.

More about storyboards: intersecting shapes

2012-02-20T06:44:00.000+10:00

They say that the squeaky wheel gets the grease, so David, here's your new tutorial :)

David would like to know about creating storyboard designs using intersecting shapes. Here's the layout I'll demonstrate for this article:

It's very easy. It just requires a bit of planning (as usual) and careful utilisation of these important little buttons:

(Note: If you're new to storyboards design, this article is not the right place to start. First, you must read this tutorial, then this one.)

As always, grids are vital. After planning your design, and creating your blank file, turn on the grids and set them to an appropriate size.

Choose the Shape Tool (1), set it to "Shape layers" (2), and "Create new shape layer" (3):

(Elements users, you don't have #2. Don't worry, Elements is stuck fast on the correct setting, so you're good to go!)

Draw the first shape:

Then choose "Add to shape area" in the options bar:

And draw the rest of the main shapes:

Then choose the circular Shape Tool, and make sure it's set to the third option - "Subtract from shape area":

Use that tool to cut out the "hole" in the design:

(Tip: For something like this, I always start by clicking right in the centre of the desired area, then hold down the Alt key and drag outwards. The Alt key makes the shape draw from its middle point. Once you try it you'll see what I mean.)

Then choose "Add to shape area" again ...

... and draw the final shape:

Then turn off the grid to see your finished template:

Then go ahead and add your photos!

If you can imagine it, and plan it, you can do it. Using grids, and those five important little buttons, there's no design you can't create.

The best way to learn is to play. Invest half an hour of your time to familiarise yourself with the role and behaviour of each of those five buttons, and you'll be rocking your storyboards in no time.

The Camera Calibration tab in ACR

2012-02-05T05:58:00.000+10:00

If you own an X-Rite ColorChecker Passport, or a similar tool, you can create your own custom colour profile for your camera's Raw files. I'm not a photographer, so I've never done this (nor would I even if I was, probably), so I can't discuss that.

But what I can discuss is the range of canned profiles that come with Adobe Camera Raw. You find them in the "Camera Calibration" tab.

In Elements, it looks like this:

In Photoshop, it's more involved:

By default, the profile is set to "Adobe Standard". Let me make something very clear - Adobe Standard is an excellent profile. Most people never change from it, and never need to. If you want to keep your life simple, just stick to that default, and go and have a beer or something.

But if you're of a dabbling nature, you might like to take a look at the other options available, and there are quite a lot:

As you scroll through them, you'll find that some make your photo look like complete tripe, others make it a bit darker, or a bit richer, or a bit flatter, or whatever. All of these different profiles are different interpretations and representations of the ones and zeros of your Raw data.

If you are interested in choosing one of the other profiles as your default setting, you're welcome to do so, but please heed these three points of advice:

Only view unedited files while making judgements about the profiles. Don't use ones to which you've already made adjustments. SOOC data is what is relevant here.
View LOTS of your photos. Do not make a hasty decision based on one file, or even a handful of files. You need to see the profiles acting upon a vast range of your images - indoors, outdoors, people, scenery, dark, light, etc, etc. Take plenty of time for this process, because it's a very important one.
Be perfectly happy to return to the Adobe profile at the end of it all, if nothing better appears. Like I said, it's a good profile.

If, after exhaustive testing, you find a profile that you like, just choose it and set it as your new default:

PS: In Photoshop's ACR, as well as the menu of profiles, you also have a bunch sliders which can further tweak the colours of your camera's data. You're welcome to play with those as well, but you're totally on your own, because I've never used them, and they scare me a little. The same three points of advice apply as above - unedited files only, lots of them, and be prepared to walk away if you make a mess :)

What is XMP data all about?

2012-02-04T17:33:00.000+10:00

XMP data is the data that Adobe software (ACR and LR) creates to store the editing that you do to a Raw file. The truth is, we really don't need to think about XMP data at all - it just does its thing in the background, and we get on with our lives. But it confuses some people, especially when they see those little XMP files in their folders, so let me explain a bit more.

It's very important to understand that Raw files themselves are absolutely, impenetrably, rock-solidly, unchangeable. You can't modify a raw file, no matter what you try. The ones and zeros of Raw data that come out of your camera will never be altered.

So, when you're editing a Raw file, you're not actually changing the file itself. Rather, you're making cosmetic changes to its appearance - much the same as dressing yourself. You might be dressing in work clothes, or casual clothes, or gardening clothes; but underneath, your body is the same. Your external appearance is being modified by your garments, that's all.

So it is with Raw files. You might "dress" a Raw file with a warmer white balance, or reduced exposure, or whatever. But underneath, it's still exactly the same Raw data, and therefore it's as easy as pie to "undress" it again later if you wish to start your editing all over (more about this later).

The visual modifications you make to the Raw file have to be saved somewhere, of course, and since the Raw data itself can never be altered in any way, the editing information has to be saved beside it. That's the XMP data.

When Adobe Camera Raw users edit their Raw files, then press "Done" or "Open" or "Save", the XMP data is saved as a separate .xmp file alongside the Raw file in its folder. When I look at a folder of Raw files on my computer, I can see at a glance which ones I've already edited, because they have the XMP file beside them:

In Lightroom, the XMP data is constantly updated as you edit, and saved inside LR's catalogue, unless you change the preference to tell it to save externally. That change seems like a good idea to me, so I encourage LR users to read this article. Once you change that preference, LR behaves the same as ACR, and puts XMP files beside the Raw files in their folder on your hard drive.

The exception to all of this is the DNG format. DNG files have an extra "pocket" to contain the XMP data inside them, so no additional file is necessary. (Some people love this about DNG, others loathe it. I am in the latter category, and I wrote in my last post about why I eschew DNG files.) So, if you're a DNG user, you won't see XMP files in your folders, the way I can in mine.

Please understand that you don't need to do anything to save XMP data. There's no File>Save As, or File>Export, or anything like that. The XMP data just gets saved automatically while you're doing your work. It's a wonderfully automatic process that we don't even need to think about 99% of the time.

I must also mention that you can't open an XMP file, and nor do you need to. If you double-click on one, you'll just get a "Can't open" error message, or maybe it'll open in a text editing program as a whole bunch of gibberish. I can't read gibberish, and I bet you can't either ... but your Adobe software can, and that's all that matters.

Needless to say, XMP files are very important, so don't delete them carelessly, and don't move them into a separate folder from their Raw files. If you do either of those things, then open your Raw file, all of your previous editing will have disappeared, and your image will be "naked" again.

That brings me to my last topic of discussion - things you can do with an XMP file. (Again, this doesn't apply to DNG users - they can play with their XMP data, but it's more complex, and I won't discuss it here.)

The most obvious one is when you don't like your edit, and want to start again. If you've got the file open in your program at the time, then you can just reset your edits. But if it's not open, and you're browsing the folder of Raw files, just delete the XMP file. Bam! Next time you open that Raw file, it'll be back to its original state.

More commonly, though, you will probably decide not to delete it, but just to rename it. This has the same effect of returning the Raw file to its unedited state, but with the added bonus that if you don't like your second attempt at editing better than your first, you can simple re-rename it back to the correct file name, and there's your edit again. If renaming isn't your thing, you can just move the XMP file into another folder (eg your desktop), and move it back again if necessary.

(This isn't necessary with Lightroom, I think, because LR has a good system of "virtual copies".)

So, that's about all there is to tell you about XMP. I hope it has cleared up a few things for you. Of course, if you have more questions, visit us at Ask Damien.

To DNG, or not to DNG

2012-02-03T06:27:00.002+10:00

People sometimes ask me "Should I convert my Raw files to the DNG format?" I honestly don't know.

Well, let's clarify - some people have no choice. If their Adobe software is older than their camera, then the only way they can open their Raw files is to convert them. No ambiguity there. (More info)

But lots of other people have Photoshop or Lightroom as new or newer than their camera, so they can open their NEFs, or CR2s, or ORFs, etc, just fine. For those people, conversion to DNG becomes a matter of choice.

Plenty of people much smarter than me recommend DNG conversion. It's the generic raw format, and it's meant to be "future-proof", so that you'll be able to open and edit your files forever, regardless of software evolutions over the coming years. Please don't ask me for an opinion about that, because I don't have one, nor enough knowledge to form one.

Other people convert to DNG because the files are slightly smaller, and the disk space saving is worth it. That sounds sensible.

But despite all of that, I don't convert to DNG format. Why? Because I love XMP files!

XMP files are the little separate files that are created when you edit a Raw file. The XMP files (known as a "sidecar file") holds all your edit data. They're very small, but very important. When you see XMP files in your folders, please don't delete them - they're your friends :)

DNG files don't have sidecar files. The XMP data is stored within the DNG file. Just one file instead of two - that probably sounds very sensible to you. But not to me.

In the next post I'll write more about XMP files, and why I like having them around when I'm editing. For now, I'll just concentrate on the one big reason why I prefer not to convert to DNG format - speed of backups.

Backups are darn important, of course. Everyone has a different backup strategy - mine is quite simple. One or more times a day I run a backup of my internal hard drive onto my external one. The SyncToy backup software scans my internal drive, and copies all new files (either newly-created, or recently modified) onto the external drive. That way, my external drive is an up-to-date copy of my internal.

"What does this have to do with DNG??" you're asking yourself.

Well, let's say that on Tuesday I downloaded one hundred new raw files to my computer. Tuesday's backup would take a little while, right? Because 100 raw files is a lot of data to transfer. Tuesday's backup would take some time, regardless of whether the files were DNG or not.

On Wednesday, I edited those raw files. How long would Wednesday's backup take? If the raw files were DNG, the backup would take the same (long) amount of time that it took on Tuesday. But since I keep regular raw files, and therefore have separate XMP files, the backup is lightning fast, because the only things that need to be copied over are the 100 tiny XMP files. The raw files themselves haven't changed.

So yeah, speed of backup. That's why I don't convert to DNG. Yet, anyway.

If you have an online backup system, what I've described in this article will probably make lots of sense to you. Not having to upload entire raw files again after editing? Priceless.

View your filename extensions, Windows peeps!

2012-02-02T06:23:00.000+10:00

Occasionally - well, surprisingly frequently, actually - somebody will come to me with laments like this:

"I saved my photo, and now I can't re-open it!"

"My image files have no thumbnails!"

etc.

If the person is a Windows user, the reason for the problems is often that the file has been saved without the filename extension - ".jpg" or ".psd" or whatever.

Windows need filename extensions. If it doesn't see the "dot-something" after the name, it has no idea what program to associate it with; and therefore can't preview it, or open it. I'm not sure what causes files to be saved without their extensions. It just seems to happen sometimes. Very annoying.

What's more annoying is that Windows, by default, hides all filename extensions. I wish I knew which genius at Microsoft decided this was a good idea, because I'd like to punch him in the nose. When all extensions are hidden, you can't see at a glance if any filenames don't have an extension at all.

So, I urge you to follow these steps and make your filename extensions visible.

First, go to the Start Menu, and open the Control Panel:

In recent versions of Windows, the default view of the Control panel is "by Category", which is silly, and deserving of another punch in the nose. Change it to "Small Icons":

Choose "Folder Options":

Then go to the View tab, and make sure "Hide extensions for known file types" is unchecked:

Thanks for doing that. Now it'll make troubleshooting much easier for all of us. You can see at a glance if a problem file is missing its extension; and if so, you can simply rename the file to add the extension back on (as long as you know what it's meant to be), and it should preview and open just fine.

PS: Make sure you don't add extra letters to a filename extension! You might think you're being clever by saving your file as "Image3849.jpg-BW". But your computer won't recognise the "-BW" part, and it'll choke on it. Don't mess with anything after the dot!

Painting on some skin

2012-01-28T21:14:00.000+10:00

This is one of those tutorials that I'm writing in the hope that you'll never have to use it.

Sometimes, things don't quite go right, do they? I'm talking about some very bright light hitting the skin of your subject, or some accidental overexposure, causing complete blowout in one or more areas.

Here is one such image that I was able to (at least partially) rescue from the brink of catastrophic blowout. I know it's not perfect, but at least it's better than nothing. The method I'm about to show you is not a miracle cure, by any means. But it's a way of subtly adding some fake skin colour where none exists.

Obviously, you should be shooting Raw, which allows some latitude to rescue difficult lighting. But even Raw is not infallible, as you can see in this gorgeous example image, kindly provided by Terri:

There was some bright light hitting Mum's fingers which caused a loss of detail which wasn't quite recoverable even in Raw. So Terri needs to add a little bit of colour there, to prevent the viewer's eye being drawn to that area, instead of the beautiful child.

So, here's how to go about it. (By the way, this method will work in both Photoshop and Elements.)

First, use the Eyedropper tool to sample a darker area of good skin colour. In this example, I sampled the thumb:

Then, add a Solid Color layer ...

... which will, of course, fill the image with that sampled colour:

Next, change the blend mode of that layer to "Darken" (sometimes Multiply works too, but I've had better success with Darken):

Then invert the layer mask to hide the colour altogether:

Choose the Brush Tool, and choose white as the foreground colour (if it isn't already). Set your brush opacity very low - I used 5% in this example, but sometimes I even use as low as 2% or 3%:

Then, start painting. I have no particular words of wisdom about this. Just very gradually build up the colour in the areas which need it. Don't go over the top - the darker you go, the more visibly fake it will look. The best you can hope for is to add just enough colour to stop that bright area being an eyesore.

Here's the before-and-after of my job on the fingers. I hope you agree with me when I say that it's strong enough to be effective, but not strong enough to look dodgy:

I hope this little tutorial helps you out of a tricky situation once in a while.

Easy, funky, crazy clipping masks

2012-01-28T15:54:00.000+10:00

I'm prompted to write this short post by a great question from David on the Ask Damien page.

"How would you go about making a storyboard with some 'effect' to its edge?"

This question follows my posts about storyboards and clipping masks (here and here). If you don't know how to make storyboards and collages and things, I encourage you to read those tutorials.

Ok, David, let's have some fun. This is the easiest way I know to do this, and I hope you love it.

Of course, the first step is to decide on the size and shape of your document. I'm making a small one today. Here it is, my plain blank transparent document:

Brushes. That's where the fun is, my friend. Choose your brush tool, then find all the presets:

By default it'll probably be on "Basic Brushes". You can see there are lots of other brush libraries to choose from. And inside each of those libraries are dozens and dozens of different funky brushes. Honestly, you could use a different one every day, and have fun for a whole year.

I choose the one called "Dry Media Brushes".

When you choose a new one, it checks to make sure you want to load them, and remove the Basic ones. Of course you do. You can re-load the Basic ones afterwards. Press OK:

Then scroll through the brush options available in the library you've chosen. If none appeal to you, try another library, and so on.

I've chosen one called "Heavy Smear Wax Crayon":

Then, just paint. Scribble away and fill the space. Mine looks like this:

You know what to do from here. File>Place and place your photo, then clip it. Voila!!!

Of course, it needs a background, so I add a gradient layer behind:

And finish it off by adding some layer styles to the painted layer:

I hope you enjoy playing with this as much as I enjoyed writing it.

Creating, saving and using a watermark (PS & PSE)

2012-01-27T22:35:00.000+10:00

Putting a watermark on your photos is important if they're going online. In this tutorial I'm going to give you some simple guidance for making and using your own.

(I've captured this tutorial in Elements 9, and it's relevant to all versions of Photoshop and Elements. Please also note that I have zero design skills, so please don't laugh at my feeble watermark!)

PART ONE: PLAN IT

As always, a bit of forethought goes a long way. Take a little time to conceive your design. This might involve surfing the web to look at other people's watermarks for inspiration; or it might just mean scribbling some ideas with pencil and paper.

PART TWO: MAKE IT

I've always found it's best to actually make your watermark on a photo. This might seem plainly logical to some of you, but I'm a bit surprised at the number of people who design their watermark on a plain white background, and don't really know how it looks on a photo until they try it out afterwards.

Also, it's a darn good idea to make the watermark bigger than you'll ever need it. Therefore, I make mine on full-res photos, even though I'm only likely to use it on a low-res one.

Pick any old photo, really. I've chosen this one of a snoozy koala:

So, I start. I type my name ...

... then try a few fonts and choose one I like:

Then I add the next layer of text, and choose another font:

I want to add a little extra embellishment, so I choose the Shape Tool ...

... and choose a custom shape from Photoshop's extensive library:

(If you've never really looked at all the shapes that come with Photoshop, do yourself a favour. There's lots of fun stuff in there.)

Then I draw the shape beneath the watermark. Now I have three layers comprising the design - two text layers and a shape layer:

If I want to move the watermark around, without the different elements separating, I select all the layers at once. I do this by first selecting the top layer, then holding down the Shift key, and selecting the bottom layer. This highlights all the layers, and I can use the Move Tool (V) to move the watermark around.

Time to modify things and "jazz it up". I select the "WATERMARK" text layer, then go to Layer>Layer Style>:

(The above menu slightly differs between Elements and Photoshop, but you'll get the idea.)

I add a bit of a stroke to the text:

And I lower its opacity a bit:

(Please note that there's little point in playing with the Layer Blend Modes when making a watermark, because they don't transfer onto the image when used.)

Now I select my name layer, and change the colour to blue:

On the shape layer, I add a Bevel layer style (I've zoomed in so you can see it a bit better):

(At this point, the design-conscious among you will be feeling nauseous at my hodge-podge of styles. I'm just throwing a few things together to give you some ideas of what can be done.)

I re-arrange the layer order a bit, and move the shape layer upwards so it's overlapping the text:

At this point, I think I'm done. But there's one more important thing to do, and I urge you not to forget this step. I have to move the watermark around to various parts of the image, and check its visibility and clarity.

When I place it over a dark area - the tree trunk - I see that the blue text gets a bit hard to read:

I have a few options. I could just not worry about it - that's the easy way :) I could dabble with different text colours. Or I could play with more layer styles.

I decide on the latter. I add a bit of soft white "outer glow" to make the text stand out from the image a tiny bit more:

Finally I'm satisfied. The last step is to delete the background layer so that the watermark is on a transparent background:

Then crop away the excess space:

File > Save As ...

... and save it as a PSD file:

Then I close it. I don't need it open any more.

PART THREE: USE IT

Now I open a new photo. This is the one I want to upload somewhere. I shrink it to web size, flatten it, sharpen it, yada yada yada ...

Then, File>Place ...

... and choose the watermark file, and it appears in the centre of the photo:

The transform handles are active, so I can resize it if I need to, and move it around to where I want it:

I press Enter to finish the transformation (I can Ctrl-T to do it again later if I need to).

I can stop there if I wish. Or, I can play with blend modes ...

... or opacity. Whatever suits my style ... or mood ... or whatever.

Then I save for web, and it's ready to go!

SOME NOTES

Now that you've learned how to make and use a watermark, learn how to build it into an action (PS only): http://damiensymonds.blogspot.com/2010/03/tip-for-watermarking.html

Some people say you have to save a PNG file to be a watermark. This is absolute nonsense. PSD is the best format - not only can it have that transparent background which is so important for a watermark, it is also completely easy to come back and edit it later (eg if I decide I want my name in red not blue).

Other people turn their watermark into a brush. This is really restrictive, and I don't recommend it. A brush can only be one colour, and it can't be batch-processed. Use this PSD/Place method instead.

GO FORTH!

I hope this helps. Have fun making your own, and don't forget to visit me here if you have any questions or problems: https://www.facebook.com/groups/195567190503489/

The light around your computer

2012-01-23T20:48:00.001+10:00

Be careful not to underestimate the influence of the ambient light while you're editing your photos.

I've seen many people comment that when they edit photos at night, then look at them during the daytime (or vice versa) they look markedly different.

Of course, when you edit some photos (especially if you're tired), then look at them again a few hours or days later, you're always going to notice some things you'd like to do differently or better. That's human nature, I guess, and in a way it's a good thing - while you're perpetually critical of your own work, you'll keep striving for improvement. And as long as you do all your Photoshop editing with layers, it should be very easy to make whatever adjustments you need.

But I'm not here to pester you about layers again. I want to discuss light. The light that's around you while you're doing your post-processing.

Some people say you should edit in the dark. I mean, really dark. Other people don't say "dark", exactly, but they say that the computer screen should be the brightest thing in the otherwise dim room. Me, I think that's completely ridiculous. What kind of idiot stares for hours directly at the brightest light source in the room? How is that good for you? And what about if you need to jot something down, or read something you jotted down earlier? And how on earth are you supposed to compare your prints to your screen to see if your calibration is accurate? (Actually, the answer to that last question is: you're meant to buy a fancy viewing booth. I've got better things to spend my money on, frankly.)

Ok, I realise that last paragraph was a bit of a rant. If you like to edit in the dark, and it works for you, go with it. But any time somebody asks me, I always give the same response: "I think you should only edit in light that you'd be happy for your kids to do their homework in." You wouldn't like them to be risking eye strain by writing in too-low light, and I don't want you to risk it either. I hasten to add I have no scientific basis for this - it just seems sensible to me.

Anyway, what's MOST important about your light is that it's consistent. It's got to be the same every time you edit. And that's not always an easy thing to achieve.

Me, I'm incredibly lucky. The bloke who owned our house before us was a musician, and he built himself a windowless (and soundproof!) room downstairs. When selling, the poor old real estate agent must have been wondering "how the heck am I going to market a windowless room??". But it's perfect for me! I've got some nice "daylight" (6500K) bulbs in the ceiling, and no matter when I'm editing - day or night - the light is exactly the same. It plays a huge role in the consistency of my work, and that's important.

Actually, the middle of the day isn't such a problem - not here in Queensland, anyway. That light is pretty "white". But it's the golden tones as the afternoon goes by, and turns into evening, that are problematic. Lovely for taking photos in, not so good for editing.

Not many people will be as lucky as me, I guess. Some of you might be able to find a windowless nook in your home, away from that nasty changing sunlight. But most of you will have windows to contend with. So you have to do the best you can.

Heavy curtains are the best answer, of course. Cardboard is also good, but doesn't flatter your house! Anyway, you've got to block out that outdoor light, and install some good indoor light. As I said, I had no trouble going to the lighting store and asking for 6500K bulbs. They're very standard. Then it doesn't matter when you're working - it's all the same light.

One more thing to mention. Often people ask me what I think about laptops for editing. I don't like them. Not because of the laptops themselves (yes, some have really bad screens, but others have perfectly good screens), but because of the very thing that makes them attractive - their portability. If you've got a laptop, you're tempted to edit on the couch, or in the hammock, or wherever. Please be careful about this. Give a thought to the light you're in, and what impact it might be having on your edits.

A Pinterest idea

2012-01-22T14:55:00.000+10:00

I've been thinking ...

I know a lot of my readers sometimes struggle with skintones, because you feel uncertain about what looks good and what doesn't. Sometimes when someone shows you how it should look, it gives you some clarity of mind, and you can continue editing with more direction and confidence.

I also know a lot of you are on Pinterest. So, here's what I'm thinking. I think you should create a board called "Skintones I love". And whenever you're cruising around, checking out other photographers' blogs, or whatever, and see a photo with skintones that take your breath away, pin it!

Then, when you're doing your own editing, and feeling a bit uncertain about "is this too yellow?" or whatever, just browse your board. Immerse yourself in great skintones for a few moments, then re-visit your own photo with fresh perspective.

Web browsers and Facebook

2012-01-16T14:50:00.000+10:00

I get SO irritated every time I see somebody say "Facebook changes the colour of images".

NO IT DOESN'T.

Facebook makes a mess of the clarity of images, we know that. Its compression settings are pretty strong, and it plays havoc with our sharpening, in particular. Nothing we can do about that. With all the billions and gazillions of images on their servers, can you blame them for compressing them a bit? Of course not.

But it doesn't change colour. If you see that your images' colour looks different on Facebook than it did in your editing program, it's your fault, not Facebook's. It's because you're not using the right browser.

Let's discuss the four main browsers. We'll start at the bottom of the pile, with Internet Explorer and Google Chrome. These two have no colour management at all.

(There have been announcements about colour-management features for both of them, but I can't get them to work, and all my googling hasn't yet found me proper instructions to do so. If anybody knows how to turn on colour management in either Chrome or IE, please let me know.)

If you're browsing in Chrome or IE, you just don't see correct colours. End of story. Even if you have a properly calibrated monitor, those two programs still don't honour colour profiles. If Chrome or IE are your favourite web browsers for whatever reason, that's great - keep using them for most things. But do NOT offer critique to anybody about the colour of their images in them; or you'll end up embarrassing yourself.

Next up the ladder is Safari. Safari is colour-managed ... almost. It honours the embedded colour profiles in images, and shows them accurately. Wonderful! Safari is a great option for most of your colour-critical web browsing.

But Safari has one little quirk. For images on websites which don't have colour profiles embedded, it doesn't attempt to colour-manage them at all. It just shows them however your monitor wishes. In other words, it displays them the same way Chrome and IE do.

All photographers should know that it's vitally important to save their web images with the ICC profile embedded. (Best practice for web photos) But not everyone has got the message yet, so some people still save "untagged". If you're viewing those images, Safari will show them wrong.

More relevant to this thread is the issue of Facebook's thumbnails. Facebook shows large versions of photos with the ICC profile embedded, but not the thumbnails. I'm sure the Safari users among you have noticed that the small photo you see in the newsfeed is a different colour than the large photo you see when you click on it.

So, if you're a Safari user, never offer CC on the colour of a photo just by looking at the thumbnail. Make sure you view the full-sized file.

At the top of the browser ladder is Firefox. Now, I know some people have speed issues with Firefox - I have no advice to offer in that regard. It's always run really fast for me. In this article I'm just talking about colour. And it's the king of colour.

However, I need to make something very clear: At its default settings, Firefox behaves the same as Safari. That is, it shows tagged images correctly, but untagged images incorrectly. Once you make a simple setting change, it then assumes that untagged images are in the sRGB colour space, and therefore gets them right nearly all of the time. (Obviously in cases where somebody has foolishly uploaded an Adobe RGB or ProPhoto RGB image without the ICC profile embedded, Firefox won't show it correctly. Fortunately that doesn't happen often.)

If you're a Firefox user, but haven't changed the colour management settings, you'll observe the same issue that Safari users see - the thumbnails on Facebook don't match the large images.

Time to rectify that! Here goes.

1. Launch Firefox, and type "about:config" in the address bar:

2. Say you'll be careful:

3. Type "color" in the Filter field:

4. Double-click the color_management.mode line:

5. Its default setting is 2. Change it to 1, and press OK:

The next bit is optional. I don't know how much difference it makes, to be honest. Ignore it if you want. But this is what I do (to make Firefox completely match Photoshop).

6. Double-click the rendering intent line:

7. Change it from 0 to 1 and press OK:

8. Restart Firefox.

That's all there is to it! Now, your Facebook browsing will be completely and utterly colour-managed.

As always, visit me at Ask Damien if you have any questions or problems with these instructions.