After taking a photo every day for a year, I thought it would be worth a few extra hours putting it all together in to a nice looking mosaic.

I think the easiest way to do this is using Google’s Picasa program.

The problem with this is that you need to have all your photos in on folder, which I didn’t have. Rather than spend ages going through each folder and finding the photo I had chosen for that day, I just downloaded mine from Flickr. Since I had them all uploaded in one set and all named with their correct numbers, this seemed like it would be the easiest method.

I downloaded the Saleen Flickr Downloader (which didnt seem the most elegant solution, but worked well enough) and told it to download my Project 365 set. I set it to download just the Large size, rather than Original, which would take hours. It still took quite a while and one photo didn’t download properly (lucky I noticed that!) but after that I had the folder of every image.

The next problem was that the downloaded images had the Flickr ID in the file names, so they weren’t being sorted properly. They also had no EXIF data, so I couldnt arrange by date taken. To fix this, I used the trusty old Bulk Rename Utility.

The files were named like this:

5794243281_ee16c5544c 121_365_L.jpg

So I just told Bulk Rename Utility to remove the first 22 characters from all the file names. And just to neaten them up, remove the last 2 characters too.

Which left much nicer looking file names which also could be sorted by name:

121_365.jpg

Note: This is only useful for the Picasa mosaic which is in chronological order. For the final mosaic, I didn’t need to do all this as the photos are arranged by colour.

Now the photos are in a folder and sorted by date, we can find the folder in Picasa and click on “Create Photo Collage” (make sure you only have the folder selected, rather than any of the individual photos within the folder, or it will just make a collage with the selected photo).

There are a lot of options here, so play around and see what you like. Since all my photos are square and I wanted an orderly, linear mosaic, I chose “Grid”

This was my result (after adding some space around the edges and text in Photoshop):

This was good, but it looked a bit messy. I wanted to see if I could arrange the photos by colour.

The closest thing I could think of was using a Photomosaic program. AndreaMosaic seems to be one of the best and is also free.

So after installing that, I found the image I wanted the images arranged on – I went for a rainbow type image. Basically a colour picker box.

Note the settings in the above image: “no duplicates”, no colour change, only use the original tiles. This ensures that only the 365 photos we have will be used, won’t be used multiple times and the colours won’t be modified to match the rainbow image.

Another thing to watch out for is having half-images on the edges of the frame. The first mosaic I made, 5 photos were right on the border and only showing about 5 pixels of the photos, so I almost missed them. I just needed to change the mosaic size so they weren’t being cut off.

Now we need to tell the program to use the images in the 365 folder. Click on the number 2 (at the top right, the dolphin photo)

Click on “Add Folder” and select the 365 folder. Then at the top click “Save List”. AndreaMosaic will go through and analyse the photos, ready to put it all together. Once that is done, click OK.

Finally, click number 3 (the Van Gogh) and see how it turns out.

There will be a warning saying there will be gaps (because we turned off duplicates and colour changing).

Mine had quite a few gaps, but I realised the photos would all fit.

After some cutting, pasting and rearranging in Photoshop, the mosaic was done!

Flange focal distance

Or: why you should try a Nikon lens on a Canon body

Coffeeground and simon.hucko made an interesting discovery here about the focal distance of the different camera manufacturers lenses and bodies. You can get a more technical explanation here on Wikipedia (as well as a list of all the other manufacturers and their focal distances), but basically Canon lenses are designed to focus on the Canon sensors which are 44 mm from the mounting flange (the metal ring on the camera and the rear of the lens). Nikon lenses are made to focus on Nikon sensors, which are 46.5 mm from the back of the lens.

This is interesting to freelensers because it means if you use a Nikon lens on a Canon body, you will be able to focus on things that are much further away, while still getting the beautiful bokeh and other freelensing effects. You don’t even need an adapter, since you’re just holding the lens in place.

Help – nothing is in focus!

Freelensing works best with prime lenses (50 mm is a favourite). Telephoto and zoom lenses can be used, but they will have more vignetting and seem to be harder to focus. Also, because most zoom lenses only start at f/4 or f/5, they can’t let as much light in (compared to an f/1.8 50 mm, for example), which may result in darker photos.

Before removing the lens it is best to set the focus on “infinity” – for two reasons. Firstly, you are more likely to drop the lens if you’re fiddling around trying to change the focus while holding the camera and lens separately. Secondly, setting the focus on infinity will let you control the focus more finely by using the tilt of the lens and will let you actually be able to focus on things more than a few feet away.

Freelensing doesn’t work with my Nikon!

If you use a Nikon camera for freelensing, you may need to set the camera to Manual or Aperture Priority mode. (Thanks to Eddie Barksdale for discovering this)

Nikon G lenses also close down the aperture blades when disconnected from the camera (the D lenses don’t do this – thanks kristopherlundborg), so if your photos are coming out very dark, this might be the reason. You will need to make something that will hold the aperture lever on the back of the lens in the open position. A small piece of blu-tack or cardboard seems to do the trick. (Thanks to ted @ndes and Joseph Yarrow for discovering this)

Narflet also reports that freelensing can make the metering on her Nikon a bit unreliable: “I often have to boost +1-2EV”. If your photos are still coming out dark after fixing the aperture level problem, this might be the next step to take.

Will freelensing work with my Sony?

In your camera’s menus, look for the “Release w/o Lens” option, and make sure it is enabled. If you can’t find that option in any menus, try setting the camera to Manual mode (M on dial) and also set the camera to manual focus (thanks ted @ndes).

Similar to the Nikon lenses, you may also need to make something that will hold the aperture lever on the back of the lens in the open position. If your photos are very dark while freelensing, this is likely the reason. A small piece of blu-tack or cardboard should be able to hold the lever and keep the aperture blades open.

Olympus, Pentax and others

I haven’t heard any problems from those using Olympus or Pentax cameras. It seems the Olympus and Pentax lenses, like the Canon ones, don’t auto-close the aperture blades when disconnected.

Anyone using a Panasonic or other camera for freelensing? Did it work? Any problems? I’d particularly like to know if anyone has tried it on any of the Micro Four Thirds system cameras. Let me know in the comments here or on the Flickr group discussions page.

I thought it would be interesting to see how hard it would be to achieve in Illustrator. With one small, free Illustrator script, it turns out it is quite easy.

The script we’ll need is “vary_hues” from wundes.com. Save the .js file on to your computer somewhere. It’s well worth checking out the other scripts too.

Step 1

Create a new document, choose the Rectangular Grid Tool and click on your canvas to open the options for the Grid Tool.

I chose the same width and height to keep the grid as a bunch of squares and have set the rows and columns to 10×10. Unless you’re going for a Mondrian style effect, leave the skew at 0% on each.

Step 2

You may notice at this point that the grid is just a rectangle with lines through it, rather than a bunch of individual squares.

To fix this, select the grid and open the Pathfinder palette. All you need to do now is click “Divide” and this will cut the grid up in to a bunch of squares for you.

To make it easier to see, fill the squares with a colour.

Step 3

Now we have a grid of squares, but it’s still all in a block. Ungroup the grid so you can select individual blocks – right click > ungroup or CTRL + SHIFT + G.

Now we can get some colour in to them.

Step 4

This is where the vary_hues script comes in to its own. Select all the squares and go to File > Scripts > Other Script, then browse to where ever you saved the vary_hues script.

When the dialog box comes up, enter 100 and click OK. 100 will ensure that there’s a wide variety of colours.

Step 5

The random colours will, most likely, look fairly ugly at the moment. The secret to the Gerhard Richter and “Yes” examples is that the colours aren’t completely random. So to get some order and harmony to the colours we’ll use the Live Color tool.

With the squares still selected, choose the Recolor Artwork button to open Live Color.

Step 6

This is where you can have some fun choosing colours. Click on the little grid button to open the swatches menu. This will give you a large selection of colour swatches to choose from. Art History > Pop Art works well, but experiment with others to see how it works.

The Pop Art colours will be applied to the squares, replacing the random colours with more harmonious and hopefully attractive colours.

The default Pop Art colours are a little dull, so let’s jump over to the Edit panel.

This will give you more control over the individual colours. The slider at the bottom (under the little sun icon) adjusts the brightness of the colour, so we’ll turn that up to make it a bit more vibrant. You can also grab and adjust individual colours on the palette to adjust them.

Click OK and you’re done!

That’s the basic idea covered, so hopefully it’s a useful starting point for some interesting designs. If you use it, I’d love to see what you come up with.

MSN Live Messenger drawn-conversation in progress

The problem with these drawn-conversations is that it can be difficult to keep a record of what has been drawn. Sometimes whole comics or detailed pictures are drawn with no way to save the images… unless you have Adobe Illustrator. Here’s how.

Next time you have a drawing in MSN that you want to save, open up Illustrator and simply drag and drop the drawing into the Illustrator workspace.

Drag and drop the drawing into Illustrator

The beauty of using Illustrator (rather than just taking a screen shot) is that it saves the drawing in full vector detail.

So whether it’s a comic that you’ve been able to get just the right expression on or just a funny picture you want to save, you’ll be able to retain all the detail and enlarge or refine the image further in Illustrator. Or even  export it to Flash and animate it, or print it out and give it to the friend you were talking to (things seem funnier in print).

Right: Vector outlines in Illustrator. Left: A custom brush applied to the strokes.

Illustrator is the only program that seems to be able to capture the vector data properly. If you know of any other programs that work, let me know!

Occasionally for a print or website, you might require images with a white background. Usually these are easy to find, take or mask yourself, but for some things it can be more difficult.

Fire is one example. It would be very difficult to try and mask out flames from a fire photo using the pen tool or a brush, because of the organic shapes and indistinct edges.

As you will see in moment, it doesn’t have to be difficult. Using blend modes and layers, it’s very easy! So easy that I’ve turned it into an action which you can download at the end of the article if you so desire.

One

Take or find a photo of some fire (other images with a similar black background might work too – let me know in the comments if you’ve found something else on which this technique works!)

I started out with a rather nice Creative Commons licensed photo found here on Flickr.

Two

In Photoshop, you should have the default “background” layer. Duplicate it and call it something like “inverted”. Then, do just that – invert the image (CTRL + I)

Now set this “inverted” layer to the “Luminosity” blend mode.

As you can see, the whites of the fire have been dulled and it doesn’t really look very fiery. We’ll fix that in the next and final step.

Three

Duplicate the “background” layer again and drag it above the “inverted” layer. Rename it something like “Lighten”.

Finally, set this layer to the “lighten” blend mode and that’s it. You should have a nice white background on your fire photo.

Download the action

This technique is very useful in general. For example you can use it for blending objects into a background or for making a grungy text font.

Watch below or in high quality.

Below are some examples I’ve made previously.

I thought this idea was too good to let slip away into obscurity, so here it is and here’s one way to do it in Photoshop.

As you can see in the image, they have used balloons to outline the negative space in the shape of the OXO Tower. The design is simple, light and colourful and I really like it. The easiest way I can see to do this in Photoshop is to use a brush and play with the scatter and size settings.

To start off, I’ve created a balloon shape. This was done by drawing a circle with the Ellipse Tool, selecting the two middle points and moving them up a bit to make a balloon shape. Then I drew a little knot in the end and combined that with the balloon. Finally, I drew a bit of ribbon tied on the end using the Pen Tool. It took less time to create than it did to write this sentence, actually. But if you like, you can download the balloon Custom Shape, balloon AI vector or just go right ahead and get the balloon brush shape.

Next, I flattened the layers (ALT+L+F) and made it into a brush shape (ALT+E+B). It’s good to have the shape as a decent size, so it doesn’t get too blurry or pixilated at different sizes as a brush, so mine is about 800px tall.

Now, you’ll want something to be in the negative space so that the brush shape (balloons in this case) have something to form around. I happen to have a hot air balloon shape handy, which seems appropriate enough, so I’ll use that.

To set this up, I’ve created a blank canvas and have my hot air balloon shape in a new layer and have set it to slightly darker than white (so I can see the outline) and have a white background. I’ve also put a slight radial gradient on the background to make it show up a bit more.

With that set up, it’s back to the brush. Select the brush tool and go to the brushes palette. The newly created balloon brush should be at the bottom.

Brush Tip Shape
Depending on the size of the document, you will need to adjust your balloon size. The balloons look best far apart, so set that quite high. I’ve done about 300%

Shape Dynamics
If you’re fortunate enough to have a graphics tablet, then you can have size jitter at 100% and control it with the pen pressure. Otherwise, you can set it to fade or just manually adjust the brush size using the [ and ] keys. It’s a good idea to set a minimum diameter so you don’t get lots of tiny 1px balloons. I’ve chosen about 20%. To give the balloons a bit more of a natural look, adjust the Angle Jitter – not too much though, unless you want them going in all directions. I’ve also set the brush to Flip X Jitter so they’re not all facing the same way (Y Jitter will flip them upside down).

Scattering
To add to the natural, random look of the balloons, adjust the Scatter slider to your liking. 100% is good for me.

Color Dynamics
Finally, colour dynamics will adjust the colour of the balloons. I’ve set foreground/background jitter to 100% – this way, the balloons will be set to a colour somewhere between the colours you have chosen in the palette.

Other Dynamics
You might also like to adjust the opacity of the brush to make the balloons slightly transparent.

Now, we can finally paint some balloons! To start off with, it’s good to get some smaller balloons around the shape of the object (hot air balloon in this case). Set the foreground/background colours to black and white (press D) and paint around the shape until you’re happy with it.

Once you have a rough outline, you can go nuts – fill the entire canvas with balloons if you like! Keep in mind that it does tend to look better with some white space.

A moment ago I said to set the colours to black and white. This wasn’t to depress you or deprive you of colours – just the opposite. We can now do my patented (not really) Gradient Map trick.

Make a new Gradient Map Adjustment Layer and start scrolling through the gradients and seeing how they look – you might like one of the colour schemes. Personally, I’d like it to be lighter and brighter, so I set the Gradient Map layer to Soft Light. This will keep the background light while embiggening the balloons with colour. The default Photoshop gradients are okay, but there are some really nice ones in “Pastels” and “Simple” or you can even go to Colourlovers and get some from there.

Here are a couple of quick examples of what I came up with (Click to view 1600x1200px versions).

If you make something yourself using this method, let me know in the comments – I’d love to see what you create.

This guide will be a (hopefully) simple guide to creating skins correctly for Launchy 2.0 based mostly on my own trial-and-error in making this Elegant Brit skin.

The 5 main skin files

Launchy skins are made up of 5 files.

1. alpha.png – is the main part of your skin and should contain all the colours, patterns, etc and have a transparent background. It should also have the areas for the input, output and icon removed (transparent).

2. background.png – will be the background for the input, output and icon areas and should contain nothing else. The rest of this image should be filled in red (255, 0, 0).

3. mask.png – should be based on your background.png, but with the red areas filled in white (making them transparent) and the input, output and icon areas being in black.

4. misc.txt – defines the areas for the input and output text, location of the icon, alternative suggestions area and the size of the above images (all the same size).

For example:

• input = 14,49,276,19 (input text where you will type the program you want. This box will start at 14 pixels from the left edge of the image and 49 pixels from the top. It will be 276 pixels wide and 19 tall)
• output = 14,14,276,20 (will be the title of the program selected. The box will start at 14 from the left and top and will be 276px wide and 20px tall)
• boundary = 0,0,311,83 (the overall image size of alpha.png and the others which are the same size. 311 being the width and 83 the height)
• icon = 264,9,32,32 (position of the icon for the selected program. 264px from the left, 9 from the top, in a 32x32px box)
• alternatives = 7,74,296,100 (location of the alternative suggestions box. 7 from the left, 74 from the top, 296px wide and 100px tall)

Some skins also define a few font related variables in this file because it is not possible in the qss file. These related to the path details in the drop down list:

dropPathColor = colour in RGB
dropPathFamily = font name
dropPathSize = font size
dropPathWeight = bold
dropPathItalics = italics

5. style.qss – is basically a CSS file and controls the fonts and borders for the skin.

<QLineEdit#input or output or alternatives>
font = font size and name
background-color = text background colour, including alpha value %
alternate-background-color = background colour for the alternative suggestions box
color: font colour
border-style = whether border is a solid line, dotted, dashes, etc. Borders are usually off
border-width = border size in pixels – usually 0, meaning off

Other files

author.txt – information about the skin such as title, author, date, version, etc.

Sometimes other skin features such as option button, close button and skinned scroll bar are included.

–

Thanks to sierra13 and this thread in particular on the Launchy SourceForge page.

When I started using Lightroom, I thought it would be very handy to be able to import all my Camera Raw presets into the developing room. Afterall, I’ve spent the past couple of years developing them and they’re the same options as far as I could see. Wrong.

I thought it would be a matter of copy and pasting the files from the Camera Raw folder into the Lightroom one. Which is what I initially tried. It didn’t work. So I saved a setting from Lightroom, had a look back in the folder and it has some strange .LRTEMPLATE file to store the settings.

Not sure who thought that was a good idea. It’s probably just the same Camera Raw .XMP file but with some changes to make .XMP unreadable.

As far as I can tell, the following is the only way to import the Camera Raw settings into Lightroom (at the time of writing).

Essentially, we need to get a bunch of photos, apply one preset to each photo, click done to save the XMP sidecar file, then open those files with Lightroom, go to Develop on each one (it will read the sidecar files with your Camera Raw settings) and click the plus sign to save as a Lightroom preset.

It shouldn’t take very long, depending on the number of presets you have. Personally, I had about 25 Camera Raw presets that I use and tweak further for my photos. I’m not sure if that’s a lot or not. Took me about 10 minutes to go through them all and make the Lightroom files.

Step by step

Open Photoshop and import a Raw photo, then go and check to see how many preset files you have. You’ll need to open the same number of files to apply one preset to each.

Cancel Camera Raw and go and select the number of raw files you need. In my case it was about 25. Import them into Camera Raw (don’t worry – we’re not going to be opening them all with Photoshop)

Now select the first photo, then go to the presets tab and select the first preset. Then select the second photo, apply second preset and so on, until you have one preset for each photo. Shouldn’t take too long.

When you’re done, press the Done button – not the okay button which will open all the photos in Photoshop (which, if you have a fairly slow computer like mine, could take ages). Pressing the Done button simply writes the sidecar file with the settings.

Close Photoshop and open Lightroom. Go to the folder where those Raw files and sidecar files are saved. Select the first photo and Press D to go to the develop section.

The photo should look pretty much the same as in Camera Raw (if it doesn’t, wait a few seconds for Lightroom to read the sidecar file and apply the settings).

Now click on the little plus sign to save a preset. It’s not really well labeled. I thought the plus sign would grow and shrink that palette when i first saw it.

Give it a name or a number – done. Your first preset has been created. Now, press the right arrow key to move to the next photo. Again, click the plus sign, name it, and so on through the folder.

Before too long you should have all the presets saved into Lightroom’s own strange little .LRTEMPLATE preset files.

The photo I started with was okay, but nothing particularly interesting. The exposure was good, but the real scene was more dramatic. I’ve tried to recreate that scene the way I saw it. And, admittedly, enhance it to the way I wish I had seen it.

Step One – Shadows/Highlights

Duplicate the photo layer. Photoshop doesn’t yet have the ability to have Shadows/Highlights as an Adjustment Layer, so I like to keep a backup of the original photo for fine-tuning later on.

To give it a bit more of an HDR look (which is how our eyes see), I used the Shadows/Highlights tool . I kept the settings subtle to just bring a bit more detail into the clouds and to brighten the dark areas of the road and trees.

Step Two – Darken the Sky

Ensure your foreground colour is set to Black (press D to reset the colours to their Defaults)

Now add a Gradient Adjustment Layer . Set the gradient to “Foreground to Transparent”. I want the gradient to be going from black at the top to transparent at the bottom, so I’ve ticked “Reverse”.

Set this Adjustment Layer to Soft Light blend mode and now we have a nice dark sky. A similar effect could have been done by using a Graduated Filter on the camera when taking the photo, but a) I like the control and fine-tuning this offers and b) I don’t have a graduated filter (at least not one that fits my main lens).

Step Three – Super-Sharpen

Finally, to give the image more impact, I used some super-sharpening. My preferred method of sharpening (everyone seems to have their own way) is using the High Pass filter.

To use this, I needed to first flatten the image into a new layer above all the others (Ctrl+Alt+Shift+E – a very useful shortcut!).

Then to Filters > Other (at the bottom) > High Pass. The amount depends on the size of your photo and the effect you want. A small number (0.5 or 1) is good for normal sharpening. But for this super-sharpening effect, I went with 50.

Now, you should have a mostly grayscale image with a bit of smudged colour. I prefer to keep it all grayscale, but sometimes the extra colour looks good. To remove the colour I use Ctrl+Shift+U.

Set this layer to Hard Light blend mode to see how it looks. Now, turn the layer off and on a few times to compare. The unsharpened version almost looks blurry by comparison! The difference is dramatic. Some people will like it, others might think it’s too much. You can (and I usually do) turn the opacity down to keep it subtle.

I don’t usually use that super-sharpening but for this photo it seemed to work well. The method also works well on more texture based photos.

In my original I used my vintage colour technique to adjust the colours further and tuned the opacity of layers to get it looking good to my eyes.

Hopefully this tutorial has given you a couple of new techniques to try out on your own photos. If you’ve used these techniques on one of your photos, I’d love to see what you have come up with – feel free to send me the link in the comments.