Blogging Wi’ Shiny Pictchurs!
Ah love takin’ pictures and posin’ fer some too but y’see it takes a wee bit o’ pokin’ aboot tae get the results ye want.
What we have here is a rare creature on this blog: a post about blogging itself. Not going too far, of course, it’s about something in particular I like to use on my blog (as do many others), namely images.
Images, pictures, portraits, doodles, sketches or representations - whichever type or flavour you pick it will generally enhance your posts. I have always been a very image-centric person - I like art, animation and video as well as creating websites and ‘graphics’. It was only natural for me, then, to add a lot of art and imagery to my blog when I began it, though it took a little while to get the kind of pictures I wanted.
Why use images?
Although bolding some of your sentences, adding headings, italics and using various ·lists as well as plenty of s p a c e s can give structure to your posts I find that it is images which can make me stop and really look at a post. I could insert a cheesy line about a picture saying a thousand words but, to be honest, I believe that the way I (and perhaps blogs in general) use them is more in line with adding pepper to soup - it enhances the ‘flavour’ of the surrounding text and makes it tastier to eat. Read. Devour?
Images can also be used as visual clues - when I write a post about my shaman I usually include a picture of him - same for my druid, rogue and hunter. A lot of people use this method if they have multiple authors so it’s no big leap to utilise it if you’re a frequent, chronic sufferer of the dreaded altitis. Using Aurik as a model I hope I can illustrate reasonably simply how I make shiny images for my blog assuming basic knowledge of image editing.

Before I begin I will note that this isnae a ‘how to make shiny images of awesome dhoom’, a guide on how to upload / capture pictures and it also definitely isnae the only way to do things. It’s simple few tips / short guide about how to make some half-decent, quick images for your blog - in my opinion, unless your blog is centred around the images you use then making them should not take longer than writing the post. *grin*
There are two types of images I will cover: Screenshots and WoWModelViewer based images. I don’t really like non-game images in my WoW blog so I’m afraid that’s all yer getting from me.
Tools
- Image Editor: Adobe Photoshop, Paintshop Pro, Gimp
- Wowmodelviewer
Personally, I usually use Photoshop as I’m most comfortable with it, however I know not everyone can get their hands on it so I poked around with Gimp for a bit (there is now a version for windows) and, although a little clunky, it does have most, if not all, of the editing facilities that I use in Photoshop for my blog images. If you’re willing to play about with it, it’s a vast improvement on many simple image editors.
ModelViewer Images
I use modelviewer extensively. Despite some hiccups with items (some do not exist in the basic version of the modelviewer, though they are able to be added by anyone who can use notepad!) it’s a very nice tool for getting the exact pose that you want.
You can download modelviewer here. Once you’ve downloaded it, open it up, and you can create almost any character with most gear in the game and pose them how you like. I grabbed a quick screenshot of Aurik to show how I make most of the random images for my posts. Try to make sure the model is as big as it can be - if you only want a head and shoulders then zoom right in - the bigger your original image the better.
First, crop the image down so that it is surrounded by whichever background colour you chose in modelviewer (using the crop tool or selecting an area and going to Image -> Crop), then use the selection wand (I use it at around tolerance 12, rather than the preset 32 and uncheck the ‘contiguous’ box) to select all of the blue areas (or whichever colour you choose to use - pick one not found on your character).
You could simply stop here and skip forward to the second last step but the next few steps are the ones which I use to give the basic models a little bit of a softer, brighter look.
The next stage is simply to copy the layer (right click on the layer in photoshop) and blur the picture by about 2-4 pixels.
Once this is done, copy the blurred layer and stick this new copy underneath the original base layer. Looks a bit messy at the moment but the wonder of layer styles will save us!
Set the original blurred layer layer style to Hard Light (sometimes Soft Light will look better, it depends on the colours involved and how bright a picture you’re aiming for). Of course, now Aurik looks like he decided to sit in a gnomish nuclear reactor a bit too long…
This is quickly rectified by toning down the layer we have styled by putting the opacity to anywhere between 50-70%. I tend to find lower than this is barely different from the base and higher is a bit too day-glo in style - again though, this depends on the look you want to create. If you were going for a mystical, glowing night elf, for example, you might want to have increased the blur (on the first blurred layer only) a bit more, use soft light and leave the opacity high.


The reason we took the image as large as we could, earlier, is so that we could resize it. As you can see above, resizing an image drastically reduces the quality however we’re actually going to use it as a tool. When you resize something in photoshop it will end up smoothing out all of the sharp pixels. Although the second image here looks terribly blurry I should point out that it’s doubly blurry since I scaled it back up to show the basic effect of resizing. As you will see in the end image, the actual picture doesn’t look quite that blurry.
Last thing to do is to un-check the background visibility. This will mean no backgrounds so that if your site changes colour (new style, for example) or when viewers read your site in a feed reader, for example, the images will not look out of place. When I started doing the images for /hug I simply put the character, resized, on a background the colour of my site in jpg format. When I looked at this in a feed reader I realised how utterly terrible it looked and started using transparent-background pngs.
Ta-Da! *ahem* If you click this image you’ll see a comparison between Aurik half with and without the blur / glow effects. It’s a minimal difference but I think it makes hair and some other details look better without spending a lot of time on them (around 1-3 minutes per image depending on complexity).
Shiny Stuff
The thing about wow is, though, that the developers like shiny stuff as much as we all do. Shiny stuff is not so easy to auto-select, though and sometimes comes with see-through bits (which show that horribly bright background colour you chose rather too well. A simple way to deal with this is to turn them off in wowmodelviewer (View -> Model Control -> Select parts from the dropdown box and uncheck ‘particles’) and then add your own glow but you can also use the select tool to remove the ‘excess’ glow and blur the rest.
For example:
Soft light, opacity 85%, self-added glows using a very soft brush (and very quickly… usually I just remove all of the glows and leave it at that…) Again, clicking this will show a comparison with an image with and without the glow / blur effects.
Some of the best image editing based on modelviewer I’ve come across is created and shown off at the wonderful Flinthammer Hall and Musings from an Angry Bansidhe. Puts my stuff to shame!
Screenshots
I love screenshots (you may have noticed if you’ve been kind enough to hang about and read my blog for a while) and usually I do very little post-processing on them as, for me, it takes a little of the ’spirit’ away from a screenshot to go away and play with it too much - like taking a holiday snapshot and then coming home and editing in some sunshine. The essence of some screenshots is simply that you noticed a moment of beauty and tried to capture it. Some are carefully set up, others are random pics that turn out nicer than you could have hoped.
In fact, post-processing can leave you with damned ugly screenshots like this one. I hate it, but I was on my laptop, with settings turned down and I only grabbed one or two shots before hurriedly running here to squee about the dragons. It has been sharpened to help show detail but lost a lot of quality in doing so. Even blurring / relighting can’t save screenshots like this!
Tips for Cleaner Screenshots:
- Make sure you have your graphics settings as high as your computer can handle
- Turn off player / guild / pet / the kitchen sink’s name tags
- Unless the picture is of you, be zoomed into first person view
- Check the light and make sure you or the object you’re taking the screen of doesn’t have it’s back to it - unless, of course, this is the effect you want or it’s an immovable object
- Try mounting up or getting somewhere high if a ground viewpoint doesn’t show all of what you’d like to capture
I know these are mostly obvious but I am not a fan of screenshots with player names (with the exception of, say, bosskills or those where the names illustrate something) and scenes blocked by the inclusion of the person taking the shot (though I’m horribly lazy and do this myself sometimes)! *ahem* Of course, this is all an issue of taste and if you think those do look better then you should just do what you like the look of best.
So where do my images go?
An especially pertinent issue on my mind at the moment as, some of you many have noticed, the newer versions of wordpress mess horribly with image positioning in feed readers - with side aligned pictures being booted underneath text and suchlike due to ‘compliance with web standards’ apparently. Damn standards! *ahem* There are several places your images can go and it all depends on the type, shape, size, context, etc. of the image and your post. Multiple pictures of the same size can be aligned in the centre together to make a neat ‘banner’ with the advantage that each is separately linkable.
Banners, large images and screenshots are usually best if centred above the text they represent. Smaller images - especially full / half body character pictures which are much taller than they are wide work well aligned left or right.
I prefer to mix and match the images but sometimes you have a post full of banners or smaller images it’s hard to vary them.
‘Less is more’ sometimes. Images are usually best if spread out and limited in number. They are best placed to break up big blocks of text - though not if they are incongruous to said text. Of course a tutorial type post, like above, breaks this convention horribly. Bugger.
It can be a little hard to post something like this without knowing exactly what people might want to knw so poke me in the comments if there’s something that needs greater clarity or if you think I’ve missed a vital step - I just felt that a big part of my blog was these images and I wanted to get something out there about how I use images and why.
10 Comments so far
Leave a reply














\o/ yay for a post about blogging… yer actually kinda good at it! :P
I like the idea of the transparent background… should try that out at some point. I just wish they’d make a mac version of WMV as the one that’s available is suckage in comparison. *sigh*
There’s a few issues with transparent PNGs.
1. They don’t work in IE 6. Instead, they show up with a mid-grey background. This is because Microsoft couldn’t be bothered to implement the standard properly. Whilst I personally feel that people still using IE 6 should be shot for crimes against mankind (or at least web-designerkind) there’s still a depressingly large number of them.
2. They’re big. PNG files are lossless, which is why they look so good, but that also makes them big. You can generally make a JPEG with a fixed background that’s an order of magnitude smaller.
Yes, you have to change the image when you change your site’s background colours and yes it doesn’t look fantastic in a feed reader.
That’s really the trade off you have to make: do you want people to view your page quickly, or do you want the people with feed readers to have images that fit nicely into the background? :P
The *ideal* way to fix this would be to use SVG. Save the colours from an image as a JPEG so you get good compression, and then define a mask using either a greyscale PNG or a hard-edged shape. That’s not really an option because despite being a seven year old standard, Microsoft STILL refuses to support it.
A few other points:
* I really loathe the GIMP’s interface, but one awesome thing it has is a “select colour” tool which will select a specific colour from everywhere in an image. I use fluorescent pink for my model viewer background, making it easy to blast away the background in a click and a keypress.
* For compositing, I use Inkscape (http://www.inkscape.org/). It’s a vector image editor, which makes it much easier to edit images at any size, and then output them as fixed-size PNGs later on. If I later decide that a particular shadow isn’t quite right, I can select it and re-adjust it without having to re-create it.
I played around with something similar to your “blurred copy behind” trick before. I felt it made things look too blurry, and so I just ended up making sure I always had a VERY high resolution source image, and then scaled it down at least twice.
The only suggestion I can think to make on your actual images would be to drop a shadow underneath characters that stand there, and possibly put a border on edges where you cut parts of a character off.
1. Thanks!, I hadn’t realised that I hadn’t implemented the ‘fix’ for that on my new style! Don’t know how that slipped through - though, you’re right, I should add a link to the fix if I’m going to advocate the use of pngs. I’ll see if I can find the right one - it can take a few seconds but the grey should not show when the page is finally loaded in i.e. 6
2. It’s not just a matter of them ‘fitting into the background’ it’s a matter of them looking like I want them and, I have to admit, yes I’d rather the page was a little slower loading than it looked crap in a feed reader or if I decide to change my theme. My other half disagrees with me on this as well, and I know it’s a very valid criticism for web-pages and especially blogs but I do prefer this way.
I have tried inkscape - but, like most vector editors, I find it oddly uncomfortable to use. It was a long while ago, though, so I might give it another shot.
The blurred effect really is a personal taste thing, I know not everyone is fond of it :)
Drop shadows I’m not a fan of as they often make things look more flat, I have tried adding a shadow at the feet but I find it takes a lot longer and never looks as good as I’d like (lack of inherent skill on my part :() though I might give it another shot sometime.
Borders I did try and they look terrible on the images which are cut off. The cut-offedness annoys me a bit, too, but borders just make it very seperate from the rest of the content which I also don’t like.
Thanks for the fantastic critique :) <3 your comments.
Awesome post! I think I might use modelviewer to grab a bunch of different expressions and poses to include in my blog (once I get it recombobulated) It really does at flavor.
I’ve used GIMP (Windows) for a few years now, and I love it. You’re right in that it doesn’t have every single feature and tool that Photoshop does, but for the armchair image editor like me, GIMP provides more than enough functionality. Plus, it’s FREE!!
Itsnoteasy makes good points, especially about the file size of PNGs. I’m stubborn and refuse to switch to something that is pre-IE7, but PNGs can be rather large. One way of reducing the file size of an image when large file sizes can result in a large loss of space or even bandwidth is through posterization. You can typically reduce a file by half or more with some savvy effort.
GIMP is also awesome because you can typically specify what the transparent background is when making a PNG, allowing people with IE6 to see the preferred background color (hopefully to match your site!) without having that grey slate color instead. I haven’t found how to replicate that on Photoshop yet, but I haven’t put much effort into it.
Thanks for the kind words. I always like to see techniques other people use because different things work for different folks! I’m not doing anything you aren’t already doing yourself other than cutting out wowmodelviewer characters and putting a background behind ‘em. I couldn’t get the mapviewer to work for a long time and once I got a new computer, boy, did it help me out. I didn’t have to levitate or get on a mount or do stupid flips to get juuust the right angle. Oh, yeah, I do shade the characters. I don’t know how to explain it for GIMP, but in photoshop, I create a layer over the characters and set it to multiply. With a gray color brush (you can use other colors, experiment), I just add shadows to the sides, under the neck, wherever the light is not falling to help them blend in with their environment and to keep them from looking so flat.
It’s a lot of work, really. I can spend hours working on a story that took my husband 10 minutes to write. That’s OK. I enjoy it when I get into it. Now to find the time to add pictures to the older stories! Oy. With WoLTK coming out soon, that’s a tall order. :D
Keep up the good work with screenshots. It’s that extra flavor of why I keep coming back and reading your blog. It’s just a very warm and enjoyable place to visit.
Einz: Thank you :) GIMP is, indeed, a powerful tool and I really would love to use it more (free + open source is a big plus in my books) if they just sorted out the interface!
Cynra: Posterisation, whilst giving a cool effect is sort of the opposite of what I like for my blog - It often ends up giving big plains of colour and is less smooth / soft focus.
In photoshop the png background can be set when using the save-for-web dialogue but it’s a bit clunky - there may be a better way I’ve not found yet. That would be a nice compromise if people aren’t happy / confidant about adding the i.e. ‘fix’, thanks for the suggestion! ^^
Talking of better ways, I was prodded into the realisation that the wand tool in Photoshop also happily selects all of one colour if it is not set to be contiguous so I’ve edited to reflect that - it’s even faster that way, too. /bonk
Beli: I really do love Flinthammer Hall and the illustrations you make are wonderful - you really make the scenes and characters come to life! Thanks, once again, for your kind words - I’m glad you enjoy this little corner of the blogosphere <3
/hug
Yay Aurik for a great post /hug! And to all the commentators with their great advice /more hugs! I was just today wondering how to create images with no backgrounds without having to spend hours in Photoshop (I find it hard enough to make time to write posts >.>) since I’m quite the noob, and will definitely be downloading the model viewer and giving these techniques a shot.
I also agree that anyone still using IE6 needs to be charged with crimes against humanity. Especially with so many other great browser options out there. The need to spend hours adding work arounds and hacks to make things look ok in IE6 drives the designers where I work bonkers heh.
Hi Aurik ! Nice post, thanks for these infos. :)
I would like to know how you host your pictures ?
Thanks again.
Sellia: I use a self-hosted wordpress and so host my images on my own webspace. If you’re looking for somewhere to store and load images from I can highly recommend photobucket.com, though, as I used it prior to getting my current web-space and found it to be fast, reasonably reliable and, most importantly, free ;P
/hug