In what scenarios CSS sprites would be good to use and when not? Is it time saver or only server request saver?
I think bandwidth use will be same because image is same like ong big image but it's in a parts?
When and where use of css sprite is a time saving(in work) option ?
For navigation it's good for rollover pre-loading effect but not good for images disabled people?
What are other good usage which can save our time once and in future (if changes comes in design) both?
Edit: Sprites is only for css background so should we use images in background as much as possible to save sever request, is it good idea?
Update:
To implement takes more time then regular method and mostly client do not much worry about some slowness like http request. My question is can we save time in site making process and future maintenance of website using css sprite. or regular method is enough.
In nutshell my question is: “can CSS sprites save our designer and xhtml css coder time (I'm not talking about server request)?”
It reduces the number of HTTP requests which will enhance site performance.
CSS Sprites are the preferred method
for reducing the number of image
requests. Combine your background
images into a single image and use the
CSS background-image and
background-position properties to
display the desired image segment.
In Minimize HTTP Requests
CSS sprites are a time saver because it is a server request saver, as server requests are notably time-consuming. Using CSS sprites usually decreases your webpages' load/render time dramatically. There are times when they cannot be used, such as with background images repeating in two dimensions, but when you can use them, it's almost always worth the effort.
Of course you shouldn't sprite groups of images that are very big, especially if they're not very likely to be shown. Don't sprite an entire photo gallery into one big image, for instance =)
Other measures which amount to pretty much the same thing would be minifying, compressing and combining your scripts and styles into only one js file and one css file.
EDIT
With regards to your clarification, i'd say no, CSS sprites will always mean more work, never less, compared to just using the separate images as they are. I still wholeheartedly endorse their use, tho =)
CSS sprites are best used for elements that have a fixed width and height. Otherwise, you need large empty spaces in your sprite image, which can (depending on file type) increase the size a bit.
Due to the way different file formats compress images, sometimes a CSS sprite image will have a noticeably smaller file size than the total file size of separate images. That’s a nice bonus.
As mentioned, sprites reduce the HTTP request overhead, which can help load time. I’ve never seen any numbers on the magnitude of this effect.
Sprites will add a bit of time for your CSS developers. It shouldn’t affect your designers. But bear in mind that your developers code the site up once; the benefits of sprites apply every time someone looks at the site.
It will only reduce the number of requests but that will benefit both the server and the client. The server will not need to handle as many requests. The client, because it is limited in the number of parallel requests that it can make, will render faster as many of it's previous "requests" for the image will be served from its cache, allowing it to make the requests that it does need more expediently.
Using sprites reduces the number of requests and thus also the network overhead. Loading a few sprite image is faster and uses less bandwidth even if the image data is the same (or even a bit more) than the individual images.
It needs a bit more work and some planning to combine the individual images into sprite collection images, so the development time is somewhat longer. The difference is less if you have it in the plan to start with rather than combining the separate images afterwards.
Any scenario where you have several same size background images that replace each other (or complement each other) is ideal for sprites.
As long as you have something like a dynamic photoshop PSD file in the back, then the designer's maintenance won't be an issue. But if it is a static file like PNG/GIF, then maintenance will take more time as you cannot control the individual images separately anymore.
Overall, sprites is a great idea. Use it for fixed width & height images that are less likely to be updated frequently.
Sprites are always good to use. They help speed up the loading of web pages and prevent the blinking effect on navigation hovering.
Related
I been working on rollover images with pure css and i am aware of 2 solutions to achieve the desired result. I wonder if there are any benefits which i am unaware of with those methods and if there is a preferred method or is it just a preference (I just want to make sure before i finish a giant project so there are no surprises later)
first method
The first method i run across is to save multiple images as one image.
then to set the image as the background of a divider (that is smaller then the whole image)
then on :hover you set background position to negative and the other images appear
Some cons and pros i thought of is that:
pros
you do not need to preload the images manually (by placing code at the end of css file)
It might be easier to organize your image files as there could possibly be fewer of them
cons
it can be difficult to know how much exactly move the image especially if you have seperate team designing the graphics
second method
The second method i run across is to set an image background of a divider and then to change the background url on :hover
Some cons and pros that i can think of on top of my head:
pros
possibly less time in development
can exchange rollover images with more ease and dont need to specify the size of the divider
cons
need to preload images at the end of css file
possibly more mess in your image folder if you have a lot of files
Is there any pros or cons that i am missing? Is one method generally preferred over the other and why? Or is it personal choice and there is no problem with using either one??
Update:
Here is a link to simple preload for the second method with pure css. I thought there was an even simpler way (of just placing the links) but i cannot find it. Still this is pretty basic.
Update(2):
I have learned that sprites have additional
benefits of making websites faster by reducing the number of http requests.
and benefit of rollover being as fast as possible with no flicker
And a blunder of making a less accessible site, and would limit the potential benefits of the title and alt attributes in the HTML. (if used incorrectly for all images instead of just decorations and buttons)
CSS sprites (the first method) is generally favored due to the fact that pre-loading isn't necessary to maintain a quality user experience.
the ALA link Wex posted is a good primer on the benefits of sprites, and Smashing Magazine recently published an article which addresses your concerns about background-position when using a sprite with many images
http://coding.smashingmagazine.com/2012/04/11/css-sprites-revisited/
The first method is much better than the second one. You don't always have to bundle up every image into one giant image - that usually makes it hard to add new images - but it does make sense to put every image and their hover state in the same image so you don't see a flicker on hover, and so you don't have to rely on some preloading script.
Here's a good article that talks about the benefits of using CSS sprites: http://www.alistapart.com/articles/sprites
Will an imageless design (using CSS to generate icons) load faster than a traditional image based design?
NOTE: I completely understand the flexibility of being able to change an icon's colour, size etc. will a few changes to the CSS file is much more efficient, but I am unsure about the loading time.
Background:
I am looking into completing a project for a client that is an imageless design using CSS to generate the page icons, which was a project requirement to decrease loading time on mobile devices. Normally, I would just slice out the icons in an image sprite and not worry about the load time seeing the browser would only have to load and cache on image.
Thanks!
Yes, if site icons can be generated entirely in (relatively efficient) CSS, this will load faster than images. The rub is that most icons are far too complex to be generated using CSS, and the CSS might in some edge cases become so complex that the maintenance issues associated with maintaining complex code across browsers might outweigh the benefits. But in general, if you can do it in CSS without having to bend over backwards, it would be optimisation—and in the scenario you describe might well be worthwhile optimisation.
I don't think the question is whether or not it's faster, but whether or not it's a good idea at all. I wouldn't sacrifice the flexibility of using images by making everything into pure CSS. You guarantee cross-browser compatibility with images, something that CSS3 cannot. Plus images are easier to maintain.
Think it in this way .. the size of css icon(the size of the code required to produce an icon) is an advantage and the other major advantage is the server hits. for ever image icon there is a hit which means there is a waiting time, downloading time, and rendering time, though its only in milliseconds the waiting time and downloading time are directly proportional to your server load.
In that case a css3 icon will not only save time and band-width but also reduce your server hits as the css file downloads only once and its contains all the style declarations for the entire site.
The advantages are enormous to choose a css icon. But there are some cases where a css icon is not suffecient, so its up to the developer to choose from a css icon or an image based on the requirements.
The typical rule of thumb when it comes to using CSS sprites for images is that you should only do it for smaller images (like icons) and that actual image content should always be represented through <img> elements. My question is: why? Aren't the advantages of spriting worthwhile for content images as well?
One reason I have read is to enable the use of alt text, to be more syntactically correct and accessible to screen-readers. However, when that is a concern, couldn't you just as easily use a single tiny transparent image with all the syntactical sugar atop a sprite that presents the real visual content?
You could, but:
Content images don’t tend to be re-used as much as UI-type images like icons. Imagine a newspaper site: if every content image they used in every story was part of a sprite, that sprite would very quickly get huge, and would be downloaded even by users who only looked at one story.
Website content is generally expected to be maintained by people who don’t know CSS. It’s a bit unreasonable to expect content editors to edit a master sprite image file, re-save out to a JPG, and pop in some CSS just to put an image on a page.
If you sprite a lot of large image files, the sprite file will get really large. It might take an unacceptably long time to download when the user first visits the page, or it might use up too much bandwidth on users who only end up seeing one of the images within the sprite.
Obviously, those are generalisations — in a specific situation, it might make perfect sense to sprite larger/more content-y images.
On using an <img> tag with a tiny transparent image file for sprites, you can do that for any sprite images — it’s often useful for clipping and positioning the sprite image beyond what background-position allows.
One additional reason I can think of, is search-engines; if you have an image with a descriptive alt-tag or a figure element with a figcaption, search engines will be able to find, classify and show it.
Sprites are usually usefull for static contents (images that are common on many pages).
Content pictures often appears in only one page, so you can't add it to your sprite file.
If you want to have real-time sprite generation, making custom sprite file with all your pictures, I think the generation cost is far more expensive than the duplicated HTTP calls it saves.
Sprites are here to save HTTP requests, but you should not waste server-side computation time to make it, nor put expensive and unusefull pictures in your sprite file.
Sprites should be used for common icons across the whole website and not for page specific content. When you use sprites for big images you get a lot of white space that is wasted.
According to http://blog.vlad1.com/2009/06/22/to-sprite-or-not-to-sprite/ this is also a problem:
Another (but much less important) downside is that if a sprite-using
page is zoomed using the full-page zoom feature that’s present in many
browsers, the browser may need to do extra work to get the correct
behaviour at the edges of these images — basically, to avoid adjacent
images in the sprite from “leaking in.” This isn’t a problem for
small images, but can be a performance hit for bigger ones.
From the research I've continued to do on this, another potential issue is memory consumption. Despite the fact that sprites may be compressed enough to download quickly, the footprint they fill in the client machine's memory is after the browser has rendered them, meaning it can be quite large, even for sprites with fairly small file sizes.
I haven't seen a definitive answer on whether or not this memory footprint is any larger than what one would see when loading the same number of images without sprites. I will be testing that in the coming weeks for the project that prompted this question, so I will return and update this answer once I have a conclusion.
Sprites are used to reduce the amount of requests to a server. The impact of lots of small requests is slowing the server more than one bigger request. But if the image of sprites, I like to call it sprite-map, is too large, it will also slow down the performance. The other thing of importance is like you sad the possibility for each single picture to give some name, to manage and to index by search engine.
1) First, should I order the images in my spritesheet a certain way?(like biggest to smallest images, or images that appear at the top of the page to bottom of the page?)
2) Say I have a css spritesheet that contains before and after images. Like the image shows a cow, but when the image is hovered, it shows a cow with wings. Is it in my best interest to not use css spritesheet for that then? Does the css load all the pictures in my spritesheet at once?
3) Is a spritesheet better in terms of caching? Unrelated, but what does it take for a browser to cache something? I mean if it's only after a single page view, perhaps it's not worth it.
4) Lastly, I want to start a forum. I don't know anything about forums yet, but I plan to start one soon. I'm thinking of just having like a default set of 40 images that people can only choose from as their avatars. Should I even make a spritesheet for those images (if it's even possible)?
I know this is a lot of questions, so please answer any that you have knowledge of. Thanks!
A 'spritesheet' is just one large image. So...
1) Doesn't matter.
2) Again, it's just one image. If not all users will want to activate the 'after' feature, then you can save them some bandwidth my making the afters a separate sprite. If most people will want to use the after feature, then you can save them bandwidth by making it all one sprite. (Though note if we're talking really large images, there will be a practical limit to how much you want to stick into one sprite. No one is going to wait to download 1mb file, for instance).
3) Again, a sprite is just an image. It has the same caching pros/cons as any image.
4) 40 hits on the server is a lot compared to 1 sprite. So, based solely on that, a sprite would be useful. But if it's rare that you'll get more than 10 or so of those avatars on one page, then the sprite would be a detriment, as it's loading such a large file.
Just on the ordering of images, I have a sprite file for a site I’m working on that contains various browser logos with version numbers added to them. As such, there’s quite a lot of repetition in visual information in the file.
I was quite surprised to find that the direction of the repetition could have a big effect over the file size I could achieve for the image when saving as a PNG. When I had similar logos in columns, the file came out at about 120 KB; whilst when I arranged them as rows, it came out at 41 KB.
Once the project’s live I’ll post the actual images. It’s probably quite rare to have such similar images repeated in a sprite file; normally your images would be different, or you’d just use the same image repeatedly. (Indeed, I might end up refactoring my sprites so that the varying bits are in their own file.) But I hadn’t realised that such supposedly similar images could be encoded in two files of such varying sizes, purely based on the geometric arrangement of the elements in the image file.
1) Not sure, but I don't think it would matter much, if at all.
2) The best way to do this is with CSS image rollovers.
3) Spritesheets would be better for caching, since it's only a single image, instead of the web server having to connect, send an image, disconnect, send again, send another image, disconnect, etc...
4) I would just use single images. There's really no reason to use a spritesheet in that situation.
How do you approach the use of image sprites in css?
Should I take all the images in my website and combine them to one image sprite? Is it really that beneficial?
How hard is it to maintain those images and change them later on?
Should I take all the images in my website and combine them to one image sprite?
Of course not. You're taking it too literal.
I find sprites are best used for groups of similar images. Examples include:
All states of a graphical button
States of icons
All permutations of a background (unless it needs to tile two ways)
Is it really that beneficial?
If you have a lot of them on a busy site, very. It saves a request for each image, saving the user time and your server a whole bunch of concurrent connections.
How hard is it to maintain those images and change them later on?
If you've used them logically, pretty simple. If you need to add another navigation item, you open up your nav sprite and expand it. For things like navigation it can actually be easier to maintain because you have like comparisons right next to you in the same document.
Edit, having seen one of the more extreme examples, I'll add that I would never go that far because:
It's 60k to download. Not huge but on slow connections, that's 60k that has to be downloaded before anything shows. If all your visual assets are tied up, it can make the load time seem longer.
Your CSS becomes a nonsensical mish-mash of background-position commands. If you do want to make changes you have to go back to the sprite and measure everything. Again and again and again.
God have mercy on your soul if you need to enlarge something in the top-left of the sprite. You'd probably just add a new sprite below the current ones.
And that might lead to bloat. Indeed, just loading all these images might be loading a whole lot of material that some users will never actually see. Loading unused data is probably worse than a connection overhead (considering how easily static content can be served by multiple cheap servers or a CDN)
The other examples are a lot more simple and worthwhile (IMO).
Sprites are a great way of cutting down load-time on graphics (sometimes), and always a way of cutting down requests to the server. Generally speaking, they may take some serious planning as you don't simply want to drop a bunch of images onto a canvas and export as a jpeg. I would suggest you study some sprites currently in use by larger companies like Amazon. Get an idea for how they layout their elements, and what types of images they even consider for use in sprites.
You'll also want to evaluate your site and be sure whether you can successfully implement them or not. If you weren't planning on using them to begin with, it may require a lot of back-tracking and updating to prepare for them.
Amazon Sprite
Ebay Sprite
Current.com Sprite (Whoa)
Google
Sprites work well when you’ve got an element with at least one fixed dimension (width or height), and you want it to have a different background image in different circumstances.
When I’ve tried it, I’ve found that sprite image files tend to be smaller than the total size of the individual images files they’re made from, so you can get bandwidth savings as well as the other two benefits:
fewer HTTP requests
no delay waiting for another image to download when an image state changes on hover
That does depend on the contents of the images though.
Personally, I wouldn’t put unrelated images together in one sprite image, as I think it makes maintenance too non-obvious. Also, as mentioned in To Sprite Or Not To Sprite, really big sprite images can use quite a bit of browser memory. (Whether this is actually a bad thing depends on the context.)
The idea is to avoid unnecessary HTTP requests. This is especially an issue if you have a lot of small icons (say, for a WYSIWYG editor like the one used on this site). If you have twenty 16x16 pixel icons, that won't amount to much bandwidth, but it will still mean twenty extra requests each time the page is loaded.
Other candidates for sprites are button states and anything that's purely decorative but part of the layout.
If you use roll-over background image changes, you'll also find that you'll either have to preload the roll-over state image (either with JS or with silly hardcoding) or you'll encounter some latency as the browser requests the previously unused image. Sprites can alleviate that.
Things you probably shouldn't be making sprites of are pictures that are NOT just graphical elements (e.g. graphs, illustrations, avatars, ads) or that will change a lot (e.g. avatars or ads).
It's not impossible to change sprites, but depending on how much thought you put into the arrangement of the sprite sheet, it may be very hard to do. There's nothing forcing you to make the sprite sheet ultra-condensed, but it's obviously better for the file size if there's not much unnecessary whitespace in it (see Google).
Note that the extra requests may not be a problem for you if you have a relatively low traffic site (which almost everybody has, unless you're Google or Amazon). Sprites may still improve performance for mobile devices, though, as it means less chances for errors and thus lower latency.