I'm looking to use CSS gradients for an element of my web app that is drawn repetitively, as it would allow me more flexibility in dynamically changing its appearance. However, my question is whether gradients are more expensive for the browser to render than bitmap images. Or do images use more processing power? (And I'm not concerned with cross-browser compatibility- the app will only be used on the iPad)
According to an article on the Webkit Wiki, images perform better:
Sometimes it's tempting to use webkit's drawing features, like -webkit-gradient, when it's not actually necessary - maintaining images and dealing with Photoshop and drawing tools can be a hassle. However, using CSS for those tasks moves that hassle from the designer's computer to the target's CPU. Gradients, shadows, and other decorations in CSS should be used only when necessary (e.g. when the shape is dynamic based on the content) - otherwise, static images are always faster. On very low-end platforms, it's even advised to use static images for some of the text if possible.
Source: https://trac.webkit.org/wiki/QtWebKitGraphics#Usestaticimages
Of course, you have to balance that CPU time with the extra time it would take to load the image from the server. Also, for Internet Explorer, filters are extremely slow, especially if you have many on one page.
This above answer is directly lifted from How does the performance of using background-gradients in CSS vs using images?
These links can be useful read for you
Browser Repaint/Reflow performance: using CSS3 Gradients vs PNG Gradients
CSS gradients are faster than SVG backgrounds
http://leaverou.me/2011/08/css-gradients-are-much-faster-than-svg/
Runtime Performance with CSS3 vs Images
http://jacwright.com/476/runtime-performance-with-css3-vs-images/
Related
Use of gradient images is very common among developers for styling a page. Gradient images are used in many places from styling the navigation bar to styling the background. Technique like repeating a small image in either direction is also common.
One more way to style and give this effect is by using multiple div's one below another with different color, the latter being a little lighter than the former. In the most simple scenario doing so would include only a small script. So, no problem of writing a lot of markup, just some simple code.
The only concern that remains is speed and performance.
Speed
The script, more precisely the function would be much shorter in size than a image. So, in matters of speed the latter method seems more good.
Performance
Today's browsers are very powerful, so the difference between displaying an image and executing a function is negligible.
Css management
Obviously, problems like positioning would be another concern but we do struggle with such problems in every day life. The problem is no greater than overlapping two div's and setting their z-index. The whole gradient div's can lie inside one parent div.
So having addressed the issues of performance and speed isn't using Gradient div's a much better approach than using images?
It's an alternate approach, yes. But not a good one. You get zero points for:
Maintainability
Scalability
SEO
Separation of presentation from content
Furthermore, to say that we needn't worry about performance since today's browsers are more powerful is a gross assumption.
Actually, I think the second option you describe (creating multiple divs with atering colour) is downright terrible.
You're altering markup for the sake of styling. That's a no-go.
It's a common thing among users to disable JavaScript. What happens then?
As you said yourself, positioning mayhem.
When it comes to performance, I would be more cautious than to state it's no longer an issue. Especially with the dawn of mobile browsers in mind.
Such styling would be harder to understand and maintain. Particularly when your team changes someday.
Also, there are two other ways to implement gradients.
CSS gradients - limited to simple variants and requiring a lot of CSS to provide decent cross-browser capabilities. You can try this generator get a taste of these: http://www.colorzilla.com/gradient-editor/
SVG backgrounds. These allow you to create just any gradient you wish. You can use an svg file in your CSS just like any other image. However, some browsers don't support this feature. Here's a table showing when it's an option
Using images is the most reliable option, while combining SVG with normal images (for these browsers that don't support SVG) seems the most flexible approach.
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.
Ok I want to know what is the best practice for performance regarding CSS background images and http requests.
1. Use many different 1px png background images resulting in several individual http requests
OR
2. Use one large image sprite with big gradient block chunks for use as background image. This will increase file size but save on http requests.
Love to hear you opinions...
I think it would be better to use data:uri technique for small images (like 1px-backgrounds).
background: url(data:image/png;base64,....) top left repeat-x;
It works for all modern browsers. For old IE browsers (like IE6, IE7) you can overwrite styles by conditional comments.
background: url("path/to/background.png") top left repeat-x;
Of course this way you have to re-encode background, if it has changed. But it saves a lot of requests.
If you're talking about using these images for gradients, I'd suggest using CSS gradients, then you won't need any images at all.
Of course, the obvious problem with CSS gradients is that it isn't supported by older versions of IE. The good news is that there is a fix for IE called CSS3Pie that allows it to support the standard CSS gradients feature (along with several other useful CSS features.
No more images required; just one HTC file (which only gets downloaded by IE).
saving http requests is always the better solution. But nevertheless you should watch the file size, so that it does not get to big. Then you should consider making two images or more.
Look at the following tool which allows the easy creation of sprite images from unsprited images:
http://spriteme.org/
I'm working on a site that uses images diagonally aligned. The images are fairly simple and have some text inside so I can achieve the effect using either images or the CSS3 transform: rotate and border-radius properties, granted, with all the proprietary extensions.
However I'm wondering if there are significant advantages for using one over the other. I'm thinking maybe the CSS-only alternative would load faster, but a downside would be the use of more code for each image. Although using images I'd still have to do a lot of positioning for each element. Also, the CSS in the first case would not validate using browser-specific extensions, how important is this if I still make sure the site is correctly displayed in most browsers?
Is there a general approach for using either one of these options? What would you recommend?
The use of CSS3, versus images, is best answered by your need to support Internet Explorer (or other older, or CSS3-unaware, browsers): if a large portion of your user-base browses with IE then you should use images.
If you're only interested in the later, more standards-compliant, browsers (possibly including IE9, but I've no experience with it as yet) then CSS 3 is likely the better option, since it might involve a larger CSS file, but it does allow you to switch your layout more easily in future, without having to create/recreate a new set of images for the new design.
You could, of course, combine both approaches: use CSS 3 for the compliant browsers, but include an IE-specific stylesheet, with conditional comments, to supply the images as, perhaps, background-images to build up the design. This is, of course, likely to be a lot of work, though.
I would honestly recommend an image for this. CSS3 is cool, but browsers that don't support it might break your layout. Also (possibly more importantly), text really needs proper anti-aliasing to be readable when rotated or scaled, and you can get finer control over that with an image.
Two things I think you should consider here:
Images are a killer for those on mobile devices. So if you are targeting those with mobile or expect a large mobile audience, you may want to reconsider using a lot of images (and having your users pay a ton in bandwidth).
The text in your images (if you use images) will not be searchable by search engines and will not be accessible for those with accessibility issues, unless you are good at filling in your alt tags. ;-)
I've been reading a lot of ways recently of how to avoid having to use Photoshop for things like gradients and shadows on buttons, when you can use CSS3 Box Shadow for such features. Now this is great, but obviously legacy browsers and most IE browsers don't yet implement CSS3 features, so my question is, why save yourself extra work in Photoshop when you can use CSS3 but then HAVE to use Photoshop for other browsers to see the desired effects? Isn't that just extra work?
you can use CSS3 but then HAVE to use Photoshop for other browsers to see the desired effects?
Often you don't really care whether the other browsers see the desired effects. If it's just a subtle background gradient effect, the site's going to work fine and still look reasonable with a flat background in downlevel browsers.
Compatibility was more of a problem when CSS 2 Positioning was at this stage: an unsupported positioning feature could lead to an unusable layout. But for essentially frivolous effects like backgrounds, borders and shadows, it's not critical if the effect goes missing.
There are a few reasons. From a developer/designers perspective, using CSS3 can be more flexible. An image is more or less just a static image. With CSS you can use script or CSS to animate or adjust the styling, and have advanced integration between the various technologies. This allows you to be able to do some things that are not possible or difficult just using images.
The page weight is also lower usually when using CSS, which allows the page to load faster, especially on slow connections or mobile devices. It also reduces HTTP requests (although you can use data uris too) which reduces latency and thus speeds up page loading.
Using CSS is often more maintainable and quicker, as you can adjust say a background colour or a shadow quickly just by changing the CSS, while with images you have to open up photoshop, edit the file, resave, etc. This is fantastic for quick tweaks.
There are also some people (myself included) that are much more comfortable in code than they are using graphics packages. For me it is easier to use CSS, but your milage may vary.
Also from the browser perspective we want to move the web platform forward and increase the capabilities. SVG is an example of another technology that doesn't work across all browser versions yet, but adds a lot of power. Although it may not be possible to deploy sites using certain CSS3 properties or SVG without fallbacks right now (depending on your audience or client) there will be a time in the future when old browsers are dead and what we are adding today works for all users. There was a time when most of CSS2.1 wasn't supported for example, while now you can rely on most of it if you don't need to design for IE6.
There are some parts of CSS3 that you can happily use without fallbacks, if you have a client that doesn't require all browsers to look the same. border-radius is an example, or giving a solid colour fallback to old browsers and applying transparency for newer browsers.
The world is moving forward, you can't ignore it. as long as old browser will show a button and new browser will show a more nice one, you are ok.
Otherwise, we were still be working on VT100... as why using "Windows" if most monitors doesn't support more than 4 color graphics and has 80X22 resolution....
If your bosses accepts the idea of having two versions of looks and feel and a warning. CSS3 is a good idea.
So, you use CSS3 features in your code and forget about the browsers that are not compatible -- they will see the "low-graphics version", you might want to issue a non-intrusive warning (like the one we get as alerts in StackOverflow), saying, "Your browser is not compatible with CSS3. To see this website in high graphics please upgrade your browser. Click the link here to update now."
But, depends on where you work. Big companies with lot of resources may like to stick with Photoshop and smaller once may like a light-weight CSS3 code that degrades silently in non-compatible browser.
Well you can always use ChromeFrame, to impose nice behavior on IE. Assuming that a lot of products will insist for it the near future (yeah we can argue against this), chances are that most IE users will have chromeFrame even before your application would insist for it. Once you take a decision on this, decision on using CSS3 properties or HTML5 elements is - well - hell yeah!!
Marvin