It seems almost incredulous that it could be as bad as it is, but a simple H.264 stream inside a flash container (ie, no fancy extra stuff, just video in a box) is a painful hog in OS X. It's not just "resource intensive", it's outrageously dog slow for seemingly no reason. In the short term, I suspect that Apple just doesn't care all that much. Long term, there isn't any particular reason why HTML5, which offers vector objects and bitmap canvases with javascript control, should be markedly slower than Flash, which offers vector objects and bitmap canvases with Actionscript control.
ADOBE FLASH CS5 GAME MAC
I suspect that their intentions for HTML5 basically boil down to "Achieve broad enough adoption for video purposes that, for any random video website our customers go to, they'll get a lump of h.246 for our hardware decoder and a couple of vector widgets, rather than a 'you don't have flash, so sad' embed box." and "Achieve performance decent enough that, if web designers and their idiot customers simply have to have their fancy flash-based menu effects, they can implement them in HTML5 and not break the experience for iPod users."(and, presumably, in the not so distant future, Mac users). By virtue of their market share(and their customers' willingness to buy widgets), the "If you want performance, make an App and shut yer trap." argument has worked pretty well for them. Flash, by contrast, is however Adobe wants it to be.įurther, I suspect that Apple doesn't really need Flash-level performance out of HTML5. I would suspect that Apple considers HTML5 to be "better", regardless of what benchmarks say today because they have the power to improve it, subject only to the limitations of their engineering resources and any fundamental defects in the spec(which, because the process is at least moderately open and consensus based, and one where Apple has a fair seat at the table, they have some hope of ironing out).
We will steal (with HTML5 and SVG and Canvas) part of these value, to make the web AWESOME. I serve no one to ignore that Flash add value to the web. But with a better model, these will be more easy to control, limit, optimize.Īnd people want these Flash features. Animated banners in HTML5 are not better than in Flash.
Some people will argue that "Flash-like" features in the web are bad news. Needing Flash is *mucho* wrong, and we DO NOT WANT. So Flash will not be required for some things. But there will be a "leak" of the good features of Flash into the web, so the web will get whatever good we have learn from Flash. I don't think Flash game dev's will move to HTML5 in 5 or 8 years.
ADOBE FLASH CS5 GAME UPDATE
Updating the browser will update the rendering of such canvas thing, or svg thing. Any optimization on the memory handling will affect all. Having everything following the document model (dom), any optimization made will touch all. And is not that good either, Linux users have bad experience with Flash banners that take the 100% of the CPU. A greasemonkey script is fun, google page rank search engine is fun.Įven if Flash is fast, a what price?, you have to support a separate things, with his own memory management and probably bugs. The points is a document model, that is easier to examine by bots and archivers, that can be modified by external tools, that can be linked, and all the good and cool features we have learn a Hyper Text have.Ī binary stream of bits that render vectorial stuff is not fun, because you can't do much with these bits. Scripting is as bad if not worst, than a binary stream vectorial format. And maybe it will still be a bit slower than Flash. It will take years to optimize HTML5 to something comparable to Flash.