I've used the beta on my MBA and I have to say I'm pretty impressed. Around 20-30% CPU usage when watching a 1080p video. Sure, it could be better but since I very rarely even use Youtube, let alone 1080p videos, it's not that big concern for me.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Adobe's Flash Player 10.2 to Offer Up to Ten-Fold Improvement in CPU Efficiency
- Thread starter MacRumors
- Start date
- Sort by reaction score
Ok by HTML5 I meant something like Canvas + SVG, which HTML5 supports. Can I do that drawings and animation with jquerry? Is there an example? Could you please post it on that thread so that others can comment? Thanks.
This sounds great for certain situations where other elements have been overlaid on top of video, like when your playback controls arent hidden.
But Id love to see some acceleration for the REST (non-video) of Flash... to fix the way my Mac fans crank up simply due to a banner ad that has long since stopped moving. Static graphic... CPU overload. Something is amiss. And it cant just be that the ad creator carelessly looped some complex script that does nothing visible, because that would be rare--and the Flash CPU overload is the norm.
Until then, I have Flash off (I turn it on when neededmaybe once a week) and I use the YouTube5 Safari extension (which works with other sites too). Sometimes, when a site still says I need Flash for video, I spoof an iPad (one click in the Develop menu) and it plays.
Dont get me wrongIm a Flash developer by trade and its a great tool for developing certain things cheaplybut some of what Adobe shows just seems like too little too late. That COULD change in a big way, and I very much hope so! Meanwhile, a little progress is still nice.
But Id love to see some acceleration for the REST (non-video) of Flash... to fix the way my Mac fans crank up simply due to a banner ad that has long since stopped moving. Static graphic... CPU overload. Something is amiss. And it cant just be that the ad creator carelessly looped some complex script that does nothing visible, because that would be rare--and the Flash CPU overload is the norm.
Until then, I have Flash off (I turn it on when neededmaybe once a week) and I use the YouTube5 Safari extension (which works with other sites too). Sometimes, when a site still says I need Flash for video, I spoof an iPad (one click in the Develop menu) and it plays.
Dont get me wrongIm a Flash developer by trade and its a great tool for developing certain things cheaplybut some of what Adobe shows just seems like too little too late. That COULD change in a big way, and I very much hope so! Meanwhile, a little progress is still nice.
Wirelessly posted (Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_2_1 like Mac OS X; en-us) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8C148 Safari/6533.18.5)
Hah, so ironic. Only OS X crashes frequently because of Flash. Windows 7 hasn't crashed on me for that (or any other) reason in the year and a half I've had it. I like Apple products too, but this forum is hilarious.
Hah, so ironic. Only OS X crashes frequently because of Flash. Windows 7 hasn't crashed on me for that (or any other) reason in the year and a half I've had it. I like Apple products too, but this forum is hilarious.
Flash is out-dated, but not evil
All I hear is people saying it's "about time" that Adobe started optimizing Flash, and that they should have implemented these new optimizations "a long time ago."
By that reasoning, Mozilla and other browsers should have implemented JIT Javascript engines like Tracemonkey "a long time ago."
This is the flaw in your reasoning. For Flash's original niche (cute web animations and stuff you couldn't do with HTML and Javascript), Flash was JUST FINE. This is because the demands placed on it were low. It wasn't until Flash became the only fully cross-platform video player platform that it became a problem. Flash wasn't bad for video because it was evil. It was bad for video because it was not originally intended to be used so heavily.
Similarly, it's only recently that Javascript performance has become a major bottleneck. Prior to Google Apps, Javascript was used for relatively mundane stuff like making your buttons change color when you mouse over. Only when we started doing large-scale AJAX stuff did anyone's Javascript engine become insufficient. Once that happened, the browsers started adding sophisitcated just-in-time compilation to accelerate Javascript performance massively.
Adobe may be a bit SLOW on the uptake, and I've never been terribly happy with their tendency to ignore bug reports, but what's happening here isn't a matter of evil or even incompetence. Everyone is straining to adapt to accelerating changes to the way people use the Web. It's that simple. How people have used the web has changed RADICALLY in the past few years, with massive advancements in the kinds of things we do in browsers and therefore the amount of CPU demand. Back when HTML pages were lightweight, Javascript performance didn't matter, because the bottleneck was the Internet. Now, most of the work is done in the browser, making the JS engine the bottleneck. So we adapt. Adobe is struggling to keep up, and I firmly believe they needed Apple giving them the hell to get a move on, but their only sin is failure to be proactive, which is the same sin committed by all browser developers.
Because we have HTML5 as an alternative (a good thing, BTW), Adobe risks obsolescence. This is not evil or incompetence. It's lack of crystal ball. Now they know they're in danger of losing dominance, and they've started to adapt. Will it be too late? Possibly, possibly not.
All I hear is people saying it's "about time" that Adobe started optimizing Flash, and that they should have implemented these new optimizations "a long time ago."
By that reasoning, Mozilla and other browsers should have implemented JIT Javascript engines like Tracemonkey "a long time ago."
This is the flaw in your reasoning. For Flash's original niche (cute web animations and stuff you couldn't do with HTML and Javascript), Flash was JUST FINE. This is because the demands placed on it were low. It wasn't until Flash became the only fully cross-platform video player platform that it became a problem. Flash wasn't bad for video because it was evil. It was bad for video because it was not originally intended to be used so heavily.
Similarly, it's only recently that Javascript performance has become a major bottleneck. Prior to Google Apps, Javascript was used for relatively mundane stuff like making your buttons change color when you mouse over. Only when we started doing large-scale AJAX stuff did anyone's Javascript engine become insufficient. Once that happened, the browsers started adding sophisitcated just-in-time compilation to accelerate Javascript performance massively.
Adobe may be a bit SLOW on the uptake, and I've never been terribly happy with their tendency to ignore bug reports, but what's happening here isn't a matter of evil or even incompetence. Everyone is straining to adapt to accelerating changes to the way people use the Web. It's that simple. How people have used the web has changed RADICALLY in the past few years, with massive advancements in the kinds of things we do in browsers and therefore the amount of CPU demand. Back when HTML pages were lightweight, Javascript performance didn't matter, because the bottleneck was the Internet. Now, most of the work is done in the browser, making the JS engine the bottleneck. So we adapt. Adobe is struggling to keep up, and I firmly believe they needed Apple giving them the hell to get a move on, but their only sin is failure to be proactive, which is the same sin committed by all browser developers.
Because we have HTML5 as an alternative (a good thing, BTW), Adobe risks obsolescence. This is not evil or incompetence. It's lack of crystal ball. Now they know they're in danger of losing dominance, and they've started to adapt. Will it be too late? Possibly, possibly not.
Let me give you an answer to that. I care so much about my CPU cycle because I need my Laptop running on its battery for the whole day. I lost over 1:30 hour of battery life with flash enabled if I forgot to close any web page that contain Flash ads.
Once again, the real problem with flash is not the video, but the Actionscript interpreter, and no hardware acceleration can fix that beside recreating the flash platform from scratch. So the question is: Why spending time and effort with patch and fix on a turd. The HTML5 + CSS3 is already the evolutive response to Flash.
I still wonder how much people see Adobe has good guys and Apple has the bad one. All of those should do their homework and trace back the origin of FutureSplash and how Adobe took apart the competition (Macromedia) and killing everything they done beside Flash. Adobe only want to protect the only revenue coming from Macromedia aquision 10 years ago.
Just disconnect from the internet.
I know what you're trying to say, but I think a great deal of it was and still is incompetence. Even the flash development environment is unstable. It has been for years. If you strip out iPhone support from Flash CS5, the feature list is ominously short. Adobe reacted to the iPhone's success. Had they taken proactive approach and made game development tools specifically Phone/iPad, Apple would be singing their praises.
Adobe has not blazed a path like Macromedia did. I remember when Macromedia added video support to Flash - that drew the ire of Apple and even got them sued. But Adobe has not innovated in a while.
They react to the market instead of anticipating it.
- The latest Premiere was a reaction to Final Cut - their original reaction was to pull their product from the Mac when FCP came out.
- The latest 3d engine from Flash Max was wonderfully impressive, if it wasn't a reaction to Unity3d and other similar tools eating their cake.
- The optimizations to the Mac version would be great, had it not been a pain for so many years and now people are publicly flogging them for it.
- The iPhone exporter for Flash is a great idea, but it was a reaction to the phone's success and produces slow running apps.
Basically, I have to wait for some outside factor for Adobe to feel motivated to fix their products and move forward. Customer feedback doesn't help because the guy in India you get on the phone doesn't know what you're talking about. It's a bit like dealing with Autodesk.
Last edited:
So basically you say that an embed plugin decoding a file on the fly is theoretically faster than a native solution... Mmm interesting concept...
Last edited:
Ive had a few times where a Youtube video just becomes a black box with the message 'An error occured, please try again'.
But a simple refresh will fix this. Is Flash that bad on OS X? I haven't really used it enough to notice.
Good for them.
Wow this is really great. Hopefully this will eventually trickle down to their mobile implementation of Flash. Despite the poor performance of Flash on my Droid X I still like having it available when I need it.
It would be great if this somehow made it much more efficient.
Wow this is really great. Hopefully this will eventually trickle down to their mobile implementation of Flash. Despite the poor performance of Flash on my Droid X I still like having it available when I need it.
It would be great if this somehow made it much more efficient.
promising, but adobe like microsoft has a tendency to have carefully controlled demos/previews that fall well short of the shipping product. case in point flash 10.1 on mobile - when adobe demoed it, it was pretty good; real world not so much.
it's also endemic of adobe/macromedia culture. they're are so focused on creating/extending their platforms and experimenting with multiple ways to do the same thing they lose focus on the the core product lines. they simply don't have the resources to support excellence in their product line. adobe has 9,000 employees to support 100 or so products. apple by contrast has 35,000 non-retail employees to support around 50 products. yeah their product lines involve a lot more people on a single product, are arguably more difficult to create than adobe products and many of those employees are retail, but in gross numbers you're still talking ~700 people per product compared to ~90 people per product.
i'd love to see adobe pull an apple circa 2006 and strip out a ton of products, focus their engineering talents and then slowly re-introduce product segments. it'll never happen, smaller companies will start to slowly eat away at their product lines. i think they'll eventually head down the same road as SGI.
it's also endemic of adobe/macromedia culture. they're are so focused on creating/extending their platforms and experimenting with multiple ways to do the same thing they lose focus on the the core product lines. they simply don't have the resources to support excellence in their product line. adobe has 9,000 employees to support 100 or so products. apple by contrast has 35,000 non-retail employees to support around 50 products. yeah their product lines involve a lot more people on a single product, are arguably more difficult to create than adobe products and many of those employees are retail, but in gross numbers you're still talking ~700 people per product compared to ~90 people per product.
i'd love to see adobe pull an apple circa 2006 and strip out a ton of products, focus their engineering talents and then slowly re-introduce product segments. it'll never happen, smaller companies will start to slowly eat away at their product lines. i think they'll eventually head down the same road as SGI.
don't install crap like browser tool bars. except for issues with IE, i can't remember when i've had flash crashes on Windows 7
Battery life is relevant to a relatively small group of people. I am not aware of people other than college students sitting in outdated rooms (not equipped with power outlets) who do not have access to power. If Flash is 10x faster than HTML5+CSS3, it's not a good response, is it?
If Apple made it, it'd be a piece of crap. Apple still has a thing or ten to learn about optimization. Like somebody said in another thread infested with fanboys who can't fathom the fact that Apple is far from perfect, Zune software makes iTunes look like a failed 8th grade programming project.
There is a major flaw in this logic. Advertisers do not care how much Apple users spend on their computers. They advertise everything - not just computers. If anything, the advertisers might be less interested in Apple users because those have less money left after purchasing Apple product than Windows users
You should go into advertising.
The battery life will be more and more relevant in the future has the web will be more and more portable. If this is true that Flash suck lot more power than HTML5 on my macbook, it will be the same on a Blackberry or a Motorola Droid.
If your goal is to do the job with the least amount of processing work like Apple, Google, Firefox, Opera is trying to, it will be always better to only get the job done approach of Microsoft and Adobe.
Have you ever wonder why according to the Moore Law computer should double their capacity-processing-integration every 18 month and still waiting about the same time for booting your computer and launching your software has it was 15 years ago? Why developer back then we're able to fit the Mario bros game into 5ko and now it took megabytes only to get an hello word on the screen?. No one care about those thing anymore, they all want to get the job done.
BTW, everything I've tested so far pointing HTML5 to be faster than flash. You 10x claiming is absurd
Last edited:
What are you talking about? The improvements made here are for both Windows and Mac. Adobe could've and should've done this a long time ago with or without hardware acceleration.
Anyway it still doesn't perform better than HTML5 video and HTML5 ads will be just as easy to block as image ads.