Adobe Releases Preview of 64-bit Flash For Linux 329
Rinisari writes "Finally, the day has come. Adobe has released a pre-release version of the 64-bit Flash player. It is available at the Adobe Labs Flash Player 10 download site immediately.
Where are the Windows and Mac versions? 'Release of this alpha version of 64-bit Flash Player on Linux is the first step in delivering upon Adobe's commitment to make Flash Player native 64-bit across platforms. We chose Linux as our initial platform in response to numerous requests in our public Flash Player bug and issue management system and the fact that Linux distributions do not ship with a 32-bit browser or a comprehensive 32-bit emulation layer by default. Until this pre-release, use of 32-bit Flash Player on Linux has required the use of a plugin wrapper, which prevents full compatibility with 64-bit browsers. With this pre-release, Flash Player 10 is now a full native participant on 64-bit Linux distributions.' Windows and Mac OS X 64-bit versions will follow, and the final versions all will be released simultaneously. Tamarin, the JIT compiler in Flash, is now capable of producing 64-bit code and nspluginwrapper is no longer required. There are, however, no plans to release a debugger version of the 64-bit plugin."
Re:Silverlight (Score:5, Interesting)
Java is free but Flash is not (gnash and swfdec are getting there, but still not good enough for everyday use). And x86_64 is not the only 64-bit platform; what about Sparc and Itanic users, for example?
A binary blob for x86_64 is nice, I guess, but better would be for Adobe to give a bit of help to the projects trying to make a free implementation of Flash. So please continue to bitch, if you think that helps.
More Importantly (Score:1, Interesting)
While native 64 bit support is good news, it's not that big of a deal. 32 bit operation on 64 bit Linux has long been feasible so everyone was more than capable of poisoning their systems with Flash all along.
What is desirable and still missing is a Flash player that can run a video without putting dual core processor utilization at 60-90% for the duration of the video. It amazes me that there isn't a great outcry over this as it is an issue that has been around since circa Flash 7 and it only seems to be getting worse.
On the subject of 64 bit software, Skype is the more urgent need. Presently, it is not possible to run any version of Skype on a 64 bit Linux system. 64 bit Skype is a much bigger deal that is still absent.
Why no 32 bit browser? (Score:3, Interesting)
Why don't they anyway?
Given that many if not most users will want to install some plugins, installing a 32-bit browser by default would seem logical.
The browser would only have 2 gb address space, and it could not make use of new 64-bit registers or processor extensions.
Both seem like a very minor disadvantage for a browser, especially compared to being able to run 32 bit precompiled plugins.
Is this just one of these ideological things where the actual advantages for the user are disregarded?
At last!!!! (Score:4, Interesting)
I was one of many that put my name to asking for a 64 bit version. Now if only there was a 64 bit Skype and that ever elusive 64 bit browser plug-in for the 64 bit Sun Java.
Will be downloading the 64 bit Flash to test it out, hopefully it is easier than playing around with nspluginwrapper to get the 32 bit version working, and with a lot less processor power being eaten up just to run a Flash video.
Re:Silverlight (Score:3, Interesting)
How many Sparc and Itanic users need a 64-bit Flash for web browsing needs?
Adobe isn't going to GPL Flash just because people complain it isn't "free". Nor do I need every app in the world to be OSS. I prefer OSS apps, but at the end of the day, I'll take the best app for the task.
I don't think the GPL was the way to go for Java either, as Java needs to be fairly standardized. I would like to see a new OSS license that prevents forking, but allows me to read the source code, submit patches upstream, and recompile. However the license would prohibit people from distributing altered/forked versions. Such a license might very well convince Adobe to release the source code, as well as Nvidia for their drivers.
Re:Just tested it (Score:2, Interesting)
Make that 4. No problems here on gentoo 64.
It sure did feel good to emerge -C nspluginwrapper
one more time (Score:4, Interesting)
The complaint is not directed at Adobe, but against the idiots that are happily destroying the open nature of the web by embracing a closed product.
If you think it's impossible for anyone to make any money with software based on open standards, then you haven't been paying any attention to the development of the web.
If you don't get the problem that closed architectures are vulnerable to manipulation by the agencies that own the architectures, then you need some history lessons (start by searching for "Microsoft").
But this is all a fucking waste of breath, I'm afraid. On the one hand, you have an abstract argument about the long-term advantages of open architectures, and on the other, you have "look at the fonnny kittens!".
The web is doomed. Oh well, it's been an interesting few decades.
Re:Why linux first (Score:3, Interesting)
You've also been asking for 64-bit nvidia drivers for a good while as well?
The real solution is to give the community what they need to do it themselves. It's the easy, cheap, lazy, and better way out.
Re:Silverlight (Score:5, Interesting)
Might I suggest open standards so anyone who cares can implement their own?
You mean like this one? [whatwg.org] Yes, good idea. ;)
Re:doinitrite? (Score:2, Interesting)
Re:Coming up in 6 months (Score:3, Interesting)
using Intel processors as a baseline-
4 bit - 1971
8 bit - 1972 (1 year)
16 bit - 1978 (6 years)
32 bit - 1985 (7 years)
64 bit - 2004 (19 years)
128 bit - 2026 (22 years by my VERY crude estimate)
128 bit Flash some time after that.
Ubiquity of flash (Score:5, Interesting)
Out of interest, what do you consider the smallest possible user base that any concession should be made with regard to support?
Given that flash is starting to get ubiquitous, any online-capable platform is going to need it badly soon. Restricting a (closed) technology to only a couple of most widespread platform is really bad.
If we exit the world of desktop computers the most widespread CPU are embedable RISCs such as ARM and MIPS. Yet, none of them can currently get even a free-as-in-beer plugin downloadable from adobe's website.
Thus embed platforms (such as PDAs, MID) have either to count on gnash (lots of F/LOSS BusyBox/Linux based PDA & Smartphone projects) or do without flash at all (iPhone currently).
There are lots of other possible creative usage for other machines. But, by lacking flash some of them will suddenly be not so useful for experiencing the web.
That's why the whole flash thing was a bad idea to begin with. In an open system as the internet, where everything is documented and free to get re-implemented, where everything was built in this open way, suddenly there comes a new "standart" which will only run on a couple of processors, preventing anything else to use it and stoping any creative use of it.
How much effort should be put into supporting either Sparc or Itanic, or indeed any other minority platform?
Well not much is asked from Adobe. Just help the open source enough to have a descent open source implementation.
That should be that much difficult, taking into account that adobe makes no money on the free-as-in-beer plugin, instead their main income comes from the creation suites.
Adobe has done it in the past (PDF is a published standard, with numerous alternative implementation existing - Adobe makes money on the Acrobat suite).
Concurrent of flash have done it (Silverlight vs. Moonlight)
And I'm not just talking about closed source apps here, I'm also talking about open source projects and the stance they take, and the whole range of possible support options
Usually, after the first couple of ports have revealed all the hidden platform dependent bugs, lots of additional posts come almost "for free", generally only a recompile away.
In addition, we don't expect the Gnash developers to maintain port for every fucking platform under the sun.
Gnash is free/libre opensource software. If the developers don't have the resources to port the application themselves, others are open to do it.
In fact that's what's happening : gnash is mainly developed on x86 and x86_64 architecture, but that hasn't stopped other enthusiasts to port it to PS3 (MIPS).
Flash is getting popular. As long as there's some interest for some platform there are bound to be enough interested developers (for that platform) to port it instead of gnash's own developers.
Popular game engines on portable device is a nice example : ...as long as gnash is compatible enough to flash.
ScummVM is available on a dozen of hardware platforms, some not even POSIX compatible. Not all of them are maintained by the main ScummVM developer. But the popularity of ScummVM and the coolness to be able to plas scumm-based (mostly LucasArt) games on whatever pocketable machine you have has nonetheless attracted enough motivated people to port it to a wide array of machines.
Same goes for several other game engine (Doom and Duke3D are nice examples. Usually its the second thing that get ported to any new hardware platform, right after Linux).
Given the rising popularity of flash, Gnash will probably follow the same trend...
And for that, cooperation from Adobe will help immensely.
Re:flash on linux (Score:3, Interesting)
has always sucked, and in all likelyhood will always suck, 32 or 64 bit.
Your opinion (and mine) of Flash doesn't matter. The important part is that a major software developer just told the world that Linux is a good platform for testing and to support. Even if I'm indifferent to Flash outside of Youtube, I can be excited about Adobe's announcement.
Re:Silverlight (Score:5, Interesting)
Release the source, or at least an open API/documentation/something, and then let us do the work.
Yeah I've heard this before. Must have been last week that people were bitching that Google hasn't ported Chrome to Linux yet (here on slashdot). It just seems to me that the more companies give in to the Linux community, the more they demand. It usually goes:
Release the specs! -> Release the source! -> Port it for us!
Re:Why no 32 bit browser? (Score:3, Interesting)
Yes.
It's actually more about power than ideology. The key people behind Ubuntu, Debian, Fedora and the other major distros all fervently believe that they and they alone should be the gatekeepers of all software run by their users. They see no reason why somebody should be able to get software from somewhere other than them. Thus they see no reason why binary compatibility is needed or useful. Thus being binary compatible with 32 bit binaries is simply a waste of disk space (which is extremely cheap but logic doesn't play into this).
This position is moronic but ubiquitous in the Linux world. It leads to absolute frustration on the part of any developer who actually cares what bits make it into their users hands - the almighty Firefox/Debian smackdown is rooted in this problem. There have been many other problems though: proprietary software like video games or Google Earth rely on huge hacks to make their stuff work, upstream developers like Inkscape get tired of their users running versions of their software obsoleted months ago because no upgrade is available from their distro. It leads to bugs like this one [launchpad.net], which boil down to "our packaging policies are too inflexible to give you the features you want". And so on.
It always fascinates me how much criticism Apple has received over their control over the iPhone app store and the arbitrary policies they impose on developers there, but when Linux distributors do exactly the same thing there is only stony silence. Software gets dropped, changed arbitrarily, held back to obsolete versions or denied entry into the repositories entirely, forcing bizarre hacks (try writing down the instructions for running a 3rd party loki installer some time!).
The lack of 32 bit support is just one facet of this ridiculous ideology.
Re:doinitrite? (Score:3, Interesting)
32-bit Tortoise doesn't even run on 64-bit Windows - so I'm stuck.
It may not show up in the 64-bit explorer shell, but it does install and run. I believe a lot of people install both so that they can see TSVN options in the open dialog of their 32-bit apps.
Re:Why no 32 bit browser? (Score:2, Interesting)
It depends who is more important really, doesn't it? Microsoft goes through the pain because supporting binaries is something they have to do in a new OS otherwise people won't upgrade to it. Actually the whole OS is architected so that most 32 bit stuff just thunks through to 64 bit code so the amount of duplication is probably quite small.
The Linux world makes acid comments about closed binaries and does not. Users don't pay for uprades, so it doesn't matter how inconvenient the upgrade to 64 bit is. Of course that means they end up blaming Adobe for not porting to 64 bit whereas the Windows world doesn't much care, but then again flash is an evil closed binary too so Adobe deserve it.
Linux Hater ranted about this, in his inimitable style.
http://linuxhaters.blogspot.com/2008/07/my-browser-needs-16-exabytes.html [blogspot.com]
What's funny about this is that even now they've done it people are complaining that flash is still a closed binary, they haven't released a debug version and don't support Sparc64 or Itanium and what they really should do is to release the source code.
Re:Flash specification (Score:2, Interesting)