Flash 9 Beta for Linux Available

DemiKnute writes "According to the official Penguin.SWF blog, the a beta release of the long-awaited Flash 9 for Linux is available for download, a mere year after the release for Windows." From the blog: "While we are still working out exactly how to distribute the final Player version to be as easy as possible for the typical end user, this beta includes 2 gzip'd tarball packages: one is for the Mozilla plugin and the other is for a GTK-based Standalone Flash Player. Either will need to be downloaded manually via the Adobe Labs website and unpacked. The standalone Player (gflashplayer) can be run in place (after you set its executable permission). The plugin is dropped into your local plugin directory (for a local user) or the system-wide plugin directory." Report bugs here.

Re:AMD64 version?

[Anonymous Coward]

Adobe have said no in the past, just install a 32-bit web browser instead of your 64-bit one.
Yeah, it's a pain, but you only need to do it once.

Why not say something into adobe.com/go/wish ?

Re:AMD64 version?

[Anssi55]

You could try nspluginwrapper [gibix.net].

Re:AMD64 version?

[Anonymous Coward]

You can use nspluginwrapper. It seems to work with flash 9 also. You just need to get 0.9.90.3 from a mandriva x86_64 cooker mirror and alien -cv *.rpm, and follow instructions in:

http://www.gibix.net/dokuwiki/en:projects:nsplugin wrapper [gibix.net]

0.9.90.1 that's available in the official site doesn't work with new firefoxes, so you really need to get 0.9.90.3 from mandriva.

gentoo ebuilds

[kswtch]

here [gentoo.org] and here [gentoo.org].

Good news!

[bioglaze]

Even being beta version, Flash 9 for GNU/Linux works very well when compared to previous player.

Some flash movies that hogged Firefox UI with old player work flawlessy now. Audio is now in sync with video.

While not perfect, this release makes me wonder when the free software Gnash player reaches a usable state. Being a free software enthuasist, i generally don't like the idea of using a proprietary plugin, but being also pragmatic, i use it. I also think that the official Flash plugin could be faster and more bug-free, if the source code were available and under a GPL compatible licence.

That being said, i still think it's important that GNU/Linux users, especially Average Joe, have a lot less hassle with badly designed, flash-dependent websites.

Re:gentoo ebuilds

[Anonymous Coward]

Already in portage, old chap...

Re:right

[dylan_-]

That's a bug with your distro: report it to them.

Compiling bugs

[thebluesgnr]

There are a few problems with this release that I hope will be fixed in the future (I can only hope, since it's not open source yet).

The plugin will search for libssl.so and libasound.so; that's broken. They should dlopen the actual library or build it statically, but a hack like that is certainly going to cause problems. (btw, in Ubuntu/Debian you need the libssl-dev and libasound2-dev packages to use all the features of this plugin).

The most annoying bugs I had with Flash (believe it or not) are still there. If the mouse is hovering a Flash content inside a browser window, the browser won't recognize keyboard or even mouse events. This is annoying when you're scrolling through a page with Flash ads or when you want to Ctrl+L but the damn mouse is in the wrong place.

The other problem is that Flash ads that have the "point your mouse here to see the full ad" will always display the "full ad", and you have to choose between the Flash Block extension and not reading that damn page at all.

Re:Good news!

[sarathmenon]

While not perfect, this release makes me wonder when the free software Gnash player reaches a usable state. Being a free software enthuasist, i generally don't like the idea of using a proprietary plugin, but being also pragmatic, i use it. I also think that the official Flash plugin could be faster and more bug-free, if the source code were available and under a GPL compatible licence.

gnash is usable enough for me. Most ads works (sigh), and from what I've seen pretty much everything is rendered fine except for the flv videos. Now that ffmpeg and xine have full flv7 support, its only a matter of time before we can start to see gnash support youtube in its full glory. The best part is that it "works" on an x86_64.

Re:AMD64 version?

[CastrTroy]

Yes, but you also have to have a lot of other 32 bit libraries installed just for the browser to run. I think that one of them is glibc. I'm running mandrake 2007 rc1 (haven't downloaded final yet, but i've installed all the updates), and when I tried using 64-bit, even isntalling a 32-bit browser didn't work. Firefox would crash every time flash tried to start. So, we could either install only the 64 bit libraries, or install 64 and 32 bit libraries, and the 32 bit browser and hope it works. However, I'm still running full 32 bit linux on my AMD64. I tried 64 bit for a while, but I found that a lot of stuff still isn't stable enough for me on 64 bit. For one thing, the 3D desktop on Mandriva 2007 wouldn't work on my Radeon X550 when I had 64 bit. With 32 bit, no problems at all. I guess i'm going to have to wait until Mandriva 2008, when hopefully 64 bit linux will be ready. I also tried out other 64 bit distros (Fedora, Suse) and found that they weren't any better.

Re:AMD64 version?

[jascat]

Look into dchroot and setup a small 32bit chroot environment. On my AMD64 desktop running Ubuntu, I have Firefox, Adobe Acrobat, a 32bit JDK and Mplayer installed and it works like a champ. HOWTO here. [debian.org]

Sad state of affairs

[Anonymous Coward]

Note all the people installing both flash and flash block! First they contract the disease and then they grab something to make it bearable. That says as much as the oblig technical derision.

BTW: there's a youtube downloader listed on freshmeat so you don't actually need to taint your OS with flash in order to watch the funnies on youtube.

Re:AMD64 version?

[bug1]

"Why would running a web browser in 32bit mode have any negative effect on uptake of 64bit OSes?"

It makes the distribution much more complex to have combinations of 32 and 64 bit applications and libraries.

I assume all the libraries of a 32 bit app on a 64 bit system would haveto be 32 bit as well, look at all the libraries effected...

# ldd /usr/lib/firefox/firefox-bin
libmozjs.so => /usr/lib/libmozjs.so (0x00002b566e625000)
libxpcom.so => /usr/lib/libxpcom.so (0x00002b566e7bb000)
libxpcom_core.so => not found
libplc4.so => /usr/lib/libplc4.so (0x00002b566e9a5000)
libnspr4.so => /usr/lib/libnspr4.so (0x00002b566eaaa000)
libpthread.so.0 => /lib/libpthread.so.0 (0x00002b566ebe6000)
libgtk-x11-2.0.so.0 => /usr/lib/libgtk-x11-2.0.so.0 (0x00002b566ecfb000)
libgdk-x11-2.0.so.0 => /usr/lib/libgdk-x11-2.0.so.0 (0x00002b566f124000)
libX11.so.6 => /usr/lib/libX11.so.6 (0x00002b566f2b9000)
libpng12.so.0 => /usr/lib/libpng12.so.0 (0x00002b566f4c2000)
libjpeg.so.62 => /usr/lib/libjpeg.so.62 (0x00002b566f5e5000)
libz.so.1 => /usr/lib/libz.so.1 (0x00002b566f708000)
libsmime3.so => /usr/lib/libsmime3.so (0x00002b566f81e000)
libssl3.so => /usr/lib/libssl3.so (0x00002b566f948000)
libnss3.so => /usr/lib/libnss3.so (0x00002b566fa6e000)
libcairo.so.2 => /usr/lib/libcairo.so.2 (0x00002b566fbf0000)
libXinerama.so.1 => /usr/lib/libXinerama.so.1 (0x00002b566fd59000)
libXt.so.6 => /usr/lib/libXt.so.6 (0x00002b566fe5c000)
libXp.so.6 => /usr/lib/libXp.so.6 (0x00002b566ffbc000)
libXft.so.2 => /usr/lib/libXft.so.2 (0x00002b56700c4000)
libfontconfig.so.1 => /usr/lib/libfontconfig.so.1 (0x00002b56701d9000)
libxpcom_compat.so => /usr/lib/libxpcom_compat.so (0x00002b567030c000)
libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0x00002b567042e000)
libm.so.6 => /lib/libm.so.6 (0x00002b567062e000)
libc.so.6 => /lib/libc.so.6 (0x00002b56707b0000)
libplds4.so => /usr/lib/libplds4.so (0x00002b56709ec000)
libgdk_pixbuf-2

Re:AMD64 version?

[Anonymous Coward]

On my AMD64 machine, I have an i386 install of Debian in a chroot. Debian supports this really well - search for "dchroot". Inside that install, I can use any standard i386 Debian package, including multimedia stuff that only runs on i386 (e.g. Win32 codecs). It's very convenient.

Re:Compiling bugs

[cortana]

Please report the incorrect use of dlopen to Adobe. I already filled in a bug report, hopefully if enough knowledgable people do it then it will get past their QA people to someone who will understand the problem.

The most annoying bugs you mention are actually bugs in Mozilla(tm): https://bugzilla.mozilla.org/show_bug.cgi?id=95541 [mozilla.org] https://bugzilla.mozilla.org/show_bug.cgi?id=87383 [mozilla.org]

Re:Inaccurate.

[Anonymous Coward]

Flash Player 9 for Windows was officially released on June 28, 2006.

This is not the official release, only the first beta. Please compare beta to beta release date, or official to official.

Re:Inaccurate.

[illegalcortex]

I have to agree with you on this one, even if the other replies are jumping on you. At best, that line is just poor grammar. At worst, it's plain wrong.

According to the official Penguin.SWF blog, the a beta release of the long-awaited Flash 9 for Linux is available for download, a mere year after the release for Windows.

According to the official Penguin.SWF blog, the a beta release of the long-awaited Flash 9 for Linux is available for download, a mere year after the beta release for Windows.

Makes a world of difference in readability.

Bigger virtual address space

[Nicolas MONNET]

The much larger address space allows for more leeway in memory management, even if you don't have over 4G.

For instance, nptl threads get a performance boost from not having to juggle around to save on stack space.

There are also advantages with prelinking.

Finally, even if you have "just" 4G in 32 bit, you won't be able to use all of it in one process, as the kernel needs some address space too.

Re:AMD64 version?

[CommandNotFound]

For mortal users (most of us), the benefit is an instant CPU performance boost of around 20-30% at least on Athlon 64 units when using the 64-bit instruction set vs the 32-bit instruction set. I have a dual-core AMD64 now, but I'm running everything 32-bit as the performance is more than acceptable. However, in a couple of years I will upgrade everything to 64-bit once all these glitches are solved and I should get a free upgrade in speed.

This PCStats article [pcstats.com] has some benchmarks on the topic. Anandtech had some too, but I couldn't find them immediately.

Should be much faster

[mjbkinx]

Previous revisions of Flash Player for Linux preformed very poorly compared to the win32 versions (even the win32 verison in crossover office did a better job).

Yeah, Tinic ranted about that on his blog a while ago, saying he used wine for Flash on Linux (before v9, obviously) -- and he's a FlashPlayer engineer. His entry [kaourantin.net] about this beta release addresses performance. He says he's not happy with the current state of font rendering speed yet, but that it beats the Windows version by 20% with other stuff. They're still working on it.

Over all, you should see better performance of existing content, thanks to the new rendering engine introduced in v8. This is especially true for SWFs (competently) written for v8 and using cacheAsBitmap -- not rerendering vectors every frame seems to improve performance. Who would have thought...

The second performance increase will probably take a while to become common: FP9 comes with a new, JIT compiled VM. The old one is still included for backwards compatibility, but once FP9 has a good install base and is supported by developers making scripting-heavy stuff, you should definitely notice the performance increase -- it's much, much faster.

If somebody feels like playing with it, there's the free (beer) Flex SDK on the Adobe site somewhere. However, I'd like to recommend haXe [haxe.org], a Free (capital F) compiler for a very fine language, with a great type system, that I really enjoy coding in. It supports Flash 6 to 9, the Free NekoVM [nekovm.org], and can generate JavaScript (Yes! Typed!). Windows users can use the FlashDevelop [osflash.org] plugin [haxe.org], for the rest of us there's Eclipse with EHX [osflash.org].

Tried it

[spitzak]

On my Mandrake machine. I got no sound from YouTube, and sound works in the FlashPlayer7.

Notes:

Biggest problem is no sound from YouTube (or probably from anywhere). Sound works for me with FlashPlayer7 and switching back to that makes it work without any restarting (so it did not permanently mess up sound, like some programs can). This is a Mandrake machine, 2.4.22-10mdkenterprise, I really have no idea how I have sound set up, but it works for me in most software.

Yes it fixed places that check for the version number of the flash player.

Popping up the menu with the right button (which I did to check that it reported 9 or 7) would cause Firefox to crash somewhat later. Does not seem to happen with 7. May indicate an overflow of some malloc'd data buffer.

To use, put libflashplayer9.so into ~/.mozilla/plugins and don't rename it. Apparently if it exists it will be loaded in preference to libflashplayer.so. (I wasted some time making a flashplayer.so symbolic link that switched between 7 and 9 before I finally figured out that 9 was being used no matter how I set it. Instead, to switch back to 7, rename libflashplayer9.so to libflashplayer9.so.hidden).

Removal instructions in the readme.txt say to remove libflashplayer.so, not the correct file of libflashplayer9.so.

ldd shows it links in far more libraries than 7 did, lots of gtk stuff. I suspect this is due to Pango (which does I18N text layout) using the gobject library, not because any gtk widgets are being used. This has also been complained about on Cairo (which is supposed to be a drawing library *used* by toolkits like gtk, but because good font layout requires Pango, there is a circular dependency back to gtk!)

Re:AMD64 version?

[Curtman]

64-bit CPUs have been out for almost 2 years now

2 years [wikipedia.org]?

1991: MIPS Technologies produced the first 64-bit microprocessor, as the third revision of their MIPS RISC architecture, the R4000. The CPU was used in SGI graphics workstations starting with the IRIS Crimson. However, 64-bit support for the R4000 was not included in the IRIX operating system until IRIX 6.2, released in 1996.