Adobe Flash Player 10 Release Candidate

[Vadi]

According to Ubuntu's popcon (which is disabled by default, needs to be activated to report anon. data) under 15% of the installations are 64bit.

I didn't check debians, but I don't expect it to be much higher either.

[Dragoran]

http://smolt.fedoraproject.org/static/stats/stats.html (also opt in to send data)
20.7% if you add ppc64 its 21%

[TechMage89]

That's the chicken and the egg speaking, to mix a metaphor. So few people install 64-bit versions because it's such a flipping pain to get them to work with all the 32-bit only software.

[RealNC]

Adobe is too poor to spend dollars for a 64-bit port. They barely get enough money together for their daily bread, how to you expect them to raise the funds for a 64-bit version of Flash? Geez, you people should be thankful for the awesome piece of software called Adobe Flash(TM) rather than ranting about 64-bit support.

[deanjo]

According to Ubuntu's popcon (which is disabled by default, needs to be activated to report anon. data) under 15% of the installations are 64bit.

I didn't check debians, but I don't expect it to be much higher either.

That doesn't suprise me as debian was very late to the 64-bit game (4 years behind other distro's). However if you take a look at the the more mature 64-bit distro's such as openSUSE 1 in 3 systems is 64-bit on the latest release and with every release grows faster and faster (5% for the last release).

[Svartalf]

Adobe is too poor to spend dollars for a 64-bit port. They barely get enough money together for their daily bread, how to you expect them to raise the funds for a 64-bit version of Flash? Geez, you people should be thankful for the awesome piece of software called Adobe Flash(TM) rather than ranting about 64-bit support.

Heh... Write it 64-bit clean in the first place and you'll take care of both, actually. FYI- I wouldn't trust any code that wasn't 64-bit clean because there's a lot of intrinsically BAD coding practices that fall under the "unclean" category (Cardinal rule #1 of coding: an "int" is NOT the same as a "void *" in C/C++ terms... I'm pretty sure Flash's code breaks that rule... Anything that doesn't marshal that void pointer to something proper with a way other than typecasting is going to be iffy code to begin with.) and it might be a source of nasty transients.

[Kano]

Well you know what is different, why don't you help em then

[greg]

Svartalf, did you ever port a JIT/dynamic recompiler? If no, please shut up.
I hope the Flash guy will make a version available that has at least libcurl statically linked in. It's also possible to make libnss optional, so that without it you simply won't be able to use SSL.

[Svartalf]

Well you know what is different, why don't you help em then

I offered back when I was one of the early adopters for x86-64 (I happen to still have the Solo2 motherboard AMD gave me...)...they didn't seem too interested back then, Kano. Not my problem, really, I'm trying to get other fights fought nowadays...

[Svartalf]

Svartalf, did you ever port a JIT/dynamic recompiler? If no, please shut up.

Have you? If so, please enlighten us as to why it would be radically different than what I commented on? If not, please spare everyone the commentary on the subject.