Michal Kubecek composed on 2015-08-28 10:30 (UTC+0200):
Felix Miata wrote:
I remember seeing plenty of threads around where this sort of topic has come up, and seeing lots of people still using 32 bit by preference on their 64 bit systems.
That's exactly what I suspected: most people using 32-bit distributions these days actually do so because of their beliefs, not because they have to. Such use is certainly legitimate - but way less relevant for the question whether we should support the architecture.
Maybe there are more "have tos" than you think. Did upgrading an existing 32 bit installation to 64 bit ever become a supported option? I don't recall any such thing. I find upgrades far less painful than fresh installations. Probably other 32 bit preferrers do too.
It seems to me any acknowledgement would primarily be that the majority of developers don't want to bother with less than the newest and fastest machines,
Newest and fastest? 64-bit CPU's are widely available since ~2003 and prevailing since ~2005, for last 5 years, it's almost impossible to buy a 32-bit one.
The point was that devs who might otherwise wish to at least test on older systems find plenty reasons not to spend the time, among which waiting on slower RAM, CPU and I/O busses, not to mention logistics of keeping older equipment functional and pragmatically available.
And again, it's not about "fastest". The tricks kernel has to do to cope with 32-bit architecture are quite ugly. There are even problems that can't be resolved on i586 (I remember a guy having over 60% of his 2GB RAM unused but unable to add a netfilter rule because of memory allocation failure). You have fewer registers, leading to much less efficient function calling convention etc.
largely I'll bet to compensate for software bloat, driving the cycle that makes vendors happy, but not so much users, particularly those on tight budgets.
Seriously?
I never asked for compositing, the bling it enables, or the speed and RAM they depend on. KDE3 and TDE still match my needs as well as ever, unlike KDE4 and Plasma5. Did the kernel really need to more than double in size in 8 years[1], plucking so much out of userspace? Or was that growth a consequence of software bloat needing compensatory assistance to keep newer, faster PCs from seeming slower? There do still exist people who don't like having fixed what ain't broke. :-) [1] https://en.wikipedia.org/wiki/Linux_kernel#/media/File:Lines_of_Code_Linux_K... -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org