"Subway steel" <foo@xxxxxxx> writes:
> "Roy Schestowitz" <newsgroups@xxxxxxxxxxxxxxx> wrote in message
> news:1733603.KUf3eGiOSO@xxxxxxxxxxxxxxxxxx
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> ____/ Subway steel on Friday 12 September 2008 14:29 : \____
>>
>>>
>>> "Roy Schestowitz" <newsgroups@xxxxxxxxxxxxxxx> wrote in message
>>> news:1311060.HUKxdvdlQ3@xxxxxxxxxxxxxxxxxx
>>>> -----BEGIN PGP SIGNED MESSAGE-----
>>>> Hash: SHA1
>>>>
>>>> 64-bit Linux, is it time?
>>>>
>>>> ,----[ Quote ]
>>>> | On the Linux platform, there are still some issues with software that
>>>> haven't
>>>> | been compiled for 64-bit. However, they are closed source applications
>>>> so
>>>> | getting them to work for 64-bit is up to the software company. Adobe
>>>> has
>>>> not
>>>> | yet released a 64-bit version of their venerable Flash plugin, however
>>>> its
>>>> | been said on one of Adobe's blogs that Flash 10 may be released
>>>> 64-bit.
>>>> | Another package is a 64-bit version of Sun's Java interpreter. 64-bit
>>>> Java is
>>>> | available for Windows and (surprise) Solaris, but not for Linux. It's
>>>> unsure
>>>> | if Sun is not offering 64-bit Linux binaries because of its Solaris
>>>> operating
>>>> | system.
>>>> |
>>>> | Hopefully the points discussed here will start to take shape in the
>>>> Linux
>>>> | world. Distributions should start marketing 64-bit. Programmers should
>>>> start
>>>> | adjusting their code for 64-bit, while not taking away from 32-bit.
>>>> Software
>>>> | companies should provide 64-bit binaries when the source packages
>>>> aren't
>>>> | available.
>>>> `----
>>>>
>>>> http://blog.jeffanderson.us/64-bit-linux-is-it-time/
>>>
>>> The guy who wrote this blog is entitled to his own opinion but that
>>> doesn't
>>> make it right. I'm a developer and we have a very large enterprise level
>>> application that we sell. About 95% of our customers use the 32-bit
>>> version
>>> because the 64-bit version of the app generally runs between 5% to 10%
>>> slower and it requires more resources (memory). The few customers that do
>>> run the 64-bit version do so because they have such large amounts of data
>>> that they need the extra address space.
>>>
>>> So unless home users are typically using multi-terabyte databases then
>>> what
>>> is the point of having a 64-bit OS. For general purpose computing it's
>>> going
>>> to be about 5-10% slower and it's going to take more memory. I don't see
>>> how
>>> a 64-bit OS (any 64-bit OS) is more "sensible" for desktop computing
>>> needs.
>>
>> True. I think that it's still quite a good option for 'production'
>> desktops
>> that do something like rendering and run just a couple of dedicated
>> applications anyway. Chip designers too seem to favour 64-bit GNU/Linux
>> laptops (seen it in some publications over the past few years).
>
> I'll agree that a 64-bit desktop is good for specialized use. For the
> general user who surfs the web, writes documents and reads email there
> really isn't much benefit. But if someone is doing so very CPU
> intensive
And some good reasons not to use it. Like reduced functionality with
some closed source apps you might need. And up to double the memory
usage in certain cases. Ask Koehlmann. We schooled him in that recently
and Tim even posted some numbers to shoot Peter down in flames.
> work, video production, CAD-CAM, etc then the ability to move 64-bits of
> data around at a time would benefit those specific users.
In some cases.
|
|
