longest without a reboot
allen....@gmail.com
reboot?
A friend of mine went over a year
DFS
> what is the longest you have had your linux system running without a
> reboot?
A day or so.
spi...@freenet.co.uk
> what is the longest you have had your linux system running without a
> reboot?
> A friend of mine went over a year
I've had a year's uptime before now, I think.
It's rare though cos I tend to shutdown when I'm away from the house for
more than a couple of days.
--
______________________________________________________________________________
| spi...@freenet.co.uk | |
|Andrew Halliwell BSc(hons)| "ARSE! GERLS!! DRINK! DRINK! DRINK!!!" |
| in | "THAT WOULD BE AN ECUMENICAL MATTER!...FECK!!!! |
| Computer Science | - Father Jack in "Father Ted" |
------------------------------------------------------------------------------
spi...@freenet.co.uk
> A day or so.
That's doofus for you.
--
______________________________________________________________________________
| spi...@freenet.co.uk | |
|Andrew Halliwell BSc(hons)| "The day Microsoft makes something that doesn't |
| in | suck is probably the day they start making |
| Computer science | vacuum cleaners" - Ernst Jan Plugge |
------------------------------------------------------------------------------
Antonio Murphy
<allen....@gmail.com> wrote in message
news:4fa41a12-51e4-4161...@r60g2000hsc.googlegroups.com...
> what is the longest you have had your linux system running without a
> reboot?
>
> A friend of mine went over a year
On most linux installs "X" (the gui) locks up very frequently because of
poor driver support with modern video cards. Now people will argue that
linux is still alive and running and all I need to do is to telnet or ssh
in from another machine and I can restart X. Great except that I don't have
another machine to do this with. The fact is that if the mouse, keyboard
and screen are frozen solid then for all practical purposes, the OS is
locked up solid and needs a hard reset. Just hope the filesystem survives
the hard reboot without corrupting itself.
--
Posted via a free Usenet account from http://www.teranews.com
DFS
> DFS <nospam@dfs_.com> did eloquently scribble:
>> allen....@gmail.com wrote:
>>> what is the longest you have had your linux system running without a
>>> reboot?
>
>> A day or so.
>
> That's doofus for you.
That's as long as I can stand the stench.
JAG CHAN
@bignews7.bellsouth.net:
doofus I think you like the stench.
Otherwise there is no reason for you to come here.
Stephen Fairchild
I have found recent NVidia drivers (last 6 months) do not have this problem.
--
Stephen Fairchild
Rick
You are SUCH an asshole.
--
Rick
Richard Rasker
My desktops usually get rebooted daily, because of noise, power savings, and
wear -- although some of my users sometimes have 'em running for weeks on
end; the only reported stability problems had to do with X sometimes
choking on a game or so, and a crappy wireless card with crappy Windows
drivers.
Servers under my control get rebooted on average once a year, because that's
how often the mains power fails over here. Apart from this, I've only had
Linux servers failing because of hardware trouble (bad disks, fans, and
mainboard capacitors, and one case of lightning damage). The longest uptime
I recall was some 450 days or so, after which a kernel update forced a
reboot.
Richard Rasker
--
http://www.linetec.nl/
Hadron
> allen....@gmail.com wrote:
>
>> what is the longest you have had your linux system running without a
>> reboot?
>>
>> A friend of mine went over a year
>
> My desktops usually get rebooted daily, because of noise, power savings, and
> wear -- although some of my users sometimes have 'em running for weeks on
> end; the only reported stability problems had to do with X sometimes
> choking on a game or so, and a crappy wireless card with crappy Windows
> drivers.
What game or so? As for the crappy wireless card with crappy Windows
drivers, why not load the excellent Linux drivers?
> Servers under my control get rebooted on average once a year, because
> that's
I would bet you are a liar. A word constantly thrown around here by the
"advocates" I notice. I thought it would be nice to throw it back. You
just pulled that number our of your arse. Any server room worth its salt
has back up power and smoothers/regulators/UPSs to stop that kind of reboot
being necessary.
> how often the mains power fails over here. Apart from this, I've only had
> Linux servers failing because of hardware trouble (bad disks, fans, and
> mainboard capacitors, and one case of lightning damage). The longest uptime
> I recall was some 450 days or so, after which a kernel update forced a
> reboot.
>
> Richard Rasker
I am interested - when you sell your services to companies using
Windows, do you insult them and laugh at them like you do behind their
backs here?
--
Endless Loop, n.:
see Loop, Endless.
Loop, Endless, n.:
see Endless Loop.
-- Random Shack Data Processing Dictionary
ray
A couple of months. It's hard for me to go much longer without doing some
hardware tinkering.
Linonut
> <allen....@gmail.com> wrote in message
> news:4fa41a12-51e4-4161...@r60g2000hsc.googlegroups.com...
>> what is the longest you have had your linux system running without a
>> reboot?
>>
>> A friend of mine went over a year
>
> On most linux installs "X" (the gui) locks up very frequently because of
> poor driver support with modern video cards.
Antonio, you are-a full-a da shit!
You Lentini muthah-da-fuckah!
> Now people will argue that
> linux is still alive and running and all I need to do is to telnet or ssh
> in from another machine and I can restart X. Great except that I don't have
> another machine to do this with. The fact is that if the mouse, keyboard
> and screen are frozen solid then for all practical purposes, the OS is
> locked up solid and needs a hard reset. Just hope the filesystem survives
> the hard reboot without corrupting itself.
Antonio, why-a you make-a uppa diss shit!
Nobody believe-a da bullshit! Maricon! Fa sito, ostia, you crazy
half-breed!
Roy Schestowitz
> what is the longest you have had your linux system running without a
> reboot?
>
> A friend of mine went over a year
Over half a year on two of my desktops. A colleague shut one of them down,
thinking that he was 'helping'.
Unless you install new hardware or make no use of wake-on-LAN (and the likes of
it, which save energy), there's never a reason to reboot really.
--
~~ Best of wishes
Roy S. Schestowitz | "Nothing to see in this sig, please move along"
http://Schestowitz.com | GNU is Not UNIX | PGP-Key: 0x74572E8E
http://iuron.com - proposing a non-profit search engine
Richard Rasker
> Richard Rasker <spam...@linetec.nl> writes:
>
>> allen....@gmail.com wrote:
>>
>>> what is the longest you have had your linux system running without a
>>> reboot?
>>>
>>> A friend of mine went over a year
>>
>> My desktops usually get rebooted daily, because of noise, power savings,
>> and wear -- although some of my users sometimes have 'em running for
>> weeks on end; the only reported stability problems had to do with X
>> sometimes choking on a game or so, and a crappy wireless card with crappy
>> Windows drivers.
>
> What game or so?
Nexiuz comes to mind -- at least in combination with some Intel-based
graphics cards. And oh, on at least one laptop I found that closing and
subsequently opening the lid caused the screen resolution to switch to some
weird mode. Last I heard, that one's fixed after the latest update.
These problems are few and far between, and get even more rare every year.
Most of my ~100 users have no stability problems.
> As for the crappy wireless card with crappy Windows drivers, why not load
> the excellent Linux drivers?
Ah, if only the manufacturers of wireless hardware could be bothered to a)
supply Linux drivers, or, failing that, b) supply stable Windows drivers(*)
which can be used with ndiswrapper, or, failing that, c) supply sufficient
data on their chipsets for the OSS community to create proper Linux
drivers. But alas, the only thing provided are Windows drivers which
sometimes even have issues when used under Windows -- in particular with
the cheaper products.
>> Servers under my control get rebooted on average once a year, because
>> that's
>
> I would bet you are a liar. A word constantly thrown around here by the
> "advocates" I notice. I thought it would be nice to throw it back. You
> just pulled that number our of your arse. Any server room worth its salt
> has back up power and smoothers/regulators/UPSs to stop that kind of
> reboot being necessary.
I have no server room. I manage a dozen simple server boxes for groups of
6 - 15 users (mostly non-commercial), spread all over town. These machines
usually feature Internet connectivity (firewall/router) and several minor
servers (DHCP, caching nameserver, NTP server, file server, backup server
and sometimes Web server).
In most cases UPS for those servers is useless, because in case of a power
outage all normal PC's, switches and Internet connections hooked up to it
are down as well.
>> how often the mains power fails over here. Apart from this, I've only had
>> Linux servers failing because of hardware trouble (bad disks, fans, and
>> mainboard capacitors, and one case of lightning damage). The longest
>> uptime I recall was some 450 days or so, after which a kernel update
>> forced a reboot.
> I am interested - when you sell your services to companies using
> Windows, do you insult them and laugh at them like you do behind their
> backs here?
I don't mind what they hook up behind my servers; they know I'm only paid to
keep the servers up and running, and that any trouble with their Windows
machines isn't my problem. For al the rest I don't insult Windows users; I
feel sorry for them -- they're just victims of the criminal assholes who
created Windows and foisted it onto the world.
DFS
There is a reason: smack down lots of lying cola idiots.
7: Linux is 50x faster than Windows
Homer: Half of Europe has dumped Windows
Roy Lying Spammer: 90% of IT pros reject Vista
Rasker: Vista requires 10x-20x the resources of Linux
Ballard: 70% of Dell's recent PC sales are Ubuntu
Hadron
> I don't mind what they hook up behind my servers; they know I'm only paid to
> keep the servers up and running, and that any trouble with their Windows
> machines isn't my problem. For al the rest I don't insult Windows users; I
> feel sorry for them -- they're just victims of the criminal assholes who
> created Windows and foisted it onto the world.
Yet you happily take money translating stuff which is created for that
very platform.
Richard Rasker
You couldn't be more wrong. Ten years ago, when I dumped Windows in favour
of Linux, I told my customers that I could no longer take on any
translation jobs for Microsoft products, or even Windows-only products, as
I wouldn't be using Windows any more.
This decision (which cost me a significant percentage of my turnover in the
first year) was actually prompted by several translation jobs of books
dealing with Microsoft products. I was expected to tell readers that
Windows was a great platform, offering a stable and secure environment for
both application development and daily use. I was expressly forbidden to
make mention of problems, bugs or other shortcomings, other than in
suggestions along the lines of "It's always the user's fault." After a
steadily growing loathing for Microsoft and its products in general, and a
particularly harrowing experience with NT4, I was so disgusted by this
whole world of lies, deception and vaporware that I never wanted to promote
it in any way ever again. It's been one of the decisions in my life which I
haven't regretted for even a second.
So no, I don't take money supporting Windows in any way. The opposite is
true: dumping Windows at first cost me money; and now I take money while
helping other people get rid of Windows.
DFS
> This decision (which cost me a significant percentage of my turnover
> in the first year) was actually prompted by several translation jobs
> of books dealing with Microsoft products. I was expected to tell
> readers that Windows was a great platform, offering a stable and
> secure environment for both application development and daily use. I
> was expressly forbidden to make mention of problems, bugs or other
> shortcomings,
Forbidden by who? This sounds like a real cola bozo lie. Every book I've
ever read about Windows (or any software) mentions problems or bugs or
shortcomings or limitations, or all four.
> other than in suggestions along the lines of "It's
> always the user's fault."
Good practice for being a cola 'advocate'.
> After a steadily growing loathing for Microsoft and its products in
> general,
> and a particularly harrowing experience with NT4, I was so disgusted
> by this whole world of lies, deception and vaporware
What lies and deception and vaporware? Specifically, please.
The rest of the world enjoys using MS products, so much that they're the
world's largest software company.
> that I never wanted to promote it in any way
> ever again. It's been one of the decisions in my life which I haven't
> regretted for even a second.
I dunno about that. All you can think and talk about is Windows. You don't
even talk about Linux on comp.os.linux.advocacy.
Plus, you recently spent days trying to find a Vista machine you could use
for free (nutjob!) so you could write a 22-screenshot idiotic harangue about
DVD burning problems on Vista, which you later admit was partly due to your
wrong-headedness (I give you credit for that, few cola idiots admit it).
So you do have some kind of regret, since you continue to follow Windows.
> So no, I don't take money supporting Windows in any way. The opposite
> is true: dumping Windows at first cost me money; and now I take money
> while helping other people get rid of Windows.
Congrats on at least trying to walk the walk, even if Windows occupies your
thoughts. Now if you could only get rid of the bad habit of lying about
Windows and Linux.
Tim Smith
<4fa41a12-51e4-4161...@r60g2000hsc.googlegroups.com>,
I find a year hard to reach. On our desktops at work, and my server at
home, power failures are the main cause of rebooting. We get a few
storms a year that bring high winds, and there are a zillion trees
around here, and a lot of overhead power lines, so a couple times power
goes for a day or so.
On work servers that are on the internet (which are generally in a data
center with good power, so the wind storms don't cut power), kernel
updates get them, because (apparently unlike most "advocates" here,
judging from other responses) we actually think it is a good idea to
apply kernel updates that close remote exploits. :-(
On work servers that are not on the internet, power failures keep them
under a year usually, because the servers that aren't on the internet
are at our offices, not the data center.
We also sometimes do reboots if we've made major changes to the software
configuration on a server. E.g., if you take a server that had been a
web server, and change it to a database server, it is a good idea to do
a reboot, to make sure that it will come back up correctly. If you have
made a mistake that will prevent the server from coming back up
correctly, you want to find that out then, not months later. Imagine if
you take the server down six months later to install more RAM or a
bigger hard disk, and it doesn't come up correctly? There's a good
chance you'll assume that you broke it with the hardware update somehow,
and that will direct your investigation down fruitless lines.
--
--Tim Smith
Hadron
> Hadron wrote:
>
>> Richard Rasker <spam...@linetec.nl> writes:
>>
>>> I don't mind what they hook up behind my servers; they know I'm only paid
>>> to keep the servers up and running, and that any trouble with their
>>> Windows machines isn't my problem. For al the rest I don't insult Windows
>>> users; I feel sorry for them -- they're just victims of the criminal
>>> assholes who created Windows and foisted it onto the world.
>>
>> Yet you happily take money translating stuff which is created for that
>> very platform.
>
> You couldn't be more wrong. Ten years ago, when I dumped Windows in favour
> of Linux, I told my customers that I could no longer take on any
> translation jobs for Microsoft products, or even Windows-only products, as
> I wouldn't be using Windows any more.
Wow. How incredibly silly.
>
> This decision (which cost me a significant percentage of my turnover in the
> first year) was actually prompted by several translation jobs of books
> dealing with Microsoft products. I was expected to tell readers that
> Windows was a great platform, offering a stable and secure environment for
> both application development and daily use. I was expressly forbidden to
> make mention of problems, bugs or other shortcomings, other than in
> suggestions along the lines of "It's always the user's fault." After a
Pinocchio has nothing on you. I have been developing SW at various
levels for years and we never, ever boast about the platform. Only our
particular SW which obviously needs to gain a market niche so we all get
paid.
> steadily growing loathing for Microsoft and its products in general, and a
> particularly harrowing experience with NT4, I was so disgusted by this
> whole world of lies, deception and vaporware that I never wanted to promote
> it in any way ever again. It's been one of the decisions in my life which I
> haven't regretted for even a second.
I don't believe that either.
>
> So no, I don't take money supporting Windows in any way. The opposite is
> true: dumping Windows at first cost me money; and now I take money while
> helping other people get rid of Windows.
How do you help people dump windows? Seriously.
Night0wl
>{snip}
> How do you help people dump windows? Seriously.
Hadron? Don't be an idiot. Reformat their hard drives, load Linux
distro of choice, and spend maybe a whole 15 minutes showing them how to
use it. That's twice as long as it took me to learn the gui side. The
rest takes a bit longer, but, if the user is a non-tech, he/she just
wants something that works. And Linux DOES sell to frustrated Windoze
users, make no mistake!
Cheers;
Ed
--
History is three men in a room agreeing to a lie.
Jerry McBride
> In article
> <4fa41a12-51e4-4161...@r60g2000hsc.googlegroups.com>,
> "allen....@gmail.com" <allen....@gmail.com> wrote:
>> what is the longest you have had your linux system running without a
>> reboot?
>>
>> A friend of mine went over a year
>
> I find a year hard to reach. On our desktops at work, and my server at
> home, power failures are the main cause of rebooting. We get a few
> storms a year that bring high winds, and there are a zillion trees
> around here, and a lot of overhead power lines, so a couple times power
> goes for a day or so.
>
Buy yourself an UPS... it'll do wonders for your linux uptimes. If you're
not an aggressive kernel updater, you can find your linux boxes staying up
for years without a reboot... NO PROBLEM...
> On work servers that are on the internet (which are generally in a data
> center with good power, so the wind storms don't cut power), kernel
> updates get them, because (apparently unlike most "advocates" here,
> judging from other responses) we actually think it is a good idea to
> apply kernel updates that close remote exploits. :-(
>
> On work servers that are not on the internet, power failures keep them
> under a year usually, because the servers that aren't on the internet
> are at our offices, not the data center.
>
> We also sometimes do reboots if we've made major changes to the software
> configuration on a server. E.g., if you take a server that had been a
> web server, and change it to a database server, it is a good idea to do
> a reboot, to make sure that it will come back up correctly. If you have
> made a mistake that will prevent the server from coming back up
> correctly, you want to find that out then, not months later. Imagine if
> you take the server down six months later to install more RAM or a
> bigger hard disk, and it doesn't come up correctly? There's a good
> chance you'll assume that you broke it with the hardware update somehow,
> and that will direct your investigation down fruitless lines.
>
The point of all of this is... about the only time a linux box get's
rebooted is for security or feature updates for the kernel. In windows
land... reboots are REQUIRED to keep the os on an even keel. That's a BIG
difference.
--
Jerry McBride (jmcb...@mail-on.us)
Hadron
> On Sat, 22 Dec 2007 23:11:57 +0100, Hadron wrote:
>
>>{snip}
>
>> How do you help people dump windows? Seriously.
>
> Hadron? Don't be an idiot. Reformat their hard drives, load Linux
> distro of choice, and spend maybe a whole 15 minutes showing them how to
> use it. That's twice as long as it took me to learn the gui side. The
Don't be a total ninny. I was of course referring to real world
migration. Not how to remove it and install Linux.
Night0wl
> Night0wl <horne...@gmail.com> writes:
>>> How do you help people dump windows? Seriously.
>>
>> Hadron? Don't be an idiot. Reformat their hard drives, load Linux
>> distro of choice, and spend maybe a whole 15 minutes showing them how
>> to use it. That's twice as long as it took me to learn the gui side.
>> The
>
> Don't be a total ninny. I was of course referring to real world
> migration. Not how to remove it and install Linux.
How much do you know about Linux in general, and Ubuntu in particular?
I've used a single ISO disk to wipe a system and install Linux in about
30 minutes. In a corporate environment, with the right commands set, a
complete network push, including reboots should take about an hour; and
then, updates can be accomplished at night when the systems are on, but
logged-out. That could take 2-3 hours. The human factor would involve
about three hours transitional training -- about what I currently give my
XP/Office 2003-to-Vista/Office 2007 students, if that long...
Hadron
> On Sun, 23 Dec 2007 00:02:34 +0100, Hadron wrote:
>
>> Night0wl <horne...@gmail.com> writes:
>
>>>> How do you help people dump windows? Seriously.
>>>
>>> Hadron? Don't be an idiot. Reformat their hard drives, load Linux
>>> distro of choice, and spend maybe a whole 15 minutes showing them how
>>> to use it. That's twice as long as it took me to learn the gui side.
>>> The
>>
>> Don't be a total ninny. I was of course referring to real world
>> migration. Not how to remove it and install Linux.
>
> How much do you know about Linux in general, and Ubuntu in particular?
A fair amount.
>
> I've used a single ISO disk to wipe a system and install Linux in about
> 30 minutes. In a corporate environment, with the right commands set, a
> complete network push, including reboots should take about an hour; and
> then, updates can be accomplished at night when the systems are on, but
> logged-out. That could take 2-3 hours. The human factor would involve
> about three hours transitional training -- about what I currently give my
> XP/Office 2003-to-Vista/Office 2007 students, if that long...
You still fail to understand what I mean.
Real world migration. Real businesses.
This means work processes, existing libraries of macros, format issues,
customised SW, networks, VPNs, shared printers etc etc etc.
That type of thing.
Yes, anyone who just uses computer to surf the web or knock up "home
accounts" on a simply spreadsheet can be up and running in no time.
Tim Smith
Jerry McBride <jmcb...@mail-on.us> wrote:
> > I find a year hard to reach. On our desktops at work, and my server at
> > home, power failures are the main cause of rebooting. We get a few
> > storms a year that bring high winds, and there are a zillion trees
> > around here, and a lot of overhead power lines, so a couple times power
> > goes for a day or so.
> >
>
> Buy yourself an UPS... it'll do wonders for your linux uptimes. If you're
> not an aggressive kernel updater, you can find your linux boxes staying up
> for years without a reboot... NO PROBLEM...
Well, generally, the time it takes them to restore power from a downed
line is longer than a UPS will keep a system running. Keep in mind that
when high winds are bringing down trees, there are generally multiple
outages.
Last one that got me at home, for example, it took them 24 hours to get
to it, and they had something like 100 crews out restoring power.
--
--Tim Smith
chrisv
>>
>> Yet you happily take money translating stuff which is created for that
>> very platform.
WTF is it with you pieces of shit who claim to object to advocates
working with Microsoft products? There is no dilemma there. Jobs and
advocacy are two different things.
You're a piece of shit, Quack. With corn in it.
Hadron
>> Hadron puked:
>>>
>>> Yet you happily take money translating stuff which is created for that
>>> very platform.
>
> WTF is it with you pieces of shit who claim to object to advocates
> working with Microsoft products? There is no dilemma there. Jobs and
> advocacy are two different things.
Poor chrisv. Still struggling with reality I see. The problem is quite
simple - COLA advocate says "windows users are windiots and morons" and
then go and use it themselves to make money while all the time
maintaining that MS have an "illegal monopoly". And yet you milk it
yourselves!!!
Hypocrites.
>
> You're a piece of shit, Quack. With corn in it.
Glass actually. I eat glass. That's how hard I am.
Tim Smith
chrisv <chr...@nospam.invalid> wrote:
> WTF is it with you pieces of shit who claim to object to advocates
> working with Microsoft products? There is no dilemma there. Jobs and
> advocacy are two different things.
It depends on what kind of advocate the person promoting Microsoft
products is. If they are a Free Software advocate, then they should not
be promoting non-free products. According to RMS, helping people run
non-free software (even so little as telling them where to get it) is
unethical. You would not argue that unethical behavior is fine if done
for work, would you?
But for those advocates who aren't Free Software advocates, but just
advocate Linux, and don't mind running proprietary software on Linux,
then there's no problem with them using and even promoting Windows for
work.
--
--Tim Smith
Rick
> In article <6NSdna69HZlBOvDa...@giganews.com>,
> chrisv <chr...@nospam.invalid> wrote:
>> WTF is it with you pieces of shit who claim to object to advocates
>> working with Microsoft products? There is no dilemma there. Jobs and
>> advocacy are two different things.
>
> It depends on what kind of advocate the person promoting Microsoft
> products is. If they are a Free Software advocate, then they should not
> be promoting non-free products. According to RMS, helping people run
> non-free software (even so little as telling them where to get it) is
> unethical. You would not argue that unethical behavior is fine if done
> for work, would you?
RMS cannot decide what is, and isn't, ethical for other another person
unless he knows what code of ethics that person subscribes to.
>
> But for those advocates who aren't Free Software advocates, but just
> advocate Linux, and don't mind running proprietary software on Linux,
> then there's no problem with them using and even promoting Windows for
> work.
Does Free Software advocacy not mean you promote the use of Free
Software? Can a person not promote the use of Free Software and use CSS
at the same time?
--
Rick
Jim Richardson
Hash: SHA1
On Sat, 22 Dec 2007 06:22:10 -0800 (PST),
allen....@gmail.com <allen....@gmail.com> wrote:
> what is the longest you have had your linux system running without a
> reboot?
>
> A friend of mine went over a year
At home? few months, we have power issues, even with the inverter.
At work? over two years, (720 days IIRC) then we moved data centers :)
most of the rest of the systems tend to go about 4-8 months before we
reach a "reboot needed" upgrade of kernel or hardware. Since we have a
redundant load balanced setup it's not a big deal to drop one out of
rotation at a time and handle it.
Uptime per se is not the thing, unplanned downtime, and downtime that
makes you dance around it is the problem. Like our exchange "cluster" at
work :(
Thank the heavens that Q2 next year is "dump exchange for something that
fscking *works* well" That and the WSUS server are the last bastions of
MS-Windows in the office server room, there's no MS-Windows in the DC of
course.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
iD8DBQFHbbLDd90bcYOAWPYRAowYAKCo4C/UU7ILQeeuS8PaokJI/KcXNACePqNb
oUoauijw8Us+TYucLg5m0fw=
=3oQd
-----END PGP SIGNATURE-----
--
Jim Richardson http://www.eskimo.com/~warlock
Disclaimer: Elvis would agree with me, but he's got dirt in his mouth.
Hadron
> On Sat, 22 Dec 2007 16:45:59 -0800, Tim Smith wrote:
>
>> In article <6NSdna69HZlBOvDa...@giganews.com>,
>> chrisv <chr...@nospam.invalid> wrote:
>>> WTF is it with you pieces of shit who claim to object to advocates
>>> working with Microsoft products? There is no dilemma there. Jobs and
>>> advocacy are two different things.
>>
>> It depends on what kind of advocate the person promoting Microsoft
>> products is. If they are a Free Software advocate, then they should not
>> be promoting non-free products. According to RMS, helping people run
>> non-free software (even so little as telling them where to get it) is
>> unethical. You would not argue that unethical behavior is fine if done
>> for work, would you?
>
> RMS cannot decide what is, and isn't, ethical for other another person
> unless he knows what code of ethics that person subscribes to.
And yet you and the COLA loonies can?
Bwahahahahahahaha.
>
>>
>> But for those advocates who aren't Free Software advocates, but just
>> advocate Linux, and don't mind running proprietary software on Linux,
>> then there's no problem with them using and even promoting Windows for
>> work.
>
> Does Free Software advocacy not mean you promote the use of Free
> Software? Can a person not promote the use of Free Software and use CSS
> at the same time?
Of course one can. But you seem to confuse the meanings.
You seem to think here that OSS is "Linux SW" (its not, more OSS runs on
Windows), and that CSS is "payed for SW on Linux". For sure
you think anyone using Windows is a moron. get your story straight.
Handover Phist
> what is the longest you have had your linux system running without a
> reboot?
>
> A friend of mine went over a year
Over 400 days, cant remember the exact number, but that was a server. I
reboot my desktop a couple of times a week to play a game or two on
Windows.
--
Coito ergo sum
Linonut
I'm kind of in the middle. I agree with RMS that non-free products tend
to enslave the user, and aren't as fixable as Free products. On the
other hand, while RMS would tell you to do without functionality rather
than use a proprietary product, I'm okay with using one if you really
need it or want the functionality. As long as you keep in mind that you
are on a leash of unknown length and unknown termination date.
Now, as far as unethical behavior done for work, it depends in part on
the seriousness of the effects of the behavior. And even then, you may
have to do the work if the only other choice is to starve. Obviously,
there's no formula you can apply to a situation.
In any case, DFS and Hadron are using the dilemma to support their
jeering. That is all.
--
Tux rox!
Hadron
> * Tim Smith fired off this tart reply:
>
>> In article <6NSdna69HZlBOvDa...@giganews.com>,
>> chrisv <chr...@nospam.invalid> wrote:
>>> WTF is it with you pieces of shit who claim to object to advocates
>>> working with Microsoft products? There is no dilemma there. Jobs and
>>> advocacy are two different things.
>>
>> It depends on what kind of advocate the person promoting Microsoft
>> products is. If they are a Free Software advocate, then they should not
>> be promoting non-free products. According to RMS, helping people run
>> non-free software (even so little as telling them where to get it) is
>> unethical. You would not argue that unethical behavior is fine if done
>> for work, would you?
>>
>> But for those advocates who aren't Free Software advocates, but just
>> advocate Linux, and don't mind running proprietary software on Linux,
>> then there's no problem with them using and even promoting Windows for
>> work.
>
> I'm kind of in the middle. I agree with RMS that non-free products tend
> to enslave the user,
*snip*
"enslave"
Ridiculous.
Rick
> Rick <no...@nomail.com> writes:
>
>> On Sat, 22 Dec 2007 16:45:59 -0800, Tim Smith wrote:
>>
>>> In article <6NSdna69HZlBOvDa...@giganews.com>,
>>> chrisv <chr...@nospam.invalid> wrote:
>>>> WTF is it with you pieces of shit who claim to object to advocates
>>>> working with Microsoft products? There is no dilemma there. Jobs
>>>> and advocacy are two different things.
>>>
>>> It depends on what kind of advocate the person promoting Microsoft
>>> products is. If they are a Free Software advocate, then they should
>>> not be promoting non-free products. According to RMS, helping people
>>> run non-free software (even so little as telling them where to get it)
>>> is unethical. You would not argue that unethical behavior is fine if
>>> done for work, would you?
>>
>> RMS cannot decide what is, and isn't, ethical for other another person
>> unless he knows what code of ethics that person subscribes to.
>
> And yet you and the COLA loonies can?
>
> Bwahahahahahahaha.
Laughing at yourself?
>
>
>>
>>> But for those advocates who aren't Free Software advocates, but just
>>> advocate Linux, and don't mind running proprietary software on Linux,
>>> then there's no problem with them using and even promoting Windows for
>>> work.
>>
>> Does Free Software advocacy not mean you promote the use of Free
>> Software? Can a person not promote the use of Free Software and use CSS
>> at the same time?
>
> Of course one can. But you seem to confuse the meanings.
That is your incorrect inference.
>
> You seem to think here that OSS is "Linux SW"
I do not..
(> its not, more OSS runs on Windows),
Does it? Are you sure?
> and that CSS is "payed for SW on Linux".
I do not think that at all. That is, again, your incorrect inference.
> For sure you think anyone using Windows is a moron.
Do you never tire of being wrong?
> get your story straight.
It is straight.
--
Rick
Sinister Midget
> allen....@gmail.com <allen....@gmail.com> did eloquently scribble:
>> what is the longest you have had your linux system running without a
>> reboot?
>
>> A friend of mine went over a year
>
> It's rare though cos I tend to shutdown when I'm away from the house for
> more than a couple of days.
I have a server in the basement that reboots itself every hundred days.
I don't know why. I don't really care enough to bother looking for the
cause. It serves, but only certain things which are only important
other times of the day. It reboots itself at a time that doesn't bother a
thing because the only activity at 3:30 in the AM is this machine
gathering the mail from that one.
If not for that oddity it would have passed 1000 days at one point.
Currently it would be around 300 days instead of 11.
--
Behind a great man, there's a woman preventing him from being greater!
Rex Ballard
> 7: Linux is 50x faster than Windows
It depends on which version of Windows you are comparing to which
version of Linux. Microsoft published a benchmark in their "Fast
Facts" that compared a Linux 2.2 kernel with Windows NT 4.0 using 4
ethernet cards and for SCSI drives. This peculiar arrangement was
chosen to force Linux to spinlock more frequently, making it appear
slower.
Depending on the benchmark, the configuration, and the type of load
used, Microsoft has been able to show Windows faster in Bizarre
benchmarks in which Linux is deliberately crippled while Windows is
carefully tweaked.
In a "standard" configuration, running "standard" applications, Linux
can be as much as 50% faster for things like servers, because it does
not dedicate memory to graphics consoles and terminal servers. Linux
has faster memory management as well. In "Real World" environments,
most Windows machines are only assigned a single function, while Linux
machines typically perform multiple functions on a single machine.
Since Microsoft's Licenses expressly forbid the publication of
benchmarks without Microsoft's prior written permission, and Microsoft
often rewrites and revises the benchmark parameters, settings, and
statistical windows to get a more favorable result, Windows appears to
be faster in these "doctored" benchmarks.
In Germany, where such benchmark restrictions are treated as a form of
fraud, benchmarks have been published, and often show Linux as
substantially faster, as much as 50% faster in certain industry
standard benchmarks.
> Homer: Half of Europe has dumped Windows
I don't think you have yoru facts correct there eather.
> Roy Lying Spammer: 90% of IT pros reject Vista
Many corporate IT departments have taken the official stance NOT to
deploy Vista at this time. This doesn't mean that they will never
deploy Vista, but for now, they are ordering machines with Windows XP,
which meets their needs.
The bigger obstacle to Vista is not the IT department, but the Legal
department. Many of the new restrictions on Vista, even Vista
Business Edition, are a big concern. In addition, a license that
gives Microsoft permission to completely disable the machine if they
SUSPECT piracy, just doesn't play well, especially with companies who
had Volume License Manager licenses disabled when support programs
were not renewed.
How many companies are looking at Linux as a Desktop and Laptop
solution is anybody's guess. A computer world survey suggested that
85% of the respondents (self selected) were implementing plans for
Linux migration as a result of Microsoft's attempt to "Force Feed"
them XP. Many companies are still using Windows 2000.
When Microsoft attempted to force corporate customers to upgrade to
Windows 2000 or Windows 2003 by declaring Windows NT 4.0 to be
"obsolete and unsopported", they found that many of these older NT
systems were being converted to Linux or Unix, especially of the
applications were written in Java.
> Rasker: Vista requires 10x-20x the resources of Linux
Again, this depends on the version of Linux used, the configuration
used, and the applications being run. Slackware Linux running
standard Linux applications such as FireFox, OpenOffice, and other
common applications can run with as little as 256 Megabytes since
Linux makes more efficient use of swap and paging space. Vista
Ultimate, in 3D mode, running Office 2007, and new applications
written specifically for Vista - can push the memory requirements to 2
gigabytes. Almost 10x more memory.
> Ballard: 70% of Dell's recent PC sales are Ubuntu
You got mine totally wrong. I said 70% of Dell's PC sales are "Linux
Ready" machines that are CAPABLE of running Ubuntu.
DFS
> On Dec 22, 2:47 pm, "DFS" <nospam@dfs_.com> wrote:
>
>> 7: Linux is 50x faster than Windows
> In a "standard" configuration, running "standard" applications, Linux
> can be as much as 50% faster for things like servers, because it does
> not dedicate memory to graphics consoles and terminal servers.
Show me the config, the study, the benchmark, the hardware, the software,
who did it, etc.
> Linux
> has faster memory management as well. In "Real World" environments,
> most Windows machines are only assigned a single function, while Linux
> machines typically perform multiple functions on a single machine.
I don't believe this.
> Since Microsoft's Licenses expressly forbid the publication of
> benchmarks without Microsoft's prior written permission, and Microsoft
> often rewrites and revises the benchmark parameters, settings, and
> statistical windows to get a more favorable result, Windows appears to
> be faster in these "doctored" benchmarks.
How do you know they're doctored? Why wouldn't MS just lie about
everything, the way cola idiots claim they do?
> In Germany, where such benchmark restrictions are treated as a form of
> fraud, benchmarks have been published, and often show Linux as
> substantially faster, as much as 50% faster in certain industry
> standard benchmarks.
Show me the config, the study, the benchmark, the hardware, the software,
who did it, etc.
>> Homer: Half of Europe has dumped Windows
> I don't think you have yoru facts correct there eather.
Homer sure doesn't. It was his claim.
>> Roy Lying Spammer: 90% of IT pros reject Vista
> Many corporate IT departments have taken the official stance NOT to
> deploy Vista at this time.
Virtually every corporation in the world has an official stance against
desktop Linux.
> This doesn't mean that they will never
> deploy Vista, but for now, they are ordering machines with Windows XP,
> which meets their needs.
More are ordering Vista systems than XP, as MS has sold 80,000,000+ licenses
in under a year.
> How many companies are looking at Linux as a Desktop and Laptop
> solution is anybody's guess. A computer world survey suggested that
> 85% of the respondents (self selected) were implementing plans for
> Linux migration as a result of Microsoft's attempt to "Force Feed"
> them XP.
The old "Linux contingency plan" eh? My guess is you'll be spouting this
for the next 25 years.
> Many companies are still using Windows 2000.
How many? Which?
> When Microsoft attempted to force corporate customers to upgrade to
> Windows 2000 or Windows 2003 by declaring Windows NT 4.0 to be
> "obsolete and unsopported", they found that many of these older NT
> systems were being converted to Linux or Unix, especially of the
> applications were written in Java.
Who's 'they'?
>> Rasker: Vista requires 10x-20x the resources of Linux
>
> Again, this depends on the version of Linux used, the configuration
> used, and the applications being run.
All that matters is you compare two systems with similar functionality and
apps - that means at a minimum KDE. And no KDE system will run with 1/10th
of the 512mb Vista minimum.
> Slackware Linux running
> standard Linux applications such as FireFox, OpenOffice, and other
> common applications can run with as little as 256 Megabytes since
> Linux makes more efficient use of swap and paging space. Vista
> Ultimate, in 3D mode, running Office 2007, and new applications
> written specifically for Vista - can push the memory requirements to 2
> gigabytes. Almost 10x more memory.
In DFS math 2gb/256mb = 8x. Regardless, it's horseshit of the highest
order. Vista runs OK in 512mb. Vista/Aero and Office 2007 runs fine in
1gb.
>> Ballard: 70% of Dell's recent PC sales are Ubuntu
> You got mine totally wrong. I said 70% of Dell's PC sales are "Linux
> Ready" machines that are CAPABLE of running Ubuntu.
I was wrong. Your claim was 70% of recent Dell shipments were XP.
Rex Ballard
> RexBallardwrote:
> > On Dec 22, 2:47 pm, "DFS" <nospam@dfs_.com> wrote:
>
> >> 7: Linux is 50x faster than Windows
> Show me the config, the study, the benchmark, the hardware, the software,
> who did it, etc.
The original Mindcraft benchmark published in 1997, showed Linux as
being about 50% faster than NT 4.0.
There is a fundamental design principle of Microsoft which assumes
that MIPs are Cheap, Memory is Cheap, and Disk drive is relatively
cheap. Microsoft looks for ways to utilize as much of each as they
can, and offer some sort of value for the extra effort. As a result,
Microsoft throws away extra CPU time, memory swaps, and drive accesses
for things like animated puppy dogs, dancing paper clips, Animated
backgrounds, animated screen savers, and other memory and CPU
consuming activities based on the assumption that people who want the
latest version of Windows/Vista will also want the newest and fastest
hardware.
Microsoft has slower context switching as well. This works well for
desktops that are mostly designed to run one foreground application
which has threads waiting for events triggered by mouse activity. It
works against more complex applications and systems where an
application may be split into multiple different independent
processes. This changes the programming style required to get
performance. Microsoft applications have to be written as huge
monolithic applications where a single program has to do everything
and call several huge DLLs and load them into memory, then switch
between threads within the same huge process. You have to load
everything that program could ever want to do and initialize it,
before you can actually start doing work.
Microsoft applications also work with large application objects, such
as COM objects for spreadsheets, word documents, powerpoint slides,
and so on. As a result, these huge monolithic applications have huge
blobs of memory associated with application data. The result is some
heavy memory usage. For a given amount of memory, Windows will begin
to thrash and swap memory much earlier than Linux, and garbage
collection can be very time consuming.
Linux on the other hand, has a heritage of being oriented toward older
and therefore slower and smaller PCs. It follows the UNIX tradition
of attempting to be as efficient as possible with memory, cpu cycles,
and hard drive, as possible.
Linux has a very fast context switching system, which makes it
possible to have much smaller and simpler processes doing fundamental
workloads as services rather than having to load up unneeded
functionality. The applications do take a bit longer to start,
because Linux applications initialize their registry (X11 Resource
Database) from text files, but once they are initialized, they are
quite fast and responsive.
Linux is also based on Open Source libraries, which means that there
is less tendency to "roll your own" or write wrappers to be compatible
with other systems, or because you don't understand how a library
works. There are some very standard toolkits available, and the
library programmers work to make their programs as secure and stable
as possible. Since many of these libraries are also used in
production UNIX environments, they are often as stable and secure as
production UNIX systems. In addition, they are also optimized for
performance, since, in server environments, a 10% savings in
performance by optimizing a few loops and memory accesses can lead to
an increase in the number of customers that can be served on a web
site.
Linux needs much less memory for execution because it uses pipelines,
streams, and message queues as their primary interfaces between
programs. Even the interface between graphical applications and the
display is through a stream of "X-wire" commands, each very small,
simple, and independent of other commands. The result is that
applications don't have to allocate and free huge chunks of memory.
If there is free memory available, Linux uses it to accelerate disk
accesses. When Linux is used with SCSI or SAS hard drive,
applications can queue multiple requests and have them satisfied by
reading as much as an entire cylinder in a single rotation. Linux
will even determine the optimal access size and adapt it's reading to
this size. This is especially useful when reading sequential files
such as applications, text files, or XML files, since the drive
controller can have the memory queued up while the main processor(s)
are working on the stream.
Of course, one of the problems with doing benchmarks between Windows
and Linux is that standard versions of Windows are incapable of
running the standard applications used to benchmark Linux and UNIX
systems. On the other hand, Windows applications use Microsoft's
proprietary APIs, and therefore benchmarks used to compare Windows
systems can't be used to run benchmarks of Linux systems.
When Microsoft wants a favorable result compared to Linux or Unix,
they use such benchmarks as Web Server benchmarks, but they compare
ISAPI on Windows to CGI calls on Apache Linux or Unix. Rarely do they
do comparisons to Module interfaces.
Java based benchmarks can be run on both machines, but the VM does so
much of the work, that it really doesn't show the strengths or
weaknesses of the host operating systems. Still, the extra disk
buffering in Linux or Unix, the extra fast context switching, and the
efficient use of memory tend to make it possible to run multiple Java
processes on a single machine.
> > Linux
> > has faster memory management as well. In "Real World" environments,
> > most Windows machines are only assigned a single function, while Linux
> > machines typically perform multiple functions on a single machine.
>
> I don't believe this.
You don't have much experience with using Linux and/or UNIX in real
world production environments either, do you?
You have already stated that the bulk of your career has been focused
on writing VB front-ends to SQL Server back-ends. It's a good skill
to have, and it pays well, but it does give you a limited view into an
entire corporate architecture.
> > Since Microsoft's Licenses expressly forbid the publication of
> > benchmarks without Microsoft's prior written permission, and Microsoft
> > often rewrites and revises the benchmark parameters, settings, and
> > statistical windows to get a more favorable result, Windows appears to
> > be faster in these "doctored" benchmarks.
>
> How do you know they're doctored? Why wouldn't MS just lie about
> everything, the way cola idiots claim they do?
There was a very extensive COLA thread related to the MindCraft
benchmarks, and several other threads related to several other "Fast
Facts" benchmarks, discussing exactly how Microsoft's benchmarks were
altered to be different from the industry standard benchmarks. I
don't think I need to re-hash them here.
> > In Germany, where such benchmark restrictions are treated as a form of
> > fraud, benchmarks have been published, and often show Linux as
> > substantially faster, as much as 50% faster in certain industry
> > standard benchmarks.
>
> Show me the config, the study, the benchmark, the hardware, the software,
> who did it, etc.
Perhaps someone a little better versed in German can do this. The
problem is that the benchmarks cannot be translated into English
(German law only requires that uncensored benchmarks be written in
German).
> >> Roy Lying Spammer: 90% of IT pros reject Vista
> > Many corporate IT departments have taken the official stance NOT to
> > deploy Vista at this time.
>
> Virtually every corporation in the world has an official stance against
> desktop Linux.
Really? Care to give me some examples?
Many corporations have official stances as to what is required to use
desktop Linux. Many corporations have a migration plan, which is in
various stages of implementation. Many corporations have restrictions
on HOW Linux can be used. For example, IBM permits the use of Linux,
they even provide a standard "Corporate Images" (Client for E-Business
aka C4EB), and if an employee wants to install a different
distribution he has to install certain applications and auditing tools
to be sure that the laptop doesn't get configured as a firewall
breach, but they have a very aggressive pro-linux policy.
Other companies have certain applications that need to be run when you
are running Linux, such as the CISCO VPN application. You have to
download it from CISCO, and get the proper keys from the company.
Many companies now encourage the use of virtualized desktops for those
who want to use Linux. There are several vendors who now use VMWare
Player virtual images to demonstrate their products.
> > This doesn't mean that they will never
> > deploy Vista, but for now, they are ordering machines with Windows XP,
> > which meets their needs.
>
> More are ordering Vista systems than XP, as MS has sold
> 80,000,000+ licenses in under a year.
But Microsoft has publicly admitted that the way that OEMs are
permitted to offer XP Professional is by "upgrading" from Vista Home
Edition, to Vista Business Edition, then shipping the "Downgrade" to
XP Professional. This means that each copy of XP is actually 2 Vista
licenses.
Most of the major OEMs are reporting that a much higher proportion of
the machines they sell are sold with XP than with Vista - compared to
the proportions reported by Microsoft.
Keep in mind that the OEMs are purchasing huge bulk quantities of
Vista Home edition, and getting various discounts on the upgrade to
Vista Business edition, which is what they have to license in order to
ship Windows XP Professional - which is what most corporate customers
seem to be ordering.
> > How many companies are looking at Linux as a Desktop and Laptop
> > solution is anybody's guess. A computer world survey suggested that
> > 85% of the respondents (self selected) were implementing plans for
> > Linux migration as a result of Microsoft's attempt to "Force Feed"
> > them XP.
>
> The old "Linux contingency plan" eh? My guess is you'll be spouting this
> for the next 25 years.
Or until Linux is adopted on the desktop, or until Microsoft realizes
that it's position is not as solid as it used to be and stops trying
to "force feed" their newest stuff even when it isn't wanted.
> > Many companies are still using Windows 2000.
>
> How many? Which?
You want a hit list for the Microsoft' marketing department? I don't
think so.
I'll just say several major banks, several insurance companies, and
several government agencies that I have worked with in the last 3
years are still using Windows 2000 instead of XP. All were on the
Fortune 500.
> > When Microsoft attempted to force corporate customers to upgrade to
> > Windows 2000 or Windows 2003 by declaring Windows NT 4.0 to be
> > "obsolete and unsopported", they found that many of these older NT
> > systems were being converted to Linux or Unix, especially of the
> > applications were written in Java.
>
> Who's 'they'?
Let's just say, the orders weren't up to expectations, because many of
these machines were running WebSphere, Java, and other portable
applications which were now available on Linux. The Linux licenses
were cheaper than the Windows Enterprise Licenses, so the machines
were flipped to Linux.
These days, when we tell the customer, "You can use Either Windows or
Linux servers, we support both", they almost always say "Linux", often
without hesitation.
More and more companies are insisting that the desktop interfaces
support FireFox as well as IE, because they want to reach Linux users
as well as Windows users who use FireFox, and Mac Users.
> >> Rasker: Vista requires 10x-20x the resources of Linux
>
> > Again, this depends on the version of Linux used, the configuration
> > used, and the applications being run.
>
> All that matters is you compare two systems with similar functionality
> and apps - that means at a minimum KDE. And no KDE system will
> run with 1/10th of the 512mb Vista minimum.
Have you actually run Vista with 512 MB? I think even Microsoft is
now recommending 2 gig for best performance.
And why do you assume that Linux MUST have a 3D full-featured KDE
interface? There are "performance" settings on KDE that can reduce
memory requirements, there is also Gnome, which has settings to make
it very memory efficient. FVWM will run on very small machines.
There are versions of Linux such as Damn Small Linux, Puppy Linux,
Embedded Linux, and Slackware Linux, that can run on very small memory
and disk drives.
I have a friend - Fan Ying Jen, who is still running FreeBSD on his
80486 with 6 SCSI drives, and 32 megabytes of memory (it's all his
motherboard will hold). He uses it as a web server. FreeBSD isn't
actually UNIX, but it's based on the same principles, Open Source
kernel, Open Source Libraries, and Open Source applications. Since it
doesn't need a cooling fan, he has had up-times of over 2 years. He
has been using this machine with Linux and FreeBSD since 1995.
> In DFS math 2gb/256mb = 8x. Regardless, it's horseshit of the highest
> order. Vista runs OK in 512mb. Vista/Aero and Office 2007 runs fine in
> 1gb.
I will leave others to challenge this one? Anyone out there having a
wonderful time with Vista on 512 Mb? Is it a bit like trying to run
XP on 128 Mb?
> >>Ballard: 70% of Dell's recent PC sales are Ubuntu
> > You got mine totally wrong. I said 70% of Dell's PC sales are "Linux
> > Ready" machines that are CAPABLE of running Ubuntu.
>
> I was wrong. Your claim was 70% of recent Dell shipments were XP.
I think all I said was that according to Dell's statement to CNBC
shortly after Microsoft reported their earnings was that "Most" of the
machines they were selling were being shipped with XP. That would
be some number over 50%.
I did notice that about 70% of the models being offered by Dell were
"Linux Ready".
The same is true of PCs made by HP, Lenovo, Toshiba, and Acer.
It may be that those "Linux Ready" machines are selling better than
the "Windows only" machines, which means that actual shipments of
Linux-Ready machines could be much higher.
How many of these machines will actually be converted to Linux?
Who knows?
Speculation ranges from 1% to around 50% of these buyers will either
add Linux as a VMWare client, or will convert Windows to a VMX file
and install Linux as the client, or will convert the older machine
being replaced by the newer machine to Linux, or install numerous OSS
applications on their new machine.
We know that they get shipped to the customer with some version of
Windows XP or Vista pre-installed. Microsoft will not hesitate to
tell us that ALL of these licenses are in fact VISTA Licenses (XP
users can upgrade to Vista Business for free if they want?).
Microsoft's goal is to make sure that every home has 1 PC that is
licensed for Windows, and that each Office worker has 1 PC that is
licensed for Windows. It doesn't really matter with it is running
Windows or Linux, because Microsoft got their money, and their
bragging rights.
Linux on the other hand, is available, for free, without restrictions,
to anyone who wants it. The 12 year old kid who can't afford a Vista
machine, but his dad's hand-me-down Windows 95 machine or Windows 98
machine just doesn't have the glamour. The 15 year old kid who wants
to learn more about how a computer works and takes dad's old Windows
XP machine and converts it to Linux in Dual-boot mode. Or the
corporate office worker who has learned that a few PERL scripts or
shell commands can save a lot of time and has installed CYGWIN or has
Linux as a virtual client.
And then there are the people who purchased a dual-core 64 bit
processor, with higher resolution WSXGA display (1080p). They bought
the machine to run Linux, but that XP license will also be put to good
use as a virtualized client. They might use VMWare Converter to
create a vmx file and vmware player to run XP when they need it. They
want "bang for the buck".
And meanwhile, there are all those people out there who are choosing
not to wait for Dell, HP, or Acer to come up with what they want, and
have decided "I'd rather have a Mac".
Rex Ballard
wrote:
> what is the longest you have had your linux system running without a
> reboot?
>
> A friend of mine went over a year
I've had a few servers that went 2 years without a reboot.
There are reports Linux servers going as long as 4 years without a
reboot.
The biggest problem is that Linux kernel upgrades are more frequent
than BSD, and since the system needs to be rebooted to install the new
kernel, it's hard to resist the temptation to reboot to get a 20%
increase in speed, or 200% increase in memory capacity or file size.
Linonut
>> > On Dec 22, 2:47 pm, "DFS" <nospam@dfs_.com> wrote:
>>
>> >> 7: Linux is 50x faster than Windows
>
>> Show me the config, the study, the benchmark, the hardware, the software,
>> who did it, etc.
>
> The original Mindcraft benchmark published in 1997, showed Linux as
> being about 50% faster than NT 4.0.
1. What a difference a "%" versus an "x" makes.
2. Mindcraft's 1999 benchmark test concluded that
Microsoft Windows NT Server 4.0 is 2.5 times faster than Linux
as a File Server and 3.7 times faster as a Web Server
Although that study had serious problems (e.g. see
http://badtux.org/home/eric/editorial/mindcraft.php), it did
chagrin Linus quite a bit, and a lot of improvements were made to
update kernel 2.2 to 2.4, and then to 2.6 (where Linus seems to
feel pretty satisfied with performance even relative to the
BSDs.)
3. How about a Mindcraft test this year, hmmmm? Got another tilted
comparision for 2003:
http://www.linuxtoday.com/news_story.php3?ltsn=2003-05-08-019-26-NW-MS-RH&tbovrmode=1
"Windows Server 2003 delivered significantly better File server
throughput compared to Red Hat Linux Advanced Server 2.1 and
Red Hat Linux 8.0 on the configurations we tested.
But not for 2007.
> There is a fundamental design principle of Microsoft which assumes
> that MIPs are Cheap, Memory is Cheap, and Disk drive is relatively
> cheap. Microsoft looks for ways to utilize as much of each as they
> can, and offer some sort of value for the extra effort. As a result,
> Microsoft throws away extra CPU time, memory swaps, and drive accesses
> for things like animated puppy dogs, dancing paper clips, Animated
> backgrounds, animated screen savers, and other memory and CPU
> consuming activities based on the assumption that people who want the
> latest version of Windows/Vista will also want the newest and fastest
> hardware.
What Intel giveth, Microsoft taketh away.
(Even Linux code does that, too, though.)
<other wide ranging, sometimes iffy stuff snipped>
> You don't have much experience with using Linux and/or UNIX in real
> world production environments either, do you?
>
> You have already stated that the bulk of your career has been focused
> on writing VB front-ends to SQL Server back-ends. It's a good skill
> to have, and it pays well, but it does give you a limited view into an
> entire corporate architecture.
Noted.
>> Virtually every corporation in the world has an official stance against
>> desktop Linux.
>
> Really? Care to give me some examples?
Noted.
>> More are ordering Vista systems than XP, as MS has sold
>> 80,000,000+ licenses in under a year.
>
> But Microsoft has publicly admitted that the way that OEMs are
> permitted to offer XP Professional is by "upgrading" from Vista Home
> Edition, to Vista Business Edition, then shipping the "Downgrade" to
> XP Professional. This means that each copy of XP is actually 2 Vista
> licenses.
Odd.
> I'll just say several major banks, several insurance companies, and
> several government agencies that I have worked with in the last 3
> years are still using Windows 2000 instead of XP. All were on the
> Fortune 500.
I doubt these guys have a real need for the multimedia advances of Win
XP, anyway <grin>.
> I have a friend - Fan Ying Jen, who is still running FreeBSD on his
> 80486 with 6 SCSI drives, and 32 megabytes of memory (it's all his
> motherboard will hold). He uses it as a web server. FreeBSD isn't
> actually UNIX, but it's based on the same principles, Open Source
> kernel, Open Source Libraries, and Open Source applications. Since it
> doesn't need a cooling fan, he has had up-times of over 2 years. He
> has been using this machine with Linux and FreeBSD since 1995.
That must be hard on the uptime ;->
> Microsoft's goal is to make sure that every home has 1 PC that is
> licensed for Windows, and that each Office worker has 1 PC that is
> licensed for Windows. It doesn't really matter with it is running
> Windows or Linux, because Microsoft got their money, and their
> bragging rights.
Let 'em brag.
--
Tux rox!
SW
>
> The biggest problem is that Linux kernel upgrades are more frequent
> than BSD, and since the system needs to be rebooted to install the new
> kernel, it's hard to resist the temptation to reboot to get a 20%
> increase in speed, or 200% increase in memory capacity or file size.
Are you saying moving to kernel 2.6.23.12 from 2.6.18.8 will result in ..."a
20% increase in speed, or 200% increase in memory capacity or file size."?
Rex Ballard
No, but upgrading from a 1.2 kernel to a 1.4 then to 2.0 then 2.2
kernel can give some substantial improvements in performance - in
specific environments. 2.4 kernel added the queue based scheduler and
virtually eliminated the bottleneck of spinlocks. The 2.6 kernel
improved 64 bit performance and 64bit memory addressability, as well
as 64 bit file addressing.
The latter changes wouldn't make a word processor run much faster, but
if you are doing a database of photographic images, medical images, or
financial records, it can provide a substantial performance
improvement.
Keep in mind that there were similar evolutions in the Windows kernel,
as Windows 3.0 went to Windows 3.1 with improved support for larger
memory models, and then to Win32 with it's 32 bit memory model and 32
bit registers, and the transition from FAT to FAT16 to FAT32 and
NTFS. Each upgrade provided access to larger storage capacities and
larger file sizes.
There are still fundamental design differences between Windows and
Linux which makes Linux perform substantially faster on industry
standard benchmarks such as bytemark benchmarks.
Windows narrows the gap when benchmarks are based on a single process
multi-threaded application (IIS or SQL Server) using standard "Cluster
Sized blocks" (typically exact multiples of 4096) and compared against
functionally identical Linux application that uses standard
multitasking and standard file systems designed to optimally store and
retrieve sequential files and chunks of data in random lengths.
Windows has scalability to about 8 Intel processor cores, Linux has
scalability to 64 core ZVM Mainframes. Vista needs a minimum of 512
meg, 30 gigabytes of hard drive, and at least a 1 billion instruction
per second processor. Linux can scale down to systems the size of an
ethernet connector, with a few megabytes of memory and a few megabytes
of flash storage. There are even specialized Linux systems that can
perform such dedicated functions as terminal servers or PPP gateways.
Linux powered gateways, routers, WiFi hubs, Firewalls, Appliances,
Storage controllers, printer controllers, and DVRs all function very
well on very light horsepower.
If you have HDTV, there is a good chance that Linux or Unix is the
brains behind it. If you have a DVR, it's probably powered by Unix or
Linux. If you have a cable modem, it's probably powered by Linux or
Unix. If you have an "electronic picture frame" it's probably powered
by Linux or Unix. Even some USB drives, as well as most SAN drives,
and SAN controllers are powered by Linux or Unix.
It's becoming more and more common to see Linux devices in the home,
in the office, and in the computer room.
If you think about it, the amount of things you do that are done by
Windows, only by Windows, and can only be done by Windows, is very
rapidly shrinking.
Even such things as adding Visual Basic instructions to your Word
Processing or Spreadsheet program is also possible with Open Office on
Linux.
This is reducing the value of premium priced applications such as
Microsoft Office, Microsoft Project, and Microsoft Visio.
Handover Phist
> On Dec 26, 11:14 am, "SW" <s...@invalid.invalid> wrote:
>> Are you saying moving to kernel 2.6.23.12 from 2.6.18.8 will result in ..."a
>> 20% increase in speed, or 200% increase in memory capacity or file size."?
>
> No, but upgrading from a 1.2 kernel to a 1.4 then to 2.0 then 2.2
> kernel can give some substantial improvements in performance - in
> specific environments. 2.4 kernel added the queue based scheduler and
> virtually eliminated the bottleneck of spinlocks. The 2.6 kernel
> improved 64 bit performance and 64bit memory addressability, as well
> as 64 bit file addressing.
Rex:
I haven't read many of your posts simply because they're long and I have
little time. I'm glad I took the time to read that. Well said. You've
spanked your trolls well.
--
Conquest is easy. Control is not.
-- Kirk, "Mirror, Mirror", stardate unknown
SW
>
> Rex:
>
> I haven't read many of your posts simply because they're long and I
> have little time. I'm glad I took the time to read that. Well said.
> You've spanked your trolls well.
linux-1.2.13.tar.gz 02-Aug-1995
linux-1.3.100.tar.bz2 10-May-1996
no 1.4.x kernel**
linux-2.0.8.tar.gz 25-Jul-1996
linux-2.2.12.tar.bz2 26-Aug-1999
linux-2.4.0.tar.gz 04-Jan-2001
linux-2.6.0.tar.bz2 18-Dec-2003
Rex uses 500 words to say, "Software evolves over time." and you call it
spanking. You're a freaking goofball.
spi...@freenet.co.uk
> <allen....@gmail.com> wrote in message
> news:4fa41a12-51e4-4161...@r60g2000hsc.googlegroups.com...
>> what is the longest you have had your linux system running without a
>> reboot?
>>
>> A friend of mine went over a year
> On most linux installs "X" (the gui) locks up very frequently because of
> poor driver support with modern video cards. Now people will argue that
> linux is still alive and running and all I need to do is to telnet or ssh
> in from another machine and I can restart X.
Not even that. The kernel intercepts some keyboard combinations and acts on
them no-matter what state the machine's in (apart from the most severe
complete hardware lockup).
alt-magic sysrq
You can kill the process on the current tty, which means, totally kill X
with extreme prejudice and then recover. Or even kill all active processes
apart from init. (then recover from that later whilst maintaining your
uptime)
Of course, it can also be used to SAFELY shutdown or reboot by forcing a
disk sync and remount before doing so.
Great except that I don't have
> another machine to do this with. The fact is that if the mouse, keyboard
> and screen are frozen solid then for all practical purposes, the OS is
> locked up solid and needs a hard reset. Just hope the filesystem survives
> the hard reboot without corrupting itself.
--
-----------------------------------------------------------------------------
| spi...@freenet.co.uk | Windows95 (noun): 32 bit extensions and a |
| | graphical shell for a 16 bit patch to an 8 bit |
|Andrew Halliwell BSc(hons)| operating system originally coded for a 4 bit |
| in |microprocessor, written by a 2 bit company, that|
| Computer Science | can't stand 1 bit of competition. |
-----------------------------------------------------------------------------
DFS
> On Dec 24, 2:22 pm, "DFS" <nospam@dfs_.com> wrote:
>> RexBallardwrote:
>>> On Dec 22, 2:47 pm, "DFS" <nospam@dfs_.com> wrote:
>>
>>>> 7: Linux is 50x faster than Windows
>
>> Show me the config, the study, the benchmark, the hardware, the
>> software, who did it, etc.
>
> The original Mindcraft benchmark published in 1997, showed Linux as
> being about 50% faster than NT 4.0.
I don't need another restatement of your wacko interpretation. I want all
the details you claim to remember, but shown to me by a 3rd party. You're
extremely dishonest and unreliable.
>>> Linux
>>> has faster memory management as well. In "Real World" environments,
>>> most Windows machines are only assigned a single function, while
>>> Linux machines typically perform multiple functions on a single
>>> machine.
>>
>> I don't believe this.
>
> You don't have much experience with using Linux and/or UNIX in real
> world production environments either, do you?
No. I still don't believe it, and you still can't show me any evidence of
either of your claims.
> You have already stated that the bulk of your career has been focused
> on writing VB front-ends to SQL Server back-ends. It's a good skill
> to have, and it pays well, but it does give you a limited view into an
> entire corporate architecture.
I do more Oracle work than SQL Server. And my biggest client doesn't
support Oracle on Windows (only on Solaris).
>>> Since Microsoft's Licenses expressly forbid the publication of
>>> benchmarks without Microsoft's prior written permission, and
>>> Microsoft often rewrites and revises the benchmark parameters,
>>> settings, and statistical windows to get a more favorable result,
>>> Windows appears to be faster in these "doctored" benchmarks.
>>
>> How do you know they're doctored? Why wouldn't MS just lie about
>> everything, the way cola idiots claim they do?
>
> There was a very extensive COLA thread related to the MindCraft
> benchmarks, and several other threads related to several other "Fast
> Facts" benchmarks, discussing exactly how Microsoft's benchmarks were
> altered to be different from the industry standard benchmarks. I
> don't think I need to re-hash them here.
cola idiots have not one time, ever, disproven any MS 'Get The Facts'
claims.
>>> In Germany, where such benchmark restrictions are treated as a form
>>> of fraud, benchmarks have been published, and often show Linux as
>>> substantially faster, as much as 50% faster in certain industry
>>> standard benchmarks.
>>
>> Show me the config, the study, the benchmark, the hardware, the
>> software, who did it, etc.
>
> Perhaps someone a little better versed in German can do this. The
> problem is that the benchmarks cannot be translated into English
> (German law only requires that uncensored benchmarks be written in
> German).
If you know these German benchmarks show Linux to be up to 50% faster than
Windows, you should be able to direct me right to them. Why can't you?
>>>> Roy Lying Spammer: 90% of IT pros reject Vista
>>> Many corporate IT departments have taken the official stance NOT to
>>> deploy Vista at this time.
>>
>> Virtually every corporation in the world has an official stance
>> against desktop Linux.
>
> Really? Care to give me some examples?
http://money.cnn.com/magazines/fortune/fortune500/
> Many corporations have official stances as to what is required to use
> desktop Linux. Many corporations have a migration plan, which is in
> various stages of implementation. Many corporations have restrictions
> on HOW Linux can be used. For example, IBM permits the use of Linux,
> they even provide a standard "Corporate Images" (Client for E-Business
> aka C4EB), and if an employee wants to install a different
> distribution he has to install certain applications and auditing tools
> to be sure that the laptop doesn't get configured as a firewall
> breach, but they have a very aggressive pro-linux policy.
I'd love to know the exact percentage of 'only-Linux' users at IBM. It's
not very many, that's for sure. IBM can't afford to have field sales and
marketing people, or system architects like yourself, show up at client
offices with OSS crapware wasting everyone's time.
> Many companies now encourage the use of virtualized desktops for those
> who want to use Linux.
Which companies? How many? How do you know?
> Most of the major OEMs are reporting that a much higher proportion of
> the machines they sell are sold with XP than with Vista - compared to
> the proportions reported by Microsoft.
Give us one major OEM that's reporting that. You can't, because none are.
It's another Ballard-esque claim that you pulled out of your ass.
>> The old "Linux contingency plan" eh? My guess is you'll be spouting
>> this for the next 25 years.
>
> Or until Linux is adopted on the desktop, or until Microsoft realizes
> that it's position is not as solid as it used to be and stops trying
> to "force feed" their newest stuff even when it isn't wanted.
The next 500 years, then. 'Cause Linux will never be adopted in significant
numbers on the desktop. It's not good enough, and MS isn't stupid and lazy
enough to let their business be taken over by free sludgeware.
>>> Many companies are still using Windows 2000.
>>
>> How many? Which?
>
> You want a hit list for the Microsoft' marketing department? I don't
> think so.
> I'll just say several major banks, several insurance companies, and
> several government agencies that I have worked with in the last 3
> years are still using Windows 2000 instead of XP. All were on the
> Fortune 500.
So 'many' really means 'a tiny few'. Thanks for admitting it. Finally.
>>>> Rasker: Vista requires 10x-20x the resources of Linux
>>
>>> Again, this depends on the version of Linux used, the configuration
>>> used, and the applications being run.
>>
>> All that matters is you compare two systems with similar
>> functionality and apps - that means at a minimum KDE. And no KDE
>> system will
>> run with 1/10th of the 512mb Vista minimum.
>
> Have you actually run Vista with 512 MB? I think even Microsoft is
> now recommending 2 gig for best performance.
I have. It runs decently, but I also recommend 2gb - which is an excessive
requirement for a desktop OS. Luckily, memory is dirt-cheap.
> And why do you assume that Linux MUST have a 3D full-featured KDE
> interface?
Because only KDE is close to Vista in features and functionality.
>> In DFS math 2gb/256mb = 8x. Regardless, it's horseshit of the
>> highest order. Vista runs OK in 512mb. Vista/Aero and Office 2007
>> runs fine in 1gb.
>
> I will leave others to challenge this one? Anyone out there having a
> wonderful time with Vista on 512 Mb?
Doubtful. I said OK. Some things are quite snappy, others lag.
> Microsoft's goal is to make sure that every home has 1 PC that is
> licensed for Windows, and that each Office worker has 1 PC that is
> licensed for Windows.
They aimed high, and basically achieved it.
> It doesn't really matter with it is running
> Windows or Linux, because Microsoft got their money, and their
> bragging rights.
Cool. They deserve to brag.
> Linux on the other hand, is available, for free, without restrictions,
> to anyone who wants it. The 12 year old kid who can't afford a Vista
> machine, but his dad's hand-me-down Windows 95 machine or Windows 98
> machine just doesn't have the glamour.
It won't be an enjoyable experience, no matter what lightweight OSS apps you
put on it.
> Or the
> corporate office worker who has learned that a few PERL scripts or
> shell commands can save a lot of time and has installed CYGWIN or has
> Linux as a virtual client.
I don't think the corporate office worker will be a perl programmer.
> And meanwhile, there are all those people out there who are choosing
> not to wait for Dell, HP, or Acer to come up with what they want, and
> have decided "I'd rather have a Mac".
Anything but 'free' Linux.
DFS
> What Intel giveth, Microsoft taketh away.
>
> (Even Linux code does that, too, though.)
See OpenOffice.org: 10 seconds to launch on a supercomputer.
>> Microsoft's goal is to make sure that every home has 1 PC that is
>> licensed for Windows, and that each Office worker has 1 PC that is
>> licensed for Windows. It doesn't really matter with it is running
>> Windows or Linux, because Microsoft got their money, and their
>> bragging rights.
>
> Let 'em brag.
They will, and they should.
What do you have to brag about?
Linonut
> Linonut wrote:
>
>> What Intel giveth, Microsoft taketh away.
>> (Even Linux code does that, too, though.)
>
> See OpenOffice.org: 10 seconds to launch on a supercomputer.
And you wonder why they brand you as DooFuS?
I read once where a computer compiled the whole kernel in 7 seconds.
OpenOffice Writer comes up ready to edit in two seconds on my desk
machine.
>>> Windows or Linux, because Microsoft got their money, and their
>>> bragging rights.
>>
>> Let 'em brag.
>
> They will, and they should.
>
> What do you have to brag about?
I don't believe in bragging, nor do I like it when others do it,
"justified" or not. The world is full of a vast range of combinations
of abilities and accomplishments, and it is unseemly to exaggerate the
importance of one or the other data points in that range.
--
Tux rox!
bugbuster
> Linonut wrote:
>
>> What Intel giveth, Microsoft taketh away.
>>
>> (Even Linux code does that, too, though.)
>
> See OpenOffice.org: 10 seconds to launch on a supercomputer.
>
Wow! I had no idea I had a supercomputer in the palm of my hand. My Eee
PC launches OpenOffice in 10 seconds on the first try and 7 seconds after
that.
Bug
DFS
> * DFS fired off this tart reply:
>
>> Linonut wrote:
>>
>>> What Intel giveth, Microsoft taketh away.
>>> (Even Linux code does that, too, though.)
>>
>> See OpenOffice.org: 10 seconds to launch on a supercomputer.
>
> And you wonder why they brand you as DooFuS?
They're lazy and dimwitted?
> I read once where a computer compiled the whole kernel in 7 seconds.
Tell it to ubergeek Jerry McBride, who desperately "wanted to spend time"
with a fast computer.
> OpenOffice Writer comes up ready to edit in two seconds on my desk
> machine.
It takes 12 seconds to cold launch on my P4, 3.0ghz system w/ 2gb of RAM and
a fast SATA drive. It's pure OSS crapooolllaaaa.
> I don't believe in bragging, nor do I like it when others do it,
> "justified" or not. The world is full of a vast range of combinations
> of abilities and accomplishments, and it is unseemly to exaggerate the
> importance of one or the other data points in that range.
Don't call it bragging then. What do you have to be proud of, the way MS
has their software empire to be proud of?
DFS
Amazing, isn't it? Slow To OpenOffice is as bogus on your toy as it is on
BlueGene/L.
bugbuster
What's amazing is how dense you are. How long does it take for MS Office
to start on BlueGene/L? My guess is about 50 years. Do you really think
office apps are optimized to run on massively parallel supercomputers.
Do you really think that massively parallel supercomputers are optimized
for running office apps? Do you really think that office start up is cpu
intensive? And why is it that OpenOffice starts up faster on my "toy"
running 630Mhz and 1GB of RAM than your 3GHz P4 with 2GB RAM? What
version of OpenOffice? Was that running on Windows or Linux? Or have
you once again forgotten whether you even tried it before making the
claim.
Bug
Linonut
> It takes 12 seconds to cold launch on my P4, 3.0ghz system w/ 2gb of RAM and
> a fast SATA drive. It's pure OSS crapooolllaaaa.
Maybe you should try running it on 64-bit Linux, then.
--
Tux rox!
Sinister Midget
> What's amazing is how dense you are. How long does it take for MS Office
> to start on BlueGene/L? My guess is about 50 years. Do you really think
> office apps are optimized to run on massively parallel supercomputers.
> Do you really think that massively parallel supercomputers are optimized
> for running office apps? Do you really think that office start up is cpu
> intensive? And why is it that OpenOffice starts up faster on my "toy"
> running 630Mhz and 1GB of RAM than your 3GHz P4 with 2GB RAM? What
> version of OpenOffice? Was that running on Windows or Linux? Or have
> you once again forgotten whether you even tried it before making the
> claim.
DuFuS hasn't forgotten whether he tried it. He already knows he didn't.
He's forgotten whether he previously lied about it, and fears he'll get
caught telling a totally different "experience" this time around.
--
When I was a child we had a quick-sand box in the backyard.
Rex Ballard
> RexBallardwrote:
> I don't need another restatement of your wacko interpretation. I want all
> the details you claim to remember, but shown to me by a 3rd party. You're
> extremely dishonest and unreliable.
Let's see what I can find:
Here's one
http://www.kegel.com/nt-linux-benchmarks.html
<quote>
"Samba 3 extends lead over Win 2003
By Roger Howorth [14-10-2003]
The latest Samba release shows Windows a clean pair of heels in file
and print peformance
Tests by IT Week Labs indicate that the latest version of the open-
source Samba file and print server software has widened the
performance gap separating it from the commercial Windows alternative.
The latest benchmark results show an improvement over [Samba 2], which
performed twice as fast as Windows 2000 Server when it was tested by
IT Week Labs last year. Overall, it now performs 2.5 times faster than
Windows Server 2003.
In terms of scalability, the gains of upgrading to Samba 3 are even
more striking. Last year we found that Samba could handle four times
as many clients as Windows 2000 before performance began to drop off.
This year we would need to upgrade our test network in order to
identify the point where Samba performance begins to fall in earnest.
</quote>
http://www.kegel.com/nt-linux-benchmarks.html#benchapril1999
<quote>
Sm@rt Reseller's January 1999 article, "Linux Is The Web Server's
Choice" said "Linux with Apache beats NT 4.0 with IIS, hands down."
The companion article said unequivocally "The bottom line, according
to our hands-on analysis, is that commercial Linux releases can do
much more with far less than Windows NT Server can." ... "According to
ZDLabs' results (see test charts), each of the commercial Linux
releases ate NT's lunch."
</quote>
<quote>
c't Magazin, June 1999
c't Magazin ran very interesting benchmarks of Linux/Apache and NT/IIS
on a quad Pentium 2 Xeon system. These tests used custom benchmark
software written by c't (available for download). Like WebBench, this
test used a small document tree (10,000 4KB files); unlike WebBench,
these tests also used a second document tree (1,000,000 4KB files)
that was too large to fit in main memory, which tests the disk
subsystem and caching behavior of the operating system.
See also IT Director's summary of the c't tests.
Here's their graph of performance on a single-CPU system on small sets
(10^4) and large sets (10^6) of files:
</quote>
Roy Schestowitz
Don't forget the notorious 'studies' from around that time where Linux and
Windows were tweaked to make Windows appear like a better choice. This started
a riot. Always follow the money. People are not earning money merely by
conducting studies. Watch the ignition and find out who started the fire. It's
rarely the free as in free beer side.
--
~~ Best of wishes
Roy S. Schestowitz | "I think I think, therefore I think I am"
http://Schestowitz.com | RHAT GNU/Linux | PGP-Key: 0x74572E8E
run-level 2 2007-12-10 11:12 last=
http://iuron.com - help build a non-profit search engine
Hadron
> Rex Ballard :
>> On Dec 26, 11:14 am, "SW" <s...@invalid.invalid> wrote:
>>> Are you saying moving to kernel 2.6.23.12 from 2.6.18.8 will result in ..."a
>>> 20% increase in speed, or 200% increase in memory capacity or file size."?
>>
>> No, but upgrading from a 1.2 kernel to a 1.4 then to 2.0 then 2.2
>> kernel can give some substantial improvements in performance - in
>> specific environments. 2.4 kernel added the queue based scheduler and
>> virtually eliminated the bottleneck of spinlocks. The 2.6 kernel
>> improved 64 bit performance and 64bit memory addressability, as well
>> as 64 bit file addressing.
>
> Rex:
>
> I haven't read many of your posts simply because they're long and I have
> little time. I'm glad I took the time to read that. Well said. You've
> spanked your trolls well.
How has he spanked anyone? All he has done is cut and paste the readme
from the past few kernels. I am yet to see any upgrade to anything which
doesnt include "improved performance in X,Y or Z".
I note that Rexx is doing the typically "high level overview" thing again.
SW
>
> No, but upgrading from a 1.2 kernel to a 1.4 then to 2.0 then 2.2
> kernel can give some substantial improvements in performance - in
Where can the 1.4 kernel be found Rex?
Paul Hovnanian P.E.
>
> On Dec 27, 11:17 pm, "DFS" <nospam@dfs_.com> wrote:
> > RexBallardwrote:
>
> > I don't need another restatement of your wacko interpretation. I want all
> > the details you claim to remember, but shown to me by a 3rd party. You're
> > extremely dishonest and unreliable.
>
> Let's see what I can find:
>
> Here's one
>
> http://www.kegel.com/nt-linux-benchmarks.html
> <quote>
> "Samba 3 extends lead over Win 2003
> By Roger Howorth [14-10-2003]
> The latest Samba release shows Windows a clean pair of heels in file
> and print peformance
Oh yeah? Well, here's something Windows can do that Samba doesn't:
--
Paul Hovnanian mailto:Pa...@Hovnanian.com
------------------------------------------------------------------
f u cn rd ths u r usng unx
Handover Phist
> Rex Ballard wrote:
>>
>> On Dec 27, 11:17 pm, "DFS" <nospam@dfs_.com> wrote:
>> > RexBallardwrote:
>>
>> > I don't need another restatement of your wacko interpretation. I want all
>> > the details you claim to remember, but shown to me by a 3rd party. You're
>> > extremely dishonest and unreliable.
>>
>> Let's see what I can find:
>>
>> Here's one
>>
>> http://www.kegel.com/nt-linux-benchmarks.html
>> <quote>
>> "Samba 3 extends lead over Win 2003
>> By Roger Howorth [14-10-2003]
>> The latest Samba release shows Windows a clean pair of heels in file
>> and print peformance
>
> Oh yeah? Well, here's something Windows can do that Samba doesn't:
>
> http://www.computerworld.com/action/article.do?command=viewArticleBasic&taxonomyId=154&articleId=9054178&intsrc=hm_topic
Application Error
An error has occured with the page you requested. If you continue having
problems, you can E-mail he...@computerworld.com for assistance or submit
a Help Desk inquiry by clicking here.
--
I can't decide which WRONG TURN to make first!! I wonder if BOB
GUCCIONE has these problems!
DFS
> On Fri, 28 Dec 2007 20:40:40 -0500, DFS wrote:
>> Amazing, isn't it? Slow To OpenOffice is as bogus on your toy as it
>> is on BlueGene/L.
>
> What's amazing is how dense you are. How long does it take for MS
> Office to start on BlueGene/L? My guess is about 50 years.
What's your guess based on?
> Do you
> really think office apps are optimized to run on massively parallel
> supercomputers.
> Do you really think that massively parallel supercomputers are
> optimized for running office apps?
ROFL! Is that the latest excuse for Slow-O?
"OpenOffice isn't optimized to run on computers! Why would you expect it
to launch in less than 12 seconds?"
> Do you really think that office start up is cpu intensive?
Quit giving them excuses. They're already lazy enough.
> And why is it that OpenOffice starts up
> faster on my "toy" running 630Mhz and 1GB of RAM than your 3GHz P4
> with 2GB RAM? What version of OpenOffice? Was that running on
> Windows or Linux? Or have you once again forgotten whether you even
> tried it before making the claim.
I've been trying that OpenOffice crud for years, on Windows and Linux.
Doesn't matter where you run it - it's a pale imitation of MS Office.
Except for its cost, it doesn't do a single thing better than the MS gold
standard.
Gregory Shearman
> bugbuster wrote:
>> On Fri, 28 Dec 2007 20:40:40 -0500, DFS wrote:
>
>>> Amazing, isn't it? Slow To OpenOffice is as bogus on your toy as it
>>> is on BlueGene/L.
>>
>> What's amazing is how dense you are. How long does it take for MS
>> Office to start on BlueGene/L? My guess is about 50 years.
>
> What's your guess based on?
>
>
>
>> Do you
>> really think office apps are optimized to run on massively parallel
>> supercomputers.
>> Do you really think that massively parallel supercomputers are
>> optimized for running office apps?
>
> ROFL! Is that the latest excuse for Slow-O?
>
> "OpenOffice isn't optimized to run on computers! Why would you expect it
> to launch in less than 12 seconds?"
12 seconds???? Fuck... what is it running on? An 80386?
A 4 second start max on my 2.4G P4... running a compiled Gentoo version of
OO.
It only takes about 2 seconds after the first start.
> I've been trying that OpenOffice crud for years, on Windows and Linux.
> Doesn't matter where you run it - it's a pale imitation of MS Office.
> Except for its cost, it doesn't do a single thing better than the MS gold
> standard.
Your opinions are irrelevant. If you don't like OO then don't use it... I
don't use it, but my partner swears by it.
--
Regards,
Gregory.
Gentoo Linux - Penguin Power
bugbuster
> bugbuster wrote:
>> On Fri, 28 Dec 2007 20:40:40 -0500, DFS wrote:
>
>>> Amazing, isn't it? Slow To OpenOffice is as bogus on your toy as it is
>>> on BlueGene/L.
>>
>> What's amazing is how dense you are. How long does it take for MS
>> Office to start on BlueGene/L? My guess is about 50 years.
>
> What's your guess based on?
>
Oh, just a wild guess. First you have to get the source for Windows and
Office. You'll have to to spend a few years saving up enough money to
convince MS to fork it over. Then you have to port a few million lines of
code. Good luck with that.
>
>
>> Do you
>> really think office apps are optimized to run on massively parallel
>> supercomputers.
>> Do you really think that massively parallel supercomputers are
>> optimized for running office apps?
>
> ROFL! Is that the latest excuse for Slow-O?
>
So you think MS Office can exploit thousands of processors? BlueGene/L
gets its speed from large numbers of processors running in parallel, not
from some extremely fast single processor. You keep harping on BlueGene
as if you expect it to be blindingly fast starting an office app. It
won't be. It's not designed for that. Trying out OpenOffice on
BlueGene/L my have been a mildly interesting thing to do but no one would
seriously consider running any office apps on it. You seem to have no
concept of the design point of BlueGene/L.
> "OpenOffice isn't optimized to run on computers! Why would you expect
> it to launch in less than 12 seconds?"
>
Uh, maybe because it launches in only 10 seconds on my Eee PC with a max
disk speed of about 23MB/sec. On my dual core 2.4GHz AMD 4600 with a
3Gb/s Sata drive OpenOffice launches from a cold boot in 7 seconds.
After that it only takes 2 seconds. Or how about remote logging in to my
dual core AMD 4600 from my Eee PC and launching OpenOffice using the
wireless link, not a high speed wired connection. 3 seconds after first
launch. That's with exporting the display back to the Eee PC. How long
does it take to do this using Windows and MS Office?
>
>
>> Do you really think that office start up is cpu intensive?
>
> Quit giving them excuses. They're already lazy enough.
>
I see you are refusing to answer the question.
>
>
>> And why is it that OpenOffice starts up faster on my "toy" running
>> 630Mhz and 1GB of RAM than your 3GHz P4 with 2GB RAM? What version of
>> OpenOffice? Was that running on Windows or Linux? Or have you once
>> again forgotten whether you even tried it before making the claim.
>
> I've been trying that OpenOffice crud for years, on Windows and Linux.
> Doesn't matter where you run it - it's a pale imitation of MS Office.
> Except for its cost, it doesn't do a single thing better than the MS
> gold standard.
Why? Why do you keep trying OpenOffice? And if all you are doing is
timing OpenOffice startup then you really haven't tried it.
Bug
![[H]omer's profile photo](http://lh3.googleusercontent.com/a-/ALV-UjWMQWyYwHeIwxhG3nGUeGWmLQD0R8TeZrvdWDiv5-2S_z3rHg=s40-c)
[H]omer
> On Tue, 01 Jan 2008 17:29:32 -0500, DFS wrote:
>> bugbuster wrote:
>>> What's amazing is how dense you are. How long does it take for
>>> MS Office to start on BlueGene/L? My guess is about 50 years.
>>
>> What's your guess based on?
>
> Oh, just a wild guess. First you have to get the source for Windows
> and Office. You'll have to to spend a few years saving up enough
> money to convince MS to fork it over. Then you have to port a few
> million lines of code. Good luck with that.
DooFy is almost as thick as Hardon "The PS3 can run Windows under
VMWare" Quirk.
--
K.
http://slated.org
.----
| "[Microsoft] are willing to lose money for years and years just to
| make sure that you don't make any money, either." - Bob Cringely.
| - http://blog.businessofsoftware.org/2007/07/cringely-the-un.html
`----
Fedora release 8 (Werewolf) on sky, running kernel 2.6.23.8-63.fc8
04:44:40 up 12 days, 2:20, 2 users, load average: 0.42, 0.57, 0.40
Ewok
> It takes 12 seconds to cold launch on my P4, 3.0ghz system w/ 2gb of RAM and
> a fast SATA drive. It's pure OSS crapooolllaaaa.
>
Your computer P4, 3.0 GHz, 2 GB RAM and fast SATA
My computer AMD Duron, 1,3 GHz, 1 GB RAM and IDE
We could guestimate that your computer is roughly 3 times as fast as
my is. You say that it takes 12 seconds to cold launch open office on
your computer. It took 3 seconds on my computer. I just clocked it.
That is 4 times faster than on your computer. Since I use linux and
you use windows, I guess that, that would explain the difference. That
would make linux 12 times faster than windows.
I do realize that this is in no way a scientific approach to compare
the two operating systems, and that there are better benchmarks, than
to compare loading time of open office on a cold launch.
My point is simply this, that windows is very competently constructed
to run softwares created by Microsoft faster than software created by
other companys. There is a mayor loading time difference in firefox on
linux and on windows. There is also a mayor difference in the amount
of resources that firefox hog on linux and on windows. I could easily
have firefox up, with 20 tabs on linux, but if I do that on windows,
the memory usage will slowly rise and suddenly, my computer will be
painfully slow and need a restart.
One of the reasons for this, is that MS doesn't release the specs for
windows to other software vendors, giving it an unfair edge.
One question to you, what kind of stuffs do you run on that computer,
to make it so slow? I cold start open office faster than you do, even
on a liveCD with Beryl running.
/Your friendly neighbourhood Ewok
Peter Köhlmann
> On Dec 29 2007, 2:39 am, "DFS" <nospam@dfs_.com> wrote:
>> It takes 12 seconds to cold launch on my P4, 3.0ghz system w/ 2gb of RAM
>> and
>> a fast SATA drive. It's pure OSS crapooolllaaaa.
>>
>
> Your computer P4, 3.0 GHz, 2 GB RAM and fast SATA
> My computer AMD Duron, 1,3 GHz, 1 GB RAM and IDE
>
> We could guestimate that your computer is roughly 3 times as fast as
> my is. You say that it takes 12 seconds to cold launch open office on
> your computer. It took 3 seconds on my computer. I just clocked it.
> That is 4 times faster than on your computer.
Guess what: DFS is lying.
He brings this bogus "argument" since months, and it has been made up the
very first time he wrote it.
OO *was* slow to start. Way back
It isn't any more. It takes 2 seconds longer to start than MS Office on my
machine.
> Since I use linux and
> you use windows, I guess that, that would explain the difference. That
> would make linux 12 times faster than windows.
DumbFullShit would argue that it takes that long to start on his liveCD
linux "install" and therefore linux is slow
< snip >
You will not get DFS to be reasonable, as trolling and lying is all he wants
--
Microsoft? Is that some kind of a toilet paper?
Roy Schestowitz
> On Dec 29 2007, 2:39 am, "DFS" <nospam@dfs_.com> wrote:
>> It takes 12 seconds to cold launch on my P4, 3.0ghz system w/ 2gb of RAM and
>> a fast SATA drive. It's pure OSS crapooolllaaaa.
>>
>
> Your computer P4, 3.0 GHz, 2 GB RAM and fast SATA
> My computer AMD Duron, 1,3 GHz, 1 GB RAM and IDE
>
> We could guestimate that your computer is roughly 3 times as fast as
> my is. You say that it takes 12 seconds to cold launch open office on
> your computer. It took 3 seconds on my computer. I just clocked it.
> That is 4 times faster than on your computer. Since I use linux and
> you use windows, I guess that, that would explain the difference. That
> would make linux 12 times faster than windows.
OOo can be launched (from a 'cold' state) within just a few seconds on this old
PC that I use. It's very fast.
...Too bad for those who still use Windows because _even_ if they choose Free
open source, they are treated like second-class citizens.
> I do realize that this is in no way a scientific approach to compare
> the two operating systems, and that there are better benchmarks, than
> to compare loading time of open office on a cold launch.
>
> My point is simply this, that windows is very competently constructed
> to run softwares created by Microsoft faster than software created by
> other companys. There is a mayor loading time difference in firefox on
> linux and on windows. There is also a mayor difference in the amount
> of resources that firefox hog on linux and on windows. I could easily
> have firefox up, with 20 tabs on linux, but if I do that on windows,
> the memory usage will slowly rise and suddenly, my computer will be
> painfully slow and need a restart.
>
> One of the reasons for this, is that MS doesn't release the specs for
> windows to other software vendors, giving it an unfair edge.
"We should dedicate a cross-group team to come up with ways to leverage Windows
technically more."
--Jim Allchin, Microsoft
"I am convinced we have to use Windows – this is the one thing they don’t have.
We have to be competitive with features, but we need something more — Windows
integration."
--Jim Allchin, Microsoft
> One question to you, what kind of stuffs do you run on that computer,
> to make it so slow? I cold start open office faster than you do, even
> on a liveCD with Beryl running.
You're responding to Microsoft's handbook of FUD. Microsoft has all sorts of
documents (and sites) that guide those who are looking for classic (and
invalid) ways to dismiss Linux.
--
~~ Best of wishes
Roy S. Schestowitz | Run a Linux server, sit on your hands all day
http://Schestowitz.com | Free as in Free Beer | PGP-Key: 0x74572E8E
Load average (/proc/loadavg): 1.15 1.35 1.26 3/139 2683
http://iuron.com - semantic search engine project initiative
DFS
> On Dec 29 2007, 2:39 am, "DFS" <nospam@dfs_.com> wrote:
>> It takes 12 seconds to cold launch on my P4, 3.0ghz system w/ 2gb of
>> RAM and a fast SATA drive. It's pure OSS crapooolllaaaa.
>>
>
> Your computer P4, 3.0 GHz, 2 GB RAM and fast SATA
> My computer AMD Duron, 1,3 GHz, 1 GB RAM and IDE
>
> We could guestimate that your computer is roughly 3 times as fast as
> my is. You say that it takes 12 seconds to cold launch open office on
> your computer. It took 3 seconds on my computer. I just clocked it.
> That is 4 times faster than on your computer. Since I use linux and
> you use windows, I guess that, that would explain the difference. That
> would make linux 12 times faster than windows.
Actually, Linux is 50x faster than Windows.
> I do realize that this is in no way a scientific approach to compare
> the two operating systems, and that there are better benchmarks, than
> to compare loading time of open office on a cold launch.
>
> My point is simply this, that windows is very competently constructed
> to run softwares created by Microsoft faster than software created by
> other companys. There is a mayor loading time difference in firefox on
> linux and on windows.
Firefox cold launch is about 5 seconds. Warm launch is far <1 second -
virtually instantaneous (of course loading the webpage takes a few seconds).
> There is also a mayor difference in the amount
> of resources that firefox hog on linux and on windows.
With one page loaded, Firefox uses 28mb on my Windows 2003 system.
> I could easily
> have firefox up, with 20 tabs on linux, but if I do that on windows,
> the memory usage will slowly rise and suddenly, my computer will be
> painfully slow and need a restart.
Seriously? You need a new operating system - but then you already knew
that.
I'm working along just fine w/ 20 tabs
http://www.angelfire.com/linux/dfs0/20_Firefox_tabs.png.
And Rex Ballard insists he usually has "8 firefox browser windows (each with
4-5 tabs), and often 20-30 other various windows" open on his Windows work
machine.
> One of the reasons for this, is that MS doesn't release the specs for
> windows to other software vendors, giving it an unfair edge.
Who told you this Ewok fairy tale?
> One question to you, what kind of stuffs do you run on that computer,
> to make it so slow? I cold start open office faster than you do, even
> on a liveCD with Beryl running.
I [occasionally] run OpenOffice - it's by far the slowest starting program
I've ever seen. And it's no coincidence that it's open source.
Tom Shelton
> ____/ Ewok on Wednesday 02 January 2008 11:17 : \____
>
>> On Dec 29 2007, 2:39 am, "DFS" <nospam@dfs_.com> wrote:
>>> It takes 12 seconds to cold launch on my P4, 3.0ghz system w/ 2gb of RAM and
>>> a fast SATA drive. It's pure OSS crapooolllaaaa.
>>>
>>
>> Your computer P4, 3.0 GHz, 2 GB RAM and fast SATA
>> My computer AMD Duron, 1,3 GHz, 1 GB RAM and IDE
>>
>> We could guestimate that your computer is roughly 3 times as fast as
>> my is. You say that it takes 12 seconds to cold launch open office on
>> your computer. It took 3 seconds on my computer. I just clocked it.
>> That is 4 times faster than on your computer. Since I use linux and
>> you use windows, I guess that, that would explain the difference. That
>> would make linux 12 times faster than windows.
>
> OOo can be launched (from a 'cold' state) within just a few seconds on this old
> PC that I use. It's very fast.
>
> ...Too bad for those who still use Windows because _even_ if they choose Free
> open source, they are treated like second-class citizens.
>
If this ends up being a duplicate - forgive me. I had a problem posting
the original - and I wasn't sure if it had worked. So, I reconstructed
the post:
Define "a few"? Here are my system stats (fresh reboot to ensure a cold
start):
tom@bob ~ $ uname -a
Linux bob 2.6.21-gentoo-r4 #5 PREEMPT Mon Dec 3 14:02:36 MST 2007 i686 Intel(R) Celeron(TM) CPU 1400MHz GenuineIntel GNU/Linux
tom@bob ~ $ uptime
04:01:12 up 6 min, 1 user, load average: 0.12, 0.24, 0.15
tom@bob ~ $ cat /proc/cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 11
model name : Intel(R) Celeron(TM) CPU 1400MHz
stepping : 1
cpu MHz : 1392.412
cache size : 256 KB
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 2
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge
mca cmov pat pse36 mmx fxsr sse
bogomips : 2814.99
clflush size : 32
tom@bob ~ $ cat /proc/meminfo
MemTotal: 513800 kB
MemFree: 333484 kB
Buffers: 10784 kB
Cached: 92032 kB
SwapCached: 0 kB
Active: 87928 kB
Inactive: 63372 kB
HighTotal: 0 kB
HighFree: 0 kB
LowTotal: 513800 kB
LowFree: 333484 kB
SwapTotal: 500464 kB
SwapFree: 500464 kB
Dirty: 12 kB
Writeback: 0 kB
AnonPages: 48504 kB
Mapped: 40628 kB
Slab: 11352 kB
SReclaimable: 5096 kB
SUnreclaim: 6256 kB
PageTables: 1156 kB
NFS_Unstable: 0 kB
Bounce: 0 kB
CommitLimit: 757364 kB
Committed_AS: 188928 kB
VmallocTotal: 507896 kB
VmallocUsed: 69296 kB
VmallocChunk: 438020 kB
Cold start:
tom@bob ~ $ time oowriter
real 0m19.605s
user 0m0.067s
sys 0m0.087s
Now, if there is an instance already loaded, a second instance takes
about half a second:
tom@bob ~ $ time oowriter
real 0m0.582s
user 0m0.057s
sys 0m0.077s
A second start with no instances running:
tom@bob ~ $ time oowriter
real 0m19.446s
user 0m0.063s
sys 0m0.080s
Yeah, it starts in a few seconds on my old pc as well - a few being
about 20 seconds.
--
Tom Shelton
Roy Schestowitz
Very odd. Which version of OOo is it and what did you do to it? ;-)
On this 1.8GHz machine, with RAM fully occupied (therefore OOo needs to resort
to paging), OOo just takes a few seconds to launch for the first time. Don't
ask me how. I haven't a clue what you did.
--
~~ Best of wishes
Roy S. Schestowitz | "Quote when replying in non-real-time dialogues"
http://Schestowitz.com | RHAT Linux | PGP-Key: 0x74572E8E
02:20:01 up 23 days, 15:08, 4 users, load average: 0.38, 0.56, 0.82
http://iuron.com - Open Source knowledge engine project
Tom Shelton
OOo 2.3.1. What did I do to it? Nothing. I installed it from
source:
emerge -av openoffice
Here are my CFLAGS from /etc/make.conf
CFLAGS="-O2 -march=pentium3 -pipe -fomit-frame-pointer"
> On this 1.8GHz machine, with RAM fully occupied (therefore OOo needs to resort
> to paging), OOo just takes a few seconds to launch for the first time. Don't
> ask me how. I haven't a clue what you did.
>
Again - define few. I showed you my times, lets see yours.
--
Tom Shelton
Ewok
I have twice your memory, but approximately your processor, so it
should take about the same time. Mine takes 3 seconds. I googled start
up on open office and came with the following tip.
Tools menu, options
Memory under OpenOffice.org
Change:
Undo
Number of steps to 25
Graphics cache
Memory per object 2.0 MB
Remove from memory after 00:05
Cache for inserted objects
Number of objects 15
Mark OpenOffice.org quickstarter
I didn't need to do this, but it may help, to get down the start up
time.
/Your friendly neighbourhood Ewok
Roy Schestowitz
I think I once read this tip in Linux Journal (Tom Adelstein?) or Linux.com
(Bruce Byfield?). Either way, I made no adjustment to OOo. It runs with a
GNOME-ish (GTK) theme under KDE, which means it has to load some extra libs in
KDE, yet it takes just seconds to run without any preloading. As I said, don't
ask me how, but I'm not lying. It takes me a little longer than 3 seconds
because my RAM and swap is always nearly full (I multitask with little memory
available).
> I didn't need to do this, but it may help, to get down the start up
> time.
>
> /Your friendly neighbourhood Ewok
Tom Shelton is quite polite and intelligent, but he is usually here just to
pour cold water on GNU/Linux, so prejudice should be justified. Suspicion
rather...
--
~~ Best of wishes
Roy S. Schestowitz | while (sig==sig) sig=!sig;
Ewok
> I think I once read this tip in Linux Journal (Tom Adelstein?) or Linux.com
> (Bruce Byfield?). Either way, I made no adjustment to OOo. It runs with a
> GNOME-ish (GTK) theme under KDE, which means it has to load some extra libs in
> KDE, yet it takes just seconds to run without any preloading. As I said, don't
> ask me how, but I'm not lying. It takes me a little longer than 3 seconds
> because my RAM and swap is always nearly full (I multitask with little memory
> available).
About the same time it takes me. I do alot of multitasking myself, but
mine hardly takes any memory, since most of mine is done through
command line.
>
> > I didn't need to do this, but it may help, to get down the start up
> > time.
>
> > /Your friendly neighbourhood Ewok
>
> Tom Shelton is quite polite and intelligent, but he is usually here just to
> pour cold water on GNU/Linux, so prejudice should be justified. Suspicion
> rather...
>
I found that the best way of dealing with this kind of behaviour is to
be polite and kind in response. Otherwise we just give fuel to fire.
Plus, sometimes it may actually be someone who is a serious person. If
one isn't polite in that situation, it may come back to haunt you very
long. Internet trolls never forget, when it serves their purpose.
/Your friendly neighbourhood ewok
Tom Shelton
> On Jan 3, 5:27 am, Tom Shelton <tom_shel...@comcast.net> wrote:
>> On Jan 2, 7:25 pm, Roy Schestowitz <newsgro...@schestowitz.com> wrote:
>>
>>
>>
>> > ____/ Tom Shelton on Wednesday 02 January 2008 18:14 : \____
>>
>> > > On 2008-01-02, Roy Schestowitz <newsgro...@schestowitz.com> wrote:
>> > >> ____/ Ewok on Wednesday 02 January 2008 11:17 : \____
>>
>> > >>> On Dec 29 2007, 2:39 am, "DFS" <nospam@dfs_.com> wrote:
>> > >>>> It takes 12 seconds to cold launch on my P4, 3.0ghz system w/ 2gb of RAM
>> > >>>> and
>> > >>>> a fast SATA drive. It's pure OSS crapooolllaaaa.
>>
>>
<snip>
>> > On this 1.8GHz machine, with RAM fully occupied (therefore OOo needs to resort
>> > to paging), OOo just takes a few seconds to launch for the first time. Don't
>> > ask me how. I haven't a clue what you did.
>>
>> Again - define few. I showed you my times, lets see yours.
>>
>> --
>> Tom Shelton
>
> I have twice your memory, but approximately your processor, so it
> should take about the same time. Mine takes 3 seconds. I googled start
> up on open office and came with the following tip.
>
> Tools menu, options
> Memory under OpenOffice.org
> Change:
> Undo
> Number of steps to 25
> Graphics cache
> Memory per object 2.0 MB
> Remove from memory after 00:05
> Cache for inserted objects
> Number of objects 15
> Mark OpenOffice.org quickstarter
>
> I didn't need to do this, but it may help, to get down the start up
> time.
>
> /Your friendly neighbourhood Ewok
Thanks, but it didn't make any difference - but maybe because the
quickstarter thing doesn't seem to be launching. I dual boot this
machine with kubuntu 7.10 so I booted into that to see if made any
difference and the times were the same. The option for
the quickstarter is disabled on the kubuntu install so I couldn't
test it's effect on either system.
I have a hard time believing you have a cold start time of 3 seconds on
an equivalent machine.
--
Tom Shelton
Tom Shelton
> ____/ Ewok on Thursday 03 January 2008 11:22 : \____
>
>> On Jan 3, 5:27 am, Tom Shelton <tom_shel...@comcast.net> wrote:
>>> On Jan 2, 7:25 pm, Roy Schestowitz <newsgro...@schestowitz.com> wrote:
>>>
>>>
>>>
>>> > ____/ Tom Shelton on Wednesday 02 January 2008 18:14 : \____
>>>
>>> > > On 2008-01-02, Roy Schestowitz <newsgro...@schestowitz.com> wrote:
>>> > >> ____/ Ewok on Wednesday 02 January 2008 11:17 : \____
>>>
>>> > >>> On Dec 29 2007, 2:39 am, "DFS" <nospam@dfs_.com> wrote:
>>> > >>>> It takes 12 seconds to cold launch on my P4, 3.0ghz system w/ 2gb of
>>> > >>>> RAM and
>>> > >>>> a fast SATA drive. It's pure OSS crapooolllaaaa.
>>>
>>> > >>> Your computer P4, 3.0 GHz, 2 GB RAM and fast SATA
>>> > >>> My computer AMD Duron, 1,3 GHz, 1 GB RAM and IDE
>>>
<snip>
You still haven't posted your times Roy... You make vauge statements
about it only taking a "few seconds", and "a little longer then 3
seconds", but you don't give an acutal time. Why?
--
Tom Shelton
Mark Kent
>
> Tom Shelton is quite polite and intelligent, but he is usually here just to
> pour cold water on GNU/Linux, so prejudice should be justified. Suspicion
> rather...
>
Shelton's a troll, an off-topic, anti-charter poster. He's been in my
killfile for yonks.
--
| Mark Kent -- mark at ellandroad dot demon dot co dot uk |
| Cola faq: http://www.faqs.org/faqs/linux/advocacy/faq-and-primer/ |
| Cola trolls: http://colatrolls.blogspot.com/ |
| My (new) blog: http://www.thereisnomagic.org |
Jim Richardson
Hash: SHA1
On Thu, 3 Jan 2008 18:44:06 +0000,
Mark Kent <mark...@demon.co.uk> wrote:
> Roy Schestowitz <newsg...@schestowitz.com> espoused:
>
>>
>> Tom Shelton is quite polite and intelligent, but he is usually here just to
>> pour cold water on GNU/Linux, so prejudice should be justified. Suspicion
>> rather...
>>
>
> Shelton's a troll, an off-topic, anti-charter poster. He's been in my
> killfile for yonks.
>
How is posting numbers and information concerning how fast OO on Linux
starts up "off topic"?
and you have a funny definition of "troll".
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
iD8DBQFHfYBEd90bcYOAWPYRAsHOAKCC18CDqGdpTtZ33dJ5yrysMiCyzgCbB+CF
xF6C3LrtM5PItbSJEZbtmbI=
=+0Ts
-----END PGP SIGNATURE-----
--
Jim Richardson http://www.eskimo.com/~warlock
I came; I saw; I fucked up
AZ Nomad
>On Thu, 3 Jan 2008 18:44:06 +0000,
> Mark Kent <mark...@demon.co.uk> wrote:
>> Roy Schestowitz <newsg...@schestowitz.com> espoused:
>>
>>>
>>> Tom Shelton is quite polite and intelligent, but he is usually here just to
>>> pour cold water on GNU/Linux, so prejudice should be justified. Suspicion
>>> rather...
>>>
>>
>> Shelton's a troll, an off-topic, anti-charter poster. He's been in my
>> killfile for yonks.
>>
>How is posting numbers and information concerning how fast OO on Linux
>starts up "off topic"?
It is offtopic because it is anti-linux garbage. People don't use computers
to boot, start open office and shutdown over and over and over.
People start open office, and then do real work. At doing real work,
openoffice is extremely fast and capable.
The 8-15 seconds required to start open office means nothing when the computer
is then used for real work for the next several hours. What matters is that
subsequent windows open within a second.
I guess the fast initial load time for microsoft word might be important when
you have a system that crashes constantly.
Roy Schestowitz
I agree. Talking about startup time is a case of diverting attention to
non-issues, thereby escaping the great capabilities of OOo, some of which
surpass MSO.
If Tom doesn't like Linux and Free software, he has subscribed to the wrong
newsgroup.
--
~~ Best of wishes
Proprietary, lockin-based tools lead to regrets. Doc(umen)tor, heal thyself.
http://Schestowitz.com | Free as in Free Beer | PGP-Key: 0x74572E8E
Load average (/proc/loadavg): 1.31 1.08 1.09 3/147 18106
Mark Kent
... which is why it's off-topic and anti-charter. He's been in my
killfile for ages for this behaviour.