Is OS X > Linux?

MrCode · 2011-07-21 19:41:50

MrAllan wrote:

I don't believe it's fair to compare OSen by comparing their RAM usage.

Agreed, especially since different OSes have different memory usage models.

In other words, define "usage". Are we talking about "active" memory used be processes, caching, both?

As an example:

mrcode@lappy486 ~]$ fm
             total       used       free     shared    buffers     cached
Mem:          3827       1542       2285          0        130        879
-/+ buffers/cache:        532       3295
Swap:          258          0        258

Depending on how you define "usage", you could say that I'm using a gig and a half just sitting idle (with only FF and a terminal open), or you could say that I'm using just over half a gig. The former is everything (including disk cache/buffers), whereas the latter is just "active" memory allocated by processes (AFAIK).

I have no idea how OS X/Mach/Darwin handles caching, etc.

dakt · 2011-07-21 21:28:11

DISK CACHING!

             total       used       free     shared    buffers     cached
Mem:          2007        979       1027          0         13        212
-/+ buffers/cache:        754       1253
Swap:         4095         88       4007

Here it reports 979MB used, from which 754M is used for caching. And this is the machine(laptop) that has been running for 9 days now.
You want for most of your RAM to be occupied with caches...as much as possible.

ANOKNUSA · 2011-07-21 21:32:07

Inxsible wrote:

Any particular reason why this thread was revived after more than a year?

Obviously it's because Lion is now availabe, and will definitively end this debate and bury the "competition." Am I right, folks?

Beelzebud · 2011-07-21 23:41:14

I will now quote Jules from Pulp Fiction as my answer to this question:

Rat meat might taste like pumpkin pie, but I wouldn't know, because I wouldn't eat the filthy motherfucker.

I'm sticking with Linux.

pogeymanz · 2011-07-22 12:40:42

For a topic going nowhere, there are a lot of replies.

Your RAM usage may be a real effect. The apps you open in OSX are different apps than the ones you open in Linux. I don't just mean the ones that are named different things. Firefox on OSX is a different program than Firefox on Linux. It is possible that the Firefox developers somehow made the OSX version use less memory. Maybe.

But the RAM probably isn't something to be too concerned about. Like someone mentioned, the way the used RAM is calculated is different, so that might cause some error.

The slowdowns you experience are probably not Ubuntu's fault. I'm sick of all the Ubuntu-bashing that happens. What is more likely is that the Linux kernel that is installed in your version of Ubuntu prioritizes tasks differently than OSX's kernel. This may be perceived to the end user as lagginess or stuttering. Try install the CK patchset to your kernel and it may help. Many of us are using it and it makes a big difference in responsiveness.

markbabc · 2011-07-22 23:06:29

pogeymanz wrote:

For a topic going nowhere, there are a lot of replies.
Your RAM usage may be a real effect. The apps you open in OSX are different apps than the ones you open in Linux. I don't just mean the ones that are named different things. Firefox on OSX is a different program than Firefox on Linux. It is possible that the Firefox developers somehow made the OSX version use less memory. Maybe.
But the RAM probably isn't something to be too concerned about. Like someone mentioned, the way the used RAM is calculated is different, so that might cause some error.
The slowdowns you experience are probably not Ubuntu's fault. I'm sick of all the Ubuntu-bashing that happens. What is more likely is that the Linux kernel that is installed in your version of Ubuntu prioritizes tasks differently than OSX's kernel. This may be perceived to the end user as lagginess or stuttering. Try install the CK patchset to your kernel and it may help. Many of us are using it and it makes a big difference in responsiveness.

You kidding?? Ubuntu is definatly to blame for slowdowns... just look at all the programs that it runs at once compared to a typical Arch install. check your pstree in Ubuntu then check it in Arch and you should see that ubuntu is just running alot that leads to slowdowns

ANOKNUSA · 2011-07-22 23:46:32

markbabc wrote:

You kidding?? Ubuntu is definatly to blame for slowdowns... just look at all the programs that it runs at once compared to a typical Arch install. check your pstree in Ubuntu then check it in Arch and you should see that ubuntu is just running alot that leads to slowdowns

That's entirely dependent on your personal system. It's possible to run Ubuntu with DWM and lightweight apps after trimming out all the extra crap the installer sticks on your drive... though why the hell you'd bother with Ubuntu at that point is beyond me.

markbabc · 2011-07-23 02:43:23

ANOKNUSA wrote:

That's entirely dependent on your personal system. It's possible to run Ubuntu with DWM and lightweight apps after trimming out all the extra crap the installer sticks on your drive... though why the hell you'd bother with Ubuntu at that point is beyond me.

exactly...

nathan28 · 2011-07-26 17:30:23

markbabc wrote:

ANOKNUSA wrote:
That's entirely dependent on your personal system. It's possible to run Ubuntu with DWM and lightweight apps after trimming out all the extra crap the installer sticks on your drive... though why the hell you'd bother with Ubuntu at that point is beyond me.
exactly...

Hey dude, but Upstart would be, like, an 8-second boot then! That's faster than waking from suspend! Wait, Upstart moved from Canonical to Google? ONOZ!

Leonid.I · 2011-07-27 21:34:45

nathan28 wrote:

markbabc wrote:
ANOKNUSA wrote:
That's entirely dependent on your personal system. It's possible to run Ubuntu with DWM and lightweight apps after trimming out all the extra crap the installer sticks on your drive... though why the hell you'd bother with Ubuntu at that point is beyond me.
exactly...
Hey dude, but Upstart would be, like, an 8-second boot then! That's faster than waking from suspend! Wait, Upstart moved from Canonical to Google? ONOZ!

I would say, these days arch is the slowest of all when it comes to the cold boot. *buntu, fedora, perhaps opensuse use upstart/systemd which parallelize boot process. In arch it is serial. As far as i understand, systemd was inspired by apple's launchd init system which is also quite advanced.

Of course, do I really care whether my system boots in 8 or 20 sec? Meh...

markbabc · 2011-07-28 15:47:03

Leonid.I wrote:

I would say, these days arch is the slowest of all when it comes to the cold boot. *buntu, fedora, perhaps opensuse use upstart/systemd which parallelize boot process. In arch it is serial. As far as i understand, systemd was inspired by apple's launchd init system which is also quite advanced.
Of course, do I really care whether my system boots in 8 or 20 sec? Meh...

Not sure what your doing but with the default init system on my laptop and desktop I have <10sec boot-time.... with systemd i can get it to around 6sec for both

triplesquarednine · 2011-07-28 20:53:45

Leonid.I wrote:

I would say, these days arch is the slowest of all when it comes to the cold boot. *buntu, fedora, perhaps opensuse use upstart/systemd which parallelize boot process. In arch it is serial. As far as i understand, systemd was inspired by apple's launchd init system which is also quite advanced.
Of course, do I really care whether my system boots in 8 or 20 sec? Meh...

....erm, you know that Arch can use Systemd right? (no one says you have to use the regular init). Systemd takes very little time to get working too ... i take it you've also heard of things like e4rat, which can improve boot time too (for ext4);

https://bbs.archlinux.org/viewtopic.php?id=115976

i know the last time i setup e4rat - my boot time was reduced by 8sec, from that alone.

Arch can be setup to boot however you like, i don't think Ubuntu, fedora, etc have any real boot advantage over an Arch system, that an Archer has setup with fast booting in mind...

i haven't set up systemd on my new hardware / Arch install yet (it's 3 days old), but i will probably use it on this machine...

cheerz

Last edited by triplesquarednine (2011-07-28 21:05:38)

Leonid.I · 2011-07-28 22:47:40

I was just negating the statement, that Ubuntu is slow/bloated, by pointing out that it uses quite advanced init system. If it gets slow -- blame the admin

Having said that, I am totally OK with booting in 15+ sec (don't even background daemons), so I see no need in installing systemd. But I do appreciate the work done by people who port it to arch.

markbabc · 2011-07-29 05:21:29

Leonid.I wrote:

I was just negating the statement, that Ubuntu is slow/bloated, by pointing out that it uses quite advanced init system. If it gets slow -- blame the admin
Having said that, I am totally OK with booting in 15+ sec (don't even background daemons), so I see no need in installing systemd. But I do appreciate the work done by people who port it to arch.

I ran Ubuntu on my netbook for 1 year and a half. boot time like like 45sec (still faster than when i had xp on it) now with arch its 8sec... sure you could spend the time to make ubuntu not suck so much but then you would have to ask yourself "why the fuck did i even bother?" after a few hours of removing shit.

Sathors · 2011-09-14 13:15:14

dakt wrote:

DISK CACHING!
             total       used       free     shared    buffers     cached
Mem:          2007        979       1027          0         13        212
-/+ buffers/cache:        754       1253
Swap:         4095         88       4007
Here it reports 979MB used, from which 754M is used for caching. And this is the machine(laptop) that has been running for 9 days now.
You want for most of your RAM to be occupied with caches...as much as possible.

Well correct me if I'm wrong, but isn't there only 212M used for caching and 13 for buffers ?

Indeed the "cached" column of the first line plus the "buffers" column on the same line plus the "used" column of the second line gives you precisely 979M, which is the sum of the RAM used.
Which leads me to think that the RAM quantity actually used by running programms is of 754.

(I've always been too lazy to read all of the manual of free !!!)

karol · 2011-09-14 13:23:58

Sathors wrote:

Well correct me if I'm wrong, but isn't there only 212M used for caching and 13 for buffers ?
<snip>
Which leads me to think that the RAM quantity actually used by running programms is of 754.

Yes, you're right, good catch.

Arch Linux

#51 2011-07-21 19:41:50

Re: Is OS X > Linux?

#52 2011-07-21 21:28:11

Re: Is OS X > Linux?

#53 2011-07-21 21:32:07

Re: Is OS X > Linux?

#54 2011-07-21 23:41:14

Re: Is OS X > Linux?

#55 2011-07-22 12:40:42

Re: Is OS X > Linux?

#56 2011-07-22 23:06:29

Re: Is OS X > Linux?

#57 2011-07-22 23:46:32

Re: Is OS X > Linux?

#58 2011-07-23 02:43:23

Re: Is OS X > Linux?

#59 2011-07-26 17:30:23

Re: Is OS X > Linux?

#60 2011-07-27 21:34:45

Re: Is OS X > Linux?

#61 2011-07-28 15:47:03

Re: Is OS X > Linux?

#62 2011-07-28 20:53:45

Re: Is OS X > Linux?

#63 2011-07-28 22:47:40

Re: Is OS X > Linux?

#64 2011-07-29 05:21:29

Re: Is OS X > Linux?

#65 2011-09-14 13:15:14

Re: Is OS X > Linux?

#66 2011-09-14 13:23:58

Re: Is OS X > Linux?

Board footer