You are not logged in.
Please please please please don't think this is a flame thread. I'm looking for some honest feedback here, as I've been a little disheartened recently by linux. Here's what happened:
I recently installed the latest and greatest iteration of Ubuntu and dual-booted it on my Macbook (aluminum 5.1) after some fiddling with drivers and settings, I got everything working. After a while though, I noticed that when multi-tasking in Ubuntu, I was commonly experiences system slow-downs much more frequently than in OS X (10.5 leopard).
So I decided to compare the two. In brief this is what I found:
System:
Macbook 5,1 OS X 10.5.8
2ghz Intel Core 2 Duo
2GB DDR3 Ram
Ubuntu (Gnome, Gnome-do, Compiz):
Boot: 24sec
Initial *Active* Ram use: ~410mb (!)
Average Workload (Chrome 5 tabs, gedit, filezilla, Firefox 1 tab): ~1200mb ram
Ubuntu (Open-box, Gnome-panel, kupfer):
Boot: 22sec
Initial *Active* Ram use: ~320mb
Average Workload (Chrome 5 tabs, gedit, filezilla, Firefox 1 tab): ~1150mb ram (!)
OS X (10.5, smcFanControl, Quicksilver):
Boot: 72sec
Initial *Active* Ram use: ~245mb
Average Workload (Chrome 5 tabs, Coda, Firefox 1 tab): ~890mb ram
At any rate, you can see, above, that it would seem that OS X is much more thrifty in it's handling of resources. And it wasn't strictly a "numbers" thing either. I noticed that system sluggishness settled in much quicker in Ubuntu than in OS X... which sort of defeats the purpose of Linux for me.
To be clear... I'm a huge proponent of OSS and Linux. Or I wouldn't be bothering. But I don't have any (serious) opposition to Apple. I was hoping that my linux partition would make a good "work-zone" for me where I could have a lean, resource-efficient environment, for punching out websites and reserve OS X for video-editing or play. But, unfortunately, it would seem that linux, even in a light environment with no compositing, is actually less efficient than a full OS X DE with 3d.
Does anyone have any feedback for me on this? Would I be prudent to put Arch on this machine or would I be wasting my time. I really just want an optimized work environment.
Offline
On my Macbook Pro, running 7 urxvt's, Firefox with 4-5 tabs, MPD, irssi, and various other things I come in at less than 512mb of RAM. But, I run Arch, not Ubuntu, and run XMonad, not Gnome or any Gnome variant. I guess it all depends on the programs you are running.
OS 10.6 on the other hand eats up about 800 just at boot. I am seriously thinking about ditching my Macbook Pro, as I would rather run Linux anyway and almost never boot into OSX anymore.
#binarii @ irc.binarii.net
Matrix Server: https://matrix.binarii.net
-------------
Allan -> ArchBang is not supported because it is stupid.
Offline
The Linux kernel uses memory differently. All that 'used' memory can be repurposed on a moment's notice; it's just that the kernel uses it to cache things and speed up f.e. disk access. At least this is the way it was once explained to me.
Also, you were using Ubuntu, which is likely the reason behind the slowdowns. (And I question the judgement of posting this on the *Arch* forum.)
And yes, you should use a minimal distro such as Arch if you're looking to get the most out of your hardware. Ubuntu is not exactly what I'd call a speed dæmon.
Last edited by Peasantoid (2010-03-15 20:20:58)
Offline
> free -m
total used free shared buffers cached
Mem: 493 416 76 0 12 227
-/+ buffers/cache: 177 315
Swap: 956 0 956
Xmonad, Chromium (2 windows, 9 tabs in one and 3 tabs in the other), 1 urxvt running screen, and gajim.
aur S & M :: forum rules :: Community Ethos
Resources for Women, POC, LGBT*, and allies
Offline
The Linux kernel uses memory differently. All that 'used' memory can be repurposed on a moment's notice; it's just that the kernel uses it to cache things and speed up f.e. disk access. At least this is the way it was once explained to me.
Also, you were using Ubuntu, which is likely the reason behind the slowdowns. (And I question the judgement of posting this on the *Arch* forum.)
Heh, I posted this on the Arch forum for two reasons: 1. I love Arch and used it on my non-mac desktop for over a year and 2. I've had the experience of ubuntu users rabidly defending Ubuntu and getting only partially useful discussions in relation to any of Ubuntu's shortcomings. I've always seen Arch users as more objective and realistic (not that I dislike the Ubuntu community... they are very helpful when it comes to Wine and video games !)
Thank you for all your feedback though... it's true that I hardly ever noticed system slow-down before in Arch when I was using it. I think I was just a bit hesitant to put it on my Macbook because, on my first outing with Arch, I hosed my whole system. I really don't think I could do that again, because I've installed it several time... but I try to never underestimate my stupidity.
I find your stats interesting. I think I should most certainly take Arch for a spin on my Macbook before I resign myself to OS X indefinitely. Ubuntu is very useable... but it's also about a clumsy as Windows Vista when it comes to performance, imho. I guess that's the price you pay for the babying that Ubuntu provides to new users?
Ok! I'll give it another go. I too would prefer to be linux full-time if I can justify it. Video editing is the only thing I can't competently do in Linux... and there's no harm in keeping a OS X partition around just for that.
Offline
Ubuntu's your problem.
Personally, I'd rather be back in Hobbiton.
Offline
Also be aware that there may well be a difference the way the Linux-OSX kernels in the two systems are reporting used RAM and how that number is calculated so I would not put too much faith in your RAM usage outputs.
Philosophy is looking for a black cat in a dark room. Metaphysics is looking for a black cat in a dark room that isn't there. Religion is looking for a black cat in a dark room that isn't there and shouting "I found it!". Science is looking for a black cat in a dark room with a flashlight.
Offline
Also be aware that there may well be a difference the way the Linux-OSX kernels in the two systems are reporting used RAM and how that number is calculated so I would not put too much faith in your RAM usage outputs.
Hmm... good point. But then... is there a way for me to concretely find out if I'm getting a significant performance increase using Arch?
Offline
you cannot compare systems by amount of initial RAM they take.. (And for what do you have RAM, when you don't use it :-)) If the gtk/qt libraries are loaded to the memory earlier, the start of programs, which use them, is faster :-)
So compare your ENJOYMENT of working with them and choose the one, who fits the best for vou :-)
Thanks for the GNU/Linux. EDIT:Thanks for the GNU/Linux, Linux guru, for the manuals and wikis.
Offline
You are comparing apples and vintage cars. Remember that OS X is built around the hardware it is written for and as some have already said Linux and Darwin/BSD manage memory very differently.
Offline
OSX takes more than a minute too boot.
OS X (10.5, smcFanControl, Quicksilver):
Boot: 72sec
I can't see its superiority. Things are just handled differently in these operating systems(Ubuntu&Apple).
However, the RAM argument is smart, but Ubuntu in itsef isn't Linux,it's a collection of a whole bunch of software that works with eachother using a whole variety of libraries, Linux is the kernel, or in some cases, a distribution.
If you wish to compare try comparing kernel booting speeds. Now, thats the way to go.
Apple uses a modified BSD kernel that you can seek out on the internet, Linux distributions the Linux kernel. Perhaps you could even test the Ubuntu kernel, which is hefty and mighty indeed.
Up for the challenge.
Last edited by 3]) (2010-03-16 00:49:07)
“There are two ways of constructing a software design. One way is to make it so simple that there are obviously no deficiencies. And the other way is to make it so complicated that there are no obvious deficiencies.”-- C.A.R. Hoare
Offline
I hope you're not trying to distill any actual facts from 'research' that is superficial and flawed at best. Solely comparing RAM usage is not going to tell you which OS is 'best'.
I'll let this topic languish in here for a while, but this is one is nominated for TGN awards.
Got Leenucks? :: Arch: Power in simplicity :: Get Counted! Registered Linux User #392717 :: Blog thingy
Offline
OSX takes more than a minute too boot.
justaleaf wrote:OS X (10.5, smcFanControl, Quicksilver):
Boot: 72secI can't see its superiority. Things are just handled differently in these operating systems(Ubuntu&Apple).
However, the RAM argument is smart, but Ubuntu in itsef isn't Linux,it's a collection of a whole bunch of software that works with eachother using a whole variety of libraries, Linux is the kernel, or in some cases, a distribution.
If you wish to compare try comparing kernel booting speeds. Now, thats the way to go.
Apple uses a modified BSD kernel that you can seek out on the internet, Linux distributions the Linux kernel. Perhaps you could even test the Ubuntu kernel, which is hefty and mighty indeed.
Up for the challenge.
Haha... I wouldn't even know where to begin. But seeing as Arch is marketed those of us who prefer light, simple, and fast OS's... I'm inclined to trust their judgement with how this thing should be put together.
Speaking of which... I'm going under now! I'll be back when I'm on Arch (b'.')b.
Offline
I hope you're not trying to distill any actual facts from 'research' that is superficial and flawed at best. Solely comparing RAM usage is not going to tell you which OS is 'best'.
I'll let this topic languish in here for a while, but this is one is nominated for TGN awards.
I didn't assert that it was concrete =/. I was just concerned about the high ram usage and sluggish performance of Ubuntu and looking for some feedback from more experienced users. I was hoping these gaps in ram usage weren't evidence of performance gaps.
What does TGN stand for?
Oh... and btw guys... I saw this earlier today. I don't know how valid this particular test is... it was interesting.
http://www.phoronix.com/scan.php?page=a … inal&num=1
But then... most of the Archies I've ever spoken to never really singled Ubuntu out as a good performer anyway XD.
Anywho, yeah... for reals this time. Be back when I'm successfully positioned on Arch. Thanks for all your feedback!
*EDIT* forgot the link
Last edited by justaleaf (2010-03-16 01:39:17)
Offline
TGN == Topics Going Nowhere
Offline
What does TGN stand for?
#edit Waaay too slow...
Last edited by jasonwryan (2010-03-16 01:40:40)
Offline
Ah... thanks for clueing me in! Sorry if anyone thought this thread was pointless... sure helped me though! The Arch forums are great!
Offline
I would suggest that there really is a problem here and it is that you want to use Ubuntu to do stuff, and it's not as fast as it could be.
I have become of the opinion that it's silly to use an operating system just because it's a few sec faster to boot / start office suite / whatever, what's more important is to have an operating systems that provides an interface and configuration paradigm that makes sense to you; and if Ubuntu just makes more sense to you than OS X, what you need to do is figure out how to tune Ubuntu to run faster/lighter. If OS X is something you can live with, though, than it's kind of silly to use Ubuntu; just use OS X.
What you want to do is take a look at something like "top" or "htop" and see if you can figure out what's eating up your CPU cycles; if you can track that down, maybe you can tune the performance of that application. Also: figure out what starts up by default, and if you don't need it, disable it.
Edited for clarity.
Last edited by pseudonomous (2010-03-16 04:36:09)
Offline
Also don't forget, OS X uses a completely different scheduling system to Linux, I'd guess a lot better optimised as a default for desktop use. Try using Linux with the -ck kernel patchset, it optimises performance for the desktop, and it does make a substantial difference to the way the OS performs on desktop oriented tasks.
Offline
On my laptop, I find ubuntu fairly snappy at low workloads, almost comparable to Arch. Vista is very sluggish at first, but doesn't seem to get bogged down much by higher workloads (neither does Arch). Ubuntu gets noticeably more sluggish when the workload increases though.
So maybe Ubuntu isn't the best benchmark for Linux vs OS X.
Offline
actually RAM argument regarding Ubuntu is absolutely correct: this is mamory waste.
OP is takling about initial RAM use after boot. This has nothing to do witch cache.
1) So yes, Ubuntu uses/wastes more RAM than OS X.
2) linux can be customized
for example: I have KDE installed with composite enabled and initial RAM usage is 201MB
3) Ubuntu is less flexible than Arch in terms of optimization but you can do a lot to trim initial memory usage and improve interactivity of Ubuntu
4) in general linux allow to select I/O scheduler (chose best for hardware type and file system) and cpu scheduler (chose for your task)
5) memory usage can be optimized
6) boot time does not matter particularly for OS X which has much, much better than linux suspend options, so it takes 1s to have OS running
7) while optimized linux schould perform as well/better than OS X will depend on your skills
8) OS X applications in general are better quality than those for linux
Offline
I don't believe it's fair to compare OSen by comparing their RAM usage. The OS you're more productive with is the best for you. For me, it's definitely OS X (MacBook Pro 5,5 here), because under Linux, battery usage is much higher and the trackpad works a billion times better in OS X.
EDIT: *prepares to be eaten alive by fellow archers*
Last edited by MrAllan (2010-03-16 15:58:19)
Offline
Interesting feedback guys. Though, I'm thoroughly confused on ram usage now .
My logic in choosing to compare using initial active memory consumption was "since I have not interacted with the system yet, whatever resources it is consuming at this point reflect resources that would be difficult for me to reclaim and less performance overhead to keep my system running well".
At any rate... my attempt to go Arch last night was an utter failure.
I had the bright idea to try and repartition the Ubuntu side of my computer with Disk Utility instead of using a live CD.
My root/home/swap partitions were removed without issue. But there's a 900k boot/mbr partition that stores GRUB and is what rEFIt uses for "mbr"... and it wouldn't let me touch it.
In fact, I couldn't even remove it with a root instance of Disk Utility, or Disk Utility when booted from a live OS X install disk with all drives unmounted. And now, I can't launch live-cd when I try to boot to my linux side with rEFIt... all I get is the empty GRUB screen that doesn't seem to react to keyboard input.
I spent hours troubleshooting this last night with no success. I think I pretty much screwed my partition table -_-.
So, unless I completely reformat now or spend money on iPartition... I appear to be at a dead-end :<.
Offline
Not quite sure where this thread is going but I'm using OS X on a hackintosh at home and Arch on my netbook.
Not a fair comparison I know, but quite honestly the fact that OS X eats up about a gig of my RAM with a bunch of apps open or that my Arch with Openbox "weighs" ~50MB on boot means nothing to me as long as everything runs smoothly.
No OS is ever going to "win" - especially not based on a single stat, and even less so if that's colliding with your personal preference and/or needs.
And yeah, Ubuntu is by no means the "benchmark" for something like this, it might just be my worst experience with Linux - ever.
Offline
Actually you would want to use up all your RAM. Otherwise it is just a waste. Why have 32 GiB if you are only ever going to use 1 or 2 GiB? A better metric on usability is page faults or how frequently the system has to pull up memory from swap.
aur S & M :: forum rules :: Community Ethos
Resources for Women, POC, LGBT*, and allies
Offline