What is Digg?Warning: The Content in this Article May be Inaccurate
Readers have reported that this story contains information that may not be accurate.Sponsored by Activision
Band Hero view!
guitarhero.com - The biggest event music event of the year is now in your living room.
165 Comments
- t3hbagel, on 07/21/2008, -5/+149This is a terrible review, he is using two VERY different laptops, and treats these tests as credible.
He even uses two different damn web browsers to compare CPU usage while watching Youtube! What the hell, man? Do you LIKE variables in your tests?! - waydee, on 07/22/2008, -8/+79Pointless comparisons.
- addicted68098, on 07/21/2008, -2/+61Does anyone know why Flash uses so much more CPU on Linux?
If it wasn't for flash rendering pages in Linux would be super fast. - inactive, on 08/11/2008, -9/+64"lets face it... Windows Vista is a solid OS when you hack the living ***** out of it with 3rd-party programs"
Fixed. - davidmwilliams, on 07/21/2008, -6/+46Howdy artfuldodga,
The point was not to compare two super-optimised systems but to compare two typical systems. As I said in the article I had not disabled the Sidebar, Symantec Antivirus, OneNote and so forth. Yes, I could certainly streamline Vista so it had features cleaned up but that would not represent the general end user who does not have the same technical inclination or capabilities as yourselves.
I do understand Superfetch but I'm not sure why you refer to it; Superfetch will preload programs or data expected to be used but that still doesn't explain why Windows is swapping out to disk before it has used up all available RAM.
All the best! - trmanco, on 07/21/2008, -2/+391.06 GB of ram on Fedora after a reboot, yeah, what ever dude... :|
- Ademan, on 07/21/2008, -2/+37I suspect it's just an optimization issue, the windows version receives alot more attention and as a result is probably much better optimized. Another part may have to do with hardware drivers, flash, at least recently, uses 3d hardware to help it draw things, and if the 3d drivers are *****, then i suppose your performance will be ***** too.
- Eezyville, on 07/22/2008, -0/+33Because Adobe hates you.
- semvhu, on 07/22/2008, -5/+37Epic fail. Someone needs to send him a link to the wikipedia article on the scientific method.
http://en.wikipedia.org/wiki/Scientific_method - artfuldodga, on 07/21/2008, -29/+54Vista Ultimate for myself sits at 900MB RAM, 37 processes (no feature compromise), after a bit of further cleanup requires a mere 7GB for a full backup w/ 3rd party apps installed, that and normally sits idle at 1-4% CPU usage, usually because of an IM app etc, lets face it... Windows Vista is a solid OS
I don't know what kind of setup David M Williams has, but he surely has no idea how to get the most out of a Windows OS, that and he seems to have no idea about Superfetch. All that said, I admire ubuntu alot, great OS - Ademan, on 07/21/2008, -1/+25Yeah... i've got :
4 firefox windows open with a total of 28 tabs + 4 empty tabs (stupid pages wouldn't load). it's been up all night
8 instances of gnome-terminal
4 nautilus windows.
transmission with 3 torrents, one seeding, 2 sitting idle.
pidgin connected to aim with 2 screennames, and google talk
tomboy
2 instances of gvim
irssi (i was playing around with it)
xchat-gnome
tracker is running
+ all of the gnome normal stuff like gnome-panel gnome-settings-daemon gnome-keyring-daemon, metacity etc etc
ubuntu 8.04
i'm currently using 727.3mb out of 2gb ram
39.4mb of 1.9gb swap
I *sincerely* doubt fedora managed to take up 1gb on a fresh reboot.
Also, my swap usage is inconsistent with what the article said about linux always using ram first. Although I suspect the swap was used during a time when I was using a ton of memory and just was never freed, i still find it interesting. - evillawngnome, on 07/22/2008, -1/+25"a mere 7gb"
I don't remember what XP took up on a fresh install (2G?), but i can get FC8 installed on 1.2GB, and a server install on 700MB. - DarKEVA02, on 07/22/2008, -3/+26Why didn't he just used FF 3 on both OS's?
- flashpanda, on 07/22/2008, -7/+29Why the hell is this on the front page of Digg? It's absolute trash. Even if he removed all of his (many) variables, it's a pointless comparison.
Or is it anything that has Windows vs Linux in it gets dugg...this is retarded. - artfuldodga, on 07/21/2008, -3/+25These folks that do the reviews are far from experts
- inactive, on 07/22/2008, -9/+29No graphs? Stupid article, I wont read that lazy wall of text
- johnnyboy239, on 07/22/2008, -1/+20I couldn't agree more, even if he accounts for the computers being different he's then going on and comparing IE7 on Windows to FF3 on Linux.... Why not just compare FF3 on both? Same with the documents, he's using openoffice on linux but MS. Word on Windows... If this proves anything (which it does not) it is that the open programs available are better then big name microsoft programs, Not necessarily that linux is better then windows.
- inactive, on 07/22/2008, -1/+20firefox + flashblock it still allows you to play the flash you need, but when you want to (doesn't automatically load )
- Royall, on 07/22/2008, -0/+19The Linux version didn't support graphics acceleration until the latest version. Also, that one still doesn't do graphics acceleration if there's another graphical program running, such as Compiz.
- evillawngnome, on 07/22/2008, -4/+21tl;dr. Where are the charts and graphs? Why so much DENSE TEXT?
Also, why even bother comparing performance on two dissimilar machines? Why not make a backup of your machine, then do the tests on fresh XP, Vista, and 2 or 3 desktop linux distros? - TomFrost, on 07/22/2008, -1/+18The bigger deal is process management. Those who regularly use both Linux and Windows at home in a quiet room can tell you that Windows is far likelier to amp up its CPU or HDD usage while otherwise seemingly idle than Linux is. The worst part is that it significantly slows down the system when you try to use it during that period -- even if what it's doing is background search indexing, it should drop out when it's being used again.
And RAM being less expensive is NOT an excuse for shoddy coding. I shouldn't have to buy more hardware because of someone else's laziness. - f4nt0m4s, on 07/22/2008, -14/+31He should have done a gaming comparison...oh wait
- inactive, on 08/11/2008, -0/+17No, he's right about Linux always using RAM first.
- codeblue315, on 07/22/2008, -1/+15Did ya happen to catch when FF3 was released? I believe a couple ppl downloaded it.
/sarcasm - aroedl, on 07/22/2008, -0/+13Lot of people don't get Linux' memory management system. The easiest way to figure out how much memory is really used (sans buffers and cache) is to look at the "-/+ buffers/cache" line of free -tm.
- Ouze, on 07/22/2008, -2/+15why use 2 different, non-identical machines instead of.... dual booting.
Also, Symantec Antivirus is... not a part of Microsoft Windows. - DRJTower, on 07/22/2008, -6/+19Lets compare BP gas with Shell gas. Put BP gas in a mustang, put Shell gas in a focus. Hmmmm ... looks like BP gas is better. WTF!!!
- qwertycopter, on 07/22/2008, -2/+14Why didn't you compare Firefox 3 on Vista versus Firefox 3 on Linux? Same goes for Open Office.
Of course there are memory differences between IE and FF (MSOffice and OO). Are you testing the apps or the operating systems? - inactive, on 07/22/2008, -3/+14Things I'd have liked to see in the comparison:
1. Equal base hardware, if it comes to it review and benchmark one OS on a laptop then format it and install a different OS - repeat as necessary...
2. Show boot up times, application load times (for comparable apps, e.g. OpenOffice, FireFox) memory usage for identical applications on different OS's
3. Leave both machines running with various different applications running for a couple of days without a reboot and see memory usage at that point. - arcade, on 07/22/2008, -1/+12Hi there david.
I noticed what you wrote about Linux losing some memory when you closed firefox.
I suspect the application you used to monitor memory usage is faulty, or you are misinterpreting the numbers. The linux kernel gladly caches memory until it's needed by something else.
Try opening an xterm (or console). Type 'free'.
You're interested in the second line that says something about -/+ buffers/cache: .. more specifically the column which says 'free'. This is the amount of free memory when what is used for buffers and cache is substracted. Memory the kernel may free at will if it gets a tad tight.
Try typing 'free' before opening firefox, and some seconds after closing it. Does the numbers look approximately the same? Mine varies with about 10K or so - which is probably due to the ten or so other applications I've currently got open. - JD52, on 07/22/2008, -4/+14Apples vs Oranges FTL
- tomz17, on 07/22/2008, -3/+13Agreed... this article was a complete waste of time. The comparisons were absolutely worthless (at least use comparable software and the same damn computer!), and anyone drawing any conclusions based on these results is a complete moron.
- nullx42, on 07/22/2008, -0/+9...and before that Macromedia hated you.
- bratterscain, on 07/22/2008, -2/+11So much is wrong with this, I don't know where to start. He's also comparing apps mostly, not exactly the OS's themselves. Ram usage is a terrible indicator of how well a system works. Some use a lot because they're ***** programmed while some use a lot because they're efficiently programmed to make use of pre-cached ram, same with OS's. This is not a FACE OFF. There is no battle to the death. It's just a terrible comparison involving many different variables.
- FuzzyLumpkins76, on 07/22/2008, -2/+11That comparison was terrible! I guess this guy has not conducted any sort of (credible) scientific experiment before in his life. Like everyone stated above, he was using two different laptops with rather different hardware configurations - that makes absolutely no sense. He then said that both computers had 4GB of RAM. Did they really? He neglected to mention if he was using a 32 or 64 bit operating system which would change the amount of memory seen by both operating systems. He then went on to use different applications on each machine and try to benchmark them. This turned into more of a "which applications are faster/take up less ram" then which operating system. He should have standardized by using applications that are available on both systems - FireFox works on both, so does OpenOffice. Overall... terrible. This guy has no credibility.
- kayfouroh, on 07/22/2008, -2/+11Didn't know Windows Vista came with Symantec Antivirus and OneNote.
- rumblpak, on 07/22/2008, -2/+10So he tested two different operating systems, with two different sets of hardware, two different sets of software, with different amounts of programs running in the background (including an antivirus in windows which will keep programs in ram longer to run scans) and calls it a fair and balanced test? When did linux users become Fox News?
- f4nt0m4s, on 07/22/2008, -1/+9It was a lighthearted comment told in a joking mannerism, lighten up...are you that insecure about being a Linux user?
- Cr85bro13, on 07/22/2008, -4/+12I stopped reading right after: 'For this test I have two laptops. Unfortunately they’re not strictly equal,'. Buried.
- inactive, on 07/22/2008, -0/+7Macromedia's Linux flash player was a poor excuse for a port.
Adobe's Flash Player Linux 10b2 'Astro' is a decent port... could use some work... - topflight, on 07/22/2008, -1/+7David,
Your article seems to be inaccurate. You should read some "Internals" books before writing something like this. For example, Understanding the Linux Kernel, Windows Internals, or Mac OSX Internals. There are reasons why memory doesn't go back to initial state after running a program. The OS needs to keep the pages in memory to avoid having pagefaults on subsequent runs. It's smart of the OS to do this. This relates to superfetch (which doesn't seem so super) because superfetch preloads your files (executable or data) to avoid hard page faults. And, I don't know how you know that Windows moved it out to the swap, that's hard to figure out unless you hack into the kernel. Also, which Windows numbers are you using -- there's a difference between the working set, virtual memory usage, etc. What linux numbers are you using as well? For example, if you are comparing the working sets, but not actual virtual memory usage, i.e. number of pages (private + shared) that each process uses, then its not accurate. Both can be using the same amount of virtual pages, but each OS chooses what the working set should be. If your point is that Vista sucks -- it sucks, and there are lots of reasons that it sucks. Its optimized to use as much memory as possible to speed up page faults, etc, so your experiment just shows its natural behavior. And since its natural behavior is an optimization, and you are comparing apples to oranges. And yes, Vista does suck. - dkoon, on 07/22/2008, -0/+6That's just stupid, the point of doing benchmark and comparison using same hardware and software is that at the end it will reflect the real usage, it may not be exactly the same as "Real World Usage" numbers. But I'm pretty sure it will be much more accurate than the "Real World Usage" numbers, which can be affect by hundred other factors that make it unfair for testing purpose.
- Ademan, on 07/22/2008, -1/+7Well I wasn't really disputing the author, as i thought the exact same thing, i was just kinda reporting.
I asked around and what someone mentioned was that ubuntu shoots for about 3:2 for ram:swap, but it probably only does that under certain conditions... - thatsgoodkarma, on 07/22/2008, -0/+6Stop talking.
- lynx44, on 07/22/2008, -1/+6The original objective makes sense, but it would be interesting to see the two compared using FF3. I think the audience that is intended for this article is more interested in those results because they don't typically use IE.
- tomarocco, on 07/22/2008, -0/+5The real test is to take away the mouse.
- fas2, on 07/22/2008, -3/+8No idea why are are being buried, it is a legitimate point. Firefox is clearly the greatest browser in the history of mankind, but the average dumb user knows nothing about it and uses the preinstalled IE7.
- drunkinbda, on 07/22/2008, -2/+7Just to add, i think if you are going to assume a "real world usage" then you also need to use more of the "avaerage Windows end user" that really has no clue about or does not care about optimizations, tweaking, memory usages, etc etc. They just want it to work and as much as fanboys denounce it, windows does that.
- kayfouroh, on 07/22/2008, -8/+12"Consequently, I also haven’t disabled the WiFi or Bluetooth drivers on either machine, along with the various programs that run on startup. This includes Symantec Antivirus on the Windows machine as well as OneNote (Windows) and Tomboy (Linux) used for making quick notes."
... Because Tomboy is as CPU intensive as Symantec Antivirus, right? What a load of crap. Buried. - pockiez, on 07/22/2008, -1/+5Why did he even write this article? Why not just fill all those pages with "Linux wins," if that's how he was going to push it to end. Look, even the Linux fanboys don't have anything nice to say. I like Linux, but give me a break.
-
Show 51 - 100 of 167 discussions
© Digg Inc. 2009
— Content created and posted by Digg users is