The Linux Experiment

As you know I am currently running OpenSUSE 11.2 on my laptop. While I have enjoyed my time using it, I have noticed that this particular distribution tends to be on the heavy side of memory usage. This got me thinking. If OpenSUSE uses this much memory on my machine, how could it possibly run on a machine with 512MB of RAM (the lowest recommended amount)? If Ubuntu is the most popular distribution, but it is also, what I would call, a fully-fledged desktop distribution, then how does it manage given tighter memory constraints? And so the mini-experiment begins.

Points to make before I begin

• This is not a very scientific study, but rather something I did in my spare time because I was curious.
• I have picked the majority of the most popular desktop distributions. These distributions were chosen not because they were designed for minimal system specs but rather because they are popular and provide a full desktop experience out of the gate.
• What do I mean by full desktop experience? The distribution should be easy enough for a novice Windows user to install, should come with all of the standard software for desktop activities, and should not require any fine tuning.
• What you won’t find here: DSL, Arch Linux, Slackware (only because it failed at installing in VirtualBox), Gentoo, or other ‘expert’ distributions. You also will not find netbook remixes or low-resource specific distributions. This experiment is designed to see how these big distributions run on little RAM, nothing more.
• Please do not post things like ‘you forgot to test XYZ’ or other useless comments that don’t actually help the discussion. Yes I am sorry I missed your favourite distribution, but grab a tissue, clear the tears from your eyes and let’s all move on with our lives.

How I tested them

The process was identical for all tested distributions. I set up a new virtual machine inside of VirtualBox with the following specs:

• Total RAM: 512MB
• Hard drive: 8GB
• CPU type: x86

The tests were all done using VirtualBox 3.1.4 on Windows, and I did not install VirtualBox tools (although some distributions may have shipped with them) nor did I change the screen resolution from the default 800×600.

Results

I have broken the results down into a variety of categories and included fancy graphs just for you!

First boot memory (RAM) usage

For this test I installed the distribution and then on its first (post-installation) boot measured the amount of memory it used. This was to gauge the amount of resources that the stock distribution required before any updates.

Average first boot memory (RAM) usage by packaging type

This shows the average memory usage broken down by the packing type used.

Memory (RAM) usage after updates

This was a test to see whether or not system updates caused the memory usage to increase or decrease. I updated the system with all current updates and then rebooted and measured the resource usage again.

Memory (RAM) usage change after updates

This graph shows the usage difference between installation and post-updating. The formula I used was [after updates - initial installation].

Average memory (RAM) usage after updates by packaging type

Similar to above. Again this is broken down by packing type.

Filesystem layout

This is a simple graph showing the partitions that each default setup created as well as the relative size of them.

Filesystems used in partitions

This graph shows the different filesystems used for the various partitions. For example if a distribution has a value of 2 under ext4 that means that it used ext4 in two different partitions.

Occurrence of filesystem by packaging type

This graph shows the number of distributions who used a certain type of filesystem. It is broken down by packing type.

Install size after updates

This is the total OS install size after downloading and installing all of the updates. This should represent a fully updated version of the distribution.

Average install size after updates by packaging type

This shows the average install size of the distributions broken down by packing type.

Conclusion

Make you own! …well it is pretty obvious that some of these distributions would perform better than others given these low system specs. There are however other things to consider. For example which packing type you prefer, or for that matter which package manager.

        

Linux, Tyler B 512MB, distributions, Linux, RAM

As some of you already know, I am a big fan of the KDE desktop environment (or KDE Workspaces or whatever they’re calling it these days). In my search to reach Linux KDE perfection I have tested out a number of different distributions. First there was Fedora, which I happily ran throughout the length of the experiment. Once that was finished I attempted to install and try both Kubuntu and openSUSE. Unfortunately I was unable to do so after openSUSE decided not to play nice. However my search did not stop there, and once the community edition was ready I jumped over to Linux Mint KDE CE. Finally I decided to once again try openSUSE, this time installing from a USB drive. This somehow resolved all of my installation issues.

Now that I have tried out quite a few of the most popular distributions I figured I would write a little bit to tell you fine people my thoughts on each, and why I will be sticking with openSUSE for the near future.

Fedora 11

• KDE Version: 4.2 – 4.3
• Pros: very secure, not too many modifications of the KDE source, cutting edge
• Cons: could have really used some more modifications of the base KDE packages in order to better integrate GTK+, Bluetooth problems, not always stable
• Thoughts:

  I have written at length about my experiences with Fedora during this experiment. Without re-writing everything again here let me simply say this: Fedora is primarily a GNOME distribution and I could never shake the feeling that KDE got the left-over treatment.

Kubuntu

• KDE Version: 4.3
• Pros: very easy to use, nice integration of GTK+ and GNOME notifications, access to Ubuntu support
• Cons: the hardware drivers application (jockey) simply did not work, very bad sound issues, Firefox could not handle opening file types
• Thoughts:

  When I first installed Kubuntu I was thrilled. Ah, this must be what it’s like to use a real KDE distribution, I thought. Everything seemed smoother and far more integrated then it did in Fedora. For example: OpenOffice.org had a KDE theme and it’s file browser actually used the native KDE one. Furthermore the notification system was awesome. Now instead of a GNOME application, like Pidgin, generating GNOME notifications, it instead integrated right into the standard KDE equivalent.

  Then the problems started to show up. Oh I’ll just download this torrent file and… hmm Firefox doesn’t seem to know what to do with it. Why can’t I set the file type options inside of Firefox for torrents? Why doesn’t it use the system defaults? Then the sound issues came. YouTube stopped putting out audio all together and all of my attempts to fix it were futile. Maybe it’s just my hardware but Kubuntu just could not handle multimedia at all.

  While Kubuntu is definitely one of the better KDE experiences it is by no means problem free.

Linux Mint KDE CE

• KDE Version: 4.3
• Pros: excellent package manager, easy to use
• Cons: sound issues, WiFi issues, is this actually a KDE desktop? there are so many GTK+ applications in it…
• Thoughts:

  After hearing much praise for Linux Mint I decided to give the newly released KDE community edition a go. I must say at first I was very impressed. The package manager was far superior to KPackageKit and even included things like user ratings and comments. It also came bundled with many tools and applications designed specifically for Linux Mint. Sadly very few of these were re-written in Qt and so I was forced to deal with GTK+ skinning almost everywhere.

  Sound issues similar to those in Kubuntu (maybe it’s something in the shared source?) started to crop up almost immediately. Again YouTube just did not work no matter how much I tried to fix it. Finally the WiFi connection was very poor, often disconnected on what seemed like a  specific interval.

  While I think this distribution has a lot going for it I can only suggest the GNOME desktop for those who want to give it a try. The KDE version just does not seem polished enough to be recommended for someone looking for the ultimate KDE distribution.

openSUSE

• KDE Version: 4.3
• Pros: very responsive, a lot of streamlined tweaks, rock solid WiFi, excellent audio
• Cons: slower to boot, uses quite a bit of RAM, too much green
• Thoughts:

  Installing openSUSE seemed like an awful idea. After reading all of the complaints that both Phil and Dave had written over the course of the experiment I have to admit I was a little hesitant. However, I am very happy I decided to try it anyway; openSUSE is an excellent KDE distribution.

  Everything about it, from the desktop to the little helpful wizards, all seem to be designed with one purpose in mind: make openSUSE the easiest, or at the very least most straightforward, distribution possible. YaST, often a major source of hate from my fellow Guinea Pigs, does indeed have some quirks. However I honestly think that it is a very good tool, and something that streamlines many administrative tasks. Want SAMBA network sharing? Just open up YaST and click on the wizard. Want restricted codecs? Just hop on over to openSUSE-Community and download the ymp file (think of it like a Windows exe).

  My time with openSUSE so far has been wonderful. My network card seems to actually get better range then ever before, if that’s even possible. My battery life is good and my sound just plain works without any additional effort. If I had one complaint it would be with the amount of RAM the distribution uses. After a quick reboot it takes up a very small amount, around ~350MB or so. However after a couple of hours of general use the RAM often grows to about 1-1.5GB, which is far more than I have seen with the other distributions. Thankfully I have 4GB of RAM so I’m not too worried. I wonder if it has something to do with the fact that I am running the x64 version and not the x86 version. Perhaps it assumes I have at least 4GB of RAM for choosing the newer architecture.

  Whatever the case may be I think I have finally found what I consider to be the very best KDE Linux distribution. Obviously your results may vary but I look forward to hearing what you think.

This piece was cross-posted over at my person website ‘TylerBurton.Ca‘.

        

KDE, Tyler B, openSUSE KDE, Kubuntu, Linux, Linux Mint, openSUSE

Recently while browsing the Internet I noticed that Mozilla Firefox was taking up an awfully large amount of RAM. In fact it’s RAM use continued to grow even though I had not navigated to any new website or even touched it at all! Being a KDE user I decided to try the same website within Konqueror, the KDE web browser that everyone seems to toss to the side right after install. To my amazement Konqueror seemed to render the same website, using much less RAM, and better yet it’s RAM use remained static. Perhaps there was something more to this ‘throw-away’ browser than I had first thought. And thus began my idea for a series of comparisons of four of the most popular Linux web browsers: Firefox, Chromium, Konqueror and Epiphany.

Note: The numbers you are going to see below are purely anecdotal and are based on my own personal experiences, which might not represent your own.  For the record I am running Kubuntu 9.10 with the 2.6.31-19 kernel.

Packages I had to install

This is simply a list of packages I installed for each browsers to sort of give you an idea of what to expect.

Startup speed (from a cold start – i.e. from reboot)

To test this I rebooted the computer and then opened the browsers for the first time.

Startup speed (from a cached start – i.e. after opening and closing the program)

To test this I rebooted the computer, then opened the browsers for the first time and closed them. Then I opened them a second time and recorded the time.

Memory usage (about:blank)

This is how much memory the browser took to display the web page in brackets. The memory inside of the brackets is the amount of shared memory each browser uses. This of course could change depending on your system and setup. Where there are multiple numbers separated by plus signs it means that the browser spawns multiple processes which each take up memory.

Memory usage (about:blank -> kubuntu.org)

For this test I first loaded the browser to the homepage, in this case about:blank, and then navigated to the website www.kubuntu.org. In Konqueror’s case it spawned a few small KIO processes which I assume did the actually downloading of the webpage. I have averaged their values below.

Memory usage (about:blank -> kubuntu.org | google.com | bing.com)

For this test I once again started with the about:blank homepage and then opened the websites in different tabs.

Memory usage (about:blank -> kubuntu.org | google.com | bing.com) 2 minutes later

For this test I started with the above test and then waited two minutes and re-recorded the memory usage. This was to see if the browsers suffer from any sort of memory leaks, something Firefox has been famous for over the years.

Memory usage (about:blank -> kubuntu.org | google.com | bing.com | ubuntu.com)

Same as above but with one more tab.

Memory usage (about:blank -> kubuntu.org | google.com | bing.com | ubuntu.com) 2 minutes later

Same as above but with one more tab.

Noticeable rendering glitches in pages viewed

None of the browsers had rendering glitches except for Konqueror. In Konqueror Bing’s search bar was offset from where it should be and on Kubuntu’s website the header image was overlapping text that it shouldn’t have.

Acid 2 test (http://www.webstandards.org/files/acid2/test.html)

The (now old) Acid 2 test.

Acid 3 test (http://acid3.acidtests.org/)

How well did the browsers handle the Acid 3 test?

SunSpider JavaScript Benchmark (http://www2.webkit.org/perf/sunspider-0.9/sunspider.html)

The classic JavaScript benchmark test!

Conclusion

Well there you have it. Rather than ‘rate’ each browser I’m going to leave it up to you to weigh the merits of each given the above information. Even though Firefox might not be the most technically superior browser it does have the advantage of being very popular and having an excellent add-on system. Then again Epiphany can use most of those add-ons as well. Chromium, based off of Google’s Chrome browser, is also a new favourite for a lot of people and is obviously the fastest in terms of JavaScript. Being a KDE user, Konqueror might be the best browser for you thanks to its infinite customization and uniform look.

Linux is all about personalization and I think everyone should extend that to the browser they use. Who knows you might just find one you like even more.

        

Free Software, Kubuntu, Linux, Open Source Software, Tyler B Chromium, Epiphany, firefox, konqueror, Linux, Web Browsers

Since the end of The Linux Experiment I have started dual booting my laptop, switching between Kubuntu 9.10 and Windows 7 as needed. While this solves all of my compatibility issues, it does pose some more annoying issues. For example after setting up one operating system just the way I like it I now need to do the same for the other. Furthermore after becoming used to using particular applications under Linux I now have to find alternatives for Windows. Well no more! The KDE guys and gals have ported the libraries to Windows!

Installing

To install KDE on Windows all you need to do is head over to http://windows.kde.org/download.php and grab a copy of the installer exe. This will more or less walk you through the initial setup and then present you with a list of packages you can choose to install. Most applications are there including things like KTorrent, Konqueror, Konversation and more! Simply select them and watch as they are easily installed.

Image Walkthrough

The first screen you'll see when installing	The package list
kdebase-apps includes things like Konqueror	The installer downloads the source and compiles it locally
After installing the applications show up right in your start menu	The final result. Konqueror and KWrite running on Windows

        

Free Software, KDE, Tyler B KDE, windows

Well I managed to fix my compositing problem but I honestly don’t know why it worked. Basically I went into the System Settings > Desktop > Desktop Effects menu and manually turned off all desktop effects. Next I used jockey-text to disable the ATI driver. After a quick restart I re-enabled the ATI driver and restarted again. Once I logged in I went back into the System Settings > Desktop > Desktop Effects menu and enabled desktop effects. This magically worked… but only until I restarted. In order to actually get it to start enabled I had to go back into System Settings > Desktop > Desktop Effects and then click on the Advanced tab and then disable functionality checks. I am sure this is dangerous or something but its the only way I can get my computer to restart with the effects enabled by default.

I’m really starting to hate this graphics card…

        

God Damnit Linux, Hardware, KDE, Kubuntu, Tyler B ATi, desktop effects, graphics card, Kubuntu

After giving up on installing openSUSE 11.2 (the installer insisted on failing during the partition stage) I am now successfully dual booting Windows 7 and Kubuntu 9.10. On the Kubuntu side of things I must say that overall I am impressed. It is a much sleeker and more refined KDE experience than I was treated to on Fedora and I am enjoying it thus far.

That’s not to say it is without it’s own set of problems though Once again my graphics card does not play nice, although this time I place the blame firmly on Kubuntu. The KDE front end for jockey, otherwise known as Hardware Drivers to the *buntu folk, completely failed at activating my ATI driver. I was finally able to enable the driver by using the text version of jockey, “jockey-text” in the terminal, but that still didn’t solve all of my problems. Now the system uses the correct driver but refuses to enable compositing for some reason. -sigh-

Looks like I have a long afternoon of IRC ahead of me…

        

Fedora, KDE, Kubuntu, Tyler B 9.10, Fedora, Hardware Drivers, jockey, Kubuntu, Windows 7

Well it’s official, the year is now 2010 and we still don’t have flying cars.

The End

2010 also marks the end of The Linux Experiment. I can honestly say that the last four solid months of Linux use has taught me a lot. In reflection of this I decided to look back at what I had originally wrote about my goals of this experiment and see just how many of them I had accomplished.

• I will have learned enough of the ins and outs to be as comfortable within a Linux environment as I current am within a Windows one.
  • This one is a bit tricky to answer. I am far more familiar and comfortable with Linux now than I have ever been before. However I still do not understand a number of things. For example the Linux file system confuses me to no end. What is the difference between /bin/ and /sbin/? Or why do some things end up in /etc/ and others in /var/ or even /opt/? Clearly I have some room to improve here.
• My bonus goal is to have a fully functional, self-created, program that runs native to Linux.
  • This one I was actually able to realize. Not only did I have a native OpenGL program running, but in recent weeks I have even created cross-platform .NET/Mono based applications. In addition Linux has proven time and again that it is the platform for web development. I can definitely see myself utilizing it as such in the future.

Fedora

Fedora has been both a terrible nightmare and an absolute pleasure. I have had more problems getting things to just work on this distribution that I care to even remember. Yet time and time again there was something about Fedora that just kept pulling me back in. Perhaps it was the challenge of trying to master a power user’s distribution of choice. Or maybe it was just pure stubbornness. The fact remains that with the exception of Fedora 12 being incompatible with my graphics hardware there was nothing I haven’t overcome.

So would I recommend Fedora to someone? Well… yes and no. Fedora has a rock solid community and lives right on the cutting edge (what? I’m already running KDE 4.3.4??) but it does not make things easy. Now that most distributions have moved up to the 2.6.31 kernel there is really less of a reason for me to recommend the cutting edge simply as a way to get decent hardware support. Obviously if your machine is even newer than mine than perhaps Fedora is still your only stable ticket to that support, but for most users I think there are far better alternatives. Don’t get me wrong, I really do enjoy Fedora and from what I have read it has come a long way in recent years, I just don’t think I will be using it again anytime soon.

The Future

Today will bring some changes to my computing setup as I plan on removing Fedora and trying out two new KDE distributions, OpenSUSE and Kubuntu, just to see which one I prefer. In addition I will be dual booting with Windows 7 for the first time. I will be sure to keep everyone up to date with my experiences as I do so.

As we here at The Linux Experiment debate where to take the experiment moving forward, be sure to check back for updates on our new experiences!

        

Fedora, KDE, Kubuntu, Tyler B, openSUSE Fedora, Kubuntu, Linux, openSUSE, The Linux Experiment, Windows 7

In my daily RSS feeds I read about the release of the newest version of XBMC, formally the Xbox Media Center, so I decided to check it out.

While the maintainers do not specifically support Fedora with pre-built RPMs, they do offer instructions on how to build it from source here. Even so, I did run into a couple of little problems along the way. For example on the step that says to enter

*sudo ln -s /usr/lib/mysql/libmysqlclient.so.15.0.0 /usr/lib/libmysqlclient.so
*sudo ln -s /usr/lib64/mysql/libmysqlclient.so.15.0.0 /usr/lib64/libmysqlclient.so

depending on if you are running the x86 or x64 version of Fedora, I needed to change this to say

sudo ln -s /usr/lib64/mysql/libmysqlclient.so.16.0.0 /usr/lib64/libmysqlclient.so

because that is the current version of my library. In addition running

./configure

failed due to an error with OpenSSL, specifically its lack of something called “openssl/ecdsa.h”. I managed to fix this by altering the source code according to the patch found here. Then before re-running ./configure I had to run

autoreconf –force –install

(that’s two dashes in front of force and install!) from within xbmc/cores/dvdplayer/Codecs/libbnav. Once that was done the ./configure ran smoothly. From then on I simply followed the rest of the instructions and I was in business!

There really is only one word to describe this version of XBMC: AWESOME!

It picked up my pictures, videos and music from all of my network shares and local drives without issue. The user interface is absolutely stunning as well. At one point I had Star Wars playing in the background (still in view) while navigating beautifully rendered and slightly transparent menus to adjust other system settings. It can even be configured to pull down information about the movies from the Internet, including who stars in it and what the plot is. The music playback is similar and offers a variety of visualizers for your viewing pleasure. The picture options allows for very neat slideshows, accompanied by your own music playing in the background, which would be great for atmosphere at a party.

From Wikipedia here are just some of the features supported by this release:

• Physical media: CDs, DVDs, DVD-Video, Video CDs (including VCD/SVCD/XVCD), Audio-CD (CDDA), USB Flash Drives, and Hard Disk Drives
• Network/Internet protocols: UPnP, SMB/SAMBA/CIFS, XBMSP, DAAP, HTTP, HTTPS, FTP, RTSP (RTSPU, RTSPT), MMS (MMSU, MMST), RTMP, Podcasting, TCP, UDP, SFTP, RTP
• Container formats: AVI, MPEG, WMV, ASF, FLV, Matroska, QuickTime, MP4, M4A, AAC, NUT, Ogg, OGM, RealMedia RAM/RM/RV/RA/RMVB, 3gp, VIVO, PVA, NUV, NSV, NSA, FLI, FLC, and DVR-MS (beta support)
• Video formats: MPEG-1, MPEG-2, H.263, MPEG-4 SP and ASP, MPEG-4 AVC (H.264), HuffYUV, Indeo, MJPEG, RealVideo, RMVB, Sorenson, WMV, Cinepak
• Audio formats: MIDI, AIFF, WAV/WAVE, MP2, MP3, AAC, AACplus, AC3, DTS, ALAC, AMR, FLAC, Monkey’s Audio (APE), RealAudio, SHN, WavPack, MPC/Musepack/Mpeg+, Speex, Vorbis and WMA
• Digital picture/image formats: RAW image formats, BMP, JPEG, GIF, PNG, TIFF, MNG, ICO, PCX and Targa/TGA
• Subtitle formats: AQTitle, ASS/SSA, CC, JACOsub, MicroDVD, MPsub, OGM, PJS, RT, SMI, SRT, SUB, VOBsub, VPlayer
• Metadata tags: APEv1, APEv2, ID3 (ID3v1 and ID3v2), ID666 and Vorbis comments for audio file formats, Exif and IPTC (including GeoTagging) for image file formats

For a sampling of the beautiful new interface check out their official wiki here. I apologize for this sounding a lot like an advertisement but in all honesty I am floored by how impressive this application is and would highly recommend it to anyone looking for a sweet home theater setup. Try it out now!

        

Fedora, Free Software, Linux, Open Source Software, Tyler B Awesome, Fedora 11, Linux, XBMC

Thanks to incompatibilities with ATI’s catalyst drivers and the newest version of Xorg used in Fedora I am still unable to install Fedora 12 on my machine. Yes, I know there are open source alternatives but frankly they don’t work well and actually caused my system to be unstable during the brief period I did use them.

I had really hoped that I would be able to try out Fedora 12 in a meaningful way before the end of this experiment but I just don’t see that happening…

        

Fedora, Free Software, Open Source Software, Tyler B ATi, Catalyst, Fedora 12, Xorg

Finished Exams

Rather than just copy Sasha’s previous post, I will do my best to try and tell the story through the eyes of a Computer Science undergrad. Now that I have finally finished my exams for this term I can safely say that Linux has not impeded my coursework and in fact has given me quite a seamless user experience.

Web Development

Designing websites and creating server side programs has been an absolute delight in Linux. Unlike within Windows, I can easily mount a remote SSH server as a browsable folder in my file system in Linux, making additional file transfer programs unnecessary. This lets me edit the files in my favourite editor, which more often than not was just KWrite, and then watch as they updated on the remote server with a simple click of the save button.

Graphics Programming

For a different course I was required to program 3D graphics in OpenGL. On Windows my professor had recommended Dev-C++, a program I am familiar with but not exactly a fan of. Thankfully we weren’t doing anything platform specific and thus I was able to make use of the exact same OpenGL and GLUT libraries to get the job done on Linux. As a replacement for Dev-C++ I started with Eclipse but eventually settled on MonoDevelop as my IDE of choice. Even better I was able to share the exact same code with a fellow classmate for our group project, which he was in turn able to compile on Windows in Dev-C++ with no modification whatsoever!

Pretty Standard Stuff

The rest of my time spent at University was of pretty standard fare: note taking, web browsing, e-mailing, instant messaging, assignments, etc. Linux performed superbly at these tasks as well and handled everything I could think to throw at it – even our school’s insane Wi-Fi network configuration.

Three Months of School Later

And there you have it. My experience with Linux during my term at school has been, like Sasha’s, excellent. For those of you out there worried that trying out Linux will impact your school or work or have concerns that you won’t be able to find replacements for your generally Windows or Mac centric worlds, I can attest to the exact opposite being true. Give Linux a shot, it might even make you more productive! Hell, you just might even like it

        

Fedora, Free Software, Linux, Open Source Software, Tyler B Dev-C++, Eclipse, firefox, GLUT, KWrite, Linux, MonoDevelop, OpenGL, School, ssh, Thunderbird