Posted on:
August 26th, 2009
This post will go over a controversial YouTube video with a fine tooth comb. The reason why I find this so controversial is that the video is presented in such a way that it makes it seems like one operating system is a lot better then the other. My mother always told me to take everything with a grain of salt and this video is no exception.
Please read the whole article before you watch the video. I am hoping you pickup on the same items that I did on my first watch of it.
The whole setup of this video seems to be done in such a way as to make the watcher think that these two laptops are identical. They never post any of the specs for either of the laptops. In order to make a judgment we can’t assume that these are the same machine just because they look identical. This is the weakest argument and I figured I get it out of the way first.
The biggest piss off of this whole video is that the people who did this made sure to disable the 3D effects or at least use a very stripped down 3D version of Ubuntu. Unlike the Ubuntu rig, the Windows 7 machine is using the Windows default settings. Aero (the interface) is full on so you can enjoy all the 3D goodness. So what you just showed me in the video is that an OS without 3D effects boots faster then an OS with 3D. Well duh! Anyone who is anyone could probably guess that!
Another item that could play into the whole longer boot time is there is most likely an antivirus program that get loaded on Windows. Depending on the manufacturer (ahem Symantec/Norton/McAfee/fSecure) a machine’s boot time can be lengthened substantially!
The person shooting the video also opens Firefox and it SEEMS to open faster on the Ubuntu machine then the Firefox does on the Windows machine. I found this one really funny! (You could almost call this one a blooper that could actually hurt the whole “Ubuntu is better then Windows 7” theme). When the Firefox opens in Ubuntu. Look at the “Bookmarks” section. Notice anything? I will Give you a hint!
Count how many favorites the browser has! It is using the default Firefox bookmarks! Now…
The Windows Firefox takes so much longer to load. Is this a sign of Bad programming on Microsoft’s part or Firefox?
(Something to think about.)
The next thing to look at is the “Bookmarks” on the windows 7 machine! How many bookmarks does it have? It actually looks like someone actually used Firefox on this machine doesn’t it?
This is funny for 3 reasons.
1.) Obviously this machine must be better! Why would someone who finds an OS so slow even bother using it if they had an the exact same machine that would perform better!
2.) It’s been heavily documented that Firefox gets slower and slower as it is used. The Firefox team just finally released a hack to clean up the sqLite DB that Firefox uses to store it’s info to help speed things up. Here’s the info in the issues: http://tuxtraining.com/2009/04/15/improve-firefox-3-speeds-with-tmpfs
3.) Microsoft has always act knowledge that opening a program under windows might take a little bit longer the first time it is launched. This is do to Windows caching. One thing this video doesn’t show (they do it on purpose) is that if you were to close Firefox on both machines and then open it again. The Windows 7 machine would instantly open the Firefox program where Ubuntu would prob take 2x’s that the Windows 7 machine would. How many people leave their windows open the whole day? I don’t know very many. To me I’d prefer the ability to open subsequent windows faster then just the first one being a tiny bit faster.
There is one more thing I would like to point out that was purposelessly left out. Although Windows 7 does take a bit of time to boot (It’s on par with Windows XP and much faster then Vista) Microsoft has REALLY worked on “Standby/Hibernation”. I feel Ubuntu’s Hibernation/Sleep Mode is where Windows 2000 was. It just doesn’t work if you have 3D items enabled. With windows 7 my laptop goes to sleep in 4 seconds with about 12 applications open. To Turn it back on is about 2 Seconds.
Now with this knowledge watch the video. I don’t think you will be nearly as impressed.
I have a few comments about your post, some agreeing with you, some not so much. First off the hardware specs on both machines; you’re spot on with this. The viewer does not know the actual specs on the machines. Even if they were the same machine to begin with, memory could have been removed on one, (ever try to boot Vista on ONLY 1GB?).
At least in 9.04, the default state of compiz IS to have 3d disabled, so in that aspect the demonstrator IS comparing the stock setups on both OS’s. Not to mention the window compositor doesn’t actually load until the GUI starts, so it has no threshold over boot time of the OS, (or very minimal if any). I didn’t notice any antivirus boot up, and Norton definitely doesn’t go unnoticed; another could be installed, so this is within the realm of possibility.
About the browser, you have another point there. The demonstrator should have used a clean install of both default browsers for the respective OS. In this case, it would have been FF3 vs. IE8. However as to leaving the windows open, I personally have windows open all day, in fact usually I run with about 10-15 apps open at any given point throughout the day, all spread out on different desktops. Not to mention, where is the cache kept for the application? If this cache is memory, then one could argue that memory would be more useful dedicated to running applications that actually need it, not applications that the user MAY start. This affect is felt even harder when the various ‘creative’ applications are used; would you want all Photoshop, Illustrator, and Autocad in cache at once? You wouldn’t be able to use the computer! (This I know from personal experience with Vista, hopefully M$ is allowing the user to change the settings with win version 6.1, er win “7″, but I expect not.)
On your standby issue, I have to agree with you there. The Linux kernel does not quite have sleep/hibernate working quite right yet. However, one thing about sleep, the computer is not actually shutdown, as if power is removed from the device, all unsaved data is lost. I know this is extremely obvious for many of us, but I feel uneasy if my data is saved on only 1 HDD, little alone volatile memory! But, for when you do not have important data and just need to throw your laptop in your bag after class or a meeting, sleep is usually good enough.
What would be an interesting benchmark is to take some piece of hardware and time various things, particularly installing the OS from bare metal and boot up/shutdown times after installed. Another interesting feature would be how long it takes for the drivers to work on the various OS’s. How many times have you installed a winOS, only to discover the OS cannot access a basic generic 10/100 NIC…
Hey Charlie
Thanks for the comment.
Run Windows 7 first and you’ll notice why most Twitter comment are giving it praise. I run the Cs4 suite with several of the programs open at once (photoshop,Dreamweaver,Illustrator), several browsers, and coding software and the machine never seems to slow down. This is both on a year old laptop and 6 month old desktop. I’ve been using Win7 since BETA, Then RC 1 and Now RTM. It’s so much better then anything out there right now. In my opinion MS is really getting their act together… (about time)… They have even released an antivirus program that can rivial Nod32
I wrote a review and provided a link on my blog. It runs on Windows XP, Vista and now 7. It takes very little memory Under 10MB and updates with Windows update. (You don’t need another program needing to run just to update the darn machine!) http://jaredheinrichs.com/free-microsoft-antivirus-microsoft-security-essentials.html
Another cool find that you might or might not have herd about is Microsoft Live Mesh! If you have more then one PC I can’t imagine living without this service. (Pssst You can even remote desktop into any computer that is part of your mesh from anywhere using your Windows Live account ) PC and Mac support is out right now… Phone support is coming soon
“What would be an interesting benchmark is to take some piece of hardware and time various things, particularly installing the OS from bare metal…”
What a surprise, today I stumbled onto just such a test. From my quick glance, the article seems pretty unbiased, even though it IS hosted on a pro-Linux site, [and I'm not talking about the host OS, as even this site is running on *nix
]. Now, these tests are from early this year, when both Jaunty and 7 were in early development, so both would have changed since now, (Jaunty has since been released).
Bottom line I pulled from the article: Ubuntu can do much more out of the box, takes up FAR less disk space, and beta software should not be used in production regardless of publisher. Other then that, both Intrepid/Jaunty and Vista/7 were comparable in tests.
http://tuxradar.com/content/benchmarked-ubuntu-vs-vista-vs-windows-7