Microsoft Challenges Linux's Legacy Claims

Michael writes "Microsoft Corp.'s Linux and open-source lab on the Redmond campus has been running some interesting tests of late, one of which was looking at how well the latest Windows client software runs on legacy hardware in comparison to its Linux competitors. The tests, which found that Windows performed as well as Linux on legacy hardware when installed and run out-of-the-box, were done in part to give Microsoft the data it needed to effectively 'put to rest the myth that Linux can run on anything.'"

Comments

Re:Window vs Linux

[JahToasted]

Exactly. Saying windows 95 works on a 486 and KDE doesn't is stupid. I can just as easily go on about how windowmaker, blackbox, rox, etc. run fine on a 486 while WinXP won't even install.

Bait and Switch / BS

[Irvu]

In the tests run in its lab, Microsoft found that most modern commercial Linux distributions could be installed successfully on systems with a Pentium processor, with 64MB of RAM and a minimum of 2GB of hard disk space.

"Memory prevented the successful installation on a typical 1997 system, as 32MB of memory is not enough to install most Linux distributions or to run desktop applications with acceptable performance. A memory upgrade could prolong the life of such hardware, but the cost and effort of locating old memory and installing it onto all corporate clients significantly reduces the potential savings," Hilf said.

Minimum requirements for office productivity performance on a Linux system were any Pentium II (PII) system with at least 64MB of RAM, he said, adding that playback of sound and video would typically require a PII 400 or better.

The salient points are in the statment above. The claim that "most" linuix distros had limitations preventing them from accessing a 32mb system with "aceptable performance" is entirely unsurprising. I note that neither RedHat (to pick one) nor Windows XP would like such a system very much, especially for modern "desktop application performance" (read OpenOffice and MS Office). In that case it is really the apps that are the limiting factors.

They never state what distros were tested (I assume Novell and RedHat when in doubt) nor how installation was done. Rather they pull a nice switching strategy. They test some unnamed distros and then state that windows CE is better than them on legacy hardware.

That is much like saying Windows CE is better than Windows XP on legacy hardware or that MuLinux is better than RedHat on older hardware. In both cases the former was designed for such a task while the latter was not. In both cases the former have limitations that prevent them from running "Modern Desktop Apps", that is in fact the point.

This is a simple "bait and switch" comparison, and if this is all the CTO uses when comparing all distros of linux to windows for some use; fire them.

Re:Phone Exchanges

[Linegod]

And the last time I checked, each one of those (except Mitel - they've flip-flopped a couple of times) started with Windows, and is shifting to Linux...

Re:The Study didn't prove that at all

[Coryoth]

I think, to be honest, that all that was really shown is that popular modern Linux desktop distributions are targetted at modern hardware, and as a result don't run as well on older hardware. They ran Red Hat and Mandrake and Novell etc. 'out of the box' with no customisation to make it fit with the hardware - unsurprisingly the default install of such distros a targetted at modern systems and have hefty system requirements.

Pick up a distribution that actually claims to target older hardware, or just generally fit in smaller places, like say Damn Small Linux, Feather Linux or Zenwalk and I suspect you'll find much better performance and much lower system requirements all 'out of the box'. The counter-claim seems to be that Windows CE, with the right customisations, will run on older hardware too. Does anyone know if their is a release of CE set up for desktop use on older hardwre?

Jedidiah.

Re:Not reading the article?

[kfg]

Yes, I found it very interesting that they spent about the first half of the article rationalizing why they didn't actually test a distro of Linux that will actually run on anything, like the single floppy I boot my 486 laptop from, which subsequently runs the system from rather meager memory.The entire "test" is founded on misrepresenting the claim that "Linux will run on anything."

I also always get a kick out of the "poor people are idiots who can't learn to run the system" argument as well. That'll really get them on your side and buying your products.

Dear Bill,

Let me give you a hint. Poor people have more time than money and expect to have to do things the self-sufficient hard way. Many of them even take pride in being able to do so.

And the local library is full of things called "books" and people who know how to read them. Like, computer books. That's where I went to read Kernighan & Ritchie. They've got a full set of Knuth too. Not to mention that computer training is a standard part of library science these days.

Nice try.

Yours,

KFG

Re:Wow, great news....

[Hosiah]

Yeah, Linux can run on a toaster or an old 486DX. Big fucking deal. I don't think too many companies are planning on running business apps on either of those platforms.

Do you realize how much the environment is choking on throw-away tech every year? I covered this very topic http://techn0manc3r.blogspot.com/2005/12/linux-and -environment.html with links. Yes, it's a huge deal. Count me in with the other who recycles old Windows boxes I find and gets year's further use out of them. More money to donate to FOSS, less waste to the environment.

By the way, when I worked for no less an enterprise than Citigroup incorporated, you couldn't walk two feet in the processing center without tripping over a 486. This was only two years ago.

Re:Come back

[cbreaker]

Actually, his whole post was wrong. Windows x64 runs more like a nightmare. Half the stuff is broken, a lot of 32 bit apps don't run correctly, drivers are not exactly easy to come by, and it's got just as much issues with security as the other versions of Windows.

Maybe Vista will make Windows 64-bit more seamless but I somehow doubt it.

Five miles through the snow

[crmartin]

I'm just going to say up front that this is not a comment with a conclusion.

I've been in this silly business for damn near 40 years (augh); my first computer had 8K of memory (yes, 8K, not megs) but we successfully ran a whole small business accounting system on it. 100 lines per minute chain printer. TI doesn't make a calculator that small.

I went to grad school in Computer Science in 1983; we ran a whole graduate department on a PDP 11/70. Less than a megabyte of RAM, maybe 250 MB of disk total. Less than one MIP. We got a VAX in 1985; suddenly we had a WHOLE MIP, and a shared terminal in each grad student office.

I'm writing this on a G5 MAC. God alone knows how many MIPS --- thousands, certainly. I use it alone.

Frankly, I'm not sure where all the cycles go.

Re:Toastworthy Computing

[Fred_A]

The problem is indeed that there are very few decent Windows admins.

I know that in 15 years of IT, admittedly not often working in Windows shops but often visiting numerous companies to help them setup FOSS solutions and integrate them w/ their existing (often Windows) stuff, I don't think I've met more than a couple decent admins. The others didn't appear to really understand what they were doing. That is if they were doing anything beyond the usual rebooting to fix the occasional glitch.

OTOH, while there are apparently fewer Unix admins in the wild, the average level is *much* higher (although I've seen my share of really bad ones too of course).

I used to understand Windows (well, what called that in windows 3.0 days) but nowadays, it seems to be volutarily obfuscated to me. If I had to learn how the system works I wouldn't even know where to start.