At Linux trumps OS X , Jasjeet Sekhon presents benchmarks showing Mac OS X on Intel being much slower than Linux or Windows.
Hold on, says Ridiculous Fish, that's just a matter of your choice of buffer size.
I say "Baah" to all of it.
It's always possible to (deliberately or accidentally) show that X is faster than Y at Z. Operating systems approach tasks in different ways, so you can always find something that will show a specific OS in whatever light you want, good or bad.
Does it matter? Well, maybe so. If, for example, you are doing statistical work as Dr. Sekhon suggests, and your particular piece of software runs better on Linux than Mac, maybe you would want to use Linux. On the other hand, if your app is open source, and you have other reasons to want to use Mac OS X, maybe you instead want to rewrite your app to better match the OS characteristics as Ridiculous Fish suggests.
Or if speed were really important to you, maybe you'd be running something else entirely on much more expensive hardware.
Given the cost of hardware today, most of us can easily afford to run multiple machines, so if we really need Linux for a particular app to perform best, we can do that. And if we like Mac for our personal use, we can do that too.
It's the overall system that matters to most of us for our personal machine. I could use Linux or OS X for my daily work. I couldn't easily use Windows and wouldn't want to anyway, but really any Unixish OS meets my needs, and it is extremely unusual for me to be concerned about speed nowadays. Machines are so fast that it just doesn't matter: almost nothing I do takes more than a second or two anyway no matter what, and most tasks are effectively instantaneous.
What does matter to me is the desktop and graphical interface. I like the OS X interface more than I like Gnome or KDE. That is simply personal opinion, but in general Linux hasn't done well capturing desktops. Of course it is Windows, not Macs that are providing the competion there for most people.
I like my MacBook laptop. I'm sure I'd also like Linux on an Intel laptop (and in fact I'm probably going to buy one for that this year). But benchmarks are almost completely irrelevant to me.
More Articles by Anthony Lawrence - Find me on Google+
Have you tried Searching this site?
Unix/Linux/Mac OS X support by phone, email or on-site: Support Rates
This is a Unix/Linux resource website. It contains technical articles about Unix, Linux and general computing related subjects, opinion, news, help files, how-to's, tutorials and more. We appreciate comments and article submissions.
Many of the products and books I review are things I purchased for my own use. Some were given to me specifically for the purpose of reviewing them. I resell or can earn commissions from the sale of some of these items. Links within these pages may be affiliate links that pay me for referring you to them. That's mostly insignificant amounts of money; whenever it is not I have made my relationship plain. I also may own stock in companies mentioned here. If you have any question, please do feel free to contact me.
Specific links that take you to pages that allow you to purchase the item I reviewed are very likely to pay me a commission. Many of the books I review were given to me by the publishers specifically for the purpose of writing a review. These gifts and referral fees do not affect my opinions; I often give bad reviews anyway.
We use Google third-party advertising companies to serve ads when you visit our website. These companies may use information (not including your name, address, email address, or telephone number) about your visits to this and other websites in order to provide advertisements about goods and services of interest to you. If you would like more information about this practice and to know your choices about not having this information used by these companies, click here.
Click here to add your comments
Sat May 27 16:53:43 2006: BigDumbDinosaur
Machines are so fast that it just doesn't matter: almost nothing I do takes more than a second or two anyway no matter what, and most tasks are effectively instantaneous.
Tony's observation is absolutely correct. Most of what I do on my system happens so fast (at least in UNIX -- Windows is another matter) I really don't have any perception of processing time. Human beings might be able to perceive a lag of 80-100 milliseconds, but that's about it. Even the relatively slow 3600 RPM ST506 hard drives of the mid-1980's had faster seek times than that (Seagate's ST251, a 40 MB unit released in late 1987, had a seek time of 28 milliseconds -- it seemed unbelievably fast back then).
The reality of today's computer hardware is that the hard disk and external communications paths continue to be the principal performance bottlenecks. A 100Base-T network might manage a real throughput of 7-8 MB/second, and the fastest hard drives (e.g., the 15,000 RPM Seagate Cheetah U320 SCSI units) might be able to deliver 32-35 MB/second of data off the platters. On the other hand, something like a AMD Opteron 200 system can produce an effective throughput of multiple hundreds of MB/second through the front side bus subsystem. All of this is blindly fast compared to what we had even just 10 years ago when CPU clock speeds were in the mid-100 MHz range and a typically memory cycle time was 70 nanoseconds, producing an effective memory bus throughput of maybe 12-15 MB/second under ideal conditions. And *that* was ridiculously fast when compared to the systems of the late 1970's and early 1980's, a time when memory was actually faster than some processors.
One of the older books I have here on the shelf is UNIX Internals (Shaw & Shaw), which was published in the months immediately preceding the release of the first Intel 80386 processor. At that time, "UNIX" on a PC was Xenix running on 80286 hardware (which actually wasn't particularly well suited to UNIX's virtual memory model), and some of the discussion in this publication focused on that combination. Fairly early in the narrative, the authors made some interesting observations about UNIX's performance as compared to similar operating systems of the day. A particularly telling comment, which still has some validity to this day, was as follows: By "fair timesharing" they meant that all processes would be given eglitarian treatment for scheduling purposes. The reality of the situation then, as now, is that I/O bound processes are given substantial precedence over compute-bound processes. This aspect of UNIX is based on the theory that compared to machine features such as RAM and processor capabilities, I/O resources are extremely limited (they sure were 20 years ago!) and thus any process engaged in I/O should be allowed the maximum possible run time during the I/O transaction.
The point to all this is that, as Tony said, it is possible to show that any operating system is "fast" or "slow" simply by carefully choosing the right test. If I want to "prove" that Linux is faster than OS-X, I'd test the former using a benchmark that concentrates on compute-bound processing, and test the latter in a way that focuses on I/O-bound processing. Or to explain it another way, anything can be statistically proven to be true or false merely by choosing the right subset of statistics.
Thu Jun 1 20:34:41 2006: bruceg2004
"I like my MacBook laptop. I'm sure I'd also like Linux on an Intel laptop (and in fact I'm probably going to buy one for that this year)"
Wait a sec; your MacBook Pro *is* an Intel laptop :-) Why not dual boot into Linux with that?
- Bruce
Thu Jun 1 20:40:02 2006: TonyLawrence
Cuz' dual boot would leave me Mac-less.
I can and do run Linux in Parallels, but there are things I can't do there - like run Xen.
Also, it would be nice to have a spare laptop anyway..
Don't miss responses! Subscribe to Comments by RSS or by Email
Click here to add your comments
If you want a picture to show with your comment, go get a Gravatar