Slashdot Mirror
Linux beats Windows to Intel iMac
Ctrl+Alt+De1337 writes "The Mactel-Linux folks have now successfully booted Linux on a 17" Core Duo iMac. They used the elilo bootloader, a modified kernel, and a hacked vesafb to boot from a USB drive. No GUI pictures for now, just white text on a black background. The distro of choice was Gentoo, and instructions and patches are promised this weekend."
I'm not bricking my new Mac trying to run linux, I just have a horrible image of waiting on the phone with Apples tech support and them going 'no its not under waranty'.
I thought that distinction belongs to Sun SPARC boxes running Linux.
Then again, a little fancy ASM code in all of the C++ flying around really could speed things up
Great, a micro optimizer in training. Just what the industry needed.
Slashdot - where whining about luck is the new way to make the world you want.
It's sad that it's entirely possible that there's a Windows instance running on Intel Mac HW, somewhere behind closed Microsoft lab doors. OSS isn't just "open" when the source code is available for public download. The open project, the details of which are transparent and public, is another strong advantage. Particularly in the public relations arena, where the public claim is the prize, regardless of the real facts.
--
make install -not war
Of course, with Linux, comes Windows. In the form of emulating it using VMWare (which isn't supported on Mac OS X natively yet), and also with Wine (true, this isn't real Windows - but it satisfies people's needs to run some Windows programs).
Huh? The Mac's aren't early to using TPM. IBM laptops have had them for the last several years. In fact IBM said "Over 16 mllion IBM trusted clients have been shipped with Atmel TPM as of June 25, 2004"
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
I haven't seen much about the performance of the new Intel Macs, but I know the old G5s couldn't keep up with a comparably-priced PC. One advantage the PC has is that its competitive hardware market keeps prices lower.
What we need now is some solid Linux benchmarks on both systems. I'd wager that the PC would outperform the Mac on a price-for-performance scale. It would probably win overall, just because AMD has a better CPU on the market than Intel.
Of course, it all really depends on what you want to do with your system. Different architectures emphasize different things.
120 characters for a sig? That's bloody useless.
And certainly there's no need for hammers anymore now that we have advanced robots that can build cars.
A good tool is a good tool. I'd never use text-mode to browse the web (except for kicks). GUI is great for some things. CLI is great for other things. Sometimes a GUI is just too clunky, but sometimes a CLI is too confusing or difficult to get to work right. Use the right tool for the job. No need to be locked into one or the other.
I like to be able to dual-boot into linux for those Linux apps like Gnucash, which Intuit would like to charge me an arm and a leg for. I could use Gnucash in Mac, but the setup is overly hard (even with Fink and Fink commander) and then half the things don't work right, like printing without me spending half-a-day trying to figure it out. In ubuntu, I can just apt-get and forget it most of the time. I need to get work done, not configure my PC.
I don't need to run Windows, but I'd imagine some people are in a similiar situation with a must have program.
The nice thing with Macintel is that perhaps someone can get Windows/Linux may run on top MacOSX (like Inferno for various operating system), no rebooting or anything.
But 90% of the time, I work in OSX anyway.
MacOS X uses the Mach Kernel, so the initial booting environment is completely different from FreeBSD. You are getting way to hung up on the "MacOS X uses FreeBSD" thing.
Windows on a mac? That's just expensive hardware. +5 Insightful
Linux on a mac? That's just expensive hardware. -1 Troll
And likewise, Darwin may be based on BSD, but it sure as hell ain't BSD anymore.
Of course, my mom can't really go out and get a Dell, build a Linux kernel on it, and assemble an array of FOSS programs that work anything remotely like Mail/Safari/iLife. Heck, I couldn't be bothered to set that up, and I sure as heck wouldn't want to try maintaining/administrating it for my mom in another state.
In fact, iTunes is about the only thing you WOULDN'T have to live without, since there have been several stabs at getting the Windows version workable on Linux. Everything else in the Mac experience is missing.
To suggest Mac OS X is anything remotely similar to "BSD + some apps" is profoundly retarded and disingenious.
The value IBM was adding to Apple's Mac platform evaporated when Apple's PPC partners decided the desktop wasn't anything they cared about. That occured around 2000, when Microsoft completely abandoned NT's cross platform strategy plans. PPC has been on life support and in denial since PPC lost out on every desktop apart from Apple's. Since then, Apple has been leading Mac OS X development away from 68K/PPC dependance and toward a place where they could jump on the only viable platform for desktop PCs.
You can cry for PPC, but there isn't any way that Apple could continue to develop a processor platform entirely independant from the rest of the desktop PC world and remain competitive with the economies of scale enjoyed by Intel/AMD, particularly after its PPC partners gave up.
--
Linux is a very useful tool for many jobs, but its versatility is actually a major barrier for anyone trying to deploy it on the desktop. Everything is splintered to fit various different needs. Commonality and standarization is the value Apple adds with their products; the processor and underlying core OS are mere elements.
Apple can jump to Intel because they control the whole Mac world. Microsoft couldn't manage to keep Windows 2000 up on Alpha, PowerPC or MIPS because they shared control of the PC world with manufacturers.
Similarly, while Apple benefits from solid BSD foundation code, they could theoretically adapt Mac OS X Cocoa frameworks to live on top of Windows (as OSE was), Solaris or Linux (yes I realize that would not be very practical). But the point is, Apple's core competency was not PPC+BSD. It is the "Mac experience," which has little to do with individual components that might be in a Mac.
I've never noticed most of these (perhaps because I've been using Macs since, oh, '84), but these are interesting observations.
;) Command-arrows I've known forever.
* Inconsistant PageUp/PageDown use. Some programs move the cursor, some just move the screen. Very annoying when only the page moves. Now if all aps standardized even on the annoying behavior at least we'd be consistant.
How does the app know whether you just want to "look" a few pages up or down (and not lose the location where your cursor aka your current work position is), or actually "move" there? I personally hate when the cursor moves because there's no guarantee you know where it lands- and half the time I wanted to "remain where I was". But I'm a heavy mouser I guess.
* Home/End keys. If you understand the logic, it's not bad. Command-left_arrow and command-right_arrow do the trick. But if you go in and change your OS X keybindings to restore normal windows/linux home/end behavior, you only get very spotty coverage with some apps honoring the keybindings, some not.
Might be a difference between Cocoa and Carbon apps. This is just a legacy Mac thing. Since I'm a legacy Mac guy though, I've never gotten used to using the home/end keys to begin with though
* Click to focus a window absorbs that click. But not always. Depends on the app. Really slows you down if you use dual-monitors and have lots of windows spread between them.
My habit that I guess makes this not bother me is that whenever I want to bring a window to the front I click in a "non-busy" part of it. Then it doesn't matter whether the click is absorbed or not, but yes, you would still have to click where you actually want to "go". I didn't know one extra click actually bothered people. though.
* Scroll wheel can only affect a focused windows. This means you can't have your browser slightly underneath your program editor and scroll up and down through API docs without clicking away from the editor window. This one is pretty close to being a show-stopper for me. Combined with the previous problem with the focus these leads to some serious impedence of work. In essence the UI fails in this aspect because it doesn't get out of the way and let you work. Instead it is in your face.
I don't quite understand. If you arrange the windows in a non-overlapping way, it's an alt-tab to change the focus. Not very expensive to do alt-tab, roll wheel, alt-tab back. There is a UI convention that says that the frontmost window should receive all events.
You know, I wanted focus follows mouse for a long time, but then I realized that if you had focus follows mouse, you'd never be able to choose anything in the menus, unless you dragged the window to the top of the screen first to make sure it was the topmost window on your way to the menubar. So not only would you have to have focus follows mouse, but also menus tied to individual apps instead of globally. Forget about it.
For mac users the key buying points are:
No Viruses
No Noise
No Hassle
oh... and they're pretty.
They fit into the affordable luxary category, a lot like the iPod. If all you want is FIPS and MIPS, then you buy an AMD box, with water cooling and a heat sink as big as your car. Hell, why not go the whole hog and kit it out with LEDs to make it 'classy'.
Mac's are the Rolls Royce of computing, not the Ferrari. When it comes to the choice between comfort or performance, they choose comfort - but they still stick a big ol' engine in, because, let's face it, you paid for it. AMDs are the suped up Honda. Sure they get better 0-60, and are cheaper to 'upgrade', but you're still left driving a car that looks like a Honda.
If you're demanding performance specs, then either you are genuinly somebody who needs that performance (a dying breed) or your are a relic from the 1990's. Processor performance is no longer the most important factor in a desktop computer, we're still waiting for IO and memory to catch up.
Scared of flying, pointy things snce 1979!