Slashdot Mirror
Slashback: OS Xi, Sarge, Statistics
It still feels like a strange dream that they're really switching. An anonymous reader writes "With our latest Unix (MacOS-X) vendor's switch to x86, I figured now would be a fine time to revisit an old MIT Graduate Student Beer announcement from 2001."
Also, samchung writes "CoolTechZone has its latest article up that discusses the possibilities of Apple's protection on x86 hardware to prevent users from running the Mac OS X on non-proprietary hardware."More fuel: Reality Master 101 writes "Michael Robertson, CEO of Linspire posted an editorial talking about his disappointment that Apple wasn't embracing generic hardware. But the really interesting part was that he states, "My sources say that Jobs is going to use Intel's cryptographic technology called LaGrande to make sure OS X will only boot on Apple-branded hardware. This is a similar technique to the one that Microsoft used to make sure Linux could not be loaded on Xbox..." I'm still not sure how they'll do this with an open source Kernel." They're clearly part of the Linspire marketing effort, but Robertson's messages, including this one, are usually pithy and worth reading.
Hey, you could always wait for a service pack. An anonymous reader submits "Because of an error in a configuration file, Debian Sarge, released June 6th, does not have security updating enabled by default. ZDNet Australia reports that after several years of testing, the release team's error caused a significant delay in deployment. Steve Langasek, of the release team, says, 'Whoops, don't go pressing those 10,000 copies of [3.1] just yet.' Fortunately, the error may be fixed quite easily, and an update is expected within several days. OSNews also covers the story.
Sticker shock alone could defeat the other drivers. josemunizn writes "Remember the Honda FCX, from a Slashdot article in '03? Well the New York Times has an automotive review of a week-long, unsupervised test drive of the Honda. Choice quote: 'In most important ways, the FCX feels ready for prime-time combat on the world's roads.'"
Carry the one, subtract 5, voila! An anonymous reader writes "WinMX and Limewire are the most popular P2P apps? That's what NPD group claims in its research on iTunes covered on Slashdot yesterday. But as Jon Newton points out on P2Pnet and MP3 Newswire, the entire premise that more people use iTunes over the file sharing networks is 'nonsense.' With sites like Slyck.com reporting eDonkey alone has over 4.5 million concurrent users and P2P research firm BigChampagne saying in the U.S. in May an average of 6,290,327 people were logged onto the p2p networks at any given moment, how can iTunes' 1.7 million downloads over an entire month put them anywhere near the top? Zeropaid has also chimed in on these claims and even CNET is now questioning the results it reported in its original article on the NPD research."
Catching up to the 3rd parties who have caught up with the competition. An anonymous reader writes "For the impatient or those few not ready to adopt Firefox, there is now another option to get tabs. BetaNews reports, 'Users of Microsoft's Internet Explorer Web browser will not have to wait until IE7 to experience tabbed browsing. MSN has shipped a new build of its MSN Search Toolbar that adds basic tabbed browsing support to IE6. But the feature is not fully integrated into the browser, instead relying on the toolbar to create tabs.' Here's an article including a screenshot.
The most likely reason is the laptop market. AMD does not have a credible mobile chip while Intel has Pentium M which is most likely the best mobile chip out there.
Since laptops outsell desktop (I think at least the laptop macs), it would be commerical suicide not to be a player in the mobile market. Besides, the mobile market is projected to increase significantly.
" I'm going to keep this brief, so please write me with the questions you have and any tests you want run on one of the dev kits. I will have one of my own next week as well.
First, the thing is fast. Native apps readily beat a single 2.7 G5, and sometimes beat duals. Really. All the iLife apps other than iTunes, plus all the other apps that come with the OS are already universal binaries....
They are using a Pentium 4 660. This is a 3.6 GHz chip. It supports 64 bit extensions, but Apple does not support that *yet*. The 660 is a single core processor. However, the engineers said that this chip would not be used in a shipping product and that we need to look at Intel's roadmap for that time to see what Apple will ship.
It uses DDR-2 RAM at 533 MHz. SATA-2. It is using Intel GMA 900 integrated graphics and it supports Quartz Extreme. The Intel 900 doesn't compare favorably to any shipping card from ATi or nVidia. The Apple engineers says the dev kit will work with regular PC graphics cards, but that you need a driver. Apple does not write ANY graphics drivers. They just submit bug reports to ATi/nVidia. So, when we asked where to get drivers for better cards the engineers said "The ATI guys are here." He's right, they've been in the compatibility lab several times.
It has FireWire 400, but not 800. USB 2 as well. USB 2 booting is supported, FireWire booting is not. NetBoot works.
The machines do not have Open Firmware. They use a Phoenix BIOS. That's right, a Mac with a BIOS.
(I asked if the Bios had any tweaks like Memory Timing which is common for many PC motherboards, although Intel OEM motherboards don't usually have any end user tweaks like that.-Mike)
They won't tell us how to get in the BIOS. I'm sure we can figure it out when out dev kits arrive.
They run Windows fine. All the chipset is standard Intel stuff, so you can download drivers and run XP on the box.
Rosetta is amazing. (see earlier post on limitations of the Rosetta emulator - it's a G3 emulator basically - will not run Altivec code, etc. and performance isn't going to be as good as native code, but most Mac apps will run on a G3.-Mike) The tests I've run, both app tests and benchmarks, peg it at between a dual 800 MHz G4 and and a dual 2 G5 depending on what you are doing.
(I mentioned to him the limitations of Rosetta (posted below)-Mike)
It's true Rosetta does not support Altivec, but most apps run on a G3, right? Rosetta tells PPC apps that it is a G3. Apps should fall back to their G3 code tree. Everyone I tested did.
The UI tests in Xbench exceed a dual 2.7 by a large margin. (other specific tests are much lower than a G5 per Xbench site results.-Mike)
I've been talking to and watching a lot of devs. There are a lot of apps from big names running in the Compatibility lab already. Some people face more pain, sure, but Jobs wasn't kidding when he said that this transition would be less painful than OS 9 to OS X or 68K to PPC.
Game devs seem optimistic. They see porting Windows/x86 to Mac/x86 as much easier. They look forward to the day they don't have to support PPC.
I was talking to a (game Developer) that said about 1/3 of the process is handling endian issues, the rest is Win32/DirectX. For the next 3-5 years, their job will be harder since they have to port to two processor architectures and most bugs *are* endian related and that they will have a hard time making the PPC versions run as well as the x86 versions.
This transition is not about current P4 vs G5. It is about the future directions of the processor families. Intel is committed to desktop/notebook and server in a big way. Freescale/IBM are chasing the embedded market and console market. Apple would have been in a lurch in 2 years.
Also, all the cell people and the AMD people need to be quiet. Apple evaluated both. AMD has the same, if not worse, supply problems as IBM. Their roadmap is fine, but the production capacity is not.
The tested Cell as well. That processor is NOT in
The note at the top of every 3.1 download page:
/etc/apt/sources.list for "http://security.debian.org/ testing/updates" rather than an active entry for "http://security.debian.org/ stable/updates", and thus will not get security updates by default. This was due to incorrect Release files on the images.
/etc/apt/sources.list, look for any lines mentioning security.debian.org, change "testing" to "stable", and remove "# " from the start of the line.
Note: 3.1_r0 CD image problem
A bug has been discovered in the 3.1_r0 CD/DVD images: new installs from these images will have a commented-out entry in
If you have already installed a system using a 3.1r0 CD/DVD image, you do not need to reinstall. Instead, simply edit
If you installed other than from a CD or DVD (for example, netboot, or booting from floppy and installing the base system from the network), you are not affected by this bug.
These new 3.1_r0a images correct this flaw. We apologise for the inconvenience.
On another note, I wanted to start downloading the 3.1 ISO set for Sparc, but none of the US mirrors have 3.1 ISO sets, and the root server is giving out 404's. Perhaps they're all still busy updating? At this point, I don't think bit-torrent is propagated well enough to be faster than HTTP/FTP, and jigdo only puts the load on your workstation by opening 9,000 connections on your box to go download little bits of Debian.
"Ancillary does not mean you get to rule the world." --U.S. Circuit Judge Harry Edwards, speaking to the FCC's lawyer
Apple never, ever wants to have a situation where their ability to sell Macs is limited by a supply of CPU chips. That happened to them in the past and they never want it to happen again.
Intel has huge, huge production capacity, so by going Intel they can be confident that they will always have enough chips. It's that simple.
There is no reason why they couldn't use an AMD chip in a special edition Mac, other than it might piss off Intel.
If Apple has any brains at all, they will avoid the Pentium 4 family for production computers, and go with chips based on the Pentium M core. Wait, that's redundant, because Apple won't ship the new Macs until two years from now, and Intel won't be selling any more Pentium 4 cores by then.
If Apple has any brains at all, they will ship only x86_64 (or AMD64 as you might call it). Absolutely no 32-bit CPUs in Macs, so all the binaries only need to be compiled one way, and they get the extra registers and the no-execute bit for stack protection.
Intel has more to offer in terms of lower power chips
I presume you mean the Pentium M? For desktop chips, AMD kicks the ass of Intel, getting more work done per clock, and with less heat dissipation. I think the same is true of the notebook chips, but it's possible I'm wrong about that.
Intel's new dual-core chips are more or less based on the Pentium M, and thus should offer very good MIPS-per-Watt, but at this point Intel is definitely playing catchup to AMD on efficiency of chips.
There's no Open Firmware on the new machines. The developer docs say that apps requiring it won't be supported, and the developer systems from Apple just have a Phoenix BIOS on board. See http://www.xlr8yourmac.com/ for a breakdown.
Apparently, the machines boot Windows just fine. No hacking required to install it at all, it seems.
Wasting your time since 1997.
Please tell me if there are any other problems, this was the first time I heard about these 404s. Btw, saying which links will help even more, in this case I'm guessing at the powerpc isos?
Why doesn't Apple use an OpenFirmware emulation layer if the PC doesn't support it directly?
This is what Sun does on Solaris x86-- before the OS starts, the boot s/w generates the OF device tree and takes care of simulating NVRAM parameter storage and other OF goodies.
This seems more software friendly than breaking apps which expect OpenFirmware to be there and having to revamp your entire boot process.
As a total Apple n00b, I haven't been able to figure out how to turn on sub-pixel font rendering for my LCD monitor. I get the impression that the OS is supposed to be smart enough to turn it on by itself, but nothing happens. The totally stripped-down control panel dialogs in OSX don't give me anything to work with to try to fix it.
Bottom line: for me, the Linux fonts look beautiful, the Apple fonts look like ass. Oh well.
Otherwise, the "patch" would be to manually add the security.debian.org line in sources.list after installation. Just like it says in the errata in the grandparent to this comment.
I know, you probably were looking for a setting in /etc/XF86Config to change, but on the Mac you generally use the GUI to change things. For what it's worth, there's probably a command line equivalent (some .plist file somewhere has this setting)
Appearance tab in System Preferences:
Font smoothing style:
Automatic
Light
Medium (Best for flat panel)
Strong
System Preferences => Appearance => (at the bottom) Font Smoothing Style: Medium
Turn off text smoothing for font sizes X(your choice) and smaller.
I've tried all the settings: Standard, light, medium, strong. All of them == the exact same fuzzy grayscale-only antialiasing. Maybe the settings on this machine somehow got hosed before I got it, but that's not supposed to happen on a Mac, right?
The more I think about it, the more I believe the Pentium M is the single most important chip technology released by any chip maker in the past decade. More important than the P4, more import than the AMD K7 or K8 (!), more important than the G3, G4, or G5. (I'll ignore the embedded market).
:)
I'd sworn off Intel chips, especially w/ the P4 and Intel's other blundering and bullying. But you gotta have respect for the Pentium M, to so successfully fill an important market niche (lower power/lower heat) where Intel had essentially NO viable contender before. Think about it; Transmeta, Cyrix, VIA, AMD, even the various PowerPC makers all used to laugh at Intel's offering, and then the Pentium M came along. And now, Intel will probably own the laptop market for quite some time to come. Well, okay, they already did, but until the Pentium M, they never deserved to.
I'm still amazed at how versatile the descendants of the Pentium Pro have turned out to be, and as a Mac user, I'll probably end up with a Pentium M based Mac laptop in the future, and not have to cringe at the prospect.
That said, I wish AMD the best of luck, because I truly think they helped Intel to reform its ways a little (my last Intel chip was the 300Mhz Celeron w/ onboard cache; nearly everything since, until the Pentium M, has been embarrasing, IMO) And hopefully, there will also be many an application for the IBM Power descendants as the years go by.
"It's overkill, of course. But you can never have too much overkill." - Anonymous Slashdot Coward
When I first saw that I thought that maybe they did it this way so the development machines can be ready sooner and the x86 Macs that are sold to consumers will probably have Open Firmware. However, later on in that page there's a link to an Apple page that says "Macintosh computers using Intel microprocessors do not use Open Firmware." Apple also says that disk partitioning will be different. I guess the "Mac" will have a PC BIOS and the standard PC partitioning scheme.
CPUs advanced.
At the time the Pentium was a snail.
Just like AMD can claim the P4 is a snail.
But come this time next year, perhaps Intel can claim Athlons are 'snail like'.
Now they don't need to claim the P4 is a snail because they'll be using Intel's latest and greatest. And if AMD is better, well, they always have the option of selling those too.
And... where do you get that Apple claims that OS X never crashes? Can you link? Because I can't find it.
GPL Deconstructed
By the way, ever since the late 90s when Macs started using IDE/ATA, PC DIMM memory, USB instead of ADB, ATI grahpics cards, etc., etc., Mac hardware has been not all that different from PCs in the first place. Aside from the obvious CPU/motherboard differences, the other components are still pretty much commidity PC parts anyway.
IIRC virus is a 4th declension noun, so the nominative plural is "virus" also.
Even if it were second declension (most common of words ending in -us), the plural would be "viri", prononced "veery", not "vie-rie" or "vie-ry."* But "viri" already means "men" in Latin.
If it were neuter it would be "vira."
Note that none of these endings resemble "ii."
(* If you want to get technical the V isn't even supposed to be pronounced as a V, but as a U. So "virus" is pronounced "ueeroos," not "vie-russ" as an English speaker would.)
To them Intel or IBM don't make any difference. In fact, I'd bet good money that at least 80% of their customers (current and prospective) don't know about the switch and what it means.
I predict the following:
1. As sales slow down (and they WILL slow down to zero over the next year or so), Apple will heavily drop the price on G5 and G4 based models, including laptops. This won't help the sales much, but people wanting to sell their Macs will only be able to sell them for a lot less.
2. This is not the last "switch" Apple has in the pipeline. The next one will be the switch from slow and brain damaged Objective-C as the language of choice for Mac OS X programming. This one will be painful to everyone but people who kept using Carbon (that's actually a lot of people).
3. Once Intel platforms start hitting the market Apple users will be shocked by just how much _slower_ Mac OS X is on the same processor compared to Windows (or Linux if it makes inroads). That's microkernel and IPC, there's no way around this cost short of throwing microkernel architecture out of the window. Another switch?
4. Yet another switch is coming. Apple UI is currently 32 bit, so you can only run your console apps in 64 bit mode. 64 bit editions of Win XP are not castrated in this regard, and I have no reason to think that Longhorn will be. So Apple will have to convert Cocoa to 64 bit.
Each of the last three "switches" is a cost to the developer. At some point developers will just say "fuck it" and go develop software for Longhorn.
One could speculate on why they didn't, but they didn't, as the Universal Binary Programming Guidelines document (which anybody who wants to speculate on whether Apple's switching to x86 or, to use a favorite wrong guess of many folk, licensing Intel to make PowerPC chips, or on whether they're using OpenFirmware in the x86 machines, or..., should read before they speculate in public) says.
Apple has not said they will never license OS X to other PC manufacturers.
On Monday, answering questions about x86 Macs, Apple senior vice president Phil Schiller said, "We will not allow running Mac OS X on anything other than an Apple Mac."
So, while there may be some question as to whether that means "We will take technical measures to prevent it from running on non-Apple computers", or just "our license agreements will prohibit it from running on non-Apple computers", it's pretty clear that Apple won't be licensing OS X to other PC manufacturers.
Correct. PowerPC was designed as a 64-bit ISA from the start. The only difference between 64-bit and 32-bit PowerPC chips is the size of the registers, the MMU and some extra instructions for manipulating 64-bit integers natively.
x86-64, in contrast is a 64-bit hack built on top of a 32-bit cludge on top of a 16-bit ISA. As well as being 64-bit, AMD64 adds some extra registers (almost half as many as PowerPC, woohoo), which makes code faster, in spite of the 64-bit penalty (it takes longer to load a 64-bit value than a 32-bit one, and code with 64-bit pointers takes up more cache space).
OSX isn't a 64 bit operating system
OS X Tiger is a 64-bit OS. Because PowerPC64 was designed to be compatible with PowerPC32, it is possible to run 32-bit code on it. One of the most commonly used pieces of 32-bit code is the windowing system. This is 32-bit because graphical applications rarely need more than 4GB of address space[1], and so it makes no sense to slow all of them down for the few that do.
[1] They might, however, need to spawn compute processes which handle more than 4GB of data.
I am TheRaven on Soylent News
"If your Mac Tiger app is 64 bits, you're screwed. Won't even run in the emulator."
Two points.
First, there aren't any Mac app's that I know of that _require_ 64-bit CPU's, because they won't run on G3's and G4's, which means most Mac's, all laptops, etc. So app's that take advantage of 64-bit instructions also have a 32-bit version of the code.
Second, while the Universal Binary Programming Guidelines do only talk about the IA-32 instruction set, but it clearly supports 64-bit data types, and MMX/SSE/SSE2/SSE3, and I'd be stunned if it weren't possible to run 64-bit code on 64-bit x86's. Admittedly the 64-bit picture on Intel is a bit more complicated than on PPC (since the various x86 chip companies had different 64-bit stragies), but Apple's got a year to work it out. And, for what it's worth, rumor has it that Apple got MacOS X to compile on the Alpha at one point, which should have cleared up the dependencies on 32-bit code.
Enable 3D printed prosthetics!
The problem is you then have OpenFirmware emulation baked into system years after Mac users stopped caring about OF. And by 2006, EFI will be mainstream, so other than these developer boxes, BIOS is a non-issue for Mac users.
I actually expect driver code to change quite a bit, if only because vendors will want to keep their Windows and Mac drivers closer together.
Whenever I hear the word 'Innovation', I reach for my pistol.