Slashdot Mirror

For about a year, I've been patching my Intel Compiler compiled code because of this issue. I have to give credit to a poster on the comp.arch newsgroup for an explaination of ONE of the issues, and a workaround.
This is not the only anti-Athlon trick in the compiler, but it's an easy one to verify and understand.

From: iccOut (iccout2004@yahoo.com)
Subject: sleazy intel compiler trick (SOURCE ATTACHED)
View: Complete Thread (4 articles)
Original Format
Newsgroups: comp.arch
Date: 2004-02-09 14:38:40 PST

As part of my study of Operating Systems and embedded systems, one of
the things I've been looking at is compilers. I'm interested in
analyzing how different compilers optimize code for different
platforms.As part of this comparison, I was looking at the Intel
Compiler and how itoptimizes code.The Intel Compilers have a free
evaluation download from here:
http://www.intel.com/products/software/index.htm?i id=Corporate+Header_prod_softwr&#compilers

One of the things that the version 8.0 of the Intel compilerincluded
was an "Intel-specific" flag.According to the documentation,binaries
compiled with this flag would only run on Intel processors andwould
include Intel-specific optimizations to make them run faster. The
documentation was unfortunatelylacking in explaining what these
optimizations were, so I decided to do some investigating.

First I wanted to pick a primarily CPU-bound test to run, so I chose
SPEC CPU2000.The test system was a P4 3.2G Extreme Edition with1 gig
of ram running WIndows XP Pro. First I compiled and ran spec with the
"generic x86 flag" (-QxW),which compiles code to run on any x86
processor.After running the generic version, I recompiled and ran
spec with the "Intel-specific flag" (-QxN) to see what kind of
difference that would make.For most benchmarks, there was not very
much change, but for 181.mcf, there was a win of almost 22% !

Curious as to what sort of optimizations the compiler was doing to
allow the Intel-specific version to run 22% faster,I tried running
the same binary on my friend's computer.His computer, the second test
machine, was an AMD FX51, also with 1 gig of ram, running Windows XP
Pro. First I ran the "generic x86" binaries on theFX51, and then
tried to run the "Intel-only" binaries. The Intel-specific ones
printed out an error message saying that the processor was not
supported and exited.This wasn't very helpful, was it true that only
Intel processors could take advantage of this performance boost?

I started mucking around with a dissassembly of the Intel-specific
binary and found one particular call (proc_init_N) that appeared to be
performing this check. As far as I can tell, this call is supposed to
verify that the CPU supports SSE and SSE2 and it checks the CPUID to
ensure that its an Intel processor. I wrote a quick utility which I
call iccOut, to go through a binary that has been compiled with this
Intel-only flag and remove that check.

Once I ran the binary that was compiled with the Intel-specific flag
(-QxN) through iccOut, it was able to run on the FX51. Much to my
surprise, it ran fine and did not miscompare. On top of that, it got
the same 22% performance boost that I saw on the Pentium4 with an
actual Intel processor. This is very interesting to me, since it
appears that in fact no Intel-specific optimization has been done if
the AMD processor is also capable to taking advantage of these same
optimizations. If I'm missing something, I'd love for someone to point
it out for me. From the way it looks right now, it appears that Intel
is simply "cheating" to make their processors look better against
competitor's processors.

Links:
Intel Compiler:http://www.intel.com/products/software/in dex.htm?iid=Corporate+H

Wow... that's a great example, and you should gather as much evidence of it as you can, especially Intel's responses, and send it to AMD's legal team.

If I were Intel, I would respond with the product description of the Intel compiler, as found here http://www.intel.com/cd/software/products/asmo-na/ eng/compilers/clin/index.htm or here http://www.intelcompiler.com/.

The product is clearly labeled as a high performance compiler for Intel CPUs. The grandparent used the wrong tool for the job which required a generic compiler.

1) Intel's compiler does NOT support AMD's CPUs
http://www.intel.com/cd/software/products/asmo-na/ eng/vtune/220001.htm

2) Only a moron would buy Intel's compiler to develop for AMD processors (even if they didn't know about 1))

3) From TFPDF: "ISVs are forced to choose between Intel's compilers, which degrade the
performance of their software when operated with AMD microprocessors,"
How exactly are ISVs forced to choose "between" Intel's compilers? Those developing on AMD should NOT use Intel's compilers in the first place since Intel does not support that CPU.
(BTW, ISVs are not forced - they are enticed - to choose Intel's compilers. Can they prove Intel forces ISVs to buy their compilers?)

5) From TFPDF: "Unbeknownst to them, performance of their programs is degraded when run on
an AMD microprocessor not because of design deficiencies on the part of AMD, but
deviousness on the part of Intel."
Unbeknownst to them, the fucking product page does not even list AMD processors as supported. What do they expect? "Blistering" performance?

The _first_ sentence on the download page for AMD's CodeAnalyst: "The AMD CodeAnalyst(TM) Performance Analyzer is a suite of powerful tools that analyzes software performance on AMD microprocessors."

Also, note that no AMD processors are present on the list of supported processors on the Sys Reqs page for Intel's VTune.

Personally, I have no idea why on Earth you'd be downloading a trial for VTune, while using an AMD proc (as far as I can tell, CodeAnalyst, by comparison,is free). I do agree that they should have popped up a dialog saying "Hey, jerk! What do you think you're trying to pull?" or something along those lines, but in this case you blatantly disregarded: 1.) the clearly stated system requirements, and 2.) common sense.

The last time I looked, Itanium was more efficient than *PENTIUM M*

Ahahahahahaa... Good one. You guys crack me up, sometimes. Intel's own numbers shows the low voltage (aka "piss-poor performance") version as still drawing 62W. The last "real" version, the Madison core, draws an impressive 130W. Ah, but you meant to refer to the dual core one, correct? Well now, that changes everything! To realize the draw you gave (not quite half of 99W, but I'll let that slide for now), should I cut one in half with a hacksaw, or just smash a bit of it up with a really small hammer?



more efficient than *PENTIUM M*, let alone mobile athlon(64)s.

The 1MB cache LV mobile Athlon 64 has a TDP of 19W.
The ULV Pentium M-753 (Dothan core) draws... 5W. Five. Stomp stomp stomp... stomp... ... stomp.

5W<42.5W. 19W<42.5W. 42.5W<(99/2)W. 99W<130W. Please play again.



You can't fucking compare the energy efficiency of chips made with different size transistors, fuckwit.

Yes, actually, you can. I chose to measure it in terms of "Slashdot trolls I have to put up with per watt".

You could also use a more meaningful number, such as megaflops per watt, or even FPS in Doom 3 per watt. But only considering the per-gate waste doesn't tell you anything particularly meaningful - Doing so strikes me as similar to trying to compare the brightness of a 100W incandescent light-bulb to an LCD monitor based on the worst-case draw of a single pixel.

How many gates does it have total, how well does it make use of them, and does it keep idle sections of the chip active also matters rather a lot. Using half as much power per state change doesn't matter if it takes 10x as many transistors to get the same job done.

Intel's still making mother boards. Their EPSD division is still a big player in the x86 server market.

http://intel.com/support/motherboards/server/

Intel have been working on something big. It was previously rumored that this something was the Pentium V and that Microsoft would be releasing a special version of Windows specifically for the processor.

"Windows Elements"?

What the hell is that? I'm thinking that the Pentium V has something so revolutionary that it prompted:

1) Microsoft to release a special version of Windows, specifically for the processor and,
2) Apple to change sides.

I also think that Intel expected to be much further along on the Pentium V at this point. It seemed like they were expecting to use it in order to quench AMD's 64-bit lead and, when the design was set back, they scrambled to come up with EMT64 as a stop gap solution.

So just what is this Pentium V and the "stackable" design, anyway? IMHO, it will be unified processor and NVRAM (not flash, something new). There will probably be at least a few gigs of a very fast NVRAM right on the processor. This NVRAM will be as fast or faster than SRAM so there will be no need for a cache or external system memory - the operating system will be installed right in the processor. The stackable design is for expansion.

Intel's NVRAM page. Nothing to indicate that any of this is true but some interesting reading, nonetheless. This could also explain MontaVista's PRAMFS.

If the backing-store RAM is comparable in access speed to system memory, there's really no point in caching the file I/O data in the page cache. Better to move file data directly between the user buffers and the backing store RAM, i.e. use direct I/O.

I remember earrings made from the cores of Pentium chips with the FDIV bug - the ones Intel recalled. I suppose Intel had to do something with all the recalled CPUs.

"It appears Intel plans on dropping the P4 line and going to enhancing the Pentium M edition. It is expected that Apple will be going with the Pentium Ms (which apparently have dual core slated in their lineup) instead of with the Pentium EE."

I think you meant to say the Pentium D + LaGrande (DRM in silicon), not Pentium M. The Pentium D (with not-yet-released updates and fixes), does exactly what Apple is after - controlled access to media with an architecture that provides lower-power (iPod-like devices and battery-powered Powerbooks).

"It appears Intel plans on dropping the P4 line and going to enhancing the Pentium M edition. It is expected that Apple will be going with the Pentium Ms (which apparently have dual core slated in their lineup) instead of with the Pentium EE."

I think you meant to say the Pentium D + LaGrande (DRM in silicon), not Pentium M. The Pentium D (with not-yet-released updates and fixes), does exactly what Apple is after - controlled access to media with an architecture that provides lower-power (iPod-like devices and battery-powered Powerbooks).

I got sue by Intel for using in.tel .

Measuring wall power does not tell you much about the CPU's power. The only case where a multimeter would be useful for measuring actual CPU power is on motherboards where all CPU power is exclusively taken off the ATX12V connector. This figure would include VRM losses but at least excludes other 12V loads such as AGP/PCIE graphics.

Most of the numbers I quoted were from memory of articles and the actual spec/data-sheets a long time ago... have a look for yourself: http://www.intel.com/products/a_z/p_q.htm

Look at static current (no clock) figures, current chips have higher no-clock power than older chips under full load.

My P233MMX runs fine at full-load without CPU fan, my P3-1G with its larger heatsink eventually crashes at idle. The P54C-233 has a max TDP of 17W (http://www.intel.com/design/intarch/mmx/mmx.htm) while the P3's SLEEP/HALT power is 10-12W. (Islp = 6.9A / Idslp = 6.6A @ 1.65-1.75V - ftp://download.intel.com/design/PentiumIII/datasht s/24526408.pdf p.32)

So, any P3 uses almost as much power while _(deep)sleeping_ as the hottest P54C does under full-load.

Look at static current (no clock) figures, current chips have higher no-clock power than older chips under full load.

My P233MMX runs fine at full-load without CPU fan, my P3-1G with its larger heatsink eventually crashes at idle. The P54C-233 has a max TDP of 17W (http://www.intel.com/design/intarch/mmx/mmx.htm) while the P3's SLEEP/HALT power is 10-12W. (Islp = 6.9A / Idslp = 6.6A @ 1.65-1.75V - ftp://download.intel.com/design/PentiumIII/datasht s/24526408.pdf p.32)

So, any P3 uses almost as much power while _(deep)sleeping_ as the hottest P54C does under full-load.

MOD PARENT UP. Excellent comment.

Intel began self-destructing several years ago. For example, in my opinion, the Intel consumer products division released lame, unfinished products. Eventually Intel reacted to the poor sales by closing the division.

Like really, really scary horror stories? Here's one more scary than you've ever seen in film. Intel marketing has become detached from reality. Intel marketing people go to work every day, but they just pretend to have meaningful jobs and pretend to be doing something positive for Intel. They are zombies, and most of them don't appear intelligent enough to know that they are zombies. If you think this is an exaggeration, read this sentence from a recent email message from Intel Marketing (I'm talking here about Intel marketing, not Intel's advertising agency.):

"Pass any three of the four tests before July 26, 2005 and your company will get a certificate of completion - plus you'll receive an Intel BunnyPeople Character." Here's an explanation with photo: Intel Bunny People.

Intel has been giving those dolls away for 7 1/2 years. Maybe someone bought a huge number of them?

How many technically-oriented people are motivated by the idea of receiving a doll? It goes like this: 1) Give Intel marketing your company's address and phone number and email address, so that they can spam you in the future. 2) Sit through boring marketing-speak, written by people who don't know or care about Intel products, or any technical product. 3) Take a test. 4) Get a doll?

Intel management appears to have spun out of control. Apparently it is now all stock options and company politics, and nothing about actually doing well. The people in charge don't actually know what they are doing, and apparently care more about having their executive positions than making good products.

Intel is known in Portland, Oregon, where it is based, for being abusive toward its employees. I'm guessing that the present problems really began about 12 or 15 years ago, when the Intel management, just before an enormous increase in profits, pleaded broke and reduced the pay of employees by 10%. Intel is known for over-working its employees, and pressing them to work very long hours.

Once about 2 years ago, I decided to ask Intel marketing people to fix a problem with the motherboard web site. Intel's ordering model number, used when you place an order with a distributor, was nowhere connected on the web site with the marketing model number, like 845BGL. I asked them to fix that. I talked to several marketing employees, all of whom clearly did not intend to do any real work.

I could tell many, many stories about Intel's sink into depression, but that's enough for now. I will have to say, however, that Microsoft's marketing people are worse.

Notice this: It's the PX code that should be dispatched on all non-Intel processor-based systems in the current IPP 4.* versions.

In other words, the PX is the non-optimized code that the dispatcher executes on non-Intel (AMD) processors.

Weird, isn't the Xenon an intel CPU?

You also forgot one thing... The Xbox 360 CPU will have three VMX units.

But in any cases, Apple was the one that used the Gx monicker, Moto and IBM had other names for their chips (though the G3 was really code-named G3).

Microsoft just took the "G5" name prestige and ran with it.

Kinda like they used the IBM name in the 80's to sell licenses to make "IBM compatible" machines.

I guess they are also trying this with the "Xenon" name?

Let's see... the machine after the Xbox 360 will feature a processor called the G6?

For the third time, I'm still waiting for evidence of "SPEC is subject to all kinds of problems.", why Apple doesn't use your "benchmark" to advertise their products, these supposed Intel optimizations, evidence of why spec isn't accountable, and why I am "blinded by marketing".

Uh, how exactly do you get "GCC, is said to generate code that less well optimised than Intel's" from "GCC, is said to generate code that less well optimised for x86"?

I repeat, since you again seem to have poor reading comprehension skills, how is spec testing the compiler? There's no specmark for the efficiency of the compiler, nor does it matter, because the goal of spec is to provide a benchmark for the CPU under the best conditions. You arguement about standardizing compilers is equivalent to having a horse (G5) race a F1 (Opteron), under the "standard condition" of a dirt track. It's ludicrous and retarded.

Hmm, does this appear to be vector processing done by a compiler? If I "have no idea how compilers, CPU architectures, system architectures are designed", then how come I'm the one saying that hand coding is a bad idea? Even a child of 5 could understand that "out of order instruction" means that it's unlikely your "tightly written asm" is going to be executed the way you want it, and add "superscalar", and it becomes completely out of reach for humans to keep track of all the depenencies.

I also need to add two new "claims" of yours to my list: Why I "have no idea how compilers, CPU architectures, system architectures are designed", and "people are still smarter than compilers".

And if "its not worth my time to argue with you.", then quit. You're only providing more and more evidence that the G5 is crap, that stupid fanatics will use any excuse or attack to desperately try to show it isn't, and that I can refute any delusional idea that you can come up with.

Actually, if Apple wanted they could use Intel's VT to run Windows XP as guest OS.

OS/X would own the hardware and the XP kernel would think it owned the hardware, but would actually be running a virtual machine. With VT, it's quick too. All the instructions that you need to trap to virtualise Ring 0 code can be trapped and emulated. Most of the emulation is trivial - added offsets to page table bases and so on. There's no need to patch the guest OS or single step through bits of it. Interestingly, they could run Linux in the same way, at the same time. With a bit of tweaking, you could imagine having Windows applications appear on the Mac desktop.

Whether it's a good idea is another question - it wouldn't certainly put people off porting their applications.

http://appleintelfaq.com/

What did Apple announce at the Apple Worldwide Developers Conference (WWDC) on June 6, 2005?

Apple announced that it is transitioning from PowerPC processors provided by IBM and Freescale (formerly Motorola) to x86 architecture processors from Intel. The first Intel-based Macs will ship before mid-2006, and the transition will be complete by the end of 2007.

Where can I find out more official information about this announcement?

Apple press release

Intel press release

WWDC keynote address (Transcript)

Why did Apple make this change?

The following scenario likely contributed to this decision:

IBM has been unable to meet its performance commitments for the PowerPC 970 family (G5) processors. In mid-2003, IBM promised 3 GHz G5s to Apple by mid-2004. As of mid-2005, 3 GHz G5s are still not available, over two years after the initial announcement, and over one year after the promised delivery.[1]

Meanwhile, Microsoft has announced that IBM will make 3.2 GHz triple-core G5 derivatives available to Microsoft for Xbox 360.[2] IBM is also concentrating efforts on chips for Nintendo Revolution and Sony PlayStation 3.[3, 3.1] With IBM concentrating on expensive high-end server class processors and the console and embedded markets, and with Apple at less than 2%[4] of IBM's PowerPC business, it was clear IBM's priorities were focused elsewhere.

Apple is also less than 3%[4] of Freescale's PowerPC business, with Freescale focusing on embedded, communications, and automotive markets. The priorities of IBM and Freescale do not coincide with performance and other needs of the traditional desktop and portable computing marketplace.

What has Apple done to prepare for this transition?

Apple has been publicly maintaining the core OS of Mac OS X, Darwin, for both PowerPC and x86 platforms since the release of Mac OS X. Internally, Apple has been secretly maintaining Mac OS X in its entirety and all Apple applications for both PowerPC and x86 for over 5 years, since before Mac OS X's public release.[5] Mac OS X's predecessors also ran on x86.

Apple has made available Xcode 2.1, which adds the capability of creating PowerPC/x86 universal binaries. Xcode 2.1 can be used on either PowerPC or x86 systems to create universal binaries. Application developers already using Xcode in most cases need only recompile their application with an additional checkbox adding x86 architecture support.

Apple has also licensed[6] QuickTransit from Transitive Corporation for Rosetta, a realtime binary translation system to support PowerPC binaries seamlessly on x86 hardware. The current performance of Rosetta

Someone does, but it's not cheap (Pentium M chipset) and it plays Doom 3 like a slideshow. According to Intel's datasheet, the 915GM chipset has a TDP of 6.0 watts. The chipset's GMA 900 graphics displays 2048x1536 at 85 Hz.

I cannot be the only one sick of the jet engine noise and space heater performance.

Ya know, like an Mac Mini, only with high resolution.

That Aopen board is microATX, but the Mac mini look-alike that Slashdot covered uses the same chipset, so it should display 2048x1536.

Someone does, but it's not cheap (Pentium M chipset) and it plays Doom 3 like a slideshow. According to Intel's datasheet, the 915GM chipset has a TDP of 6.0 watts. The chipset's GMA 900 graphics displays 2048x1536 at 85 Hz.

I cannot be the only one sick of the jet engine noise and space heater performance.

Ya know, like an Mac Mini, only with high resolution.

That Aopen board is microATX, but the Mac mini look-alike that Slashdot covered uses the same chipset, so it should display 2048x1536.

Someone does, but it's not cheap (Pentium M chipset) and it plays Doom 3 like a slideshow. According to Intel's datasheet, the 915GM chipset has a TDP of 6.0 watts. The chipset's GMA 900 graphics displays 2048x1536 at 85 Hz.

I cannot be the only one sick of the jet engine noise and space heater performance.

Ya know, like an Mac Mini, only with high resolution.

That Aopen board is microATX, but the Mac mini look-alike that Slashdot covered uses the same chipset, so it should display 2048x1536.

I think Apple will also need to differentiate (in hardware specs) between high-end PowerMacs and high-end (and less expensive) Wintel PCs. When Apple completes the trasition to Intel CPUs, I think they can still charge a premium for the "Apple experience," but not the premium they enjoy now using a different CPU architecture. The glaring price difference between a 3GHz Dell and a 3GHz PowerMac will be more difficult to explain if they are using the same CPU. A dual-processor desktop (that's not a dual-Xeon workstation) can be the difference Apple needs.

However, I think Apple will need to design their own dual-Pentium chipset if they go this route. Currently, Intel's high-end desktop and low-end workstation platform is limited to a single high-end Pentium CPU. If you want dual Intel processors, you need to move up to the expensive Xeon workstation platform.

I have no doubt at all that you'll see dual dual core in these systems even when the leap is taken to Pentium.

I'm not that certain, but I can see it happening if Apple can design a dual-CPU Pentium chipset in time. I think it's possible we'll see single-core Pentium Mac minis, dual-core Pentium iMacs, dual-CPU Pentium PowerMacs, and dual-Xeon XServes.

I think Apple will also need to differentiate (in hardware specs) between high-end PowerMacs and high-end (and less expensive) Wintel PCs. When Apple completes the trasition to Intel CPUs, I think they can still charge a premium for the "Apple experience," but not the premium they enjoy now using a different CPU architecture. The glaring price difference between a 3GHz Dell and a 3GHz PowerMac will be more difficult to explain if they are using the same CPU. A dual-processor desktop (that's not a dual-Xeon workstation) can be the difference Apple needs.

However, I think Apple will need to design their own dual-Pentium chipset if they go this route. Currently, Intel's high-end desktop and low-end workstation platform is limited to a single high-end Pentium CPU. If you want dual Intel processors, you need to move up to the expensive Xeon workstation platform.

I have no doubt at all that you'll see dual dual core in these systems even when the leap is taken to Pentium.

I'm not that certain, but I can see it happening if Apple can design a dual-CPU Pentium chipset in time. I think it's possible we'll see single-core Pentium Mac minis, dual-core Pentium iMacs, dual-CPU Pentium PowerMacs, and dual-Xeon XServes.

Intel's implementation for the 865P/875P chipset goes through the memory hub directly http://www.intel.com/design/chipsets/schematics/25 281202.pdf while the i845 chipset has the ethernet interface connected to the ICH4 controller hub that is shared among other devices like the PCI bus http://www.intel.com/design/chipsets/datashts/2519 2401.pdf. VIA's PT894/PT880 ethernet connection goes through a "VIA Connectivity" bus much like the Intel 845 http://www.via.com.tw/en/products/chipsets/p4-seri es/pt894pro and http://www.via.com.tw/en/products/chipsets/p4-seri es/pt880. There were some value motherboards that although I recall that they use good/decent chipsets, their designers decided to connect the built-in gigabit ethernet ports off the PCI bus. I cannot recall what these were but I read about them in anandtech several years ago.

Intel's implementation for the 865P/875P chipset goes through the memory hub directly http://www.intel.com/design/chipsets/schematics/25 281202.pdf while the i845 chipset has the ethernet interface connected to the ICH4 controller hub that is shared among other devices like the PCI bus http://www.intel.com/design/chipsets/datashts/2519 2401.pdf. VIA's PT894/PT880 ethernet connection goes through a "VIA Connectivity" bus much like the Intel 845 http://www.via.com.tw/en/products/chipsets/p4-seri es/pt894pro and http://www.via.com.tw/en/products/chipsets/p4-seri es/pt880. There were some value motherboards that although I recall that they use good/decent chipsets, their designers decided to connect the built-in gigabit ethernet ports off the PCI bus. I cannot recall what these were but I read about them in anandtech several years ago.

You mean kind of like the Intel Network Processors?

http://www.intel.com/design/network/products/npfam ily/

Sorry, but that's just wrong. If they had to ramp up a PCB (printed circuit board) production line somewhere to build these things, the cost would be enormous. Bear in mind this cost is going to be shared across what, a few hundred prototypes? At most.

If they did you'd be right. But they didn't, they're using an Intel D925XEBC2 motherboard according to people at WWDC - standard PC.

Most processors are at least a little bit faulty.

I was about to mod you insightful, but then I realized what the grandparent was probably talking about... the microcode. The microcode is a lot more important for recreating details. Pentium processors have a limited ability to fix silicon bugs in software. The microcode is in volatile memory, so it must be loaded on every boot by either the OS or the BIOS. It's top-secret, but access to this lower-level info might let you write a custom instruction (or not, I don't know how much the mechanism can change the operation)

An example microcode fix is for "High Temperature and Low Supply Voltage Operation May Result in Incorrect Processor Operation"

I haven't seen an IC data sheet yet that doesn't have an internal schematic of the IC

Find me an internal schematic for this IC? That would be cool. Thanks.

Try to find internal schematics for even simple ICs like LCD controllers or audio chip. And no, block diagrams dont count as internal schematics.

Hell, you can't even find internal schematics for AC97 DACs.

What. The. Fuck?

We recently went over this in my computer organization class.

You need to leave whatever school you're going to.

Immediately.

Seriously. It's obviously a complete waste of money. Do it now before your head gets any more messed up.

Intel is reluctant to release their ISA documentation

Nooooo they aren't.

Without releasing ISA documentation, people can't program your fucking CPU. Yeah, I'm sure that's exactly what Intel wants -- nobody to code for it. Great business plan. Where do I sign up?

Interesting? Jesus christ on a cracker, the mods are fucked up today.

The article is misleading; there is no actual confirmation that apple will use this technology. It's a possible alternative, and nothing more.
For more information on this technology:

Trusted Computing Group's website:
https://www.trustedcomputinggroup.org/about/faq/

An Intel whitepaper on the TPM mentioned in the article:
http://www.intel.com/design/mobile/platform/downlo ads/Trusted_Platform_Module_White_Paper.pdf

And an interesting article from macuser (older, but still relevant):
http://www.macuser.co.uk/macuser/features/61277/tr ust-me-im-your-computer/page7.html

USB started appearing on Intel chipsets in early 1996 with the Intel 430 chipset family. You are off by "several years" plus two.

I agree that the iMac helped create a market for USB peripherals (and USB adapters) because Apple gave them no other choice. But PCs had USB in their chipsets two years before the iMac. PC makers have this strange habit of including legacy ports, so peripheral makers and OS makers put off USB support until 1998.

Actually, based on the specs you linked to, I think it's more likely an Intel D945GTP motherboard. The necessary features that the D945GTP has built-in (and the D915GUX lacks) are SATA-2 (945 chipset) and FireWire 400.

The D945GTP also accepts dual-core processors. I bet someone will needlessly benchmark a dual-core Pentium D OS X development system against a dual-processor G5 PowerMac.

LaGrande

Based upon the specs and pictures of the box the Intel PowerMac it is most likely an Intel Desktop Board D915GUX. This has the GMA900 onboard graphics adapter and DDR-2 memory that xlr8yourmac.commentioned. It also has the same layout as the photos of the PowerMac board.

From the above:

Figure 5. Verifying Hyper-Threading Technology in Windows* XP Task Manager
Hyper-Threading Technology is enabled if there are two processor listed in Windows XP Device Manager. Only one processor driver will be installed if Hyper-Threading Technology is disabled in the BIOS settings. Note: It is important to have the latest INF utility in order to optimize platform performance with Intel Pentium 4 processor-based systems using Microsoft Windows XP.

• voice interaction
• visual interaction (face recognition, identifying shape, video analysis)
• 3D graphics
• machine learning

Intel's site on the topic appears to confirm my initial assertion. To that end, I would assert that threading is rather simple to implement under most modern models.

Several years ago, this would not have been the case. Many languages did not have multi-threading implementations that were intuitive. Now the story is a bit different. C and C++ have the POSIX threads library. Java is built to be mutli-threaded, as are many of the newer languages.

I see little difficulty in most systems making use of multithreading, and therefore, capitalizing on hyperthreading technology. Therefore, if it is the case that most applications do not make good use of threading, I would assert that it is improbable that this will remain the case in the future, and if it does, that programmers become educated on the topic of multithreading.

Intel® Xeon(TM) Processor

• 64-bit
• 2MB iL2 cache, 3.60 GHz or greater, 800 MHz FSB
• Intel® E7525 Chipset

Intel® Xeon(TM) Processor

• 64-bit
• 2MB iL2 cache, 3.60 GHz or greater, 800 MHz FSB
• Intel® E7525 Chipset

Who said Apple will use BIOS in its shipping Intel-based Macs?

Just because the developer transition platform is using an Intel BIOS doesn't mean that shipping Intel-based Macs in one to two years will:

From Dean Reece of Apple http://lists.apple.com/archives/Darwin-drivers/200 5/Jun/msg00020.html:

We realize there are lots of folks that need to know what is going to be in the ROMs on these new machines, and what partition scheme will be used. Unfortunately, we are not yet in a position to make that information available, but we will communicate it as soon as we reasonably can. Don't assume that what you see in the transition boxes represents what will be present in the final product.

> I'm cautiously hoping for EFI as the firmware.

The general consensus I've heard from other developers is:
1) They don't want us to use BIOS
2) If they haven't heard of EFI, they want us to use OF
3) If they have heard of EFI, they want us to use EFI

This is not a statement about what Apple will use, just what I've heard from developers that have an opinion on the subject.

Hang in there...
- Dean

Information on EFI:

http://www.intel.com/technology/efi/
http://en.wikipedia.org/wiki/Extensible_Firmware_I nterface

Intel is big into making all kinds of chips these days, not just CPUs. For a while they've been big on this "digital home" kick, too. Clearly they see the future for the semiconductor market being in more than just PCs. The only problem there is that Intel just has absolutely no clue how to market products to end consumers. They know how to move chips, not Webcams. If a partnership with Apple means Intel can get its chips into future digital consumer devices more easily and see those products become market leaders through Apple's genius product design and marketing, that can't be all bad.

The obvious answer: These machines bear no resemblance internally to the machines Apple will release to end users in 2006. They're just dev kits hacked together to let developers get a crack at the x86 architecture and get their apps ported in time, like the Yikes machines were hacked together to ship more G4s while working out the kinks in the next-generation Sawtooth motherboards. The final x86 Macs will probably use EFI, for one thing, and other custom parts that Apple hasn't developed yet and will never be available on the open market.

Processors with Intel EM64T allow platforms to access larger amounts of memory and will support 64-bit extended operating systems.

Eh???? From the forementioned weblink:

...Tiger on Intel pre-installed, allowing you to run,...

While this interium system may be generic whiteboxes, it seems likely that this analyst is on track with Apple's true path for boxes that ship.

http://www.eweek.com/article2/0,1759,1824781,00.as p http://www.intel.com/technology/security/

LaGrande is backward compatible (i.e., windows will boot) but there are hooks for Apple to stop MacOS X (not Darwin... the full proprietary stack) from booting up. Meshes very well with what the Apple VP is quoted as saying about not precluding Windows (or Linux) from booting.

Perhaps you should read about this.....

http://www.intel.com/technology/security/

(found after reading http://www.eweek.com/article2/0,1759,1824781,00.as p) )

If you think Apple won't use that technology to lock down the proprietary libraries that run on top of Darwin I'm not sure you're very familiar with Steve Jobs' mindset. That technology has Apple's name written all over it.

With that infrastructure apparently all they'd need is some kind of "am i legal Apple box key" service and just have some of the core Apple proprietary services "check in" before completing startup.

Still may be possible to crack, but likely far more involved than install on a fresh disk partition.

If they get Intel chips with VT, then you could have Windows, Linux, and OSX running at the same time.