Slashdot Mirror

Kind of ties in to the future of anti-virus...AV is moving towards file behavioral analysis to determine a file's viral status since signature detection presents an increasingly serious zero-day problem. Virtualizing a file is a great way to determine behavior, and Intel is working on hardware-assisted virtualization:

http://www.intel.com/technology/computing/vptech/

Tie the two ideas together, and you might see one of the ways Intel is hoping to use it's virtualization technology.

They say they are price competitive with non-quantum technology.

Then they have the following prices:

1. The NvIOpRAM [ATA IDE] products have the selling price of $2.50 per 1GB.
2. NvIOpSRAM [144-pin SODIMM] products have the selling price of $4.50 per 1GB.
3. NvIOpSRAM [200-pin SODIMM] products have the selling price of $6.00 per 1GB.

Let's see, then that 2TB drive will cost $5000, and that included 1TB of RAM will cost $6000, for a minimum notebook cost of $11,000. And that's assuming the rest of the notebook (including their miracle processor) is free. While 1TB of RAM for $6000 is a good price, 2TB of hard drive for $5000 is a bit on the steep side.

Oh, and some loony hardware enthusiasts have already shown that even some current processors can run at 7GHz.

They say they are price competitive with non-quantum technology.

Then they have the following prices:

1. The NvIOpRAM [ATA IDE] products have the selling price of $2.50 per 1GB.
2. NvIOpSRAM [144-pin SODIMM] products have the selling price of $4.50 per 1GB.
3. NvIOpSRAM [200-pin SODIMM] products have the selling price of $6.00 per 1GB.

Let's see, then that 2TB drive will cost $5000, and that included 1TB of RAM will cost $6000, for a minimum notebook cost of $11,000. And that's assuming the rest of the notebook (including their miracle processor) is free. While 1TB of RAM for $6000 is a good price, 2TB of hard drive for $5000 is a bit on the steep side.

Oh, and some loony hardware enthusiasts have already shown that even some current processors can run at 7GHz.

I'v searched Intel's manuals a bit and there's no mention whether store buffers or renamed registers are used to perform STLF

It's actually a store buffer in the P4 (and most other x86 processors for that matter). Trust me, I'm paid to know this.

http://www.intel.com/technology/itj/2004/volume08i ssue01/art01_microarchitecture/p03_netburst.htm

Also, there aren't many restrictions on STLF, mostly size (a load of 128 bits can't be forwarded from a store of 32 bits, although the converse is possible) and alignment.

There are partial address match problems because the P4 doesn't use all of the bits.

In common situations and if the compiler generates proper code, STLF restrictions shouldn't be very restricting at all.

Somebody please explain this to Adobe :)

Intel's official press release and text of filing(.pdf)

Has anyone here ever seen a sticker on a hardware product that says LINUX COMPATIBLE like you see with Windows Compatible products?

My HDTV3000 tv card was marked Linux Compatible, as was my Samsung laser printer.

That's because Microsoft acts based upon their standards

Yes, that's part of the problem. When is Microsoft going to support a common standard, like EFI? Linux has supported it for a while now.

AMD don't have an incompetent marketing department they just choose not to advertise much and save the money.

AMD spends more (~17%) on Marketing/General/Adminstrative as a % of revenue than Intel (~14%).

Is Intel Good(tm) now?

No.

The new line of chips are LaGrande Compliant. LaGrande is Intel's CPU embedded implementation of the Trusted Computing Group's Trusted Platform Module.

So what does that mean?

All of the new CPUs have ID numbers again. Remember the Pentium 3 ID numbers that created so much outrage and backlash? Whell they are back with a vengance.

The new CPUs will hold crypto keys, and they are specifically designed to keep the keys (and encrypted files) secure against the owner. They are specifically boobytrapped to self destruct if you try to read out your own keys. IBM is currently using a a seperate non-CPU Trusted Computing chip and they explicitly advertize the self destruct aspect in their Man in Black Thinkpad TV commercial.

It can also act as a little spy inside your computer - this is called Remote Attestation - a spy that watches all of the software you run and send a spy report to other people over the internet. You are denied any control over this spy report. The only control you have is to turn this system off completely, and if you turn it off then you get locked out of your own files and it is impossible to run or install Trust-using software. In a five to ten years, under Trusted Network Connect, you can even be denied an internet connection unless you activate the system and send this spy report and you have an approved unmodified operating system and approved unmodified software.

It is basically a DRM enforcer CPU, but far far worse.

-

The Pentium Extreme Edition is a dual core CPU with HT. It is basically a Pentium D 840 with HT enabled.

Pentium Extreme Edition = Dual 3.2 GHz 1 MB L2 Cache, with HT

Pentium D 840 = Dual 3.2 GHz 1 MB L2 Cache, without HT

The Pentium Extreme Edition is a dual core CPU with HT. It is basically a Pentium D 840 with HT enabled.

Pentium Extreme Edition = Dual 3.2 GHz 1 MB L2 Cache, with HT

Pentium D 840 = Dual 3.2 GHz 1 MB L2 Cache, without HT

People are well aware of the scaling limits and have been for years.

There is a fundamental physical limit that puts a cap on the amount of heat that can be removed from a solid per unit time.

We are fundamentally power limited. Moore's law says the transistor density increases exponentially, but we can't switch those transistors faster because the chip gets too hot and we can't remove that heat fast enough - FUNDAMENTALLY due to the laws of physics.

So there's a tradeoff. Either put more transistors on the chip and reduce speed, or put fewer transistors on the chip and increase speed.

This is very different from the past, when we had the luxury of BOTH increased transistor speed and increased density. The total power was not yet high enough to cause a problem.

For those interested in the details, I refer them to the following paper:

http://www.intel.com/research/documents/Bourianoff -Proc-IEEE-Limits.pdf

If the chip is shipped 2MHz out of spec, than normal operating fluctuations will push it even farther out of spec. Besides, if they're "just a little" off there, what else are they "just a little outside" on? The bigger issue is that the company is doing things wrong, not disclosing those changes, and can't be trusted to do anything properly.

Rather than speculating wildly any further, though, we go to the article to see what's actually being overclocked. The P5A2 has the 200MHz FSB overclocked by ~2.3MHz. That means the processor (15x FSB) is overclocked by ~33MHz. The memory is running ~3.3MHz faster, The memory timing is also changed from 4-4-4-12 to 4-3-3-8. Those are not what's specified in the BIOS setup, it's what the board does when you tell it to do something different.

The datasheet for the i925x's memory controller is here. Quoting that data sheet, "The 82925X MCH supports a FSB frequency of 200MHz". Asus is running it at 202.3MHz. That's 1.15% out of spec, on the clock that runs the memory and graphics core, among other critical components that have precision requirements of less than one percent for communication reliability. Of course, since you know all about motherboard design, and read the article, you will undoubtedly respond that your use of 800MHz as an example was supposed to be just silly, and that everyone knows that these things run with a FSB of 200MHz. Clearly ignoring the errors introduced by multipliers was "intentional". Presumably the mention of storage temperature, which has absolutely nothing to do with a running system, was also meant in jest.

I'll leave looking up the memory timings and what most memory manufacturers spec as an exercise for someone who isn't heading for bed now. I'll submit that those timings aren't what someone concerned with stability would use, though.

As far as "The actual operating speed of your processor probably varies that much in any given day just through heat and other factors!", well, you're forgetting that pesky 15x multiplier in this particular 3GHz system. I don't think many people would believe the claim that their processor varies by more than 33MHz due to temperature fluctuations during any typical day.

Regarding the FPS example: do TV shows look smoother than movies? Movies run at 24, TV runs at 30. Video games typically reach up in to the 50+ range (or did back when I had nothing better to do than playing video games). I recall not seeing a difference when I changed video cards and gained 20FPS in quake, because I was already "seeing" over 30. A 2 FPS gain at an already good frame rate is for bragging rights, not playability - much like you can't feel a gain of less than about 15HP in a typical car.

BTW, the concept of a Faraday cage is generally taught about in introductory physics courses - even those that the non-engineers take. If you find concepts like that to be impressive, please get back to me after you've finished your first year of post-high school education, or after microwaving a frozen burrito. Whichever comes first.

And Asus doesn't make [significant] chips, they apply them (note the Intel, AMD, Via, and nVidia offerings from Asus, among others) - my employer (for whom I do not speak) is fine with that.

People should have been using FBSD for wifi access points and aggregation long ago. Glad to see the corporate folks have finally figured out how linux errodes the bottom line.

The
stupider
ones
haven't
figured
it
out
yet.
You'd
better
go
tell
them!

That's just not true. If you want to work at intel look here. A quick search of their job tools show a number of open positions for Engineering jobs in the US. None of them listing a minimum GPA. As with anywhere if you are applying as an RCG i suspect that a better GPA gets you further....

Yep.. and guess who created the 2.88 floppy?

None other than Toshiba.

I'm not sure about the BIOS, but you're correct regarding the controller. PCGuide says the 500Kbit limitation of existing floppy controllers was insufficient; the 2.88 floppies required a 1Mbit transfer rate. I'm not sure why the drives couldn't be slowed down for the sake of compatibility though. Seems easy enough to throw a jumper on there to toggle 500Kbit/1Mbit transfer rates, but I'm no EE.

The computer industry seems to be moving at different speeds. Today, for example, you can buy a 64-bit CPU that operates at 3gHz, 32-bit memory that operates at 400mHz, and a 128-bit graphics card with 300mHz RAMDAC. Nobody seems interested in designing a complete system in the PC industry -- instead all the "progress" is in optimizing or extending components and hoping they work when you throw them together.

While generally I'd agree with that statement, it's not quite as clear cut as you make out. most programs run in small loops, so while your entire system may have 1gb of physical ram, it also has a few mb in cache spread through the various chips (like the CPU, graphics chip etc) eg a P4 with HT has atleast 512k cache. Most of the time the CPU is only using that cache and not using the system memory.

In short while it's a good idea to get everything working together and talking faster, in most cases bigger cache's and improving adding / seperating the busses will produce the desired effect without the additional costs & other problems involved with getting everything communicating at the same speed as the CPU.

If HP killed the Alpha, explain this press release from Intel dated June 25, 2001, announcing that "Compag will transfer key enterprise processor technology to Intel and consolidate its entire 64-bit server family on the Itanium architecture." Not only was it announced the Alpha processor line was going to be killed, but Compaq management made sure the decision could not be reversed.

According to the press release, there was only going to be one more generation of the the Alpha, and the transition was supposed to be completed by 2004.

Note the bullet point further down in the press release that "Compaq is transferring significant Alpha microprocessor and compiler technology, tools and resources to Intel." As explained further in the press release, "Over the next couple of years, several hundred Compaq microprocessor engineers, compiler experts and infrastructure employees will be offered employment with Intel."

Compaq before the HP merger had agreed with Intel to transfer all of the engineers. There was absolutely nothing HP could have done to have saved the Alpha even if HP had had a sudden conversion and decided to reject the Itanium. HP would have had to have started from scratch with the IP and people already transferred to Intel--it probably would not have been legally possible for HP to have changed direction at that point.

"do not be surprise if it disappears"

so I'm putting a copy here for safe keeping:

Wednesday August 10, 2005

- Mac OSX x86 on PC: and now a video! [Upd] - bad_duck - 21:03:35

The Apple Developer kit version of MacOSX x86 has indeed been fully cracked!
An anonymous source has sent us a video showing MacOSX x86 booting natively on a Pc notebook Mitac 8050D (Pentium-M 735/1.6GHz).

Boot Mac OS X 86 (Mpeg4 - 1,5 Mo) - [torrent]

As you can see the boot phase is rather fast, and the error message at the end is simply due to an right/authorization error due to the kext allowing PS/2 support.

[Upd]
Here is a second video showing the boot on the same hardware, the permission error was repaired. We can see the "About this Mac" panel, Apple System Profiler and CHUD prefpane showing information on the processor (frequency, cache etc...).

Boot Mac OS X 86 v2 (.mov - 11,5 Mo) - [torrent]

[Update] - We've added torrent files for the 2 videos to relieve the stress on our server. If you use them, please keep seeding as long as possible, thank you.

[translation by Eric]

[edited - windows vista crap removed]

- Mac OSX x86 on any PC : a reality, current status - Yoc - 14:18:24

Hereafter is the current status of the OSX x86 on any PC project run by PC/Mac "bidouilleurs"

Initial problems

Several system prevent running OSX x86 on any PC:
1. TPM chips from Intel
2. SSE3-enable processor
3. GMA900-based graphic card to natively support Quartz.

First solution: VMWare

simply install VMare on ANYPC, and this application will emulate what needs to be (GMA900, SSE3 ...)
Of course this is only a short-term solution, since it is rather instable, and particularly slow.
since this morning one can find on P2P network an already configures VMWare image OSX x86.

Patches

Several solution have been found. The TPM is cracked, and no one needs it anymore. The SSE3 requirement, can be bypassed via quite complex modifications, and this case several kernels will not work since they REALLY need SSE3. A patch for those kernels is being prepared allowing SSE3 instructions to be translated into their corresponding SSE2 ones.
GMA900 can be avoided by modifying CoreGrapics, patches are also available.

The best solution: the right hardware

The best and the most secure solution is a motherboard from Intel: D915 GA, GL or GU :
http://www.intel.com/design/motherbd/ux/

Excellent results have been obtained with Gigabytes GA-8I915P motherboard card:
http://www.giga-byte.com/Motherboard/Products/Prod ucts_GA-8I915P%20Duo-A.htm

Users with such a motherboard and a Pentium 4 will be able to install MacOSX x86 with the patch for Rosetta (without the patch if you choose a SSE3-enable Pentium4).
Be careful not to use any HD in RAID settings, otherwise it will crash your system.
Use a USB keyboard, PS2 port is not really well supported.

First tests

First tests have shown that MacOSX x86 on PC is very reactive, no crash, iTunes is running perfectly, with Rosetta.
Digital camera work perfectly with iPhoto, as well as digital camcorder with iMovie.

Another solution
The last solution is based on installation of Darw

Tell that to all the people dropping hundreds of dollars for a GOOD optimizing complier (icc is $399 commercial, free non-commercial).

My point is not to be an ad. My point is that free stuff is worth what you pay for it, and when you need quality, you get out the checkbook and pay someone. VMware had most of Xen's features four years ago, and most of the difference are things VMware has invented since then. We're in for some fun times, because BOTH VMware and Xen are going to be pushing things forward, and WE benefit.

After a review of Intel's Trusted Platform Module,

http://www.intel.com/design/mobile/platform/downlo ads/trusted_platform_module_white_paper.pdf

I have to completely agree that it is entirely redundant with well established systems.
Examples include
  www.pgp.com for encrypted file systems.
  Even Encrypting File System EFS which is free will do
I know of a few system that encrypt from bootstrap to power off as well

Their are potentially hundreds of products which perform the function of TPM today(without the DRM)

Their is no functional gain for anyone but those who want to whole sale DRM content for a premium to suxxors who now pay .5c to .10c a track, but I guarantee you they will pay $10 to $20 a track after Intel promulgates this bastage of a trogon horse.

But Intel will do as Intel has always done.
Ignore all the users on the planet, brown nose to Microsoft the RIAA and government who would dearly love to enable un by passable real time tax assessments on all internet on-line purchases.

I'll bet you the same year Intel succeeds; suddenly the Federal government will waffle and legalize every sort of crap now illegal because they can then tax it.

So you want legal drugs on line tax real time epay and a DRM album that costs $160 sign on to DRM

--
After were gone who will remember what we were fighting for?

The eye-toy isn't really innovative. That is, the idea of controlling stuff on screen using video of yourself is innovative, but Sony didn't think of it. I believe that was Intel. Intel Play Me2Cam was released in 1999, and sounds like it does the same sort of stuff as the eye toy. http://support.intel.com/support/intelplay/me2cam/ sb/cs-020521.htm

Gee, i wonder what this Kernel Extension that the Rosetta Launch daemon calls is good for: 'AppleTPMACPI.kext' ..and what's with that Class contained in its .plist called 'com_apple_driver_AppleTPMClient'?

No TPM, riiight.... That Infineon-Chip is just a dummy! ;-)

We must surely all be hallucinating!

Here's Apples Transitionkit Motherboard (well - minus the COM and LPT-Ports!), and looky looky, what's that about TPM in the column on the right?

If i want BS, I'll throw a laxative party at the local farm!

...chips for a while. There's a ton of them listed here . Look for the ones called EM64T.

And Dell has been selling these 64-bit chips for long time too.

==>Lazn

The official Intel page states otherwise
http://www.intel.com/jobs/israel/sites/haifa.htm

The pioneering 8088 processor, Intel® math coprocessors, the i860® XP processor, Ethernet communication chips, and cache and memory controllers are a few of the more than 50 products designed in IDC (Israel Development Center)

So it sounds like both of them were developed there

Every time I see a post like this, I simply respond with this url:

http://www.intel.com/education/sts/winners.htm

It speaks for itself, really.

Combine this with the growth of science as a whole over just the past 50 years. The older generations discover more and more things that the younger generations have to learn in addition to the older stuff, and repeat. If the younger generations were so incompetent, we eventually would stop seeing new drugs, more efficient vehicles, (yes, this stuff is still developed in the US as well as everywhere else, believe it or not) and on and on...

Do you mean this plant?

Because of the PowerPC's architecture we've been running Mac OS X, Mac OS 9, and Linux side-by-side on the same machine using Mac-on-Linux (or Mac-on-Mac) for some time without almost zero overhead. Virtualing the processor on PowerPC is way easier than on x86. Although if you get one of the new Intels with virtualization technology then you should be able to run multiple systems with little overhead as well. Currently the only option people have had for scalable system partitioning is Xen, which requires a special kernel built for it.

What's kind of interesting is that with the VT on the newer intel chips (I wonder if AMD will ever support it), it may make it easy to run intel OSX on a non-apple hardware because with VT you can catch exceptions transparently when the hardware does not exist.

ps- this original article is dumb, because why would Apple gut a profitable division so companies like Dell can sell more PCs?

Not the newest "comparably priced branded PCs" like the HP Pavilion a1010y and the Compaq Presario SR1010Z. They use the same integrated graphics (Intel GMA 900) that's working so well in Apple's Intel developer Macs. Quartz seems to work very well sharing system and video memory through PCI Express.

This means that Linux can be installed without breaking the existing installations or screwing with the boot loader at all. The DRM is a problem but there is not too much information about if there is going to be a lot of DRM in this new bios replacement.

Trust me, Linux might work on it, but copying other software created or installed on those other operating systems will NOT be easy. This time, the DRM is in the silicon.

And there's a link on the main page of the Intel EFI page.

Read more about EFI here.

All current Pentium 4 CPU's have had EMT64 for a while now - where have you been?

http://www.intel.com/products/processor/pentium4/i ndex.htm

http://www.intel.com/products/processor/pentium_D/ index.htm

If you look at the specs you will see that most 5xx and all 6xx & 8xx models have EMT64

All current Pentium 4 CPU's have had EMT64 for a while now - where have you been?

http://www.intel.com/products/processor/pentium4/i ndex.htm

http://www.intel.com/products/processor/pentium_D/ index.htm

If you look at the specs you will see that most 5xx and all 6xx & 8xx models have EMT64

Hehe. ;-)

For others who might really not know what it is, this is EFI:

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

...and this is easy since the Developer Transition Platform is just running a generic Intel motherboard, generic Intel video chipset, an Intel Pentium 4 660 processor with HT, and a standard Intel BIOS (NOT a Phoenix BIOS as has been incorrectly reported elsewhere), but what will really be great is when someone makes a Virtual PC- or vmware-like product (perhaps even one of those products themselves) that is a virtual machine that runs under Mac OS X that allows running essentially any x86 OS at near-full speed, side by side with Mac OS X, without having to reboot.

Since it will be running on x86 hardware, processor instructions do not have to be emulated: they can run natively at near-full speed of the underlying hardware.

Further, though Apple will do nothing to stop users from installing Windows on production Intel-based Macintosh systems, it's likely that the production systems will evolve beyond the generic hardware that makes the Developer Transition Platform. Apple itself has said, "Don't assume that what you see in the transition boxes represents what will be present in the final product." This means there may be additional specialized hardware for which Windows drivers and specialized support profiles will not be maintained by Apple. Of course, this isn't stopping anyone from making them, and Intel has said that Intel-based Macs will use commodity Intel processors, chipsets, and other support components, but it might not be quite as seamless as just popping in a Windows CD and installing (though it very well could be).

Let's also not forget that the production machines may not be - and likely will not be - using BIOS, rendering useless any such conventional PC multi-boot configurations. (But even with EFI or Open Firmware, there's no reason Apple couldn't maintain a robust multi-boot system.)

The point is that a virtual machine product could offer a supported configuration for x86 OSes, including Windows, Linux variants, etc., without the headache and hassle of rebooting into another OS. Sure, dual/multi-booting has benefits, and certainly this will be possible on even the production hardware, but most users would likely prefer a Virtual PC-like environment for running x86 OSes/applications without rebooting.

On this topic, one wonders if Microsoft will be the entity that releases this first. After all, they've already got Virtual PC for Mac, and Virtual PC for Windows (and Microsoft Virtual Server) is exactly this type of virtual machine product, albeit for Windows. On one hand, you can argue that for Microsoft, it's just another copy of Windows sold, so why should they care? But on the other hand, if they make a first-class VM product for Mac OS X that runs Windows (and other x86 OSes) seamlessly at near-full speed of the native hardware, it definitely assists in the sales of more machines designed primarily to run Mac OS X, which could be a poor strategic choice...

But even if Microsoft doesn't do it, let's hope someone like EMC does with vmware.

For more general information, see http://appleintelfaq.com/.

This is insanely high. Dothan's only rated up to 100C (see the Intel thermal guide), so you're certainly pushing the envelope. Assuming that the temperature report is accurate, check the fans are spinning and check for dust. Such high temperatures, plus a delta-T of 46C, will do the other internal components no good whatsoever. On the other hand, the reported temperature could be a complete load of bollocks brought on by a buggy BIOS or faulty sensor --- which can't be ruled out given that (a) your machine hasn't died yet; and (b) you've not been physically burned by it.

1. The slowest version of one of the "ultra low voltage" CPUs from Intel, which aren't really targeted at PCs, but at embedded devices
2. A slow, quiet (A/V) disk
3. Not much memory
4. Totally passive cooling

Really, you should have a clue before posting.

Uhm, sure, this is all entirely new to me.

According to this press release, the ARM processor consumes around 8mW, though I can't dig out any real-life figures atm

Well, shows your amount of clue by linking to a press release to point out 'facts'. I mean, are you serious?

To get back to our original discussion about the Linksys board that was quoted to consume 8.6 Watt while idle. As other pointed out, they state that that's measured at the mains side. They of course also say Current: Calculated current supplied at the 5V line, while running, but who am I to quible.

A little later they show a table which is measured at the 5V side. So let's take that, and let's assume that the nsieve test is 100% load of the entire board, and let's take the 266Mhz number because that gives the best performance/Watt. So that's 0.59A * 5V = 2.95 Watt. Pretty impressive for a board with Ethernet and 2x USB 2.0.

That is still NOT _several_ orders of magnitude better power consumption than a Celeron M @ 1GHz for example, which will run at something like 15 Watt at 100% load (probably less).

Sure, it's 5x the power consumption, but it also has video, IDE and audio. Plus floating point.

In case you are wondering, the datasheet for the Celeron M at 1 GHz (TDP: 5 Watt)

So, I hope you don't mind, but I'm still going to call a bullshit on that 'several orders of magnitude' comment.

This is utter bullshit.

Eh, no, it isn't.

I was looking for some more concrete evidence on power specifically for the CPU and from the datasheet, at 266MHz they quote 1.9W max power. The STPC Elite SoC datasheet (CPU from the example you used) states 4.8W for the x86 at 133MHz. At its been said, the power for the Slug is measured at the wall while the power from the example is measured at the board, with less memory and no peripherals. Also, on the performance side, the datasheet for the STPC Elite states that it uses 486 architecture. The article you refer to contains some Dhrystone figures and they compare to P5 figures at same frequencies. There is no FPU (there is a vector unit) but as its been said, for what this box would be used for there is not much need for floating point so the performance/power ratio would be higher than for an equivalent x86 board.

This is utter bullshit.

Eh, no, it isn't.

I was looking for some more concrete evidence on power specifically for the CPU and from the datasheet, at 266MHz they quote 1.9W max power. The STPC Elite SoC datasheet (CPU from the example you used) states 4.8W for the x86 at 133MHz. At its been said, the power for the Slug is measured at the wall while the power from the example is measured at the board, with less memory and no peripherals. Also, on the performance side, the datasheet for the STPC Elite states that it uses 486 architecture. The article you refer to contains some Dhrystone figures and they compare to P5 figures at same frequencies. There is no FPU (there is a vector unit) but as its been said, for what this box would be used for there is not much need for floating point so the performance/power ratio would be higher than for an equivalent x86 board.

I suggest taking a look at this paper which discusses theoretical limits on the binary switching model.

From http://www.intel.com/technology/magazine/standards /st01041.pdf:

TCG is currently comprised of a variety of vendors, including PC platform, operating system, and TPM vendors, with the board of directors consisting of representatives from Intel, IBM, HP, Microsoft, Sony, Sun Microsystems, Seagate, Verisign, and AMD. TPM vendors include Atmel, Infineon, National Semiconductor, and STMicroelectronics. Until now, TCG has focused on specifying a TPM for the PC.

Over four million PCs have been shipped with version 1.1 TPMs installed, mostly by IBM and HP. However, Intel has also begun delivering this technology and has just released the Intel® D865GRH Desktop Board, which has a version 1.1 TPM and ships with a software suite that provides better security for users' personal information. Version 1.2 of the TPM specification was recently released, and TPMs conforming to the new specification are under development.

Now that TPM definition for the PC platform has evolved, the TCG is expanding its membership and beginning to define TPMs for cell phones, handhelds, and servers--continuing to work toward the vision where all devices can talk to one another and communicate their trust state. Work is also moving forward on defining protocols necessary for communicating and interpreting the trust state.

In other words, there are other vendors producing TPM silicon. Intel is one of the late-comers for sample hardware, not the sole driving vendor that Larry Loeb seems to think they are.

I'd file Larry's theory under "Tinfoil/Paranoia."

I hate to break it to you but the PentiumM and the Pentium4-m are not one and the same processors.

PentiumM

Pentium4-m

I hate to break it to you but the PentiumM and the Pentium4-m are not one and the same processors.

PentiumM

Pentium4-m

from: http://www.intel.com/products/processor/index.htm? iid=HPAGE+header_products_processors&

Notebook
Intel® Pentium® M Processor
Mobile Intel® Pentium® 4 Processor supporting Hyper-Threading Technology
Mobile Intel® Pentium® 4 Processor
Mobile Intel® Pentium® 4 Processor-M
Intel® Celeron® M Processor
Mobile Intel® Celeron® Processor

Thus P4-m. Please avoid correcting people without checking.

or has stolen some great ideas from DEC

Stolen? I believe the word is "purchased"

I think a settlement followed by an acquisition is sufficient to avoid being called a thief. Aside from the fact that guilt was never established in a court. You must've worked at DEC or something. :)

I think you're jumping to conclusions. Ask this simple question, does Intel sell their compilers for their architecture or for EVERY x86 processor? The answer is simple, they sell compilers for THEIR processors. They are optimized for Intel processors, it says so right on the box. AMD is going to have a hard time using the compiler argument in a court of law. If AMD wants AMD optimized compilers then maybe AMD should write some!

Raid 6!
http://www.intel.com/design/storage/raid6.htm

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