Slashdot Mirror

What is VIIV? I have scoured the internet and Intel's site to figure this out.

Considering that Intel and Micron have collaborated to create a company called "IM Flash Technologies" to produce flash memory and that Apple has already pre-purchased a substantial portion of the output, I don't see this as much of a surprise. See here and here.

I noticed who the patent holder actually is: Parkinson; Ward D. One of the four Micron founders (http://experts.about.com/e/m/mi/Micron_Technology .htm).

Also, note that Intel and Micron have had a research realtionship in the past (http://www.intel.com/pressroom/kits/bios/dbaglee. htm) with Micron holding a 2% majority.

I think that we are close to seeing why Apple *really* switched to Intel. Put on your tin foil hat because I'm about to take you for a conspiracy theory ride:

1) Intel have been working with Ovonyx since 2000 on a technology called phase change memory (or PRAM, for short). Basically, PRAM uses chalcogenide - the same material used in rewriteable optical media - in a solid state RAM, only it is manipulated electrically, instead of optically. This gives the RAM nonvolatility and random accessibility. It is several orders of magnatude faster than flash (nearly as fast as DRAM) and has a write cycle endurance of 10^12 demonstrated as of about 4 years ago.

2) Intel patent applications have led me to believe that they have made great strides in the technology, while remaining very tight lipped. Here's some insight. Note that they are discussing the displacement of SRAM, DRAM and flash with this technology. Noteworthy, is the following:

[0058] Turning to FIG. 5, a portion of a system 500 in accordance with an embodiment of the present invention is described. System 500 may be used in wireless devices such as, for example, a cellular telephone, personal digital assistant (PDA), a laptop or portable computer with wireless capability, a web tablet, a wireless telephone, a pager, an instant messaging device, a digital music player, a digital camera, or other devices that may be adapted to transmit and/or receive information wirelessly. System 500 may be used in any of the following systems: a wireless local area network (WLAN) system, a wireless personal area network (WPAN) system, or a cellular network, although the scope of the present invention is not limited in this respect.

Now, here's where it all begins:

Envision, if you will, a high-speed, nonvolatile memory with very low power consumption. This enables the following:

1) Intel Robson Technology. This would answer the question of durability. Why would Intel demo such a technology if flash memory would wear out in short order? With PRAM, you've got CMOS compatibility so you can throw the whole deal right into the processor.

2) Ultra-low power wireless devices. Add Intel's Wireless USB and you've got the perfect medium to talk to your iPod. In addition, your gonna end up using it for more than just an iPod. Store your entire "desktop" on the damn thing, add some authentication mechanisms and you can use any wireless USB equipped PC to log into your "wireless personal server".

There's more, but this should be good for now.

That's the real catch. After years in a coma, UPnP is getting more popular in NAS that can advertise and stream multimedia content to Digital Multimedia Adapters directly connected to Home Theater gear, whithout the need of a controlling PC/media server.

I recently tested Maxtor's Shared Storage Plus with a 37" Acer LCD TVwith integrated media gateway. Just plug the two devices to the same network (or hook them to the same wireless router) and you get on the TV a list of photos, videos and music you have on the hard drive. You can choose the desired content using the remote control - in a Media Center/MythTV-like interface - and enjoy the show.

This kind of technology is likely to receive a big boost after the release of the next version of the Intel's Viiv platform, which will include some media server capabilities, based on Digital Living Network Alliance and Networked Media Product Requirements. Such features should be available without the need of booting Windows XP. That's what will really make Viiv stand out from normal multimedia PCs.

I'm an Apple fanboy, and love Bonjour, but I think that it needs a big push toward non-PC and consumer electronics devices to compete with UPnP and Viiv. Or maybe Apple should try to join the DLNA and steer it toward Bonjour (unlikely).

It gets better
It has a second part about decreasing costs.
Not only will transistor size decrease, increasing transistor count exponentially,
but the cost to produce the second chip with 2x the transistors will be less than the original.
Here's my source: http://www.intel.com/technology/silicon/mooreslaw/

Where did I find the Evil Research(tm)? Where else but directly from the source of evil -- no, no, not Microsoft, the *other* source of evil -- Intel :-)

It's already in their compiler;
http://www.intel.com/software/products/compilers/c lin/docs/main_cls/mergedprojects/optaps_cls/common /optaps_pgo_sspopt.htm
(Their compiler absolutely rocks BTW)

And their excellent paper titled "Speculative Precomputation: Long-range Prefetching of Delinquent Loads" by Jamison Collins, Hong Wang, Dean Tullsen, Christopher Hughes, Yong-Fong Lee, Dan Lavery, and John Shen can be found here;
http://www.intel.com/research/mrl/library/148_coll ins_j.pdf

(Those damn delinquent loads -- GET OFF OF MY LAWN YOU DELINQUENTS! :-)

There's also
"Physical Experimentation with Prefetching Helper Threads on Intel's Hyper-Threaded Processors" by
Dongkeun Kim, Steve Shih-wei Liao, Perry Wang, Juan del Cuvillo, Xinmin Tian, Xiang Zou, Hong Wang, Donald Yeung, Milind Girkar, and John Shen which can be found here;
http://www.cgo.org/cgo2004/papers/02_80_Kim_D_REVI SED.pdf

And also;
"Speculative Precomputation on Chip Multiprocessors" by Jeffery Brown, Hong Wang, George Chrysos, Perry Wang, and John Shen at;
http://www.cs.ucsd.edu/~jbrown/papers/sp-cmp.pdf

There, that ought to cure your insomnia and answer your question: "How could a thread possibly be executed far enough in advance to make the time savings worth while, yet be sure that it is "predicting" memory accesses correctly?"

Read the papers carefully -- there will be a quiz later.

Where did I find the Evil Research(tm)? Where else but directly from the source of evil -- no, no, not Microsoft, the *other* source of evil -- Intel :-)

It's already in their compiler;
http://www.intel.com/software/products/compilers/c lin/docs/main_cls/mergedprojects/optaps_cls/common /optaps_pgo_sspopt.htm
(Their compiler absolutely rocks BTW)

And their excellent paper titled "Speculative Precomputation: Long-range Prefetching of Delinquent Loads" by Jamison Collins, Hong Wang, Dean Tullsen, Christopher Hughes, Yong-Fong Lee, Dan Lavery, and John Shen can be found here;
http://www.intel.com/research/mrl/library/148_coll ins_j.pdf

(Those damn delinquent loads -- GET OFF OF MY LAWN YOU DELINQUENTS! :-)

There's also
"Physical Experimentation with Prefetching Helper Threads on Intel's Hyper-Threaded Processors" by
Dongkeun Kim, Steve Shih-wei Liao, Perry Wang, Juan del Cuvillo, Xinmin Tian, Xiang Zou, Hong Wang, Donald Yeung, Milind Girkar, and John Shen which can be found here;
http://www.cgo.org/cgo2004/papers/02_80_Kim_D_REVI SED.pdf

And also;
"Speculative Precomputation on Chip Multiprocessors" by Jeffery Brown, Hong Wang, George Chrysos, Perry Wang, and John Shen at;
http://www.cs.ucsd.edu/~jbrown/papers/sp-cmp.pdf

There, that ought to cure your insomnia and answer your question: "How could a thread possibly be executed far enough in advance to make the time savings worth while, yet be sure that it is "predicting" memory accesses correctly?"

Read the papers carefully -- there will be a quiz later.

Despite the lack of details, it sounds quite a bit like Intel's Mitosis research:
http://www.intel.com/technology/magazine/research/ speculative-threading-1205.htm

The article has simulated performance comparisons.

From the article:
"Today we rely on the software developer to express parallelism in the application, or we depend on automatic tools (compilers) to extract this parallelism. These methods are only partially successful. To run RMS workloads and make effective use of many cores, we need applications that are highly parallel almost everywhere. This requires a more radical approach."

This is very relevant to me because my 4 year old laptop died Sunday. I now have a choice between Core Duo or Turion... what I really want is a 64 bit dual core laptop.

None of Merom/Conroe or AM2 can be purchased by users today, and Merom is late 2006 at the earliest...

My best hope is gutting it out until May 9 when AMD introduces the Turion X2 on AM2 or S1. For a dual core 64 bit laptop from Intel the wait would be too long.

http://www.channelregister.co.uk/2006/02/09/intel_ mobile_roadmap_feb_06/

AMD's Turion 64 Roadmap

Before you completely eliminate Merom, note that Merom is pin-compatible with Yonah and will work with Intel's current 945PM/GM chipsets. Therefore, a Merom-upgradable notebook is available today, dual-core Turion 64 chipset and CPU is probably coming in June, and the Merom CPU is coming in October at the very earliest.

If you are considering the upgrade-to-Merom option, a cheap interim solution will soon be available. Celeron M processors based on the Core Solo core are now selling in Japan:

'Yonah' Celeron M 420, 430 ship in Japan

This is very relevant to me because my 4 year old laptop died Sunday. I now have a choice between Core Duo or Turion... what I really want is a 64 bit dual core laptop.

None of Merom/Conroe or AM2 can be purchased by users today, and Merom is late 2006 at the earliest...

My best hope is gutting it out until May 9 when AMD introduces the Turion X2 on AM2 or S1. For a dual core 64 bit laptop from Intel the wait would be too long.

http://www.channelregister.co.uk/2006/02/09/intel_ mobile_roadmap_feb_06/

AMD's Turion 64 Roadmap

Before you completely eliminate Merom, note that Merom is pin-compatible with Yonah and will work with Intel's current 945PM/GM chipsets. Therefore, a Merom-upgradable notebook is available today, dual-core Turion 64 chipset and CPU is probably coming in June, and the Merom CPU is coming in October at the very earliest.

If you are considering the upgrade-to-Merom option, a cheap interim solution will soon be available. Celeron M processors based on the Core Solo core are now selling in Japan:

'Yonah' Celeron M 420, 430 ship in Japan

If you look at the business world they could'nt and wouldn't switch their entire system over to shiny white macs when they could have a load of Dell PC's at the fraction of the cost.
Hasn't this been disproven time and time again already ... well let me give it a shot then. Two machines similar specs, in fact after the changes below the only difference is the proc. Pentium D on the Dell, Core Duo on the Mac Dell: Dimension 9150 Stock (except for 2 changes: 1. removed modem -$20; 2. Added real copy of Windows +$10, for a net change of ... -$10) $1,416
Apple iMac Core Duo Stock: 1,299
AND the according to Intel's Price List the Core Duo 1.83 is 294, while the Pentium D 820 is 241

For home users they want something cheap that can do the basics like go on the internet, get an email or two perhaps do some work.
Mac Mini
For gamers, they want something that can the latest games and than can be upgraded cannot meet those requirements.
The hardware is there PowerMac, it's the software that needs to come over.
Macs are designed for graphics and to look nice, OS X is an excellent operating system I myself may purchase one of the new mactel machines, but when the consumer has a set budget then Apple is well out of their league,
Price again ... see above
plus the source is unreliable as its rather bias from a Mac fan news site.
No argument there
The reality is without the Ipod, Apple would be doing a lot less well, because of the shear marketing factor the ipod has had on the company.
Who knows what would have happend if they didn't come up with the iPod, but the fact is they did and it's been a success.

I'm now an avid OS X user and since the intel switch I have to ask, what's so custom about mac hardware now? The plastic casing?

It's not particularly custom, but I can tell you think all motherboards work the same way with no issues. Let me assure you that this is not the case. At the unpleasant end of the spectrum, some motherboards are known to have, e.g. broken USB chipsets that don't really work properly.

If you want to laugh this off, take a little look at Intel's motherboard selector. Try asking it to show you ATX motherboards - it returns 765 motherboards to choose from. And that's just ATX form factor boards. We haven't even started on graphics/sound cards etc.

Also, when I custom built my own PCs back in the day all the drivers I needed to run windows came with the hardware I purchased, not from microsoft, not ever.

So Windows never had drivers for your motherboard, USB, serial port, keyboard, mouse, numerous video/sound cards etc?

Regardless, it's an irrelevant point - either MS writes the drivers or the manufacturers do. Hence my point that:

All Apple would need to do is to get all the hardware manufacturers to write OS X drivers for their hardware, or do it themselves.

I put in the part about testing because [a] MS test/certify a lot of drivers, and [b] Apple are big on user experience (i.e. they like the stuff they sell to work).

In fact, as a musician, this applies to audio peripherals I buy for OS X now. As an example: M-Audio' s support for updating drivers is total shit . Do I fault OS X? No. Just like I didn't when I custom built my own PCs.

Audio peripherals other than a fairly standard sound card don't really count. We're talking about stuff that the average person would have as a result of buying a PC. If Apple tried to sell OS X as a user-anywhere x86 OS, but didn't have wide driver coverage, their image would be in tatters (see my point about the number of motherboards out there).

Mmm... kool-aid.

Was there a particular point you were trying to make there, or just trying to fulfill a meme per post quota?

> ARM and Intel are operating in very different market areas these days

how so, if intel is building so many ARM CPUs?

http://www.intel.com/design/intelxscale/xscaledata sheet4.htm

But of course i agree that ARM are in no way comparable to x86 and of no smaller importance

This is truly the stupidest post I've ever seen on Slashdot. Are you high? Almost every business, big or small, has a public fax line you retard. And they print that number on business cards, email signatures, the contact page of their website, and everywhere else they put their contact information.

Here are some fax numbers for you:

http://www.ibm.com/contact/us/
http://support.microsoft.com/contactus/?ws=mscom
http://www.intel.com/intel/location/USA.htm
http://www.amd.com/us-en/Corporate/AboutAMD/0,,51_ 52_3592_712,00.html

Obviously, the "Fortune 100" companies you've worked for were much bigger than these petty little corporations who have publicly available fax numbers. Since I work for a company that sells fax service among other things, I can tell you there are actually just a few (hundred thousand) more businesses out there who publicly publish a fax number.

Feel free to actually go to the websites of a few of the 99 or 100 Fortune 100 companies that you have never and will never work for and see if they have a fax number available online. Ass.

It's surprising that Intel would have OOOE in Core. Itanium 2 goes more the VLIW route with what Intel terms "EPIC," short for Explicitly Parallel Instruction Computing.

now that Apple have switched to using Intel x86, wont they simply be using standard Intel chipsets too?

Yes, and no.

They are using standard Intel chipsets, but not the standard 1980s PC BIOS. They are using the Extensible Firmware Interface (EFI) instead.

The only PCs which use that are IA-64 Itanic^H^H^H^H^H^H^H Itanium based (sink, you've hit the x86_64 iceburg, now is the time for you to sink); so far only 64 Bit Windows versions support booting in an EFI enviroment, no version works with EFI and a 32 Bit processor.

Even Vista dosn't support EFI on 32Bit, but I would lay odds on it working when it's released, now Apple are offically helping getting Windows booted on their hardware.

Bryn

Slashdot doesn't update a lot of logos. They refuse to update to Intel's new logo too.

With the Intel C++ Compiler? (plus some more compatibility information (PDF))

With the Intel C++ Compiler? (plus some more compatibility information (PDF))

I plan to buy a Macbook in the near future if I don't see any Intel-based iBooks soon.

If the "Celeron" brand makes you wince, "Celeron M" CPUs (based on the Pentium M architecture) have always been very good performers and an outstanding value. At launch, the new Celeron M CPUs will clock at 1.73GHz and 1.60GHz, have 1MB L2 cache, and 533MHz FSB.

I'm trying to guess what the grandparent means by hypervisor services.

I wasn't sure either, but presumed that they are simply referring to the concept of a hypervisor, which is the foundation of a multi-operating system platform. Regarding native instruction pass-thru, both Virtual Server/Virtual PC and VMWare's products do largely the same thing, passing most instructions through directly. What the article talked about regarding Microsoft's next version was support for Intel's new Vanderpool technology, which is basically additional support for virtualization on the chip itself. As it is the software expensively save and restore states between context switches to different virtual sessions, and Vanderpool offers some hardware that makes it much less costly. If you have a lot of CPU saturated virtual sessions, it can make a big difference.

The iPod/iTunes system will move into a niche with Macintosh computers because Steve Jobs has again stuck with closed architecture and total control.

If we view the iPod as just another appliance, then there no evidence to back this claim. There are millions of televisions, DVD players, stereo stytems, etc. sold every year that do not have an open architecture. Most people do not need access to the architecture of these devices. They just want them to work when they use them.

On the other hand, if we view the iPod as a Personal Server gussied up to look like an music player, then there may be something to this. The real question is how closed is the iPod platform, and how quickly could Apple open it if or when they needed to?

Intel: http://www.intel.com/business/bss/products/client/ vistasolutions/index.htm

AMD: http://www.amd.com/windowsvista

VIA: http://www.via.com.tw/en/products/vista/cpu.jsp

My problem is with the consistently mediocre reporting, when just a little bit more effort would get to primary sources, rather than this persistent blog banality culture.

Intel v. AMD???

I have had funniest experience with Intel itself.

P4 v. Pentium Dual Core: one core at 2.2GHz beats 3.5GHz P4.

P4 v. Pentium M/Centrino: at several benchmarks the notebook chip at 1.5GHz had beaten by 2 times 3.2GHz P4

If you have ever programmed in assembler and read even single spec for CPU and code optimization (Intel has good tradition of releasing such specs) you would definitely know that clock speed itself has only modest influence on overall performance. (To put it simply: exec'ing user's code isn't only task CPU is doing - all communication with peripherals goes thru CPU too. CPU normally do about 30-70% of their peek performance: interrupt latencies, memory latencies, bus access latencies, etc.)

Read on: older http://www.intel.com/design/pentium/manuals/ & newer http://www.intel.com/design/pentium4/manuals/index _new.htm

Intel v. AMD???

I have had funniest experience with Intel itself.

P4 v. Pentium Dual Core: one core at 2.2GHz beats 3.5GHz P4.

P4 v. Pentium M/Centrino: at several benchmarks the notebook chip at 1.5GHz had beaten by 2 times 3.2GHz P4

If you have ever programmed in assembler and read even single spec for CPU and code optimization (Intel has good tradition of releasing such specs) you would definitely know that clock speed itself has only modest influence on overall performance. (To put it simply: exec'ing user's code isn't only task CPU is doing - all communication with peripherals goes thru CPU too. CPU normally do about 30-70% of their peek performance: interrupt latencies, memory latencies, bus access latencies, etc.)

Read on: older http://www.intel.com/design/pentium/manuals/ & newer http://www.intel.com/design/pentium4/manuals/index _new.htm

I don't know about you but all this stuff about HT Technology and stuff is very confusing. Do I have that?

This just proves that Vista is unready for the desktop. I guess that's why they cancelled it. Har har! Har har. Har, har. *sigh*

Seriously, what exactly is DesktopBSD's website supposed to say? The thing you quote seems reasonable to me, anyone who doesn't understand it is unlikely to find any way of wording it useful anyway, unless it was worded in such a way that'd make it useless to an actual computer professional.

It's not like they'll be installing it. They'll be asking us to do it, as usual.

I don't know about you but all this stuff about HT Technology and stuff is very confusing. Do I have that?

This just proves that Vista is unready for the desktop. I guess that's why they cancelled it. Har har! Har har. Har, har. *sigh*

Seriously, what exactly is DesktopBSD's website supposed to say? The thing you quote seems reasonable to me, anyone who doesn't understand it is unlikely to find any way of wording it useful anyway, unless it was worded in such a way that'd make it useless to an actual computer professional.

It's not like they'll be installing it. They'll be asking us to do it, as usual.

I reckon the USB over WiMedia has a better chance of winning.

http://www.intel.com/technology/comms/uwb/

Mostly because USB stacks are pretty rock solid on most devices, whereas Bluetooth stacks are awful on everything.

If you had a W-USB keyboard and system, even the Bios would have a driver for it, and it would work the way USB HID devices now, which is always flawless.

For x86 assembler, Intel is a good source of information: http://www.intel.com/design/Pentium4/documentation .htm#manuals. You'll want to check out volumes 2A and 2B at a minimum for reference material.

I would be surprised if Alexander used the Visual Studio debugger; more likely he used SoftICE or one of the Windows debuggers (NTSD/CDB/KD/WinDbg). SoftICE is a commercial product sold by Compuware and provides both user-mode and kernel-mode debugging. A version of the NTSD debugger comes with Windows, but is less useful than the one that comes with Debugging Tools for Windows. NTSD and CDB provide user-mode debugging, the only difference between the applications being that NTSD opens a new console window and CDB does not. KD is the kernel debugger. WinDbg provides the same functionality as NTSD/CDB/KD but with a (spartan) Windows interface.

Maybe some day it will be possible again in some way, to get print manuals...

Printed copies of the IA-32 Intel® Architecture Software Developer's Manuals, Volumes 1-3 are not available at this time. Please check back in April 2006 to see when printed copies will be available.

You can order them from Intel using their website http://www.intel.com/design/pentium4/manuals/index _new.htm and scroll down to order the hard copy of the manual. I have not found any link on AMD's site where they mention that they will send you a hardcopy. Only a collection of PDFs is available.

Did you miss the announcement of the dual core mac mini with the GMA 950 chip? and the recent update of EyeTV that supports 1080i on the mini?

http://hometheater.consumerelectronicsnet.com/arti cles/viewarticle.jsp?id=38271

http://www.intel.com/products/chipsets/gma950/

mmm.... FUD.

A contemporary implementation of the Framework on a PC includes a CSM for supplying services to operating systems that do not boot using EFI and for supporting legacy option ROMs on add-in cards. For legacy boot the Framework initializes the platform's silicon and executes EFI drivers.

Yonah supposedly also includes virtualization support. (It was supposed to, I haven't seen any confirmation either way whether or not it actually is in there. Anyone know if it would show in /proc/cpuinfo and what specific flags I should look for? I've got a T2500-based laptop in the other room.)

Also I have not seen any confirmation yet that the new Sossaman actuall supports multiple CPU packages per system. Intel's comparison page for the Xeons lists all of the Xeons I've tried as being MP capable, while the new Xeon LVs are only listed with a system type of "DP" - the die/package already HAS two processors though.

As to the issue of shared vs. independent cache between the cores - yes, it looks like while most third-party specs for the Yonah list it as "1M + 1M", it is actually 2M of unified cache for both cores according to Intel's docs/marketing for the Yonahs. The 1M+1M spec on some vendor pages (such as NewEgg) might be a holdover from the Athlon 64 X2 (which has independent L2 caches) and the Pentium D (which also appears to have independent L2 caches).

That unified cache between the cores could provide a significant performance boost that would make up for the unpleasant shared FSB.

Yonah supposedly also includes virtualization support. (It was supposed to, I haven't seen any confirmation either way whether or not it actually is in there. Anyone know if it would show in /proc/cpuinfo and what specific flags I should look for? I've got a T2500-based laptop in the other room.)

Also I have not seen any confirmation yet that the new Sossaman actuall supports multiple CPU packages per system. Intel's comparison page for the Xeons lists all of the Xeons I've tried as being MP capable, while the new Xeon LVs are only listed with a system type of "DP" - the die/package already HAS two processors though.

As to the issue of shared vs. independent cache between the cores - yes, it looks like while most third-party specs for the Yonah list it as "1M + 1M", it is actually 2M of unified cache for both cores according to Intel's docs/marketing for the Yonahs. The 1M+1M spec on some vendor pages (such as NewEgg) might be a holdover from the Athlon 64 X2 (which has independent L2 caches) and the Pentium D (which also appears to have independent L2 caches).

That unified cache between the cores could provide a significant performance boost that would make up for the unpleasant shared FSB.

Yonah supposedly also includes virtualization support. (It was supposed to, I haven't seen any confirmation either way whether or not it actually is in there. Anyone know if it would show in /proc/cpuinfo and what specific flags I should look for? I've got a T2500-based laptop in the other room.)

Also I have not seen any confirmation yet that the new Sossaman actuall supports multiple CPU packages per system. Intel's comparison page for the Xeons lists all of the Xeons I've tried as being MP capable, while the new Xeon LVs are only listed with a system type of "DP" - the die/package already HAS two processors though.

As to the issue of shared vs. independent cache between the cores - yes, it looks like while most third-party specs for the Yonah list it as "1M + 1M", it is actually 2M of unified cache for both cores according to Intel's docs/marketing for the Yonahs. The 1M+1M spec on some vendor pages (such as NewEgg) might be a holdover from the Athlon 64 X2 (which has independent L2 caches) and the Pentium D (which also appears to have independent L2 caches).

That unified cache between the cores could provide a significant performance boost that would make up for the unpleasant shared FSB.

A contemporary implementation of the Framework on a PC includes a CSM for supplying services to operating systems that do not boot using EFI and for supporting legacy option ROMs on add-in cards. For legacy boot the Framework initializes the platform's silicon and executes EFI drivers.

Fine the subject doesn't make complete sense.. BUT... doesn't compiling code with Intel's cc result in significantly better binaries than any flag you can throw at gcc?? From http://www.intel.com/cd/ids/developer/asmo-na/eng/ 219902.htm, MySQL claims 20 percent performance improvement over gcc.

I'm not saying we all have access to icc, but if someone wants to make a binary available, I'm more liable to use that than compiling from source. Call me crazy. And I know someone will.

Well it isn't like that anymore let me assure you!

On the Intel 945 chipset, Windows installs fine if SATA is set to IDE mode which is the default. If you set it to AHCI *OR* RAID mode, then it requires drivers from Intel http://www.intel.com/design/chipsets/matrixstorage _sb.htm. This was with a clean XP SP2, and I'd imagine it'd be the same for Win2k also.

Generally SATA controllers do require drivers, unless they are in an IDE compatabile mode. It appears that the Intel 945 has a newer ICH which is likely the reason why this is the case. Going forward, I can't see the situation changing.

RAID mode isn't the issue, it is the IDE compatability mode.

Have you tried installing Windows XP on a computer with an SATA hard drive? Oh man. Pain. You actually have to kick the thing into life using drivers loaded off a FLOPPY DISK.

Some nForce motherboards I've seen have SATA RAID mode enabled in the BIOS by default, which might require a floppy. One motherboard I installed Windows XP on, the Intel D915GUX, had the SATA boot option disabled by default in the BIOS. After enabling SATA boot and choosing SATA mode without RAID, WinXP w/SP2 installed without a problem.

Maybe SP2 needs to be on the installation CD, but I doubt it because I'm pretty sure I installed Win2000 w/SP4 on a SATA drive (Intel D945GT motherboard) without using a floppy (this was a while ago).

Have you tried installing Windows XP on a computer with an SATA hard drive? Oh man. Pain. You actually have to kick the thing into life using drivers loaded off a FLOPPY DISK.

Some nForce motherboards I've seen have SATA RAID mode enabled in the BIOS by default, which might require a floppy. One motherboard I installed Windows XP on, the Intel D915GUX, had the SATA boot option disabled by default in the BIOS. After enabling SATA boot and choosing SATA mode without RAID, WinXP w/SP2 installed without a problem.

Maybe SP2 needs to be on the installation CD, but I doubt it because I'm pretty sure I installed Win2000 w/SP4 on a SATA drive (Intel D945GT motherboard) without using a floppy (this was a while ago).

Not really. Consider graphics cards.

UGA is a replacement for int 10 and the VGA registers. It gives you GetMode, SetMode and Blt. It's only used for bootup and blue screens in Windows, just like int 10 and VGA access is used at the moment.

When Windows is running, it uses a custom driver for perforemance and flexibility, so it can access the acceleration hardware which is not part of EFI.

ftp://download.intel.com/technology/efi/docs/EFI95 uga.zip

It's the same for the other EFI drivers - this is all for preboot and crashdump, not for when the system is working properly. IIRC, once the kernel starts, it calls into EFI to terminate most of the services.

So, for graphics you could use it as an OS independant driver, much in the way you can use the int 10 Vesa linear framebuffer on PCs now. But it's no substitute for a real native accelerated driver.

The fact that GPL requires a lawyer to describe what you can and can not do with software is scary enough for businesses. "If we compile our proprietary software with gcc, do we now have to distribute the source?"

You should check that with other compilers, too. The last commercial compiler I bought (a version of Borland C++) came with a list of exclusions to the license that meant any given project may or may not be legal to distribute in binary form if compiled by it, depending on what kind of software it was. I forget the details, but I recall that "operating environments" (which could have described what I was developing at the time) were not permitted, and I think anything that competed with a Borland product may have been excluded.

GCC's fairly simple: all you need to do is read the licences in question (there are two, the one for GCC itself, which is traditional GPL, and one for 'glibc', which is GPL plus an exception to the requirement to license any product linked with it under the GPL) to realise that the answer to your question is no. Anyone reasonably well educated can do it. Certainly any experienced and competent IT manager ought to be able to.

"If we include the GPL'd drivers for the left-handed USB Framis, are we compelled to release our source, or just the driver's source code?"

A slightly more complex question, but it's still reasonably simple to answer. Read the license. It outlines cases where this is necessary in very clear (although admittedly technical) terms. Also note, most businesses would probably ask a lawyer to check the license terms on any product that they purchased the right to use like this anyway.

Businesses do not like confusion. The government gives us all that we can stand, so adding in an obscure, vision-inspired license doesn't make us comfortable.

I think the only way you can describe the GPL as obscure is if you haven't read it. The GPL v2 is very easy to read, and the GPL v3 draft is even easier.

Compare it with other organisations redistribution licenses, e.g. the Microsoft .NET framework EULA, or the even harder to understand Intel Performance Primitives Library EULA.

Go to http://www.intel.com/design/mobile/platform/umpc.h tm and click on "Watch the video of possible usages" to see videos showing how a keyboard can be "extracted" from the device.
Some mechanisms are a very beautiful piece of engineering and design.

At least some versions are expected to have keyboards.

At least on of the concept models has a keyboard:

http://www.intel.com/design/mobile/platform/umpc.h tm

However I don't think this made it to the launch products:

http://www.intel.com/design/mobile/platform/device s.htm

At least on of the concept models has a keyboard:

http://www.intel.com/design/mobile/platform/umpc.h tm

However I don't think this made it to the launch products:

http://www.intel.com/design/mobile/platform/device s.htm

just provide the drivers... the community will deal with the rest...

Dell doesn't make the hardware, they integrate it and support Windows users that don't want to deal with the "community." The hardware they use (Intel chipsets/wireless, ATI/NVIDIA graphics, etc.) have Linux drivers available from the hardware manufacturers. The rest are already provided with Linux.

The "community" complains too much.

IPMI is very powerful. An IPMI session starts with a Presence ping Any machine with IPMI hardware should answer a "presence ping" on UDP port 663. This identifies an IPMI-capable machine, and returns some vendor info. Anyone can send this. This should work even if the machine is "turned off", as long as it has standby power and is on a LAN.

Then, there's a challenge-response authentication sequence. More on this later.

Once you're in, here are some of the things you can do:

• Power up the system. Power it down. Force a hard reset. Force a power cycle. Force a phony overtemperature condition (in hopes of getting a clean OS shutdown.).
• Disable front panel controls (power off, reset, and standby buttons.) Yes, that's really in the protocol. See section 28.6 of the specification. Remote control can also lock out the keyboard and blank the screen.
• Set system boot options Or, what OS do we want to run today? These include useful tools like "bypass user password".

There's more. Much more. Basically, you can remotely take over the machine, turn it on, inventory the hardware, load an operating system, boot it up, and talk to it.

IPMI's back channel can do more than this. With some help from the operating system (and yes, it's supported in Windows) you can do more remote administration functions. This is great for administering your data center remotely. But it has darker implications.

Supposedly, most machines are shipped with IPMI mostly turned off, unavailable until a program is run on the machine to load in the keys that enable it. Supposedly.

Thus, all it takes for IPMI to be a "backdoor" is for a set of secret challenge/response keys to be preloaded into the IPMI chip. There's no way to read those keys. Short of taking the chip apart, gate by gate, there's no way to tell if there's a backdoor in there. Or a set of keys might be loaded by the system integrator before shipping the system. You can't tell. So that's where to put a backdoor, where no one can find it.

There's an open source, OpenIPMI, for sending IPMI commands on Sourceforge. Send "Presence pings" to the machines you have and see if they answer.