Slashdot Mirror

Not all hardware is supported by all operating systems. Get used to it. I recently retired a Windows 2000 Pro installation. The biggest reason was ****Hardware Support****. I travel and do Power Point presentations and support others that do the same. As such often a presenter will have his presentation on a USB flash drive. Just for fun, take any modern collection of USB flash drives and plug them into a Windows 2000 laptop.. The endless search for an internet connection on the road to download drivers finally got to me. I Installed Dapper Drake and have been very happy since.

Do a little research and find a supported WiFi card. Most Intel cards are supported out of the box now so your Centrino Laptop should just work.
http://www.intel.com/support/chipsets/sb/CS-022095.htm

Many adaptors bought at Best Buy does work on Ubuntu just fine. Check your chipset first. The only way to get support for the Linux holdouts is make it a support, return and market share issue for them. If you have an incompatible card, call the company and ask for drivers. If they don't have drivers, ask for a refund. They will get the picture if it becomes a big enough issue for them.

By the way, I have an older Thinkpad T21 which doesn't have built in wireless. I use a D-link Air-Express PCMCIA card with a supported chipset. It works fine.

http://www.atheros.com/news/linux.html

Think about what you are saying... I would rather use a $200 OS instead of buying a $40 card to use a free OS. I fail to see the logic unless you just happened to already have a $200 OS just sitting unused someplace.

In the case of my laptop, it was buy something to replace Windows 2000, or buy a compatible card. Since the laptop didn't come with a card, I would need to buy one regardless. I just made sure it was Linux compatible when I bought it.

http://www.etheros.com/news/DLink2.html
http://madwifi.org/wiki/Compatibility

Gee...I guess that's why Intel's own page for the Xeon has big bold letters that say Server and Workstation Processors.

Intel has always marketed the Xeon as a high end CPU for server and workstation applications. If you're going to try to be pedantic, at least be accurate about it.

According to Intel, a laptop its harddisk plus DVD uses about 10% of the total power usage, so I'm having a hard time believing that your battery life is so much better thanks to the SSD storage. I'd more readily attribute the long battery life to a good bunch of batteries. Besides a 9 cell battery pack, the Dell D430 can pack an additional 6 cell as well.

What is interesting about the IGF competition is that most game developers enter mostly for fame and a lot less for cash prizes.
There's only one interesting money award (Seumas McNally Grand Prize - $20 000), the others ($2 500) don't stand a chance compared to some other competitions awards.

Our game (Penguins Arena - http://frogames.com/penguins_arena/) won the Intel Game Demo Contest this year (http://softwarecommunity.intel.com/articles/eng/1409.htm), the global level of competitors was lower than the level at IGF. We had a lot more chance to win $10 000 than we'll ever have to win any of the IGF prizes.
The real interest come from being selected as one of the IGF finalists, there is a lot of buzz around it.
No need to say that *both* (fame and cash) are really important for indies!

Good luck to all competitors!

DG9965WH Motherboard

This must be so new even intel does not now about it.

If you instead meant 'Intel® Desktop Board DG965WH" intel thinks it supports Linux.

Maybe the offered BIOS upgrade will do, just a wild guess.

CC.

Quoth the blogger: "With hundreds of languages and API's out there, is anyone really dumb enough to think "yet another one" will fix our parallel programming problems?"

Yet Intel touts its Threading Building Blocks library as just such a fix to many parallel programming problems. Now, TBB is a very nice product, and in many ways it is superior to a lot of existing libraries, APIs, and languages, but one gets the sense that maybe the left hand doesn't know what the right hand is doing at Intel.

I've am also actively involved with the OpenMP shared memory programming API. I've worked on all the OpenMP specifications "out there" today and I am CEO of the corporation that "owns" OpenMP (the OpenMP Architecture review Board).

Quoth the blogger: "With hundreds of languages and API's out there, is anyone really dumb enough to think "yet another one" will fix our parallel programming problems?"

Yet Intel touts its Threading Building Blocks library as just such a fix to many parallel programming problems. Now, TBB is a very nice product, and in many ways it is superior to a lot of existing libraries, APIs, and languages, but one gets the sense that maybe the left hand doesn't know what the right hand is doing at Intel.

I might also draw an analogy to the open source world, where there are often dozens of solutions to both simple/mundane problems (text editors, media players, command line shells, etc) and more complex ones (window managers, Linux distributions, etc). I wonder if the free and open source software world wouldn't also benefit from a "culling of the herd," so to speak.

http://www.intel.com/products/processor_number/chart/pentium_dual-core.htm E2140 65 nm 1MB L2 1.60 GHz 800 MHz FSB Dual core, enhanced speed step, 64bit, Execute disable support. In short it is a Core 2 Duo with only 1MB cache and not virtualization support in hardware.

http://www.intel.com/pressroom/kits/quickreffam.htm

I had to write a Perl script to count the number of processor types.

And here I thought WiMax was an Intel design. Anyways, it's not a,b,g, or n. It's 802.16.

I assume you're complaining about Intel documents such as this one that show a 1.6x to 3.8x speedup for certain HDTV encoding operations. There are still some likely reasons someone might pick SAD for encoding. One reason might be patent coverage-- the better the algorithm, the more likely it is to be patented these days. Providing a baseline fast algorithm that's part of all future Intel hardware and can be used without patent problems might be nice. Also, just because something can be done "faster in software" doesn't mean it's really faster. If it doesn't use the SIMD pipeline, it's occupying more of the int or fp pipelines and thus may have fewer potential parallelism opportunities and/or stall the pipeline more often. With SAD offloaded onto the SIMD pipeline one could do other things with the other pipelines-- perhaps even a combination of both algorithms running in parallel (assuming sufficient instruction dispatch speed) to get even more speedup.

I think there's a lot of as-yet unrealized potential here, and it will be interesting to see where it leads. It won't be the end-all be-all of encoding, but it will be another arrow in the quiver.

on modern Intel CPUs, the overwhelming majority are used to compensate for the fact that the CPU is ridiculously faster than the memory to which it interfaces. IOW, the cache. These CPUs consume between 80 and 120 watts of power. The reduced power versions use only 50.

By way of comparison, the 1 GHz AMD Geode runs on about 1 watt of power, and ARM processors can get by for even less.

By way of further comparison, a register to register transfer can be completed in 1/2 clock cycle. Contrast this with a read-modify-write memory cycle where a word is fetched from one memory location, modified, and written to a distant location, which will take 4 memory cycles (which typically runs at 1/2 or less of the clock speed of the CPU).

The power consumption problem is due more to the fact that compensating for this difference in speed requires a large SRAM cache on the die. And even then, it's not perfect - if you do things which routinely involve cache misses (such as video encoding, etc...) the CPU is stuck operating at the effective speed of the memory bus.

The key to reducing CPU power consumption is to use lower-latency memories, which require smaller on-die caches for a given performance level. We could double the throughput of DDR SDRAM by simply demultiplexing the address and data busses, similar to the way SRAM functions. There's no requirement in the underlying storage structure of DRAM to require separate row and column addresses; it's just a historical artifact. Originally, before DRAM and SDRAM became popular, computers were built with SRAM because its lower latency allowed even slow CPUs to work efficiently. But DRAM promised lower cost (via fewer bus lines) and lower power consumption (bits stored in charged capacitors, rather than cross-coupled transistors), at the price of latency, and the rest, unfortunately, is history.

This is the "tick" to be followed by next years "tock"... the same basic core in a new process (45nm, which allows more room for cache, etc.) followed by next years new core that leverages the smaller process (45nm). In other words it has a "familiar ring" because it is essentially the same core. In 2008 (late likely) we will get on board memory controller... which doesn't have such a "familiar ring".

http://www.intel.com/technology/magazine/computing/cadence-1006.htm

Wake me up when you're able to use PCC instead of GCC to do a 'make world'

I'm surprised nobody's mentioned tcc on this thread - it's lighter, faster, and close to a drop-in replacement for x86. The original poster doesn't seem aware of the wider context or they might have mentioned icc too. Even lcc is almost a drop-in, if far less optimising than gcc, and it supports more than just x86 targets.

Also check this out:

http://techresearch.intel.com/articles/Tera-Scale/1421.htm

Intel's "TeraScale" projects include a working prototype of a 80-core chip. Sure those are fairly simple cores, but they are packing decent FP oomph, and they have good branch handling and low latencies for data juggling (unlike DSPs and GPUs) -- quite the dream chip for implementing a physics engine?

They might eventually have in mind something similar to AMD's "special chips for CPU soccets" idea... Going multi-core with x86 while also shooting for more sockets on the mobo for special purposes (such as just running Havok).

This is a direct reference to 65nm vs. 45nm geometry. If AMD brings their quad core to a 45nm process, that should help yield, power and performance. If nothing else, it puts them on a level playing field with Intel (who already have product at 45nm) so that it's down to "design vs. design." Being stuck one silicon technology generation back, they need to resort to other tricks to "keep up."

In other words, to be at overall performance parity with Intel, they have to have a more advanced design in 65nm to keep up with Intel's 45nm work.

Another thing worth noting: By being 1 generation back, the quad core setup is a double whammy. The die area of a given chip roughly halves with each technology node. Not only is AMD putting twice as much on one chip, it's also making chips that are twice the size per transistor. (Remember, to double square area, you only increase your linear feature size by sqrt(2). 65/45 = 1.444... which is about sqrt(2).) Each additional sq mm of die area causes greater yield loss than the one before it (driven by defect density in the source silicon). Doubling die size has a huge impact on yield. So, AMD will potentially suffer significantly higher yield loss, and correspondingly higher costs. Even if it can keep its ASP (average selling price) up, the profit margins will suck.

It'll be interesting to see if AMD can quickly shrink this design to 45nm and get closer to parity. The benefits of the quad core design probably become much more apparent at 45nm.

Intel has nothing to lose by documenting all the instruction sets, architecture designs etc.

You mean like here or here???

They have such a big brand name - it doesn't really matter if their designs became public.

Now there you're wrong: Hasn't the competition between AMD and Intel convinced you that, at various times, one of them knew something about processor design that the other hadn't yet implemented?
A tech company giving up its core IP means giving up any edge, which translates to lower profits as competitors overtake the company.

Intel has nothing to lose by documenting all the instruction sets, architecture designs etc.

You mean like here or here???

They have such a big brand name - it doesn't really matter if their designs became public.

Now there you're wrong: Hasn't the competition between AMD and Intel convinced you that, at various times, one of them knew something about processor design that the other hadn't yet implemented?
A tech company giving up its core IP means giving up any edge, which translates to lower profits as competitors overtake the company.

Hmmm. This is awesome news. The last 40 or so systems we purchased were all Intel based purely because of the fact that they were so much less trouble due to being supported with Free drivers. This changes the equation though. It sounds from the announcement that we'll be getting better quality drivers because AMD/ATI will be releasing the full specs and not merely documenting through the use of code (which is cool and still makes Intel supportable).

Some things I still wonder about are whether or not the comparably priced AMD/ATI systems will have good Free drivers for other integral components such as wireless (which Intel have also got a lead with due to their IPW3945ABG). Intel have also got some very important work underway with PowerTOP. The upcoming Fedora 8 will be benefiting from the results of extensive testing with PowerTOP (which is written by ex-Red Hatter, now Intel employee, Arjan van de Ven). This allows monitoring of the major drains of power in laptops and can also be a major factor in server rooms.

I'm delighted by this whole move and it means that I can now make recommendations which include ATI cards as part of the specifications to purchasing. In terms of whether the AMD/ATI platform as whole will be a competitor that depends on whether the AMD motherboard chipsets will also be as open, Free and supportable. Intel have an incredible headstart in this area and possibly this will prevent them from moving into the stand-alone 3D card market (which is what I thought was going to inevitably happen). It looked as though AMD/ATI were headed for extinction, but I guess the reality of sales started to catch up with them.

All in all good news that opens up some more options for us. Perhaps we'll be seeing some interesting Dell machines soon!

Surely your toaster and coffee maker don't need 32nm? Wouldn't it be great to have all the parts that aren't CPUs down to 60nm without the costs associated with 60nm lithography?

Let's see, drive electronics, sound processors, Ethernet controllers (the ones that aren't on your southbridge), microcontrollers, any kind of embedded chip... There are lots of things that aren't 65 nm yet, or even at 90nm, and some chips aren't 130nm for that matter. Wouldn't it be great to get things that are currently larger down to CPU-ish feature sizes?

Hell, Intel's 82598 dual-port 10 gigabit Ethernet controller is 90nm. If they could make it 60nm on the cheap, that'd be great.

Just think of all the things these specialty chip designers (the ones with SSL accelerators, AES on-chip, vector processors, Forth chips, Java chips, etc) could do if they could get down to 60nm at or near FPGA prototyping prices. Hell, the Via C7 getting down from 90nm to 60nm would be great. Companies like Transmeta might bounce back into chip production. ARM9 is currently at 130 and XScale is 180. Getting those down to sizes that match Intel and AMD's current or even last-gen products at far less cost could give us really powerful handhelds.

Do not want!

No problem, Intel has motherboards for you too. I was specing out a quad core and noticed Intel has TPM and non-TPM versions of the same motherboard, for example the D975XBX2.

http://www.intel.com/design/motherbd/bx2/bx2_avail able.htm

The non-TPM version seems to have more features too, digital audio out, 8 SATA instead of 4, IEE1394/Firewire, 3 year warranty rather than 1 year.

Open Source Intel AMT Drivers and Tools. (the part that runs on the PC), Intel Active Management Technology Reference Design Kit (the part that runs on a server and remotely takes over the PC).

``The reason Linux became so successful is because of Intel's low-cost, standards-compliant, open-source hardware; but with initiatives like virtualisation, vPro, multi-threaded compilers etc. the balance gets tilted further in favour of TCPA and DRM partners;''

Err, I have no idea what you mean. Intel's hardware used to be standard-compliant and open-source? What standards? Which source? How does virtualization (and I do believe they published specs on how to use it) tilt the balance in favor of DRM? What do multi-threaded compilers have to do with anything?

Now to look at some other aspects, Intel hosts and supports a number of open-source projects, among them open source drivers for certain Intel graphics and WLAN cards. These are recent efforts, as well.

All in all, I don't think I can agree with your suggestion of Intel moving away from being supportive of open-source and towards being one of the forerunners of DRM.

I wonder how it compares to Silverthorne. According to this website http://www.intel.com/community/texas/spotlight.htm , it will consume 0.5W-2W.

How did this get modded informative?

Intel's Core Microarchitecture is not currently available in a quad-CPU platform.

Incorrect. Intel's "Core Microarchitecture" is marketed under the name "Core 2." The "Core 2 Quad" processors use the Core Microarchitecture. See Intel's product brief on the subject.

It is understandable the multithreaded performance would be poor, then.

The single threaded performance of quad core is similar to the single threaded performance of dual core, clock for clock. This should have tipped you off to the fact that quad core is using the new microarchitecture.

The current quad-cpu architecture is based on Tulsa, which a 65nm shrink of Paxville, which is essentially a Pentium 4 Smithfield, or two Prescotts shoved onto one ship.

Also incorrect. All of those processors use netburst. None of the new Quad Cores do. Your problem seems to be that you are reading dated information on server class chips and assuming all "Quad Cores" are server class. (Although new quad core Xeon's are Core as well)

Basically, it's two years ago's technology. The new Tigerton chip will be in Core based, however, it's not out yet.

Incorrect again. Although you are correct about Tigerton being a Core part.

In interests of making linux more accessible, more configuration utilities that don't require specific knowledge and in-errant editing of configuration text files.

I couldn't agree more. I hate having the find out how to edit a damn text file to share something on SMB (samba). But there's more it needs.

• Standard install method so all linux packages don't need to be repacked. .deb .rpm etc... pick one please
• Pick a WM so all apps work/look the same on all distros.
• "By developers, for developers" must stop. Even Ubuntu isn't ready for the average Joe.
• Focus on usability to make it usable by Joe.
• Hardware support. It still doesn't work with my Intel DG965WH and that's an old board.
• WINE, or Crossover office integrated for compatabilty for windows apps.

and here is the site for their research: http://techresearch.intel.com/articles/Tera-Scale/ 1419.htm

Oh I have seen crappy Nvidia GbE controllers generate more than 50,000 int/sec at around 700 Mbps...

Of course better NICs behave much better. For example the Linux e1000 driver for Intel GbE controllers configures their interrupt moderation mechanism to generate no more than about 9,000 int/sec by default when saturating a GbE link. This is the only reason why I mentionned 10,000 as the lower boundary. But yes, as you experienced it, it is of course possible to reduce this to 5,000 int/sec and probably even lower values. You don't necessarily want to do it though because 5,000 int/sec and lower means it would add, on average, 1/5000/2 = 0.1 ms to the network latency (which is about 0.1 ms on a typical GbE LAN, so it would double it).

http://www.intel.com/cd/software/products/asmo-na/ eng/compilers/279578.htm#sysreq

An ethernet switch can also revert to broadcast mode (hub-like functionality) if flooded with ARP requests, overflowing its internal ARP cache.

And, as far as checksumming goes, it's worth noting that packets are summed differently with Fragment Free and Cut Through switching technologies, which, it's atleast somewhat possible, they could be using.

Perhaps a better page here: http://www.intel.com/support/express/switches/sb/c s-014410.htm

I don't know what you're smoking, but I want some of it.

Let's start with some basic facts, that you can verify for yourself by hitting the long mode specs in AMD and Intel manuals:
1) You need PAE enabled (in CR4). Long mode uses a 4-level paging table scheme (PML4 - PDPT - PD - PT, although you can get away with only using the first three levels if you are fine with a 2MB granularity.
2) The linear address space is 64 bits.
3) The physical address space, ATM AFAIK, is 52 bits, with the other bits reserved for now. Going beyond 52-bits will likely need a PML5).
4) All registers are extended to 64-bit length, there are 8 new general purpose registers registers.
5) I am going to re-iterate - your address space is 64-bits. Your addressable memory is 2^64 - 1. Unlike PAE, where your linear address space is still 32 bits, you do not need an aperture within your linear 4GB to access physical addresses > 4GB.

I have no clue what the hell you talk about when you talk about "pointers", which are a software language concept. On EM64T/AMD64 you can perform direct and indirect MOVs to and from your entire linear (i.e. virtual) address space - and thus, through the "wonder" of paging (which you need enabled to enter long mode in the first place) - to and from your entire physical address space.

If you want a tiny piece of advice - instead of half-understanding mailing-list threads and articles written by people who know what they're talking about TO people who know what they're talking about - just hit the specs. They're free. Shit dude, if you acually bothered to try some 64-bit programming (even at the user, much less system, level) you would see that what you just wrote is just plain wrong.

Since this is Slashdot, I'll even give you links to the specs -
1) http://www.intel.com/products/processor/manuals/in dex.htm
2) http://www.amd.com/us-en/Processors/DevelopWithAMD /0,,30_2252_11467_11513,00.html

When I first saw this, I was thinking "oh, sh*t, they made another major API shift?!?," (like from 9 to 10 - a jump from 10 to 10.1 like that would orphan 10 owners) but that doesn't seem to be the case.

As far as I can tell, though, this is not really any different than in the past, except that MS is using a minor revision number instead of a major one - DX8.1 hardware is not forward compatible with DX9 any more than 10 is to 10.1. Forcing certain "minimum" support such as 4x AA isn't a big deal when all cards that I know of that support DX 10 already support 4xAA (or 8xAA) as well. The real problem with AA is when you mix in hardware HDR lighting which can cause jaggies, but nVidia and ATI support this already in all DX10 compatible cards (ATI since DX9 cards in the 1xxx series). This change may be to target Intel GMA sets, which Intel is attempting to get to being DX10 compatible by early 2008 by finally adding in HW T&L (the GMA 965 chipset supports HW T&L, but drivers are beta - see here)

OpenGL is in the midst of a move similar to DX9 to DX10 as they move from OpenGL 2.x to OpenGL 3. Long's Peak (OpenGL 3) is set to be ratified in August or early September and have a new rendering profile (once called "Lean and Mean") that is not backward compatible with older versions of OpenGL, but the older API will still be available. OpenGL 3.1(?) code-named Mt Evans is 3-5 months behind and will roll in features for modern GPUs (geometry shaders, texture arrays, etc). Originally, this was to be a "standalone" release and not include the backwards compatibility, but it now seems like they will deprecate the old API but still contain it, even with Mt Evans (gleaned from message boards, which often don't always know what's going on).

I have found that most games have issues with the integrated intel chip. There is a page somewhere on Intel's site that details what games work and what games don't...Here. Think it has something to do with no hardware vertex shader, but I could be mistaken.

Read 'em yourselves: EU's official statement and Intel's official response.

..of money for the software industry.

Ok, I'll give you that, but can you explain why this is the opposite trend in the PC hardware industry?

Case in point, in 1992 I ordered what was then a top of the line PC:

                    * 486 - 66 MHz / 8 Meg memory
                    * 240 Meg hard drive.
                    * No CDROM
                    * No sound card.
                    * No networking or modem.
                    * Diamond Stealth 64 video card (Vesa local bus)
                    * Cheap case with floppy.
                    * 14 inch VGA muti-sync monitor.
                    * Mouse/keyboard.
                    * $2,500.00

This past Sunday 7/26 I just purchased my new baby:

                    * Intel D975XBX2 (Bad AXE) mobo with a load of stuff on board.
                                http://www.intel.com/products/motherboard/D975XBX2 /index.htm
                    * Intel QX6700 (Core 2 quad core 2.66GHz)
                    * 850 Watt SLI power supply.
                    * EVGA GeForce 8800GTX 768Meg video card.
                    * 4 Gig DDR2 800 MHz RAM (PC2 6400).
                    * 320 Gig SATA 7200 RPM drive.
                    * Lian-Li PC1200B II case.
                    * 20X CD/DVD burner SATA.
                    * Windows Vista Ultimate.
                    * $2,800.00

Please tell me why inflation and rising development costs didn't have an effect on these prices?

As an engineer I can tell you that moving from the hardware technology of 1992 to 2007 was also "a lot of work" and required "a lot of resources" and "a lot of money". Yet based on inflation, I got a system that just crushed the older one into the ground.

Hmmm...

Look at the quality of hardware. A computer circa 1990 would be heavy, lots of metal, durable, and ussually higher quality in the materials and workmanship. Today Computers are flimsy, have the same or higher fail rate, and are made from elss material with poorer workmanship.

In general the same thing happening to Cars happen to Computers. They are cheaper because manufacturing advances have dropped the cost of manufacturing. They are also flimsier and generally less well made. They may contain better technology but the amoutn of effort and resources that go into one has been reduced thus the price has been reduced. A common consumer grade car has gone down in price after adjustment for inflation over it's corresponding version 40 years ago. Ditto with computers, although the rate of change is greater.

...of money for the software industry.

Ok, I'll give you that, but can you explain why this is the opposite trend in the PC hardware industry?

Case in point, in 1992 I ordered what was then a top of the line PC:

          * 486 - 66 MHz / 8 Meg memory
          * 240 Meg hard drive.
          * No CDROM
          * No sound card.
          * No networking or modem.
          * Diamond Stealth 64 video card (Vesa local bus)
          * Cheap case with floppy.
          * 14 inch VGA muti-sync monitor.
          * Mouse/keyboard.
          * $2,500.00

This past Sunday 7/26 I just purchased my new baby:

          * Intel D975XBX2 (Bad AXE) mobo with a load of stuff on board.
                http://www.intel.com/products/motherboard/D975XBX2 /index.htm
          * Intel QX6700 (Core 2 quad core 2.66GHz)
          * 850 Watt SLI power supply.
          * EVGA GeForce 8800GTX 768Meg video card.
          * 4 Gig DDR2 800 MHz RAM (PC2 6400).
          * 320 Gig SATA 7200 RPM drive.
          * Lian-Li PC1200B II case.
          * 20X CD/DVD burner SATA.
          * Windows Vista Ultimate.
          * $2,800.00

Please tell me why inflation and rising development costs didn't have an effect on these prices?

As an engineer I can tell you that moving from the hardware technology of 1992 to 2007 was also "a lot of work" and required "a lot of resources" and "a lot of money". Yet based on inflation, I got a system that just crushed the older one into the ground.

Hmmm...

Some of Intel's single and dual-port cards plug into 1x and 4x slots, respectively.

db

Yes, that is why Apple computers use EFI instead. Linux has had EFI support for a while, and Windows has it in some versions, although that page says Vista currently does not support it. According to that article, some x86 computers already ship with EFI using a BIOS legacy compatibility layer (including Macs for Boot Camp to work), and it links to an Intel page saying that they are in the process of switching over to EFI (once again with BIOS compatibility for now) for their motherboards. I suspect EFI will mostly replace BIOS on new hardware within a few years.

From the faq http://softwarecommunity.intel.com/tbbWiki/FAQ/591 .htm : FAQ
What compilers, operating systems and processors are supported?
Version 2, Changed by RANDY SMITH on 7/23/2007
Created by: JAMES REINDERS
The project is dedicated to supporting all compilers, all OSes and all processors as a cornerstone objective of the project. Up to date information on status is available on the web site.

Is this a limitation of the program itself or is it just a limitation of what Intel is willing to validate it on. I mean, I can understand Intel not testing other people's processors and only making a claim to thier current crop that is available.

If there is not technical limitation to the use on other processors and Intel just didn't warrant or claim it works on them, then it might work well with them, you just need a way to find out for sure. I ma guessing this might lead to a designed and tested for Intel processors logo or something.

What is GPL v2 with the runtime exception?
Version 1, Changed by RANDY SMITH on 7/21/2007
Created by: RANDY SMITH
GPLv2 with the runtime exception is the license under which the source code of libstdc++ is distributed (see gcc.gnu.org/onlinedocs/libstdc++/17_intro/license. html). This 'runtime exception' is therefore a standard for distributing template libraries - and that is why TBB uses it.

This blogpost talks about OpenMP vs TBB: Threading Building Blocks: Solution Looking for a Problem?

I know this comes as a great surprise, but the OSes and processors this runs on are limited. If you want your programs to run on non-Intel platforms, or on any of the BSDs, I suggest you skip it and use something else.

Processors:

• Intel® Pentium® 4 processor
• Intel® Xeon® processor
• Intel Pentium D processor
• 64-bit Intel Xeon processor
• Intel® Core Solo processor
• Intel Core Duo processor
• Intel Core 2 Duo processor
• Intel® Itanium® 2 processor (Linux systems only)
• Non Intel processors compatible with the above processor

OSes:

• Microsoft Windows Systems
  • Microsoft Windows XP Professional
  • Microsoft Windows Server 2003
  • Microsoft Windows Vista
• Linux Systems
  • Red Hat Enterprise Linux 3, 4 and 5 (when using Red Hat Enterprise Linux 4 with Intel Itanium processors, operating system Update 2 or higher is recommended)
  • Red Hat Fedora Core 4, 5 and 6 (not with Intel Itanium processors)
  • Asianux 2.0
  • Red Flag DC Server 5.0
  • Haansoft Linux Server 2006
  • Miracle Linux v4.0
  • SuSE Linux Enterprise Server (SLES) 9 and 10
  • SGI Propack 4.0 (with Intel Itanium processors only)
  • SGI Propack 5.0 (not with IA-32 architecture processors)
  • Mandriva/Mandrake Linux 10.1.06 (not with Intel Itanium processors)
  • Turbolinux GreatTurbo Enterprise Server 10 SP1 (not with Intel Itanium processors)
• Mac OS X 10.4.4 (Intel) or higher

Compilers:

• Microsoft Visual C++ 7.1 (Microsoft Visual Studio .NET 2003, Windows systems only)
• Microsoft Visual C++ 8.0 (Microsoft Visual Studio 2005, Windows systems only)
• Intel® C++ Compiler 9.0 or higher (Windows and Linux systems)
• Intel® C++ Compiler 9.1 or higher (Mac OS systems)
• For each supported Linux operating system, the standard gcc version provided with that operating system is supported, including: 3.2, 3.3, 3.4, 4.0, 4.1
• For each supported Mac OS operating system, the standard gcc version provided with that operating system is supported, including: 4.0.1 (Xcode tool suite 2.2.1 or higher)

P.S. Slashdot pulled out all the trademark symbols, and doesn't support the sup tag, so you'll just have to picture them in all the appropriate spots. :P

I see absolutely no need to continue developing the desktop (workstations are another story). Here is what is out for the laptop.Core Extreme from Intel. 250 GB hard drives from various vendors. Great graphics solution from Nvidia and AMD. Notebooks can be fitted with up to 4 GB of ram. Notebooks utilized power more efficiently which is important in this energy conscience age. I personally have no need for desktop and will never buy one again. A laptop combined with a good display is enough. It is time for the OEM to get clever and start experimenting on new designs built around mobile components. Imac is a good example of a destop built from laptop components. Maybe special docking station for the hardcore users who want more expansion that the express card slot. A eSATA for notebooks perhaps for extra hard drives.

Yes and no,

Some of the bugs will be fixed, others won't. Every CPU has bugs, it's just a fact of life. These things are designed by humans, it's just going to happen. CPU errata happens with Intel (This is the Core2 link) and AMD. None of this is a major threat to most users, and they get worked around by most people pretty quickly. Microsoft have released fixes for the Core2 issue, as have Apple. I don't know whether there has been an update to the kernel for these yet, but I am sure they would get back ported by your distribution.

There is a note here and here regarding the Core 2 bugs, I think one of these might have even become a slashdot article at one point. The two links here both are referring to Linus' comment of it being "Totally insignificant", which given that he worked for Transmeta and knows a lot more about how the industry works, I would be putting a bit of faith in his statement.

As another poster said, keep up to date on your BIOS revs, as CPU microcode does have fixes for this stuff too.

Berny

The current device that the OLPC has created is neither the last nor possibly the only unit that the project might support. If you look at roadmaps, esp since Intel is 18 months ahead of AMD on manufacturing process technology, it will have much more powerful cpus than the dated old National Semi Geode technology that AMD bought - in a similar or lower power profile. Check out Silverthorne to see what's coming.
Negroponte was probably right to use Geode when the the first OLPC unit was designed, but looking into the future, Intel cpus will make much more sense for this.
In any case, there's no reason Intel can't create a different model for other markets, but with full embrace of OLPC by Intel, the project is actually much more viable.

I don't know Apple's motivation, but you know, there are compilers that can generate better X86 code than GCC.

GCC is quite portable and generates very good code on many architectures, especially X86, but often chip vendors ship highly optimised compilers for their own processors, even when they fund GCC as well.

Also, Microsoft has some special internal compilers that they don't ship (sometimes they take a very long time to compile, but generate very tight code. I don't know if they use any of them for shipping products but I wouldn't be surprised. These wouldn't really be productisable, but are appropriate for this sort of in-house use.

I only use GCC myself, because it's really good, but I don't kid myself that it's the best in every situation.

My only hope is that Intel doesn't skew it's architecture so much that it becomes incompatible and that AMD is left behind. Would be nice if AMD could partner up as well, or create a consortium for "next gen architecture and virtualization enhancements" kinda like how MMX, SSE etc came about for graphics.

Too late - it's already happened. Intel and AMD have incompatible virtualisation technologies. Intel's is called VT with various sub-designations such as VT-d for virtualising DMA. AMD's is called AMD-V and is completely different. AMD have sub-divisions too, such as support for Nested Page Tables which Intel are still developing.

Xen supports both. Not certain about VMWare, but I'd be surprised if they didn't support both too. One interesting fact is that hardware virtualisation isn't faster than software approaches like VMWare's emulation or Xen paravirtualisation. Although this will probably change in future (and also Xen paravirt is no good for you if you want to run Windoze or other binary-only OSes).

Rich.

You mean a compiler like this?

http://www.intel.com/cd/software/products/asmo-na/ eng/266992.htm