Slashdot Mirror

The CPU runs on 4 to 8 watts, but the chipset itself consumes up to 22 watts. Sorry, but that's too much.
See http://www.intel.com/design/chipsets/designex/307504.htm
No gigabit LAN is a bummer. I'll buy if they reduce power consumption to 10 watts.

Its an Intel board and they have that in (most of) their server boards : s3000ahlx. I have 3 of those. I use the boards to control my server room temperature (amongst other things) and nothing beats the easy programming of a RS232 port. Additionally I bought some Kouwell ParPort cards to do some other jobs.

Wait... you say that installing Vista on a piece of hardware specifically designed with Vista in mind...works? And that's supposed to be some kind of amazingly good thing? I thought that was just supposed to be how it works.

If you build hardware to support specific software, and vica versa, it will work! Tune in at 11 for the rest of this amazing story of triumph!

But they get someone to help pay the bills if they ship it with XP.

"I'd really like to see intel/amd move operations back to the states just for this reason"

You mean like this or this or their sites in

Folsom, CA
Santa Clara, CA
Hudson, MA
Rio Rancho, NM
Hillsboro, OR
Dupont, WA
Irvine, CA
Fort Collins, CO
Raleigh, NC
Parsippany, NJ
Columbia, SC
Austin, TX
Riverton, UT
Chantilly, VA

AMD uses Fabs in Germany, which is much friendlier to us than China. Ireland ditto, which has at least one Intel fab, and Israel, whose Intel facility you can thank for the Core Duo revolution and the death of NetBurst.

I live in Arizona, and there are plenty of fabs down here, not just Intel. Microchip and Freescale for instance.

And frankly, I prefer Taiwan and Japan as manufacturers to Singapore, Malaysia, or of course China. Inda I consider more neutral but genuinely friendlier than the Chinese.

If we don't move to build some manufacturing in the U.S. for PC components, we need to choose our strategic partners more carefully. Taiwan, South Korea, Japan are better choices than China or most of the of the Pacific Rim.

The article points out something hopeful to me - that DARPA is looking at this and preparing to prove to the military that they are in peril. Not a moment too soon. Consider counterfeit Cisco gear as a warning.

"I'd really like to see intel/amd move operations back to the states just for this reason"

You mean like this or this or their sites in

Folsom, CA
Santa Clara, CA
Hudson, MA
Rio Rancho, NM
Hillsboro, OR
Dupont, WA
Irvine, CA
Fort Collins, CO
Raleigh, NC
Parsippany, NJ
Columbia, SC
Austin, TX
Riverton, UT
Chantilly, VA

AMD uses Fabs in Germany, which is much friendlier to us than China. Ireland ditto, which has at least one Intel fab, and Israel, whose Intel facility you can thank for the Core Duo revolution and the death of NetBurst.

I live in Arizona, and there are plenty of fabs down here, not just Intel. Microchip and Freescale for instance.

And frankly, I prefer Taiwan and Japan as manufacturers to Singapore, Malaysia, or of course China. Inda I consider more neutral but genuinely friendlier than the Chinese.

If we don't move to build some manufacturing in the U.S. for PC components, we need to choose our strategic partners more carefully. Taiwan, South Korea, Japan are better choices than China or most of the of the Pacific Rim.

The article points out something hopeful to me - that DARPA is looking at this and preparing to prove to the military that they are in peril. Not a moment too soon. Consider counterfeit Cisco gear as a warning.

The thing about dominance on the desktop is that linux is going to win it by doing an end around and using Microsoft's strategy of incompatibility work against them.

Already in this thread there are the usual Microsoft shills whining about how Ubuntu doesn't yet work with their laptop's wireless card. What they don't see is that every time they point out a piece of hardware that doesn't work in Linux it's a win for Linux now. The device manufacturer will open up the interface for development or the line of product will die because people want choice.

When the Linux in your pocket and Linux in your handbag devices just sync better with Linux on your desktop and work seamlessly with Linux on the server then the game is over. The network is the platform.

Link

Well, when is the last time you used the word? You can argue anything is technology (as demonstrated by your fire and wheel example), and that is what businesses have done. And that is why when we hear the word we need to know more, because it holds no weight. It is redundant.

This article seems to claim that there are tech lovers and tech haters. Even those who work for the RIAA have their fancy cell phones and laptops and iPods. Even music labels are tech savvy in that they are all over iTunes.

Another word? How about remove it where it is redundant, and use "feature" where it refers to something. Try it.

"IT" is the only word I would defend, since it refers to a category of jobs. I would never say "information technology" though.

US Government regulations require that Intel publish performance numbers for all of their CPUs. See the following links for the relative performance of all of Intel's CPUs. Make your own graphs if you need a pretty picture.

Intel microprocessor export compliance metrics:

• Pentium Extreme Edition, D, M, M LV, M ULV, dual-core, 4, Mobile 4: http://www.intel.com/support/processors/sb/CS-028241.htm
• Core 2 Extreme, Core 2 Quad, Core 2 Duo, core 2 Duo LV, Core 2 Duo ULV, Core 2 Solo, Core Duo, Core Duo LV, Core Duo ULV, Core Solo, Core Solo ULV: http://www.intel.com/support/processors/sb/cs-023143.htm
• Itanium, Pentium III, Pentium II, Pentium II Xeon, Pentium II OverDrives, Pentium Pros, Pentiums with MMX technology, PentiumsIntel 486s, Intel 386s, i960s, i860s: http://www.intel.com/support/processors/sb/CS-020868.htm

US Government regulations require that Intel publish performance numbers for all of their CPUs. See the following links for the relative performance of all of Intel's CPUs. Make your own graphs if you need a pretty picture.

Intel microprocessor export compliance metrics:

• Pentium Extreme Edition, D, M, M LV, M ULV, dual-core, 4, Mobile 4: http://www.intel.com/support/processors/sb/CS-028241.htm
• Core 2 Extreme, Core 2 Quad, Core 2 Duo, core 2 Duo LV, Core 2 Duo ULV, Core 2 Solo, Core Duo, Core Duo LV, Core Duo ULV, Core Solo, Core Solo ULV: http://www.intel.com/support/processors/sb/cs-023143.htm
• Itanium, Pentium III, Pentium II, Pentium II Xeon, Pentium II OverDrives, Pentium Pros, Pentiums with MMX technology, PentiumsIntel 486s, Intel 386s, i960s, i860s: http://www.intel.com/support/processors/sb/CS-020868.htm

US Government regulations require that Intel publish performance numbers for all of their CPUs. See the following links for the relative performance of all of Intel's CPUs. Make your own graphs if you need a pretty picture.

Intel microprocessor export compliance metrics:

• Pentium Extreme Edition, D, M, M LV, M ULV, dual-core, 4, Mobile 4: http://www.intel.com/support/processors/sb/CS-028241.htm
• Core 2 Extreme, Core 2 Quad, Core 2 Duo, core 2 Duo LV, Core 2 Duo ULV, Core 2 Solo, Core Duo, Core Duo LV, Core Duo ULV, Core Solo, Core Solo ULV: http://www.intel.com/support/processors/sb/cs-023143.htm
• Itanium, Pentium III, Pentium II, Pentium II Xeon, Pentium II OverDrives, Pentium Pros, Pentiums with MMX technology, PentiumsIntel 486s, Intel 386s, i960s, i860s: http://www.intel.com/support/processors/sb/CS-020868.htm

My prediction is that Skype will not only become more popular but also more profitable. Their savior will come in the form of the new mobile computing platform. UMPC or MID + 3G/3.5G/4G/WiMAX + Skype.

Once battery life increases (atom) and mobile networks improve, techies will quickly adopt this platform as their primary phones but they'll still need to make and receive calls to others with PSTN phones.

Driver support for the Intel 945.
http://www.intel.com/support/chipsets/sb/CS-020683.htm
I found graphics drivers for linux for most flavors of the 945.

https://help.ubuntu.com/community/WifiDocs/Device/Broadcom_BCM4311_rev_01_(ndiswrapper)
You have to use ndiswrapper to get it working because Broadcom doesn't provide docs or drivers. Some distros seem to have a patch that does have this card working right now. I have heard Mepis is one of them.

Of course the original post holds true. Complain to the hardware manufacture. Intel has provided drivers for there hardware. Most distros have it included and Intel has it available on it's website.

Sorry, I overstated this, according to intel http://www.intel.com/pressroom/archive/releases/20080401comp.htm the cellular functionality is integrable, not integrated.
Excerpt "Intel Centrino Atom processor technology also enables manufacturers to integrate a range of wireless connectivity options, including Wi-Fi, WiMAX and cellular data"

Intel has deployed a high-k/metal gate device architecture at 45 nm.

The proprietary edge that Microsoft has in gaming is coming to an end. Raytracing is a better solution to this problem if you have the horsepower to drive it, and that's on the way. And it will be open.

You're quite right; the Commodore 64 had an 8-bit processor, but the chip had a 16-bit address bus (link)

The ability to do 64-bit processing is not a prerequisite for accessing large amounts of memory. The Intel Pentium III Xeon chips (32-bit), for example, could address up to 64GB of memory (link, 64GB = 2^36 bytes, so a 36-bit address space).

I believe that the memory controller is that component that determines how much memory can be addressed. There are, so far as I know, three "bit" numbers that are important with processors:

1) The registers; the size of the register determines the size of values that can be processed directly.
2) The addressable space; determined by the memory controller. (IIRC)
3) The data bus width; the number of bits per bus clock that can be moved into or out of the chip.

I'm hardly an expert on this stuff though. Anyone care to give a more detailed description?

...for Linux, Mac and Windows supporting multicore and also cluster architectures.
Obviously it would be better if these worked better and were easier to use, but many people are unaware of the tools that are available right now.

...for Linux, Mac and Windows supporting multicore and also cluster architectures.
Obviously it would be better if these worked better and were easier to use, but many people are unaware of the tools that are available right now.

True, the Linux kernel does make heavy use of GCC specific directives. However, interestingly, Intel's compiler has copied enough of the GCC directives that it's actually (sort of) possible to compile the Kernel with ICC, and apparently has been for several years. See here: http://blog.janik.cz/archives/2004/03/11/T23_04_41/ or here: http://www.intel.com/support/performancetools/c/linux/sb/cs-007713.htm

Its worth noting, however, that the Linux kernel as it stands simply doesn't work with any compiler other than GCC.

May have been true once, but it hasn't been for a long time. I recall seeing news about using Intel's compiler to build a Linux kernel years ago. More recently, Rob Landley's been doing some work with tinycc to get it up to snuff for kernel compiles, with the goal of generating a system that can "...completely rebuild itself, under itself, without any gnu code on the hard drive."

My dream machine at the moment would have something like the PWRficient, a 64-bit PPC that consumes 7 watts per core at 2 GHz.

Then again, the Intel Core (2) CPUs are not half bad, as long as you use the "mobile" versions. Consider these for example:

http://processorfinder.intel.com/details.aspx?sSpec=SLA98
http://processorfinder.intel.com/details.aspx?sSpec=SLA49

They are basically the "mobile" and "desktop" versions of the same processor (Core 2 Duo at 2 GHz, 2 MB cache, 800 MHz FSB), but the desktop version consumes twice the power of the mobile one. My question is, why are these dumb and wasteful "desktop" processors produced at all? The answer is probably that they are the Celerons of today, the cheaper versions that failed some tests.

It's a good thing there are Mini-ITX boards (like this one in my server/media center) for those who want to use "mobile" CPUs in a "desktop" environment.

My dream machine at the moment would have something like the PWRficient, a 64-bit PPC that consumes 7 watts per core at 2 GHz.

Then again, the Intel Core (2) CPUs are not half bad, as long as you use the "mobile" versions. Consider these for example:

http://processorfinder.intel.com/details.aspx?sSpec=SLA98
http://processorfinder.intel.com/details.aspx?sSpec=SLA49

They are basically the "mobile" and "desktop" versions of the same processor (Core 2 Duo at 2 GHz, 2 MB cache, 800 MHz FSB), but the desktop version consumes twice the power of the mobile one. My question is, why are these dumb and wasteful "desktop" processors produced at all? The answer is probably that they are the Celerons of today, the cheaper versions that failed some tests.

It's a good thing there are Mini-ITX boards (like this one in my server/media center) for those who want to use "mobile" CPUs in a "desktop" environment.

Intel has this article about the hardware needed to run at 50fps at 1920x1080p. They're claiming you need 8 cores. In a couple of years, that could well be within reach for most gamers.

There's also this John Carmack Interview. Carmack isn't too optimistic about ray tracing replacing rasterized graphics.

Look for an internship but don't look for one programming on a large project, they are rare. But companies always need people to test code, chips, etc. Usually those testing jobs involve writing code. There are utilities to aid in the testing of the product, code to automate tests that are done manually, etc. You will end up spending lots of time working with the engineers building the product, you can learn a lot from them. (Including whether this is really what you want to do for a living.)

Last year an intern on my previous project re-wrote a debug utility that one of the engineers had created, the tool had started small and grown. The intern rewrote the thing over the summer in C#, the updated version of the tool is still in use. He got to use his C# and Windows knowledge and ended up with a very good understanding of embedded programming and wireless communications.

Don't limit yourself to software companies. The hardware and silicon companies often have software teams and associated test organizations.

Try going to the job search web sites of the companies (e.g. http://www.intel.com/jobs/jobsearch/), usually they have an option to search for intern positions. It is a bit late in the year though, many of them are already filled.

There are several major issues:

The first is the size of the packaging of the chip - the actual silicon might only occupy the space a quarter the size of the whole unit. All that extra space is just used to manage the 500+ copper connections between the silicon and the rest of the circuit board.

The second problem is that as the clock speed of these connections becomes faster, synchronisation becomes a problem. While CPU's are running in the GHz frequencies, the system bus is still running in the hundreds of MHz.

If the chip could connect to the circuit board through optical connections, then all this could be simplified. You would eliminate the need for all the copper connections while simultaneously speeding up the external clock speed.

Where is Intel in all this?

The Intel® Wireless WiFi Link 4965AGN product is an embedded 802.11a/b/g/Draft N PCIe* Mini Card network adapter card that operates in both the 2.4GHz and 5.0GHz spectrum, delivering high throughput and a host of features that enhance today's mobile lifestyle. Deploying WLAN technology in your home and business increases productivity, efficiency and flexibility by enabling faster decision making, reducing down-time, and enhancing employee satisfaction. For more information visit our WLAN ROI and WLAN Deployment web pages. http://www.intel.com/network/connectivity/products/wireless/wireless_n/overview.htm

From TFA What was even surprising to us is that Intel FORTRAN is also free of charge ... I bet Intel are surprised too. Their compilers are not that free of charge. The people at the Philippine government's official weather service are hardly "not getting compensated in any form" http://www.intel.com/cd/software/products/asmo-na/eng/219771.htm

they could take a bite out of the microcontroller segment and ARM's small but tenacious market share.

You're impugning the credibility of Wikipedia as a way of dismissing anything that contradicts your argument, rather than dealing with the matter head on. That's intellectually dishonest, and a lazy, stupid way to argue.

Also, if you'd bothered to look at the article, you'd find that the quote provides a citation, and that citation points to a PDF file of the article in which Moore made the statement in question:

ftp://download.intel.com/museum/Moores_Law/Articles-Press_Releases/Gordon_Moore_1965_Article.pdf

In short, you lose on both style and substance.

With unit cost falling as the number of components per circuit rises, by 1975 economics may dictate squeezing as many as 65,000 components on a single silicon chip

Linux has this same bug. It's in "ohci1394.c". I reported this to the Linux kernel mailing list years ago, and the reaction of the kernel developers was to make it a "feature" for "remote debugging" that's enabled by default.

Technically, here's how it works. First, see the OHCI specification, section 5.15, "Physical Upper Bound register". This determines the highest memory address into which an external device can store directly by sending a packet. If set to zero, this feature is disabled. That feature is intended for slave devices, like peripherals. On computers with an operating system, it should be zero. It's not.

In the Linux kernel, that security hole was installed in "ohci1394.c" with the comment:
/* Turn on phys dma reception.
*
* TODO: Enable some sort of filtering management.
*/

In early kernels, it was unconditionally enabled. In 2.6, it's enabled by default, but can be turned off.

Also, This patch indicates that this security hole may have been designed into some FireWire controllers, so that the "upper bound register" didn't really do anything, but read back zero.

such as "desktop" CPUs. Why are these still being produced, when the "mobile" variants of the same models are much more efficient? For example, look at these two:

http://processorfinder.intel.com/details.aspx?sSpec=SLA98
http://processorfinder.intel.com/details.aspx?sSpec=SLA49

Both of them are Core 2 Duos, 65 nm process, 2 GHz, 2 MB cache. But one of them is a "desktop" model, and I wonder what the hell it's doing to waste almost double the power of the "mobile" one.

such as "desktop" CPUs. Why are these still being produced, when the "mobile" variants of the same models are much more efficient? For example, look at these two:

http://processorfinder.intel.com/details.aspx?sSpec=SLA98
http://processorfinder.intel.com/details.aspx?sSpec=SLA49

Both of them are Core 2 Duos, 65 nm process, 2 GHz, 2 MB cache. But one of them is a "desktop" model, and I wonder what the hell it's doing to waste almost double the power of the "mobile" one.

About the scalability of ray tracing vs. rasterization. ftp://download.intel.com/technology/itj/2005/volume09issue02/art01_ray_tracing/vol09_art01.pdf

1080P doesn't sound like a reduced-quality rip... though he also said 10GB - I'd have to know whether the original movie used all the available space on the disk.

In any case, decryption (in addition to the h.264 decoding) adds to the load on the CPU, which doesn't help. Also, even laptop DVD drives use several watts during playback.

Your transfer limit is either:

1) Lack of GigE switch between the machines, or a really really poor GigE switch (80 mbps - that sounds like Fast Ethernet)
2) Poor GigE chipset(s) on the mobo.
3) Poor CPU - rsync and scp encrypt the transfer which uses alot of cpu at each end. Also, GigE generates a fantastic number of interrupts/sec. Intel's Pro/1000 GT has an "Interrupt Moderation" feature to help with that.
4) Your hard drive speed. GigE is faster than most single hard drives. Seagate's Savvio 15k could maybe push it, with a STR of 117-97 MBps. A 10k Raptor is only 84 MB/s STR.

Run iPerf between two machines and see what your hardware can really do. (Quick tutorial, run `iperf -s` on one machine and `iperf -c {ip of other machine}` on the other. Then reverse the roles. FYI for a Lan, no IP Stack tuning is necessary)

Yes, trying to get what Gig-E promises has pissed me off too.

AMD has started open-sourcing their documentation which is great, but don't forget about the competition:

Intel has built some great tools and is becoming more open-source friendly all the time. Think PowerTop and their whole LessWatts stuff, think their linux drivers.

nVidia hasn't produced any open source drivers. However, their blob drivers have always worked like a charm for me. This may be somewhat of a clash between ideology and practicability here, but I prefer the nVidia/Linus/"DO STUFF" approach to the AMD/RMS/"Ideology is important" one. While not being completely open, nVidia graphics hardware has enjoyed great linux support for quite some time and that's what counts to me.

Let us understand something. There is a free (public domain) set of software that does this very thing. This library is basically a reimplementation of some the same APIs that optimizes it for the AMD processor. Note that Intel has a similar offering already called Intel Math Kernel Library. The APL has always been a free download for evaluation purposes. Now, companies like The Mathworks can use the library for free.

Intel has a technology called TXT that would clear the memory in situations like this before handing control over to the boot medium.

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

Who on earth needs a skillset like that?

Then, he will scoop up all of the unemployable engineers at slave labor wages, laughing the entire time while sitting atop his throne made of golden skulls. In fact, this article was probably penned while he sat atop said throne.

And he would have gotten away with it, too, if it weren't for you meddling kids.

I run the blogs over on Intel Software Network, and this has been a hot topic of, erm, "discussion" there. People are REALLY mad when they buy a new laptop that says "Vista Capable", and find out later that it really meant "Sorry, you can't run the Aero theme eye candy, DVD Maker, or Movie Maker, because your Intel 915 integrated graphics chip doesn't qualify for a WDDM driver." Somehow explaining that they should have bought a machine that was "Vista Premium Ready" doesn't make them feel better.

Seriously, between the two blog posts (one with video!) on the topic that I've done, there are over 800 comments (by FAR the most visited and commented on posts on the whole blog), most of them mad at Intel for not providing a WDDM driver for 915 graphics chipsets. Problem is, we can't. It doesn't meet the WDDM spec, which is controlled by Microsoft.

Here are the posts in question:

Video: Why Intel 915 graphics don't have a WDDM driver for Vista

Update on the 915 Graphics WDDM Vista Driver Issue

I'm actually relieved to see this news story come out, not that it makes me happy to point the finger at Microsoft (it doesn't), but to at least point all those angry blog commenters at a 3rd party source that sheds some light on the problem. I maintain my naive hope that it will educate and placate them all, and they'll stop emailing me and calling my cell phone. ;-)

I run the blogs over on Intel Software Network, and this has been a hot topic of, erm, "discussion" there. People are REALLY mad when they buy a new laptop that says "Vista Capable", and find out later that it really meant "Sorry, you can't run the Aero theme eye candy, DVD Maker, or Movie Maker, because your Intel 915 integrated graphics chip doesn't qualify for a WDDM driver." Somehow explaining that they should have bought a machine that was "Vista Premium Ready" doesn't make them feel better.

Seriously, between the two blog posts (one with video!) on the topic that I've done, there are over 800 comments (by FAR the most visited and commented on posts on the whole blog), most of them mad at Intel for not providing a WDDM driver for 915 graphics chipsets. Problem is, we can't. It doesn't meet the WDDM spec, which is controlled by Microsoft.

Here are the posts in question:

Video: Why Intel 915 graphics don't have a WDDM driver for Vista

Update on the 915 Graphics WDDM Vista Driver Issue

I'm actually relieved to see this news story come out, not that it makes me happy to point the finger at Microsoft (it doesn't), but to at least point all those angry blog commenters at a 3rd party source that sheds some light on the problem. I maintain my naive hope that it will educate and placate them all, and they'll stop emailing me and calling my cell phone. ;-)

Sure, it's easy to make a pipeline divide, but to pipeline 17-60 stages for an uncommon instruction is not usually done. It also would complicate pipeline control to have such a long pipeline delay.

Intel hasn't pipelined its divide, as can be seen in the instruction throughput numbers in Appendix C of this manual: http://www.intel.com/design/processor/manuals/248966.pdf

The throughput for divide shows some CPUs having two divide units, but none have pipelining. However, they could add an abort operation to kill a pending long-latency divide as well, but I don't know if they do that either. Since Intel has been bitten very badly by a divide bug, I could see them being a bit shy adding risky optimizations in this area.

They are already doing this, take a look to Intel® Trusted Execution Technology.

Then, no one will buy a new CPU. Intel and AMD aren't stupid. they know the consumer will run if they add this crap to thier products.

I hate to break it to you, but Intel think they've solved that problem.

They're starting to integrate the TCPA technology with something which business users will find genuinely useful and are likely to enable - vPro.

In brief, it's a mechanism to remotely manage PCs from a BIOS level upwards. But when you enable it (it's disabled by default), the TCPA functionality is enabled as well.

It doesn't take a great leap to see an OS which, on seeing TCPA functionality, uses it and cannot easily be persuaded not to use it. Right now, it's not available on everything - the cheapest PCs don't tend to support it. But that's filtering down rapidly, and all it requires after that is some bought legislation and/or technology licensing terms which state "where available, DRM functionality must be used".

The Mac Pro is a nice option IF your needs are not met by a single quad-core processor and 8GB of dual-channel 800MHz ECC DDR2 memory. Unfortunately (IMO), Apple doesn't offer a decent single-processor workstation option that doesn't use FB-DIMMs.

Other replies have suggested three good sites with "system guides" (Tech Report, Ars Technica, and Anandtech), but none of them seem to have specific guides for graphic designers (Photoshop, Illustrator, etc). I think they're still worth reading because their general tips are useful for (comparative) beginners and their guides are regularly updated to keep up with current hardware trends.

In addition to Apple's Mac Pro specs page, you could use Dell's Precision workstation specs pages as a "guide" on which hardware to choose when you build your Photoshop/Illustrator workstation. Dell actually has a "Find Your Recommended Workstation Configuration" page that suggests specific workstations based on application. Their Photoshop and Illustrator suggestions are the same: the "mainstream" Precision T3400 (based on Intel's X38 chipset) and the "advanced" T5400 (like the Mac Pro, based on Intel's 5400 chipset).

If you're buying based on the Intel platform, those are probably the best and most compatible chipsets to build your workstation around. They both can use ECC memory, workstation-class video cards, PCI Express 2.0, etc. The X38 chipset might be limited by its 8GB RAM ceiling. The 5400 chipset might be unattractive because it uses FB-DIMMs and dual Xeon CPUs (when a single Core 2 Quad might do). X38 has decent cheap RAID (0,1,5,10) built into the chipset as well as eSATA and HD audio. The 5400 chipset can use up to 64GB of RAM and usually has legacy PCI-X slots for those older, expensive cards you don't want to throw away.