Slashdot Mirror

Since Intel is invested in Open Source, this makes sense.

http://software.intel.com/en-us/oss

http://newsroom.intel.com/community/intel_newsroom/blog/2013/04/16/intel-reports-first-quarter-revenue-of-126-billion

Q1 Key Financial Information and Business Unit Trends
PC Client Group revenue of $8.0 billion, down 6.6 percent sequentially and down 6.0 percent year-over-year.
Data Center Group revenue of $2.6 billion, down 6.9 percent sequentially and up 7.5 percent year-over-year.
Other Intel® Architecture Group revenue of $1.0 billion, down 3.9 percent sequentially and down 9.0 percent year-over-year.
Gross margin of 56 percent, down 2 percentage points sequentially and down 8 percentage points year-over-year.

is not the "vast majority" of computer users. The PC gamer segment is estimated at 100-300 million individuals worldwide. Sure, that's not nothing. But in comparison to estimates for total worldwide PC use (workstations, desktops, tablets) approaching 2 billion, it's not a big percentage.

There will always be some hardcore gamers. But the enthusiasts with the 1,000 watt+ power supplies and SLI GPUs are not the "vast majority of lightweights" about which we're talking. Angry Birds has 1.7 billion downloads. Show me a PC game that does that volume.

My 8 core does 0 to 100% loading websites in 5ms, my old 6 core took 7.5ms. My friend has a 4 core with turbo boost that does 0 to 100 in 4.3.

Actually running high end games on a Xeon isn't that bad of an idea these days. For some bizarre reason they are always cheaper than equivalent i7 CPUs.

eg, the E3-1275 has an tray price of of $339. It runs faster both in turbo and non-turbo mode than the i7-3840QM which has a tray price of $568.

The E3-1225 at $215 is faster in both turbo and non-turbo speeds than the i7-3610QM at $378.
The E3-1245 at $266 is faster in both turbo and non-turbo speeds than the i7-3820QM at $568.

All CPUs listed are pretty much the same. Same architecture, same generation, same internal graphics card (the P4000 is the HD4000). Yet for some reason there's a huge discount if you buy Xeon rather than i7. There's not that much difference in the motherboard prices either. And the ECC RAM is optional, not required. So bizarrely these days it's worth buying Xeon if you want a huge discount on the desktop CPU prices.

Actually running high end games on a Xeon isn't that bad of an idea these days. For some bizarre reason they are always cheaper than equivalent i7 CPUs.

eg, the E3-1275 has an tray price of of $339. It runs faster both in turbo and non-turbo mode than the i7-3840QM which has a tray price of $568.

The E3-1225 at $215 is faster in both turbo and non-turbo speeds than the i7-3610QM at $378.
The E3-1245 at $266 is faster in both turbo and non-turbo speeds than the i7-3820QM at $568.

All CPUs listed are pretty much the same. Same architecture, same generation, same internal graphics card (the P4000 is the HD4000). Yet for some reason there's a huge discount if you buy Xeon rather than i7. There's not that much difference in the motherboard prices either. And the ECC RAM is optional, not required. So bizarrely these days it's worth buying Xeon if you want a huge discount on the desktop CPU prices.

Actually running high end games on a Xeon isn't that bad of an idea these days. For some bizarre reason they are always cheaper than equivalent i7 CPUs.

eg, the E3-1275 has an tray price of of $339. It runs faster both in turbo and non-turbo mode than the i7-3840QM which has a tray price of $568.

The E3-1225 at $215 is faster in both turbo and non-turbo speeds than the i7-3610QM at $378.
The E3-1245 at $266 is faster in both turbo and non-turbo speeds than the i7-3820QM at $568.

All CPUs listed are pretty much the same. Same architecture, same generation, same internal graphics card (the P4000 is the HD4000). Yet for some reason there's a huge discount if you buy Xeon rather than i7. There's not that much difference in the motherboard prices either. And the ECC RAM is optional, not required. So bizarrely these days it's worth buying Xeon if you want a huge discount on the desktop CPU prices.

Actually running high end games on a Xeon isn't that bad of an idea these days. For some bizarre reason they are always cheaper than equivalent i7 CPUs.

eg, the E3-1275 has an tray price of of $339. It runs faster both in turbo and non-turbo mode than the i7-3840QM which has a tray price of $568.

The E3-1225 at $215 is faster in both turbo and non-turbo speeds than the i7-3610QM at $378.
The E3-1245 at $266 is faster in both turbo and non-turbo speeds than the i7-3820QM at $568.

All CPUs listed are pretty much the same. Same architecture, same generation, same internal graphics card (the P4000 is the HD4000). Yet for some reason there's a huge discount if you buy Xeon rather than i7. There's not that much difference in the motherboard prices either. And the ECC RAM is optional, not required. So bizarrely these days it's worth buying Xeon if you want a huge discount on the desktop CPU prices.

Actually running high end games on a Xeon isn't that bad of an idea these days. For some bizarre reason they are always cheaper than equivalent i7 CPUs.

eg, the E3-1275 has an tray price of of $339. It runs faster both in turbo and non-turbo mode than the i7-3840QM which has a tray price of $568.

The E3-1225 at $215 is faster in both turbo and non-turbo speeds than the i7-3610QM at $378.
The E3-1245 at $266 is faster in both turbo and non-turbo speeds than the i7-3820QM at $568.

All CPUs listed are pretty much the same. Same architecture, same generation, same internal graphics card (the P4000 is the HD4000). Yet for some reason there's a huge discount if you buy Xeon rather than i7. There's not that much difference in the motherboard prices either. And the ECC RAM is optional, not required. So bizarrely these days it's worth buying Xeon if you want a huge discount on the desktop CPU prices.

Actually running high end games on a Xeon isn't that bad of an idea these days. For some bizarre reason they are always cheaper than equivalent i7 CPUs.

eg, the E3-1275 has an tray price of of $339. It runs faster both in turbo and non-turbo mode than the i7-3840QM which has a tray price of $568.

The E3-1225 at $215 is faster in both turbo and non-turbo speeds than the i7-3610QM at $378.
The E3-1245 at $266 is faster in both turbo and non-turbo speeds than the i7-3820QM at $568.

All CPUs listed are pretty much the same. Same architecture, same generation, same internal graphics card (the P4000 is the HD4000). Yet for some reason there's a huge discount if you buy Xeon rather than i7. There's not that much difference in the motherboard prices either. And the ECC RAM is optional, not required. So bizarrely these days it's worth buying Xeon if you want a huge discount on the desktop CPU prices.

Hey, I want a pair of these new CPU chips from Intel with 10 cores each and 30M of cache in my next PC. These CPU's will now directly access a measly 12TB of ram. Heck, they're only $4616 each. Why not use them to run Far Cry 3 real fast? ;-)

http://ark.intel.com/products/53580/Intel-Xeon-Processor-E7-8870-30M-Cache-2_40-GHz-6_40-GTs-Intel-QPI

using pci-e 3.0 on the Qm77 chipset for Thunderbolt will cut video down to X8 3.0 and some boards may use switchers to get full use out of the pci-e lanes.

http://www.intel.com/content/www/us/en/chipsets/performance-chipsets/mobile-chipset-qm77.html

but only 2 lanes on the TH side still makes it useless for video cards.

Thunderbolt (codenamed Light Peak)[1] is a hardware interface that allows for the connection of external peripherals to a computer. It uses the same connector as Mini DisplayPort (MDP). It was released in its finished state on February 24, 2011.[2]

On the same day, Apple released new iMacs with Thunderbolt.

Now what are your facts?

Intel begs to differ:

http://ark.intel.com/products/codename/29890/Clarkdale

Or, use out-of-band management built into most business hardware of the last 5+ years to completely power off systems, and remotely power them up on an as-needed basis for management tasks such as backup and deployment.

Intel AMT: http://www.intel.com/content/www/us/en/architecture-and-technology/intel-active-management-technology.html
AMD DASH: http://www.amd.com/us/products/technologies/systems-management/Pages/manageability-desktops-notebooks.aspx

Power off PCs at set time via group policy, wake them up with a direct HTTP packet, shut them down afterward with whatever task they are performing. Save money now without the get-rich-quick bullshit.

ICC isn't bad. It's not open source, but you can obtain it gratis for non-commercial use.

Except that it didn't.

You'd think that, on Slashdot, people would try to know what they're talking about, before they make idiots of themselves, but I guess not.

http://download.intel.com/design/pentium/datashts/24199710.pdf

An important component of the Intel Xeon Phi coprocessor’s core is its vector processing unit (VPU), shown in Figure 5. The VPU features a novel 512-bit SIMD instruction set, officially known as Intel® Initial Many Core Instructions (Intel® IMCI). Thus, the VPU can execute 16 single-precision (SP) or 8 double-precision (DP) operations per cycle. The VPU also supports Fused Multiply-Add (FMA) instructions and hence can execute 32 SP or 16 DP floating point operations per cycle. It also provides support for integers.

Because why would you support their dishonesty when they pull shenanigans like intentionally crippling run-time performance of their code when run on non Intel hardware??
http://www.agner.org/optimize/blog/read.php?i=49#49

And of course they put the disclaimer as a "gif" so text engines won't find it.
http://software.intel.com/en-us/articles/optimization-notice/#opt-en

When Intel learns to respect their customers then I'll respect and support them.

http://ark.intel.com/ is very helpful in this regard.

Intel has been working on it.

The performance claim in the summary seems to come from page 15 of this presentation, where the speedup for a 1TB sort (presumably distributed) is 4 hours -> 7 minutes. I can't find the details for that test, but most of the speedup comes from using better hardware - faster CPU and network adapter, and SSDs instead of HDDs - while they get a 40% speedup from using their Hadoop distribution over some other Hadoop distribution, which is a fairly modest gain.

The biggest performance benefit of Spark comes from avoiding disk and network access, so improving those bottlenecks will presumably reduce Spark's lead over Hadoop somewhat. But it's hard to say how well Spark would do with this particular hardware and test setup. I would guess it's still much faster than their Hadoop distribution. (Note: I'm a Spark power user but not an expert in its performance.)

Isn't it normal for any processor to have errata? There are currently 95 bugs listed for Ivy Bridge on Intel's site. There are 120 for Sandy Bridge.

Isn't it normal for any processor to have errata? There are currently 95 bugs listed for Ivy Bridge on Intel's site. There are 120 for Sandy Bridge.

Confusingly, 32-bit x86 binaries are often packaged with a "i386" suffix. These should still be supported. Only support for the ancient 32-bit Intel CPUs before the Pentium era of the mid-90's (remember the floating point bug?) were dropped. Anyone who still has one of those should call Steve at Dell.... oops, I guess he's been dropped too. Sorry.

Communication is important, not just marketing. Many problems occur when not everyone in a company understands the issues.

You can get some free help from our web site: Futurepower.

Major points:

No one else can know your products as well as people inside your company. An outsider with experience can help, but you must guide the work. Yes, that is difficult and time-consuming, but it is necessary.

Don't be overly influenced by marketing for consumer products. Almost all consumer product marketing contains some dishonesty. Some marketing people are financially successful abusing the customer, but there are always drawbacks. For example, after 8 years of suffering from cancer, Steve Jobs died; I'm sure you will consider that a drawback.

There are many, many people who think they get away with their own methods of being abusive. It is a reasonable theory that they don't.

Every writer needs an editor. Truly professional writers want as many people as possible to search for mistakes and insufficiencies in what they have written.

Deal creatively with the tough conflicts. Resolve conflicts, don't avoid them. Often advertising that has helped readers understand conflicts has been extremely successful.

Everything is marketing. As it says on our web site, "Every public activity of a company helps form the public view of the company and is therefore advertising."

See other issues on our web site: For example, "Give one thing one name." Don't be like Intel and give one thing, a storage driver, 5 names.

You should name the companies.

Intel's ingenious self-bricking 320-series SSDs would probably qualify for one of them.

Thank you for the info. Bad day or not, rest assured that I understand perfectly how frustrating the situation with the drivers must be.

The anecdotes from places like Coding Horror are just that: anecdotes. Were early SSD failure rates higher up to 2011 than regular drives? I think they've gotten better as years pass. What about now though? Even the 2011 survey from Tom's Hardware already put SSD reliability as already higher than regular drives.

I've had plenty of spinning drives that didn't last more than a hundred days too. Hard drive controllers fail with no warning, just like SSD ones do. I think this is emphasized as more associated with SSD failures because it's the only way SSDs die.

In the middle of 2011 Intel raised warranties to 5 years on the main SSD I use in my systems. In late 2011 Seagate dropped warranties to a year. If you don't care about high capacity, it's possible for a SSD to cost less per year than a mechanical drive now. That's not a glowing statement about the manufacturers thinking SSD is more likely to fail either.

There are several smartphones with x86 CPUs in them.

There's still a need for the "gamer" PC, and that niche will continue to exist.

But for most of us, there are better alternatives. I just bought one of the Intel Next Unit of Computing systems a couple of days ago. I'm thrilled with it so far. It's totally quiet, mounts discretely on the side of my desk, supports two monitors, and is plenty fast enough for my software development needs.

I don't develop games, but I imagine that most users will be playing games on tablet-like devices in the near future, so a system like mine probably has more gaming horsepower than the average tablet as well.

For Intel chips, the only mainstream option is the Intel S1200KP. Supermicro has a few ITX boards that support ECC but they are horribly expensive as they come with built-in Mobile CPUs. Supermicro also has a new Atom board that supports ECC, but it only goes up to 8GB. (what's the point?)

Those "old" links are dated Nov 2012. If you have something more recent suggest you offer it, because you don't give any links at all. I can't see a single reference anywhere on the web that GMA 3650 supports anything other than DirectX 9. The links referred to the N2600. You made up the N450 stuff. Even Intel's own web site says it only supports DirectX 9. http://ark.intel.com/products/36331/Intel-Atom-Processor-N270-512K-Cache-1_60-GHz-533-MHz-FSB#infosectiongraphicsspecifications.

Here's a review of a Z2670 tablet which you claim runs DirectX. "One of the limitations of the W510's Intel Graphics Media Accelerator GPU is that it's not DirectX 11 compatible, so our standard 3DMark11 benchmark wouldn't run. You can forget about playing "World of Warcraft," too. Even when effects were set to low, the W510 averaged just 12 fps, and even hung up during our test flights." 'Quite Fast' my ass. http://www.laptopmag.com/acer-iconia-w510.aspx Posted Dec 28.

The spec sheets for Z2760 tablets I googled either says DirectX 9. http://www.tipidpc.com/viewtopic.php?tid=279073 http://www.pinoytechblog.com/archives/acer-iconia-w510-the-windows-8-tablet-netbook-hybrid

Or in the case of Dell doesn't give the version at all.http://www.dell.com/uk/enterprise/p/latitude-10-tablet/fs

No wonder you're posting as AC. Google Moar.

Intel really needs to get its act together: It's Atom processors are a decent low power x86 solution, but as usual Intel has delivered them with a crappy 3D graphics to the point the graphical benchmarks can't even run on them, let alone any recent computer games. For the Atom Cedar Trail release they didn't even do DX10 drivers, and sheepishly back-speced it to the now outdated DX9. ARM tablets can deliver decent 3D, so why can't Intel? Even AMD can provide 3D graphics for low-power PCs. Why can't Intel? And Intel wonders why it's becoming irrelevant to the future of computing!?

No DX10 for you!
http://semiaccurate.com/2012/01/03/intel-thinks-cedar-trail-is-a-dog-reading-between-bullet-points/#.UOY58uRJNxA

Windows must live with DX9. Linux can't do anything at all...
http://tanguy.ortolo.eu/blog/article56/beware-newest-intel-atom

Oh and did I mention it doesn't work on Windows 8.
http://communities.intel.com/message/175674
http://www.eightforums.com/hardware-drivers/12305-intel-gma-3600-3650-windows-8-driver.html
http://answers.microsoft.com/en-us/windows/forum/windows_8-hardware/windows-8-on-intel-atom-d2700dc-graphics-driver/2a6015d3-af92-453d-b0c2-20cc56b764de

So which is it, can you make me a LGA1155 socket motherboard or can't you? Or did you mean "any CPU you want, as long as it's an ancient and outdated one with open specs"?

I can make any motherboard, with LGA1155 or any other socket - or with direct attachment of a CPU that is packaged as a BGA. Why not? It's not rocket science. The pin grid is 0.91 mm, which is pretty generous today. My last BGA design involved a part with a 0.5 mm pitch; that was expensive. You may want to have Intel's reference designs, but they are obtainable today, and I have some for Atom (because that's what I need.) The DDRx routing will have to be carefully done, but that's also not an impossible task. I built 20A, 0.9V polyphase power supplies before, for a PowerPC project. There is hardly anything else that is notable.

But super-fast and super-hot motherboards of this kind are not what the digital rebel needs, IMO. He needs a small, lightweight, portable system - a tablet would be ideal, especially if it accepts external attachments like the monitor and USB. In reality all modern tablets are already suitable for the task. Communication, not data crunching, is the primary use of computers today - and any low-power system can do it just as well as a hot desktop.

Another reason for a digital rebel to not depend on Intel is that Intel can be asked (or forced) to make sure that their CPUs don't even start until they authenticate with the BIOS. You can build such a system already. For example, the CPU will refuse to access most of its address space until it issues a challenge to the BIOS (or TPM) and receives a correct response. The pre-auth mode would be just good enough to boot up, but if you need to run an OS you need the CPU unlocked. The private key to the CPU is in the mask, and the chances of getting to it are nearly zero.

In this situation it is essential to have an entirely free CPU design that is not constrained by artificial barriers. There are already lots of good CPUs that are ready for an FPGA. If there is a need, a SoC can be synthesized from existing RTL components and then manufactured as an ASIC. If that is illegal, use FPGA and program your own bitstream. Either way, computers are here to stay, and the only way to restrict access to them is not technical but social (like public beheading of underground engineers.)

Have you tried asking intel if there are any resellers in your country?
http://locate.intel.com/
https://www-ssl.intel.com/p/en_UK/techprovider/contentns?docid=483949#

Have you tried asking intel if there are any resellers in your country?
http://locate.intel.com/
https://www-ssl.intel.com/p/en_UK/techprovider/contentns?docid=483949#

This. In my, admittedly limited, hardware building experience intel boards are not only very open source friendly, but come with the best documentation and least-flaky BIOSes. And of course their SATA controllers and NICs are class leaders.

I've built a few NAS/NAS + HTPC mITX units using two of Intel's newer boards, running either debian stable or testing, and found both of these to be excellently supported OotB:
http://www.intel.com/content/www/us/en/motherboards/desktop-motherboards/desktop-board-dh77df.html.html
http://www.intel.com/content/www/us/en/motherboards/server-motherboards/server-board-s1200kp.html

And that's even with my beloved M1015 HBA's, which I've had problems with on some older AMD boards (almost certainly down to poor testing on LSI's side).

I've not done any testing of features like suspend/hibernate since these are targeting "low power always on", and the onboard graphics don't quite match up to 23.97fps at times*. My only regret about choosing the "server" S1200KP board for my NAS is that it doesn't come with mSATA or eSATA ports.

In the mITX arena, almost everyone else seems to use the ubiquitous (and ubiquitously crap) Realtek chips, so it's worth buying Intel for that reason alone. Recently Asus have made a point of using Intel NICs in their higher-end mATX and ATX boards but for small linux builds intel is the only game in town. My two pence.

* If you're really set on a good HTPC experience with e.g. linux+XBMC I would still recommend an ASRock Vision 3D with an nVidia card in it. They're ludicrously overpriced but tiny and silent and the SB models at least work perfectly with linux.

This. In my, admittedly limited, hardware building experience intel boards are not only very open source friendly, but come with the best documentation and least-flaky BIOSes. And of course their SATA controllers and NICs are class leaders.

I've built a few NAS/NAS + HTPC mITX units using two of Intel's newer boards, running either debian stable or testing, and found both of these to be excellently supported OotB:
http://www.intel.com/content/www/us/en/motherboards/desktop-motherboards/desktop-board-dh77df.html.html
http://www.intel.com/content/www/us/en/motherboards/server-motherboards/server-board-s1200kp.html

And that's even with my beloved M1015 HBA's, which I've had problems with on some older AMD boards (almost certainly down to poor testing on LSI's side).

I've not done any testing of features like suspend/hibernate since these are targeting "low power always on", and the onboard graphics don't quite match up to 23.97fps at times*. My only regret about choosing the "server" S1200KP board for my NAS is that it doesn't come with mSATA or eSATA ports.

In the mITX arena, almost everyone else seems to use the ubiquitous (and ubiquitously crap) Realtek chips, so it's worth buying Intel for that reason alone. Recently Asus have made a point of using Intel NICs in their higher-end mATX and ATX boards but for small linux builds intel is the only game in town. My two pence.

* If you're really set on a good HTPC experience with e.g. linux+XBMC I would still recommend an ASRock Vision 3D with an nVidia card in it. They're ludicrously overpriced but tiny and silent and the SB models at least work perfectly with linux.

I can hardly believe that, so far, nobody mentioned Qubes OS.

In the theoretical sense, security is possible. It's just very hard. Especially if you want to spend your time doing something other than building a secure computer system.

In practice, most people live with a reasonably amount of security by installing a reasonable alternate OS such as Debian, not installing unnecessary software such as the Java plugin, and regularly installing security updates.

But if you really want security, what you should be doing is isolating, isolating, isolating. If a program has no business using a resource, then it should not be possible for it to access that resource. Qubes is one attempt to do this while preserving application compatibility, by having applications and services isolated to their own virtual machines. Even the network card drivers are in separate virtual machines.

For maximum security with Qubes, you really need a processor with support for VT-d, such as a selected subset of Nehalem and better processors, but the AppVM security mechanism at least should work.

> SSDs are just for laptops or so, not for real data storage requirements

Yep, just for laptops

http://www.intel.com/content/www/us/en/solid-state-drives/solid-state-drives-910-series.html

http://www.equallogic.com/products/default.aspx?id=10857

SSD isn't great for bulk data storage, but where you need high IOPS a few SSDs in arrays replace a truckload of drives.

Although Intel still declare that their TDP is *not* maximum draw in their Measuring Processor Power: TDP vs ACP paper, so I am not sure whether you answered out of personal experience/knowledge or based on plain theory.

Original poster: "Amazing that it supports ECC since Intel seems committed to making you pay through the nose for stuff like that."

This article gives some insight into why Intel is doing this. Basically, ARM has been making noises for some time about getting into the server market. Intel is very concerned about this, because ARM is used to lower margins and willing to license their designs widely, and could easily undercut Intel on price. They see the writing on the wall. Sure, they would like to keep ECC and other server-type goodies as premium features, but that's no longer a realistic option. Either they have to offer something cheaper, or customers who want low-cost, high-reliability server hardware will jump ship as soon as they can. This is the market niche the Atom S1200 is designed to fill. Intel gets to tout its advantage of backwards compatibility while being able to dramatically undercut other server-grade hardware on price. With this, ARM is going to have a much harder time convincing data centers to switch.

By the way, if all you care about is ECC, you don't have to buy an expensive CPU from Intel to get that (though you do need a C-series chipset rather than the consumer-grade stuff). Many of Intel's Ivy Bridge Pentium and Core i3 processors now support ECC, though this has not been widely publicized. For example, this i3-3220 is only $119.99 at Newegg and according to Intel's official site it supports ECC.

The fastest one of the series is $64. Not hard to find... Also, it's a server chip, not mobile. http://ark.intel.com/products/71267/Intel-Atom-Processor-S1260-1MB-Cache-2_00-GHz

I wonder if CUDA can be done with intel cards.

No, never, CUDA is nVidia only. But Intel supports OpenCL.

Bear in mind that AMD leaves Intel way back in the dust in GPU performance, including embedded GPUs.

Not all XEONs have hardware virtualization. Only some of the most expensive chips have it and even then, it can be spotty.

Not true. Every Xeon since 2006 has shipped with VT-x support. Look at the Xeon 5030 for example. Absolute bottom of the line ($150 at launch) from 2006, and it supports VT-x.

You're probably thinking of Intel's desktop line where to do artificially hobble large swaths of their CPUs with respect to VT-x.

The i7 3770K has a TDP of 95W

Intel's website says 77 W while various websites say the retail packaging says 95 W. Weird.

are people going to spend close to $1000 upgrading their computers just to be wowed by some extra graphical detail? My two year old machine is still better than the higher recommended specs. I just bought a $600 system for my kids that has better specs than the recommended specs. If I can get a whole system for $600, than it shouldn't cost that much. Let's check Newegg: Intel Core i5-750 - apparently there is no such thing, but the most expensive I5 is $250. or AMD Phenom II X4 805 - apparently there is no such thing, but the most expensive AMD Phenom II X4 is $85. NVIDIA GTX 560 - The most expensive of these is about $250, but they can be had for less than $200. AMD Radeon HD5870 - No longer available, but faster cards are available for less than $100. 4GB Memory? $50, assuming your computer doesn't already have that much RAM. It is not easy to find a computer these days with less than 4 GB.

Acutally 4GB should run you less than $20. http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=100007611%20600006050%20600006066&IsNodeId=1&bop=And&Order=PRICE&PageSize=20
or
http://www.newegg.com/Product/ProductList.aspx?Submit=ENE&N=100007611%20600006050%20600006067&IsNodeId=1&name=4GB

Also, the gt500 series nvidia cards are deprecated, too. The 600's are current generation.

If the i5-750 doesn't exist, you had better notify Intel pronto: http://ark.intel.com/products/42915/Intel-Core-i5-750-Processor-8M-Cache-2_66-GHz

Automatically paralleling compilers aren't new. SGI had one for C ten years ago. Intel has it for C++ and Fortran now. It's been done for Matlab. There's plenty of theory.

Outside of the supercomputing community, it's never gone mainstream. Microsoft is at least trying to do it for C#, which is a reasonable language. An interesting point is that this Microsoft approach exploits immutable objects. Immutable objects can be freely shared without locking, so wide use of immutable objects makes automatic extraction of parallelism easier.

I'd looked at doing something similar for Python, to get rid of the Global Interpreter Lock, Python's boat-anchor. Python already makes wide use of immutable objects, but doesn't gain any performance from them. If everything is thread-local, immutable, or synchronized in the Java sense, you don't need global locking. But the politics of the Python community do not favor performance. (Potentially, Python could be up to at least the LISP level of performance, within a factor of 2 of C, if the language was restricted in certain ways.)

Another part of the problem is the pernicious heritage of POSIX/Linux locking primitives. Many programmers think that locking is an library or OS-level problem, not a language level problem. Locking via library primitives means language doesn't know what data is locked by which lock. This makes optimization and race checking very tough.

The political and social problems here are tougher than the technical ones. So the open source community can't solve them. It takes someone with a big hammer, like Microsoft, to do it.

"Who tells whom what to do?" - V. Lenin

Not a big difference between these two processors, just a bit more cache on the i7. Benchmarks show differences directly related to clock speeds among these models.