Chips, Chipsets and Damned Chipsets

Okay, so the first thing to understand is that an Intel brand, like Core 2 or Core i7, actually refers to a whole bunch of different processors. Although they generally have the same basic microarchitecture (in other words, chip design), the brand envelopes both desktop and mobile chips, chips with radically different clock speeds, that use different motherboard sockets, etc.

Because of these differences, each particular chip is given a codename, chosen for obscure geographical locations (seriously, plug just about any codename into Google Maps). For instance, the original mobile Core 2 Duo processor was Merom, and it was replaced after about two years by Penryn, which was manufactured using a new 45-nanometer process to be more efficient. Quite different, these two, but Intel pimped both as Core 2 Duos nonetheless.

View Intel in a larger map
Although Intel doesn't market chips according to their codenames, the individual chip gets a model number that gives you an idea of how it compares, spec-wise (clock speed, cache size, etc.), to other chips in the same group. So, a Core i7-950 is gonna be faster than a Core i7-920, and a Core 2 Duo P8600 isn't going to quite stack up to a Core 2 Duo P9600. The difference between a P8400 and P8600 is obviously less than the difference between a P8600 and a P9600. To match a particular chip codename to a particular model number, though, you probably have to do some Googlin' (or Bingin').

In some cases, Intel pushes chips with a ULV designator for "ultra-low voltage," which doesn't mean anything in particular in terms of chip design, since it includes several brands of chips, from Core 2 to Celeron. The point is that these chips power notebooks that are almost as portable at netbooks, but are more expensive, so computer makers (and Intel) make more money.

While we're at it, I might as well explain what the hell Centrino is. It's not a single chip, it's a platform. That is, it's a combo meal for notebooks with a mobile processor, a chipset (essentially the silicon that lets the processor talk to the rest of the computer) and a wireless networking adapter. Typically, Intel releases a new combo meal every year, though they're all been called Centrino, with the most recent making the leap to being called Centrino 2.

The reason we decided to tell you all this stuff now is that Intel is gradually phasing out the Core 2 family, like Pentiums before that, and is moving Core i7, Core i5 and Core i3 up to take its place. This is how all the families relate to each other...

Nehalem Rising: Core i7, Core i5 and Core i3

Core i7 systems use a totally new microarchitecture called Nehalem, and it's badass.

The first set of Core i7 chips, codenamed Bloomfield, launched in November 2008 for high-end desktops. They're the most outrageously fast Core i7 chips, with triple-channel memory (meaning they're able to use memory sticks in triplets rather than pairs) and other blazing accoutrements.

The new Core i7 chips, launched last month, are for desktop and mobile. The desktop variant is codenamed Lynnfield, and it more closely resembles its mobile equivalent, codenamed Clarksfield, than it does the Bloomfield monster—dual-channel memory, not triple, for instance.

You'll be seeing a lot more Clarksfield in the next couple weeks, like in the HP Envy 15, since most computer makers were holding off for Windows 7 to drop their new laptops. All of the Core i7 processors are quad-core, even the mobile Clarksfield, so you're not gonna see it in anything like Dell's skinny Adamo.

Core i5 is going to be Intel's more mainstream Nehalem-microarchitecture chip brand, and as a broader brand, the chip differentiation gets a little more confusing. Core i5 actually includes some, but not all, of the desktop Lynnfield processors. For now, the only Core i5 chip is quad-core, but you're going to start seeing dual-core Core i5 chips, and soon enough they will make up the bulk of Intel's mainstream processors. In English: Unless you're looking for a crazyfast new computer, your next machine will probably run an Intel Core i5 CPU.

Eventually, dual-core Core i3 chips will come out, and as you can guess by the number, they won't be quite as fast—or expensive—as the Core i5 or i7 chips.

Netbook's Best Friend: Atom N and Z

Atom is probably the Intel chip you hear about second only to Core 2 Duo: It's essentially the CPU that goes inside of netbooks. There are a couple of different variations out now, the N series (codename Diamondville) and the Z series (codename Silverthorne). The Diamondville chips are for nettops and netbooks (though as pointed out, nettop don't use the N prefix, just the chip number), and can handle full versions of Windows Vista and 7. Silverthrone is used in netbooks but was designed for smaller connected devices like UMPCs and MIDs. (This is why Sony shoving an underpowered Atom Z in the Vaio P, and trying to run Windows Vista on top of it, was retarded.)

The next generation of Atom is more interesting, and more confusing, in a way. The CPU is codenamed Pineview, and it's actually got the graphics processor integrated right onto the same chip, precluding the need for a separate GPU tucked into the netbook's overall chipset. The benefit is longer battery life, since it'll take less energy to crunch the same visuals. We'll start seeing Pineview netbooks sometime early next year, most likely.

Oldies But Goodies: Core 2 Duo, Quad and Extreme

Intel's Core 2 chips have been out three years now, an eternity in computer years. Because of this, and because they're the main ones used in most personal desktop and laptop systems, there is a metric shitton of different Core 2 chips.

It's also more confusing because there are way more codenames to wade through. Let's start from the top: Core 2 Solo has one core, Core 2 Duo two, and Quad has four (as does Extreme). From there, you have two distinct generations of chips within the Core 2 family.

In the first generation of Core 2 Duos, the main desktop chip was Conroe (with a cheaper variant called Allendale), while the main mobile one was called Merom. There was also a branch of Core 2 Quads called Kentsfield.

The next generation (that is, the current generation, unless you're already on the Core i7 bandwagon) arrived with a new process for making chips with even smaller transistors. Among other more technical differences, they were more energy efficient than their predecessors. With this generation of Core 2s, the mainstream desktop chips are Wolfdale, the desktop quad-cores are called Yorkfield, and the mobile chips are Penryn—if you've bought a decent notebook in the last two years, it's probably got a Penryn Core 2 inside of it.

Ancient History: Pentium and Celeron

Pentium is dead, except it's not, living on as a zombie brand for chips that aren't as good as Core chips, but aren't as crappy as Intel's low-end Celeron processors. If you see a machine with a sticker for Pentium or Celeron, run.

Okay, I hope that helps, at least a little—you should probably thank me for staying away from clock speeds and other small variations, like individual permutations of Core i7 Bloomfield processors, to hopefully give you a broader overview of what all's going on. Intel told me it'll all make more sense once their entire road map for the year is out on the market, but I have a feeling it's not gonna help my mom understand this crap one bit better.

Top image via soleiletoile/Flickr

Thanks to Intel for helping us sort all this out!

Still something you wanna know? Send questions about sweet potato chips, pumpkin pie or turduckens to tips@gizmodo.com, with "Giz Explains" in the subject line.

The heart of the matter is that the trick to actually utilizing the full power of multiple processors—or multiple cores within a processor, like the Core 2 Duo you've probably got in your computer if you bought in the last two years—is processing things in parallel. That is, doing lots of stuff side by side. After all, you've got 2, maybe 4 or even 8 processors at your disposal, so to use them as efficiently as possible, you want to pull a problem apart and throw a piece of it at each core, or at least send different problems to different cores. Sounds logical, right? Easy, even.

The rub is that writing software that can actually take advantage of all of that parallel processing at an application level isn't easy, and without software built for it, all that power is wasted. In fact, cracking the nut of parallel processing is one the major movements in tech right now, since parallelism, while it's been around forever, has been the domain of solving really big problems, not running Excel sheets on your laptop. It's why, for instance, former Intel chair Craig Barrett told me at CES that Intel hires more software engineers than hardware engineers—to push the software paradigm shift that's gotta happen.

A big part of the reason parallel programming is hard for programmers to wrestle with is simply most of them have never spent any time thinking about parallelism, says James Reinders, Intel's Chief Software Evangelist, who's spent decades working with parallel processing. In the single core world, more speed primarily came from a faster clock speed—all muscle. Multi-core is a different approach. Typically, the way a developer takes advantage of parallelism is by breaking their application down into threads, sub-tasks within a process that run simultaneously or in parallel. And processes are just instances of an application—the things you can see running on your machine by firing up the Task Manager in Windows, or Activity Monitor in OS X. On a multi-core system, different threads can be handled by different processors so multiple threads can be run at once. An app can a lot run faster if it was written to be multi-threaded.

One of the reasons parallel programming is tricky is that some kinds of processes are really hard to do in parallel—they have to be done sequentially. That is, one step in the program is dependent on the result from a previous step, so you can't really run those steps in parallel. And developers tend to run into problems, like a race condition, where two processes try to do something with the same piece of data and the order of events gets screwed up, resulting in a crash.

Snow Leopard's Grand Central Dispatch promises to take a lot of the headache out of parallel programming by managing everything at the OS level, using a system of blocks and queues, so developers don't even have to thread their apps in the traditional way. In the GCD system, a developer tags self-contained units of work as blocks, which are scheduled for execution and placed in a GCD queue. Queues are how GCD manages tasks running parallel and what order they run in, scheduling blocks to run when threads are free to run something.

Reinders says he's "not convinced that parallel programming is harder, it's just different." Still, he's a "big fan of what Apple's doing with Grand Central Dispatch" because "they've made a very approachable, simple interface for developers to take advantage of the fact that Snow Leopard can run things in parallel and they're encouraging apps to take advantage of that."

How Snow Leopard handles parallelism with GCD is a little different than what Intel's doing however—you might recall Intel just picked up RapidMind, a company that specializes in optimizing applications for parallelism. The difference between these two, at a broad level, represent two of the major approaches to parallelism—task parallelism, like GCD, or data parallelism, like RapidMind. Reinders explained it like this: If you had a million newspapers you want to cut clips out of, GCD would look at cutting from each newspaper as a task, whereas RapidMind's approach would look at it as one cutting to be executed in a repetitive manner. For some applications, RapidMind's approach will work better, and for some, GCD's task-based approach will work better. In particular, Reinders says something like GCD works best when a developer can "figure out what the fairly separate things to do are and you don't care where they run or in what order they run" within their app.

It's also a bit different from Windows' approach to parallelism, which is app oriented, rather than managing things at the OS level, so it essentially leaves everything up to the apps—apps have got to manage their own threads, make sure they're not eating all of your resources. Which for now, isn't much of a headache, but Reinders says that there is a "valid concern on Windows that a mixture of parallel apps won't cooperate with each other as much," so you could wind up with a situation where say, four apps try to use all 16 cores in your machine, when you'd rather they split up, with say one app using eight cores, another using four, and so on. GCD addresses that problem at the system level, so there's more coordination between apps, which may make it slightly more responsive to the user, if it manages tasks correctly.

You might think that the whole parallelism thing is a bit overblown—I mean, who needs a multicore computer to run Microsoft Word, right? Well, even Word benefits from parallelism Reinders told me. For instance, when you spool off something to the printer and it doesn't freeze, like it used to back in the day. Or spelling and grammar running as you type—it's a separate thread that's run in parallel. If it wasn't, it'd make for a miserable-ass typing experience, or you'd just have to wait until you were totally finished with a document. There's also the general march of software, since we love to have more features all the time: Reinders says his computer might be 100X faster than it was 15 years ago, but applications don't run 100x faster—they've got new features that are constantly added on to make them more powerful or nicer to use. Stuff like pretty graphics, animation and font scaling. In the future, exploiting multiple cores through parallelism that might be stuff like eyeball tracking, or actually good speech recognition.

Reinders actually thinks that the opportunities for parallelism are limitless. "Not having an idea to use parallelism in some cases I sometimes refer to as a 'lack of imagination,'" because someone simply hasn't thought of it, the same way people back in the day thought computers for home use would be glorified electronic cookbooks—they lacked the imagination to predict things like the web. But as programmers move into parallelism, Reinders has "great expectations they're going to imagine things the rest of us," so we could see some amazing things come out of parallelism. But whether that's next week or five years now, well, we'll see.

[Back to our Complete Guide to Snow Leopard]

Still something you wanna know? Send questions about parallel processing, parallel lines or parallel universes to tips@gizmodo.com, with "Giz Explains" in the subject line.

Grand Central Terminal main concourse image from Wikimedia Commons

Code-named "Hummingbird," this new chip is designed with smartphones in mind as it balances power consumption and high clock speeds, and matches the single core end of Qualcomm's Snapdragon platform.

Analysts suggest it could be easily dropped into a new iPhone or Pre:

Samsung could drop Hummingbird into the existing S5PC100 design with few or no changes," Halfhill said in response to an e-mail query, referring to the S5PC100 processor now used in the iPhone 3GS. "Bingo! A next-gen iPhone that could run at speeds up to 1.0GHz," he said.

But more than likely, we'll see them in Android-based or other smartphones first, just like we're seeing 1GHz Snapdragon chips pop up in other phones right now.

I just know that my heart rate matches a hummingbird's at the thought of a 3G S(uper) S(peedy). [CNET]

But because of the fact that the laws of quantum mechanics are so strange, the qubits (atoms) can be placed into a "superposition of multiple states" in order for them to store more than just the standard amount of information.

Now they're working on adding more qubits, which adds more power on an exponential scale. We're going to be Giz Explaining what's up with quantum computing soon. [TGDaily]

Even though Apple wouldn't give us the specific numbers, T-Mobile Netherlands has been a pretty reliable source for leaking things Apple doesn't quite want known, even new hardware. Besides, those specs are exactly what had been rumored, so we're just left to wonder why Apple bothered to hide them. [T-Mobile via iLounge]

Versus the previous generation Athlon X2, it starts at 3GHz, supports DDR3 memory and is part of the new AM3 packaging (which is backward compatible with AM2+). Also in the bag is the Phenom II X2 550 Black Edition, which is their first dual-core Phenom chip—so it can hit $100—but it's overclocked.

Sadly, that's it from AMD for Computex—no Atom killer or new hotness from ATI either.

AMD Introduces Next Generation AMD Athlon™ II Processor, Adds Dual Core to Record-Setting AMD Phenom™ II Processor Lineup

− AMD Athlon™ II processor delivers new native dual-core architecture, efficient 45nm technology and 3 GHz performance at an affordable price −

− AMD Phenom™ II X2 Black Edition processor combines value and unlocked potential for gamers and tuners on a budget −

COMPUTEX 2009 (TAIPEI, Taiwan) - June 2, 2009 - Bringing its acclaimed 45nm technology to new high-volume processor designs, AMD (NYSE: AMD) today announced two new dual-core desktop processors. Building on 10 years of AMD Athlon™ processor innovation, the new 45nm AMD Athlon™ II X2 250 processor gives mainstream consumers exceptional performance, efficiency and value. For enthusiasts and overclockers, AMD also announces the AMD Phenom™ II X2 550 Black Edition processor, the first ever dual-core AMD Phenom II CPU.* With this latest addition to the AMD Phenom II processor family, users can now experience the power of AMD platform technology, codenamed "Dragon," with dual-, triple- and quad-core configurations.

AMD Athlon II X2 Processor Details
The AMD Athlon II X2 250 performs exceptionally well when combined with AMD chipsets and integrated graphics solutions to create an all-AMD platform. Platforms featuring all-AMD technology can deliver up to twice the graphics performance of those with Intel integrated graphics.¹

Windows® 7 is optimized for multi-core processors like AMD Athlon™ II processors to give consumers an amazingly fast, simple and engaging PC experience.** For example, Windows 7 is tuned to make the most of the these new processors' power management features, such as AMD PowerNow!™ 3.0 technology. AMD power management technologies, in combination with Windows 7, can help OEMs and partners to build exceptionally green, cool and quiet PCs.
Based on AMD's acclaimed 45nm process technology, the AMD Athlon II dual-core processor has a TDP of 65W and can slash power consumption by up to 50 percent when doing basic tasks, up to 40 percent when running heavy workloads and up to 50 percent when at idle.²

AMD Phenom II X2 550 Black Edition Details
AMD Black Edition processors, like the AMD Phenom™ II X2 550, help users to take control and unleash the maximum potential of Dragon platform technology's unprecedented performance tuning capabilities.* The same massive headroom that set world records in recent months is at users' finger tips, offering impressive performance at a price the competition can't beat.³

Users can also maximize their overclocking experience by utilizing the new features and capabilities of AMD OverDrive™ 3.0, designed to enable quick and effective tuning of their PC experience for optimal performance.*
With dual-, triple- or quad-core processors, AMD provides platform level solutions at multiple price points, each of which exceeds expectations for virtually any user.

[AMD]

• AMD tried to kill the megahertz myth before Intel. During the Pentium 4 days Intel kept pushing clock speeds higher and higher, before it hit a wall and abandoned the Prescott architecture. The message was clearly, "more megahertz is more better." AMD's competing Athlon XP chips, while clocked slower, often beat their Pentium 4 rivals. Ironically, AMD was the first to 1GHz, as some commenters have pointed out (don't know how I forgot that). Obviously though, AMD's performance lead didn't last forever.

• AMD beat Intel to 64-bit in mainstream computers. And we're not just talking about its Opteron and Athlon 64 processors. AMD actually designed the X86-64 specification, which Intel wound up adopting and licensing—so AMD's spec is used Intel's 64-bit processors to this day.

• AMD was first to consider energy efficiency in processor designs. Okay, this is kind of an extension of point number one, but during Intel's Pentium 4 'roid rage period AMD's processors consistently used less power than Intel's. Intel's performance per watt revelation didn't really start until the Pentium M (which was actually a throwback to the P6 architecture), which set the tone for Intel's new direction in its successor, the Core line of chips.

• AMD beat Intel to having an integrated memory controller. A tech feature AMD lorded over Intel for years: AMD's processors started integrating the memory controller with its processors years ago, reducing memory latency. Intel's first chip to use an integrated memory controller is the Core i7—before, the memory controller was separate from the processor. (Here's why Intel says they held off.)

Athlon XP and Athlon 64—those were the good old days, AMD's cutthroat competitive days. The days they were ahead of Intel. I miss them—at one point, every hand-built computer in my house ran AMD processors. I felt like a rebel—a rebel with faster, cheaper computers.

Unfortunately, I don't run AMD chips anymore. Intel came back, and came back hard. But here's hoping for another resurgence, and another 40 years, guys. Share your favorite AMD memories in the comments.

The big news is the adoption of Via's C7-M chip, with clock speeds between 400MHz and 1GHz, that'll add new possibilities like HD decoding and surround sound. But that's not the end of the upgrades.

XO-1's memory should get a bump to 1GB, and its storage will get pushed to 4GB or even 8GB as an option. The upgrade, which is being referred to as the XO-1.5 rather than a totally new model, should start shipping in late August. [OLPC News]

The WSJ pulls out facts like IBM employees trying to hide their work from people from companies in cubicles next to them, helping one team out (the Microsoft team) with their design process based on knowledge they had already gained from the Sony side, and most importantly, that Microsoft received the chip from manufacturing BEFORE Sony did because they ordered "backup manufacturing capacity from a third party." [WSJ via PS3 Fanboy]

Last week, you probably noticed new computers from Dell, Gateway and others using a brand new, bizarre-sounding chip from Intel: the Core i7. You might have even seen some benchmarks and features showing that this is a real beast of processor. Well, we're pretty excited about the Core i7, so here's a quick guide to why it's so awesome:

Hokay, so the way Intel develops chips is on what it calls the "tick-tock cycle". The "tick" is the improvement of its current microarchitecture, mainly shrinking it down to make it more energy efficient, along with other tweaks. As you now can guess, the "tock" indicates the launch of a totally new microarchitecture.

Penryn, for instance, was the tick to the Core 2's tock, shrinking it down from a 65-nanometer process to 45nm. Core i7 is a tock, using a completely new microarchitecture codenamed Nehalem. Core i7 Nehalem is actually a dramatic step forward, remedying several lingering Intel architecture deficiencies that AMD actually had them beat on years ago. So, here are four things that specifically make the new chip awesome:

Bye Bye Front-Side Bus
The ancient front-side bus setup has long been a drag on Intel's chips, and they're finally ditching it. The FSB essentially carried data between the CPU and memory controller hub (which is also out the window, more on that in a sec), but that didn't work so well when you started talking buckets of cores. In its place is a new tech called QuickPath Interconnect that'll make the old bottlenecks history and running tons of cores even better. QPI uses direct point-to-point connections that have a bandwidth of about 25GB/s, way faster than what FSB could offer. The downside is that it requires a new QPI-friendly motherboard. This concept is kind of cribbed from AMD, whose HyperTransport has been doing something similar for a longass time.

Integrated Memory Controller and Triple-Channel Memory
You might notice a pattern that a lot of Nehalem's performance boosts have to do with better access to memory and fatter bandwidth. Yet another tech that AMD held over Intel's head for years is an integrated memory controller, which Core i7 finally uses. Basically this just means that the memory controller is on the same die as the CPU, cutting down memory latency. Before, with Intel chips, communication had to take place across the front-side bus, making stuff slooooow. The last memory bonus is that Core i7 supports triple-channel memory. Right now, you're probably on a computer using dual-channel memory (in English, I mean that it uses RAM sticks in sets of two). Core i7 will make three sticks of RAM the new standard—so keep an eye out for plenty of 6GB and 12GB systems running around.

The Return of Hyper-Threading
Intel abandoned Hyper-Threading after the Pentium 4, but it's back in Core i7 (and Atom, but really, psh). Basically, it's a parallel-processing tech that runs multiple threads simultaneously. In English, it divvies up tasks so they can be crunched by a processor simultaneously, instead of one after the other. It short, it makes video encoding and other parallel-friendly processes run faster. We're interested to see what kind of gains this will produce in tandem with programs coded to take advantage of threading, not to mention the next great operating systems, Snow Leopard and Windows 7, which will supposedly make better use of multiple cores and parallel processing than current OSes.

Built-In Power Management and Overclocking
Core i7 is pretty much a beast already, but whereas Intel used to say that overclocking was bad for your processor, now with the Core i7, it's built right in. The Core i7 is really aggressive with power management, more so than Core 2, so it'll sip juice when it's not busy, and then crank the power when it needs it. In the BIOS now, you can set it to overclock the CPU in certain situations, and customize that by thermal ratings so it won't overheat.

So yeah, Core i7 gets our engines running, and we're not even chip nerds. (Honest!) Sadly though, right now there are just a few Core i7 chips available, and they're all for desktops. There's not much of a downside for portables—save for the need for new motherboards and the DDR3 RAM already used by premium laptops—but before you see it in a Dell XPS notebook or MacBook Pro, you're going to see it in a lot of desktop gaming and graphics-intensive systems. Laptops probably won't appear until way into next year, but we think they'll be well worth the wait.

Something you still wanna know? Send any questions about chips, Pringles or the Hillary Swank movie The Core to tips@gizmodo.com, with "Giz Explains" in the subject line.

While benchmarking some memory, Gearlog found that processor performance is a whopping 37% lower on a MacBook Pro running on AC alone. Cinebench R10's multiprocessor test got a battery-less Pro a score of only 3,504, while with the battery it scored 5,549. They contacted Apple to find out what's going on, and Apple admitted that the performance difference is intentional, explaining that it "prevents the computer from shutting down if it demands more power than the A/C adapter alone can provide." Sounds a little fishy to me, but it doesn't really matter: Apple took a lot of care with that battery, and if you know what's good for you, you'll leave it be. [Gearlog]

Former CEO of AMD says the company could get fancy and "put software accelerators on there or maybe do something like a graphics engine." We understand what he means, but the way he said it makes us think of a kid who's making his dream car. "Put some wings and a jet engine and some flames and some missiles pew pew!" [NY Times via TUAW via Crunchgear]