Intel Takes the Lead

We’ll start off with a period of time not long ago when AMD was cheering over its shiny new X2 line of CPUs. If you’ll recall, they challenged Intel to a “Dual-Core Showdown”, directly targeting Intel’s beaten-to-death NetBurst architecture being used by the Pentium D. Benchmarks and real-life evaluations showed that NetBurst was ready for the trash compactor, for those who hadn’t figured it out already. Shortly after X2 had proven itself, Core 2 by Intel entered the game and began thrashing AMD’s prized possession to the nth degree. Since that point, there has been no full recovery. Products have been released by AMD, but the story each time has been “too little, too late.”

Intel, on the other hand, is still riding the wave of Core 2, slimming down to 45nm process, and generally enjoying the fact that they dominate the market. The good news for them is that they can continue to pressure AMD with price drops simply because they can afford to do so. They have left their competition in a position where they must sell more expensive processors that don’t perform as well. Take a wild guess at where people are putting their money.

Can AMD Recover?

So with a seemingly grim situation for the underdog, where is the light at the end of the tunnel? A little luck, combined with some very solid planning, gives AMD an extra card to play. Take the following into consideration: They just got an outside investor’s $622M, Phenom/AM3 chips work with socket AM2 (making AM2 owners like me very happy to see a viable upgrade option), the HD 38xx series of video cards are looking to be incredibly successful given the circumstances, and their losses in the last year have been cut dramatically.

Now you probably think I am going crazy, especially with that last comment. We know by now that AMD is losing all kinds of money, mostly to do with falling market share and backlash from obtaining ATI. However, looking at the charts, we can see a very clear recovery happening. Loss is indeed bad, but recovery is like a better bad. It means they haven’t given up, and that they are taking back ground.

Using this information, and basing it on the past, it is fairly safe to assume that AMD/ATI will come back. But enough about them. What is going on in the Intel world?

Well, for starters, they just opened their new 45nm chip factory. Their chips are being used by Apple (granted that this isn’t a big market comparatively, but every penny counts). Here is a kicker: they have won the heart of the enthusiast, and brought very powerful computing to nearly everyone buying a PC. Unfortunately for AMD, the enthusiast goes with what works best. The Core 2 landslide has brought enthusiasts a very reliable and exceedingly fast platform that is comparable in some ways to the earlier FX series in its day. Top that off with a massive product lineup that leads to options at nearly every price point, and you have a recipe for success. And don’t think for a second that Intel will back down; Penryn and Yorkfield are here, with more coming.

Conclusion

So now where does this leave us? Well, Intel needs to just keep on keepin’ on. They are sitting on record gains and a very fast-paced development time line that seems to be working quite well.

However, if they hope to keep the lead, they need to stay on schedule; winding up with another NetBurst-like issue will put them exactly where they were a few years ago. They’ve got the money, the advertising, and for now the time to stretch their legs and try a few things. If they get lazy and don’t keep on top of what will be aging technology, the crown will get handed back over to AMD sooner than they’d like.

What can AMD do about this monolithic company? They need to regain the touch they seem to have lost around the release of X2. The company gloated in the face of the final Pentium chips, only to get kicked in the face by Intel’s Conroe, and sand in the eyes by the rest of the lineup. Now they have rubbed the sand out (for the most part) and are sprinting to catch up. What will give them an edge is if they can focus on what needs to get done, and it appears that they are taking that direction. The recently debuted Spider platform is a direct appeal to one of AMD’s cherished market, being the enthusiast. While not quite the 1-2 knockout that is could have been, it’s strength is in AMD’s ever present scalability. If they can repeat this pattern like they have with previous architectures and extend it into their newfound graphics department and chipsets, they will likely have a formula to compete with Intel’s ‘tick-tock’.

Links mentioned in this podcast:

http://www.microsoft.com/servers/64bit/overview.mspx

http://developer.apple.com/macosx/64bit.html

Also mentioned:

If you actually have a 64-bit system (no matter what OS you use,) do you see any advantage by having it at present? Do you use any native 64-bit desktop (i.e. not server based) apps that really show a difference in speed, smoothness, ease-of-use, etc.?

Listen in and give your opinion on 64-bit.

A brief history of this article…
The original article (the “prequel”) was written by former PC Mechanic Editor-in-Chief and Forum Administrator Matthew A. Dockter. Authored back in March 2001, an otherwise boring history of microprocessor manufacturers’ marketing schemes and developing technologies was written in an informal, enjoyable story-plot style. As it was long overdue for a revision, I decided to write Part II, picking up with the development of the K7 AMD Athlons. While I tried to cover many important moves from the chip giants, unfortunately, I could not cover everything in the scope of this article.

BX Chipset — an ageless warrior
Let’s make one thing very clear. Intel unleashed a very powerful chip with the BX chip in 1998. In many ways, it was probably one of the most successful chipsets ever released, taking into account its longevity and the type of technology it was able to employ over its lifetime. BX was originally made coinciding with the success of Intel’s Pentium II line of processors. When releasing the first iterations of the Pentium III, Intel really had to question — is it worth killing something that’s so successful? BX lived on — and lived up to its current fame through most of the Pentium III days (including the Tualatin’s). While successors were planned and released (ie. the i815 chipset), it never really replaced the BX chipset in terms of what visible performance it brought to the table. The only downside of the BX was it could not run at 133 FSB without seriously overclocking the AGP bus since it had a fixed 2/3 divider. BX chipsets were also limited to AGP 2X at 3.3V and ATA33 unlike many of its later competitors.

Clockspeed Wars Picks up…
Many manufacturers practically abandoned AMD once they knew of what the BX could do. AMD had chips, but there were no motherboards for it. FIC, Micronics, and Gigabyte reaped the rewards. Asus, AOpen, Tyan, and Soyo hopped onto the AMD bandwagon. The “K7″ Athlons introduced dual-pumped front side bus. Simply, the front side bus (FSB) can run twice as efficient as the normal bus speed. Intel abandoned the slot CPUs and released an FC-PGA (Socket 370) processor using a new .18 micron fab process dubbed “Coppermine”. Smaller fabrication process generally means higher potential clock speeds. FC-PGA package equals cheaper productions. The Coppermine also halved the level two cache, but it now ran at the full speed of the processor resulting in net gained speed. AMD upped the ante and surprised many by rolling out 700Mhz. Apparently, they were able to sustain decent yields with it to the surprise of many analysts. Intel released 733Mhz, but the yields were unsatisfactory. Intel’s major customer, Gateway, publicly criticized Intel’s inability to keep up yields with the Coppermine Pentium III’s. AMD rolls out the 850Mhz and 900Mhz. Intel keeps up pace with the Coppermine 933Mhz. Computer enthusiasts and overclockers could see 1Ghz with their eyes now.

Thunderbird flashes into the high-end market
At the same time, AMD started to “change its direction” just as Intel jumped onto Coppermine. AMD’s Athlon “Thunderbird” incorporated many basic improvements that made it a rather successful revision. Compared to the early Athlons, the Thunderbirds had half the level-two cache. They had 256k L2 cache, as opposed to the 512k L2 cache on the early Athlons. Bad, you would say? Not really — while there was physically less cache, Thunderbirds had more powerful and faster full-speed L2 cache. Thunderbird, like the Coppermine, also saw a major design overhaul by ditching Slot A for Socket 462 (a.k.a. Socket A). The advantages of a PGA (pin grid array) package were obvious — cheaper to produce, easier possible cooling among them.

Celerons and Durons: battle for the low-end
While the high-end market was gaining momentum, Intel and AMD dared not abandon the low-end segment. Intel continued to produce the Pentium-II based Celerons WITH level two cache. That Celeron was limited by a 66Mhz bus speed. With decreasing pressure from AMD’s end in that market segment, Intel’s Celeron processors became largely popular. Intel originally released the Celerons with 266Mhz clockspeed and continued to scale upwards. Of course, they always lagged behind the pricier Pentium II line. They bumped the speeds to 300Mhz. Then 333Mhz. Eventually, there were Celeron 400Mhz CPU’s that were still based off the Pentium-II architecture, but on the newer PPGA (Plastic Pin Grid Array) socket design. Intel upped the clockspeed as high as 533Mhz on a Pentium-II Celeron. Intel then released a FC-PGA Pentium-III Coppermine version of the Celeron. They bumped the processor bus speeds. They raise their clockspeeds. And then AMD finally makes a move.

After the Coppermine-128 Celerons (ie. Pentium III Celeron) were released, AMD unveils a new line of processors named AMD Duron. The Durons, like the Celerons, were crippled revisions of the company’s “cream of the crop” processors. So what’s different? AMD held one major advantage over the Celeron. The original dual-pumped front side bus stuck, and so it offered a 200 Mhz FSB over the Celeron’s 66Mhz FSB speed. Intel responded when they upped the FSB to 100 Mhz. The war rages on…

If you’ve been around computers for some time, you can probably remember back to March 2000 when Intel formally introduced the gigahertz processor to the world. Touted as the “world’s highest performance microprocessor PC” (as per Intel), it was unbelievably speedy at the time. The ‘gigahertz’ became a part of common terminology. Just look for it like it’s a product review for performance. Then the processor became faster…faster…and faster. 2Ghz? No problem. 3Ghz? Here you go.

Inevitably, we closed in on a ’speed barrier’– a point where it is no longer possible to crank up the speed due to one of many conditions. Because producing newer better processors with the demanding market is essential, the manufacturers were forced to think of other improvements that will give the consumers an incentive to purchase the newer hardware. In came an emphasis on multi-tasking.

Chances are, you’ve probably heard about the new dual core processors. The corporate giants, Intel and AMD, have shifted their emphasis on multitasking using dual core as the primary driving force. However, the concept of computer multi-tasking has been with us for quite a while and it is not limited to dual-core. Multi-processor systems in servers and high-end workstations as well as Intel’s coveted Hyper Threading Technology have been with us for many years. Three technologies, one concept/goal. You may wonder, what’s the difference?

Introduction & Preliminary Mobile CPUs
Computers evolve quickly, which is very much so with microprocessors. For a long time, the industry’s focus was on getting the computers faster and faster, regardless of factors such as size, heat, and power. Now, with the ever-growing demand for portable systems, size is becoming an important factor. This especially concerns the microprocessor - one of the hottest and most power-hungry components in a system. Heat, rate of battery consumption, and processing power are the three main factors in a processor.

The major processor giants were facing a dilemma when first releasing a laptop. It consumes a large amount of power and it runs very hot. The laptop battery can only hold a certain amount of power and because of the closeness of all the components in the cramped chassis, the chip cannot produce a lot of heat. Furthermore, the chip cannot be cooled with a fan like it is in a desktop because doing so will make the laptop cumbersome and thus importable, defeating the purpose of a laptop. This was especially a problem for AMD’s early Athlon Processors that were known to heat up easily.

Initially, the processors that were fitted into a laptop were crippled desktop processors. By making the processor run below its full potential speed, it consumed less power and it gave off less heat. Intel and AMD both adopted this method - for example, Intel may have taken a Pentium III 1Ghz desktop processor and modified it into a Pentium III 800Mhz laptop processor. As such, the laptops were always a step behind in speed. AMD has the Mobile Sempron Processor for the mobile chassis which is still used today as AMD’s budget line processor - a crippled version of the AMD Sempron budget-line desktop processor.

Portable systems were becoming increasingly popular through the 1990s and the industry saw the need to tackle the limitations of a portable chassis. Several major steps forward were taken. Intel SpeedStep and AMD PowerNow! technologies were implemented, which adjusts the clockspeed of the processor when the full speed of the processor was not needed. So processing power is “on demand” - whenever the processor was downclocked from SpeedStep or PowerNow!, the processor would save energy.

Centrino: The Mobile Platform
In March 2003, Intel unveiled the Intel Centrino platform. So what makes the Centrino so special? You’ve seen it advertised by almost every laptop manufacturer and a lot of people seem to have it. The key is that Centrino is a platform, not a Processor. Intel specified that the platform include a specific chipset, a specific Intel integrated IEEE 802.11 Wi-Fi, as well as the Pentium-M processor in its platform. By allowing one company to make these specifications, power was better managed - laptops became truly “thin, light, and powerful”.

Intel Centrino Mobile Technology

The Intel Pentium-M processor was the major driving force behind the low-power low-heat portable technology. Instead of implementing the NetBurst processing architecture found in the power-hungry desktop Pentium 4 processors, Pentium-M processors are based off of the old but proven Pentium Pro technology. Specifically built from the ground up with mobility in mind, Intel allows the processor to work fast with a low TDP (Thermal Design Power). It uses a relatively low clockspeed, but efficient processing made good use of every clock cycle (including shorter pipeline architecture). The Pentium-M also implemented a fast on-die 1MB Static RAM level two cache allowing for faster access to data. More on-die cache means the processor can quickly reach for more data more frequently. At the same time, it continued the implementation of Intel SpeedStep to save power. Intel revised the Pentium-M and in 2004, they released the Dothan Core Pentium-M which boasts 400 Mhz and 533 Mhz Front Side Bus speeds and a 2MB level two cache. All these features and adjustments made the Intel Pentium-M a powerful processor without chewing much from the battery.

Is Intel’s flagship mobile technology right for you? In almost all cases, the answer is yes - power, mobility, and speed all in one package. Those looking to make the laptop a true mobile system with desktop-like powers will find what they are looking for with Centrino. There is a downside to everything, and for the Centrino, it’s the slightly high price. Most Intel Centrino laptops are range from 800 USD and up. Though pricey, the efficient technology is worth it.

AMD Turion 64 - AMD’s reply & Conclusion
AMD followed with the AMD Turion 64 processor in 2005. Unlike the Centrino, AMD Turion 64 is a processor, not a platform. In other words, the laptop manufacturer could designate which chipset to use without being bound to a single company. There are disadvantages though. When using one standardized chipset, it allows for better power management. Although processing power did not reach the heights of the Pentium-M in the Centrino platform, Turion 64 boasts 64-bit processing support whereas the Pentium-M currently does not.

AMD Turion 64

AMD has a new naming scheme laid out for the Turion 64. The numbers we’re used to seeing, such as “3200+” or “1800+”, won’t be the model numbers you’ll see. AMD Turion 64 model numbers consists of two letters followed by a number. The letters represent the class the processor belongs in, with the second letter indicating the degree of relative mobility. Relative mobility is shown greater as the second letter approaches the end of the alphabet - “Z”. The numbers denotes rated performance. So for example, a laptop with a Turion 64 MT-30 processor will have greater mobility than the same laptop with a Turion 64 ML-34. However, the ML-34 is rated to perform better then a MT-30 because the number following the letters is greater.

Despite the confusing naming scheme, for some, the AMD Turion 64 may be appealing. Because the Turion 64 is a 64-bit processor, when the new “MS Windows Vista” debuts, you’ll be prepared. However, the battery life to power ratio is not as impressive as the Intel Centrino. It is based more on the Athlon 64 than a “built from the ground up” approach that characterizes Intel’s Centrino.

The AMD Turion 64 chip and the Centrino Technology are both targeted at mainstream to high-end users. In the lower end, Intel developed the Celeron-M - not to be confused with Mobile Celeron. The Intel Celeron-M is a derivative of the preliminary Pentium-M chip, the Bainas core CPUs. Certain functions such as the power saving Intel SpeedStep technology are disabled. The processor is clocked lower and the level two cache is cut down to 512k as opposed to 1MB found in the Bainas Pentium-M. Like the AMD Turion 64 though, the Celeron-M is not a platform but rather a processor. In case you’re in the market for a budget laptop, the Intel Celeron-M is a good choice for the “thin and light” category, but you won’t see battery life like you would on Intel Centrino. AMD currently has the Mobile Sempron competing with the Celeron-M - while it’s still worth consideration even as it is an adapted version of the desktop processor, looking into either the Celeron-M or the Turion 64/Centrino category is usually more recommendable.

With size remaining an ever-important factor in a system purchase, Intel and AMD are in full blast to find faster, more efficient processors. Currently, Intel is developing the Intel Pentium-M chip with a 65nm fabrication process and two execution cores on a single die. Dual Core technology, which is already seen in servers and desktops, boosts multi-tasking abilities by allowing two threads of data to be processed simultaneously. Mobile computing won’t be the one step behind the desktop due to space constraints - it will be right alongside the big guys. Whether it be more efficient processing or faster clock speed, the future looks bright for laptop computing.

In my last guide, I told you everything there was to know about AMD’s current desktop line up. Well, it wouldn’t be fair to shut out the Intel users now, would it? So, onto the Intel technology guide!

This guide, like the AMD guide, will only focus on what is widely available on the desktop now. So don’t expect anything Pre-P4, or anything to do with the older socket 423 P4s (but if you have any sense you will avoid them anyway :).

Intel’s strategy is a little different to AMD’s. AMD had the intention of changing the world 64 bit with the release of the Athlon 64, but Intel have dragged their heels. X86-64 (which Intel call EM64t, I will refer to it by this from now on) is only just becoming available on the desktop Pentium 4’s, even though it has been in the server processor, Xeon, for a few months now. Intel have been much more interested in 32 bit performance, bumping their clockspeeds as high as they could go. They finally hit a wall, 3.8GHz, and as a result are now forced to find new ways to increase performance. One of these ways is to increase the cache, and another is to implement 64 bit into all their new processors.

Before we go much further, lets discuss the technologies available to Intel users. Unlike AMD, Intel have been making gradual changes to their processors, so gradual in fact we may as well just go straight into the cores. But first lets learn about the various processors available.

The processor world is a fast moving one. A soon as people get used to one technology, a completely new one appears to baffle you even more. So, what do you do? Come to PC Mechanic of course!

This guide will focus on the current technologies available from AMD on the desktop, and also some insight into what may be here next.

Right now, AMD are focusing on turning the world X86-64. This is a technology developed by AMD, which allows 64 bit programs to run on the chip, along with current 32 bit programs with no performance hit. This means you can run today’s 32 bit software, then when 64 bit takes hold you can simply upgrade the software with no hardware trouble. 64 bit programs are supposed to run faster, increasing performance over 32 bit programs. AMD’s mainstream and high end desktop processors are 64 bit, the server chips are 64 bit, and their laptop chips are 64 bit. If you buy a recent chip from AMD, chances are you will have a 64 bit chip unless you buy a budget processor, which I will go into later.

But, you would be foolish to think you are completely futureproof. Lets say you buy a 64 bit processor today, and later on when 64 bit software phases out 32 bit software you buy 64 bit software to replace your 32 bit software. How old would your processor be then? Very old. Consider that nearly all Windows software is 32 bit apart from beta software, and you can see why it will take so long to change the world to 64 bit. This is why Intel are only just bringing their 64 bit processors to market, there is no rush for 64 bit processing.

Ok, so into the actual processors. Basically there are 3 types of processors from AMD on the desktop, the Athlon 64, Sempron, and Sempron. That’s not a typo, there are two Semprons, and if you make the wrong decision you will regret it. One Sempron is based on the old K7 technology, while the other is based on the superior K8 technology. Lets explain the difference.

Computer technology is changing so rapidly, that in many cases, the computer you purchased six months ago is quite “old” today. In an industry that moves at blazing speeds, we like to release processor updates from time to time. This article is geared mainly towards those who are not computer gurus and don’t follow the industry everyday. If you are looking to upgrade in the near future and do not know what processors are available to you, this article will be of assistance. You’ll read about the current processors on the market, what their specifications are, how they perform, pros / cons, prices, speeds, and all other kinds of technical and practical data. This guide is divided into three sections so that each processor can be effectively covered: desktop chips, server chips, and laptop chips.

We will not be comparing one processor to another, as there are obviously infinite details to which processor is best at what. Each processor serves a purpose, and that purpose will be pointed out. Another thing we will not go into is which benchmarks what because, quite frankly, benchmark tools vary on everything, not just the processor. Also, all prices are subject to change, and probably will change week to week. When we mention “average price”, this represents the most likely range you will see. The true price range of a particular chip is somewhat larger, due to top of the line and bottom of the line outliers. We do supplement the discussions with a mention the full price range of the processor in the “cons” section of each. Here are some common definitions of terms used to describe the processors in this article so you can be familiar with them when they come up.

64 bit vs. 32 bit: Most processors and software currently on the market use a 32 bit instruction set. This means that the processor can handle up to 32 bits of information per cycle. A newer 64 bit technology is emerging fast, and has promise for the future. Essentially, 64 bit processors can handle twice as much data at a time than 32 bit processors can.

GHz: Number of cycles per second measured in billions.

MHz: Number of cycles per second measured in millions.

Front Side Bus (FSB): This is the speed at which the processor communicates with the memory on the motherboard. The faster the FSB, the better your performance will be.

Memory Cache: A very fast static RAM that stores frequently accessed data. This type of memory is generally stored on the processor, so it is very easy and quick to get to. It saves a lot of time compared to accessing data on the much slower dynamic RAM on the motherboard.

The different parts of the pipeline perform different jobs. Some parts of the pipeline are duplicated, and adds to the length of the pipeline. Also, there can be more than one pipeline, which is why modern processors are said to have a super scalar architecture. The reason parts of the pipeline are duplicated is so less work has to be done at each stage. This means more instructions are completed in the same amount of time, speeding up performance. This is one of the key reasons AMD is able to contend with Intel. AMD Athlon XP’s have 3 X86 decoders, 3 floating-point pipelines, and 3 integer pipelines. This is compared with Intel’s Pentium 4, which has only one X86 decoder, 2 floating-point pipelines, and 1 more integer pipeline than AMD’s Athlon. This leads to AMD being able to decode more instructions than Intel at the same time, and being able to perform floating-point operations quicker than Intel. Overall, AMD Athlon XP processors are able to perform 9 operations per clock cycle while Intel can only manage 6. It doesn’t sound like much, but in processors every operation is crucial. This is why I said AMD are more about getting more done per clock cycle in my AMD processor buying guide.

This doesn’t mean it’s all over for Intel though. Even though AMD manages to perform more operations than Intel in one clock cycle, Intel manages to do their operations quicker. This is because of their pipeline architecture. AMD’s pipeline is only 10 stages long. This means that because the stages in the pipeline have to do more work, they can’t run very fast. Now, with Intel, their processors have a 20 stage pipeline (Prescott core processors have 31 stages). This means that the processor can run at a higher clock speed, because less work is done in each stage of the pipeline. The reason the Prescott core has been released is because this brings yet more speed. Because there are 31 stages, even less work is done, which means even higher clock speeds.

This can be linked to CISC computing and RISC computing. CISC stands for complex instruction set computing, and RISC stands for reduced instruction set computing. RISC means having less complex instructions for the processor. Here is an example of CISC giving someone commands. With CISC it would look like this:

1. Get food
2. Get fork
3. Eat

But with RISC it would be like this:

1. Go to kitchen
2. open fridge
3. get food
4. close fridge
5. open drawer
6. get fork
7. close drawer
8. open mouth
9. put food in mouth
10. close mouth
11. chew
12. swallow

The reason the CISC is more complex is because the processor has a lot more to do in one instruction. RISC is more efficient because it is very simple instructions, meaning less “thinking” is needed to perform the instruction, resulting in faster speed. Also, using RISC means there is less transistors needed in a processor, reducing cost. This is why all modern processors are RISC processors. But there is something you should know about X86 (the way the instructions are coded). X86 is actually built using a CISC architecture. This is why the processors need X86 decoders, to convert the CISC instructions into RISC instructions.

Intel actually beat AMD to the gun by releasing Pentium IV Willamette in November of 2000 before AMD could release their Thunderbird-based Athlons. Pentium IV was exactly what Intel needed to again take the torch from AMD. Pentium IV is a truly new CPU architecture and serves as the beginning to new technologies we will see for the next several years. The new NetBurst architecture is designed with future speed increase in mind, meaning P4 is not going to fade away quickly like Pentium III near the 1 GHz mark.

According to Intel, NetBurst is made up of four new technologies: Hyper Pipelined Technology, Rapid Execution Engine, Execution Trace Cache and a 400MHz system bus. Let’s look at the first three, since they require some explanation:

• Hyper Pipelined Technology
  There are a couple of ways to increase the speed of a processor. One is to decrease the die size. Technology in this regard is developed quickly, but not quickly enough. The P5 core saw its limit quickly and so did the P6 core (which is why Pentium III was limited at around 1 GHz). The technology to move into a smaller die size was not yet ready at the time of the Willamette release, so Intel moved to plan B. Plan B is to change the design of the CPU pipeline so that it is wider, can accommodate more instructions. This is what Intel did. Hyper Pipelined Technology refers to  Intel’s expanding of the CPU pipeline from 10 stages (of the P6) to 20 stages. This effectively makes the data pipe (bad term, but descriptive) wider, and allows each stage to do actually less per clock cycle than the P6 core. The fact that each stage actually does less per clock cycle is what gives this design room for expandability. It is analogous to expanding a street highway - you add more lanes and for awhile each lane has less traffic, but eventually traffic increases and the road can handle much more traffic. The tradeoff in simply expanding this pipeline to a bunch of stages is that it takes the processor longer to recover from mistakes in the branch level prediction, being that it has to basically start over with 20 stages rather than a shorter 10-stage pipeline. The P4, though, has a newly advanced branch predictor to help with this problem.
  
• Rapid Execution Engine
  The Pentium IV contains 2 arithmetic logic units and they operate at twice the speed of the processor. While this might sound like absolute heaven, it is good to keep in mind that they had to do it this way due to the pipeline design in order to even keep integer performance up to that of the Pentium III. So, this is really a necessary design change due to the increase pipeline size.
  
• Execution Trace Cache
  Intel also did some re-working of the P4’s internal cache in order to nullify the effects of a mistake in branch prediction that can be a real lag with a 20-stage pipeline. First, they increase the branch target buffer size to eight times that of the Pentium III. This cache is the area from which the branch predictor gets its data. Secondly, Intel reduced the size of the L1 data cache to only 8K in order to reduce the latency of the cache. This, no doubt, increases the need for the 256 KB on-die L2 cache, and the latency of that has been improved on the P4 as well. Lastly, Intel added a execution trace cache. This is a new cache that can hold instructions that are already decoded and ready for execution. This means that the processor does not have to again waste time decoding every instruction when a branch prediction error occurs. Instead, it can just go to this 12K cache and retrieve the operation and go.

The early Pentium 4’s made use of the Socket 423 interface. One of the reasons for the new interface is the addition of heatsink retention mechanisms to either side of the socket. This is a move to help owners avoid the dreaded mistake of crushing the CPU core by tightening the heatsink down on it too tightly. The retention bases hold the heat sink onto the CPU. Socket 423 was short-lived, and Pentium IV quickly moved to Socket 478 with the release of the 1.9 GHz. Also, P4 was, at its launch, associated exclusively with Rambus RDRAM. Intel was stuck in this agreement with Rambus, and this was an obvious hurdle for promotion, being that most computer users to not have Rambus and don’t wish to buy any. So, early retail P4’s actually came packaged with two 64MB sticks of RDRAM. With chipset support later coming, DDR mating with the Pentium IV eventually came.

Pentium IV’s, as you might expect, were and still are on the expensive end of things. The new core was quite big when compared to other processors and the cost to produce it was innately higher. In early 2002, Intel announced a new edition of the Pentium IV based on the Northwood core. The big news with this is that Intel leaves the larger 0.18 Willamette core in favor of this new 0.13 micron Northwood. This shrunk the core and therefore allowed Intel to not only make Pentium IV’s cheaper but also make more of them. The core is still bigger than that of the Athlon XP, but this is explained by the fact that Intel increased the L2 cache from 256 KB to 512 KB for Northwood. This raises the transistor count from 42 million for Willamette to 55 million for Northwood. Northwood was first released in 2 GHz and 2.2 GHz versions, but the new design gives P4 room to move up to 3 GHz quite easily. It was recently released at 2.53 GHz using a 533 MHz front side bus. Other than that, Northwood is architecturally the same as Willamette.

Due to this new design for the Pentium IV, there is a lot of room for expansion, so the Pentium IV will not be disappearing anytime soon. Expect AMD to challenge Intel in a big way with their upcoming Hammer technology.