Intel Pentium 4 2.2A (Northwood)

Author : Wayne Date : 4th March 2002

...Product	Pentium4 2.2A
...Manufacturer	Intel
...Supplier	Intel
...Price	£557.00 approx @ Scan

 

 

 

Northwood, what it means to you :

One gripe that some people had with Intel was the switch from their Socket 423 format to the new Socket 478. The good news is that Intel has no plans to drop Socket 478 any time soon ensuring a smooth upgrade path for the foreseeable future. Users who currently own mPGA 478-pin may need to look at updating the BIOS in order to run these new speeds, but the fact that the CPU is pin compatible will no doubt come as a great relief.

As you'd expect then, the Northwood P4's come in a 478 pin package. They also feature the now trademark heatspreader which does what the name suggests, it helps to draw heat away from the core and allows for a greater contact area with your heat sink. Not only does this mean more efficient cooling, it also means the core is massively better protected than that found on the Athlon range of CPUs. If there's one thing you really don't have to worry about when you buy Intel it's cracked and chipped cores.

The image below shows how the CPU arrived in a socket 479 blank which Intel assures us is used to protect the pins only and is not a subliminal hint that Socket 479 is on the horizon.

In case you're wondering, the hole in the heat spreader is used to allow the air inside to expand when heated without causing damage due to pressure and though its primary use is during the manufacturing process it also serves to do the same thing under normal use. Because of its position you'd have to be pretty careless to get anything in there that would be likely to do any damage, even a conductive thermal paste will not do any damage unless you pipe it straight into the vent hole.

So externally the Northwood P4 is pretty much identical to every other Socket 478 Williamette core that went before it. The switch to mPGA 478-pin meant a shrink in package size and the pins, though tightly packed, are pretty robust and not easily bent.....comparatively that is.

If you've never actually seen the new 478 pin package in the flesh before, the first thing to hit home is just how small the whole deal is. Although the pictures above are a bit misleading when it comes to judging the actual size, the image below should give a better idea of the real dimensions.

In fact the real advancements offered by Northwood are hidden away under the heat spreader. In addition to being the highest clocked processor available Intel also scored a first by moving to a new 0.13micron process. While the Williamette core was built on a 0.18 micron die, all Northwoods now feature a 0.13 micron dies. Not only does this mean that Intel were able to lower the voltage requirements, and therefor the heat output, they were also able to double the amount of L2 cache from 256k to 512k. The extra L2 cache will impact just about everything you do, but primarily it will prove a massive benefit for database and server users where data is forever being regurgitated.
The voltage reductions from around 1.75 volts to Northwood's leaner 1.50 volts mean we now see the 2.2 boast an impressive 55.1 watt thermal design power, all the more impressive when we consider the various P4 1.7 steppings ranged from 64.0 to 67.7 watts.

The new 0.13 micron process, which now also uses much more efficient copper interconnects as opposed to aluminium used for 0.18 micron dies, also means a great deal more headroom when it comes to clock speeds and speculation is that we'll see these cores clocked at 3.0GHz towards the end of this year or the beginning of 2003. It also allowed Intel engineers to cram in a remarkable 55 million (60 nanometer) transistors, some 12 to13 million more than was found on the Willamette core. Despite all this Intel were still able to reduce core dimensions from around 217 square millimeters to a miserly 146 square millimeters.

Just for reference lets's take a quick look at how this fits in with the current range :

	Pentium® 4 Processors in the 423-package	Pentium 4 Processors in the 478-pin package
		Pentium 4 Processors with 256 KB cache	Pentium 4 Processors with 512 KB cache
Micro-architecture	Intel® NetBurst™ micro-architecture	Intel NetBurst micro-architecture	Intel NetBurst micro-architecture
Operating frequencies (GHz)	1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, and 2 GHz	1.5, 1.6, 1.7, 1.8, 1.9, and 2 GHz	1.6A, 1.8A, 2A, and 2.2 GHz
Manufacturing process	0.18µ	0.18µ	0.13µ
L2 Cache size	256K, on-die	256K, on-die	512K, on-die
Socket Type	PGA423 423 pins	mPGA478B 478 pins	mPGA478B 478 pins
Memory Type	PC800/600 RDRAM PC133 SDRAM	PC800/600 RDRAM PC133 SDRAM DDR 200/266 SDRAM	PC800/600 RDRAM PC133 SDRAM DDR 200/266 SDRAM
Chipset	Intel® 850 chipset Intel® 845 chipset	Intel 850 chipset Intel 845 chipset	Intel 850 chipset Intel 845 chipset
Processor Core Voltage (Vcc_core)	1.75 V	1.75 V	1.50V
System Bus Speed	400 MHz	400 MHz	400 MHz
Dual Processor Support	No	No	No

Other than the things we've covered nothing else has changed. The core remains identical in its design and function and offers the same benefits found on its older brothers. I'm not going to launch in to a full on techno-rant here so let's just cover the basics.

Features	Benefits
Processor Core Speeds Up to and beyond 2.2 GHz	Maximum performance for a wide range of emerging Internet, PC and workstation applications
Intel® NetBurst™ Microarchitecture, including: 400-MHz System Bus	High bandwidth between the processor and the rest of the system improves throughput and performance
512-KB L2 Advanced Transfer Cache	Enhances performance by providing fast access to heavily used data and instructions
Hyper-Pipelined Technology	Extended pipeline stages significantly increase overall throughput
Streaming SIMD Extensions 2	144 new instructions accelerate operation across a broad range of demanding applications
Rapid Execution Engine	Arithmetic Logic Units run at twice the core frequency, speeding execution in this performance critical area
128-Bit Floating Point Port	Floating Point performance boost provides enhanced 3D visualization and scientific calculation
SIMD 128-bit Integer	Accelerates video, speech, encryption and imaging/photo processing
Execution Trace Cache	Greatly improves instruction cache efficiency, maximizing performance on frequently used sections of software code
Advanced Dynamic Execution	Improved branch prediction enhances performance for all 32-bit applications by optimizing instruction sequences

Bandwidth -

The Pentium 4 works on a "quad pumped" internal bus. That is, although the system bus runs at 100MHz, it is "multiplied" by a factor of four. This means data is transferred internally at an incredible 400MHz. The upshot of this is that the P4 has a full 3.2GB/Second of bandwidth, totally eclipsing Athlon's maximum of 2.1 GB/Second

Hyper Pipeline -

To enable it to push processor speeds beyond the 1GHz ceiling encountered using its older P6 architecture, Intel raised the pipeline stages from 10 in the PIII to 20 in the P4. Occasionally however, data in that pipeline will need to be flushed and the more stages there are, the more data gets flushed (up to 126 instructions in fact) and the longer it takes to refill. Use of the execution trace cache aims to keep these occurrences to a minimum, but when they do occur the delays can be significant, in processor terms at least.

SSE2 -

SSE2 adds 140 new instructions to the original SSE set. Some will claim that SSE2 is nothing more than SSE should have originally been, but regardless of that its power and flexibility is now huge. This is probably just as well because one of the other parts of the P4 architecture to hit the operating theater floor when it underwent its fat reduction operation was one of the two floating point units. This is why we often see such average FPU performance when running code that can't compensate for this deficit by using SSE2.

The Rapid Execution Engine -

At the heart of the rapid execution engine lie two double pumped ALUs and two double pumped AGUs. These operate at twice the core frequency but are only able to cope with micro-ops. More complex instructions need to be channeled through the single slow ALU, and this actually accounts for the vast majority of data handled.

 

<<< Back to Review Index | Next Page, Installation and cooling>>>

Home

1 :: 2 :: 3 :: 4 :: 5 :: 6 :: 7 :: 8