Remember the good old days when processors were listed by their clock speed? In many ways, it was a pretty poor way to differentiate processors – but it was a hell of a lot easier than the mish-mash of models we have now, especially from Intel.

In a way, it’s hard to fault Intel for moving from clock speeds to model numbers. The inferior Pentium 4 “NetBurst” architecture was born of the clock speed rating system. Instead of building the fastest processor, Intel instead decided to build a processor that was designed to achieve the highest possible clock speed, even if it actually didn’t do much in each clock cycle.

Sure, the AMD Athlon was faster, but Intel figured that the manufacturer that could put the largest number next to the processor won, because the punters wouldn’t know the difference.

Thankfully, Intel has pretty much abandoned the Pentium 4 architecture. Core 2 is closer in design to the Pentium III than it is the Pentium 4. Of course, Core 2 clock speeds were actually initially lower than P4 clock speeds (though the processors were much faster), so Intel had to abandon the clock speed naming schema.

All good, except we’ve been left with a processor model naming system that’s unintelligible to even the most hard core geeks.

Nowhere is this more noticeable than in notebook processors, where it’s hard to have any idea about the relative strengths and weaknesses of a given processor model.

Is the U7700 faster or slower than the T5550? Most people probably have no idea, and are likely to assume the U7700 is faster, because it has the highest number. The T5550 is actually faster, although it draws much more power. In fact, the letter at the beginning of the model number is much more important than the actual model number itself.

Understanding model names
With all that in mind, here’s a quick rundown of Intel’s current Core 2 mobile processor naming system:

Solo, Duo, Quad: this one’s easy – it’s a reference to the number of cores in the processor. A duo processor has two cores, which means that it’s rather like running two solo processors side by side. Quad-core has four cores.

Uxxxx -- ultra-low voltage. ‘U’ model processors typically run at less than 1V, which means that their power usage is very low, typically 10W or less at maximum load. But it also means that the clock speeds are also likely to be very low – they currently max out at about 1400MHz. ‘U’ processors are the slowest but most power efficient.

Lxxxx – low voltage. Faster than ‘U’ processors (up to 1866MHz currently), but also drawing more power (up to 17W).

Pxxxx – medium voltage. The next step up from L. Current ‘P’ processors can go up to 2533MHz and consume up to 25W.

Txxxx – standard voltage. Current ‘T’ series processors can go up to 2800MHz and consume a maximum of 35W.

Xxxxx – “extreme” processor. These are the fastest processors, with speeds up to 3066MHz, but they typically suck down 44W at max load.

Qxxxx and QXxxxx – quad-core mobile processor. Intel seems to have abandoned its own voltage naming schema for these. Both suck down 45W of power at peak usage. The ‘X’ in ‘QX’ is for extreme.

SUxxxx, SLxxxx, SPxxxx – small form factor ultra-low, low or medium voltage processors. An ‘S’ at the beginning of the name indicates that the processor 22 x 22mm package, as opposed to the normal 35mm x 35mm. It means little in terms of performance.

So what does the model number (the ‘xxxx’) above mean? It’s hard to give a blanket answer to that, since the numbers are all over the place. But generally the first digit is an indication of the size of the L2 cache, as well as a differentiator between single and multi-core processors.

For example, the 9xxx processors typically have 6MB of L2 cache and two cores (four cores in 'Q' processors). 5xxx, 6xxx, 7xxx and 8xxx usually have smaller caches, while the 2xxx and 3xxx processors only have single cores. The higher the number, the more likely it is to have a larger cache and more cores.

Within those families, the last three digits are a relative indication of clock speed. For example, the T9400 and T9600 are pretty much the same, except for having different clock speeds (2533MHz vs 2800MHz). There may be some other minor differences, such as support for a faster front-side bus.

Finally we should mention processor families, which are often discussed on technical Web sites by their Intel codename. When Intel introduces a new manufacturing process or a different architecture, it gives the new processor family a codename.

‘Penryn’ is the current generation of mobile processor, and it’s manufactured using a 45nm process. Processors in Centrino 2 notebooks are all Penryn processors. The previous generation was ‘Merom’, which was manufactured using a 65nm process.

If you want a full breakdown of all the current Intel mobile and other processors, as always a good place to find them is Wikipedia. If you’re still confused about the relative specs of a given model, it’s your best bet.

Ads by Google