Dual Core Notebook CPUs Explored

Page 6 of 12:

Intel Core Duo, Core 2 Duo Models Compared

Swipe to scroll horizontally

Core Model	Model Number	Bus Speed	Clock Speed	L2 Cache	Voltage	TDP	Pricing
Core Duo	T2050/2060	FSB533	1.60 GHz	2 MB	1.1625 - 1.4 V	31 W	Row 0 - Cell 7
Core Duo	T2250	FSB533	1.73 GHz	2 MB	1.1625 - 1.4 V	31 W	Row 1 - Cell 7
Core Duo	T2300	FSB667	1.66 GHz	2 MB	1.1625 - 1.4 V	31 W	$ 241
Core Duo	T2350	FSB533	1.86 GHz	2 MB	1.1625 - 1.4 V	31 W	Row 3 - Cell 7
Core Duo	T2400	FSB667	1.83 GHz	2 MB	1.1625 - 1.4 V	31 W	$ 241
Core Duo	T2450	FSB533	2.00 GHz	2 MB	1.1625 - 1.4 V	31 W	Row 5 - Cell 7
Core Duo	T2500	FSB667	2.00 GHz	2 MB	1.1625 - 1.4 V	31 W	$ 294
Core Duo	T2600	FSB667	2.13 GHz	2 MB	1.1625 - 1.4 V	31 W	$ 423
Core Duo	T2700	FSB667	2.33 GHz	2 MB	1.1625 - 1.4 V	31 W	$ 637
Core 2 Duo	T5200	FSB533	1.60 GHz	2 MB	1.0375 V - 1.3 V	34 W	Row 9 - Cell 7
Core 2 Duo	T5300	FSB633	1.73 GHz	2 MB	1.0375 V - 1.3 V	34 W	Row 10 - Cell 7
Core 2 Duo	T5500	FSB667	1.67 GHz	2 MB	1.0375 V - 1.3 V	34 W	$ 209
Core 2 Duo	T5600	FSB667	1.83 GHz	2 MB	1.0375 V - 1.3 V	34 W	$ 241
Core 2 Duo	T7200	FSB667	2.00 GHz	4 MB	1.0375 V - 1.3 V	34 W	$ 294
Core 2 Duo	T7400	FSB667	2.16 GHz	4 MB	1.0375 V - 1.3 V	34 W	$ 423
Core 2 Duo	T7600	FSB667	2.33 GHz	4 MB	1.0375 V - 1.3 V	34 W	$ 637

Benchmarking Difficulties

It is relatively easy to compare desktop processors, due to the plethora of platforms available for AMD and Intel solutions. All you need is a consistent system setup for the AMD and for the Intel processor portfolio, both using the same graphics solution, hard drive, RAM size and appropriate timings, and the latest drivers. Then pick the fastest components available to create ideal environments for both parties. If you intend to measure power requirements as well you should also pick the same power supply for both systems - because efficiency varies from one product to the next - and that's basically it. With the system set up, all you need to do is define your benchmarks and benchmark settings, and get the processors you want to compare. In our case, creating Interactive CPU Charts for desktop processors takes approximately four weeks.

Many people have emailed us, asking that we add mobile processors to the Interactive Charts. It's a great idea and we've been thinking about it for quite some time, but not as easy as it might seem. On the one hand, we cannot use the existing desktop platforms, because mobile processors use different sockets. AMD's desktop Athlons utilize socket AM2 (940 pins), but the mobile Turion is based on Socket S1 (638 pins). Intel's desktop processors rely on Socket 775, while the mobile versions use Socket 479.

There are two reasons for the difference. First, mobile processors need to be as small as possible, which is why they come in a smaller package and even without a heat spreader (that's the metal cover on top of the processor die). Second, they require less power, which means that the processors will work properly with a smaller pin count (since most pins are used for power).

All of this means that a comparison between desktop and mobile CPUs is impossible with the existing platforms. The alternative is to think of a new set of test systems, which turns out to be difficult as well. While there are a few desktop motherboards that work with Intel mobile processors - a technique called Mobile on Desktop - we haven't found any standard or pseudo-standard motherboards that would run Turion 64 X2 processors. Our only option was getting a Turion 64 X2 powered notebook that would allow us to work with our standard components as much as possible.

As you will see on the following pages, we've assembled a Mobile on Desktop solution based on a Gigabyte board, to compare Intel Core Duo and Core 2 Duo processors. We've obtained a decent notebook platform to benchmark AMD Turion 64 X2 processors. This platform is a notebook, which we equipped with a high-performance Western Digital WD1500 Raptor drive to make results comparable to the Intel system.