I had high expectations for the fully-loaded review unit Dell sent us. The Core 2 Extreme X7900 is a rare and special processor in that it will not work in most Santa Rosa notebooks; its TDP is higher than most notebooks’ cooling solutions can handle. The standard Core 2 Duo mobile processor has a 35W TDP in comparison to the X7900's 44W. A heavy-duty heatsink is needed to handle its high heat output.
The real advantage to the Intel Extreme processors is their unlocked multiplier, so overclocking is simple. I ran each benchmark that involved scoring the processor twice; the first time with the processor at its stock 2.8GHz clockspeed, and the second time overclocked to 3.4GHz via the BIOS. The maximum overclock is 3.4GHz. Dell allows the end user to set 2.8GHz (stock), 3.0GHz, 3.2GHz, and 3.4GHz via the BIOS, but they recommend that the machine be run with 2.8GHz.
Testing notes: Prior to testing, I defragmented the hard drive and did a Windows Update. I installed the latest Nvidia drivers (169.04) for the video cards. All frames per second (FPS) benchmarks were done with FRAPS.
System Performance Benchmarks
Windows Experience Index
| 2.8GHz:  | 3.4GHz  |
Performance Difference (Stock vs. OC) (Processor): 0.00%
Oddly enough overclocking the processor did not help the WEI score at all. However, 5.7 is dangerously close to the highest-possible score of 5.9.
PCMark05
PCMark05 Comparison Results
| Notebook | PCMark05 Score |
| Dell XPS M1730 (Core 2 Extreme X7900 2.8GHz, Dual Nvidia GeForce 8700M-GT 512MB) | 7,148 PCMarks |
| Dell XPS M1730 (Core 2 Extreme X7900 overclocked to 3.2GHz, Dual Nvidia GeForce 8700M-GT 512MB) | 7,868 PCMarks |
| Sager NP9260 (2.66GHz Core 2 Duo E6700, 2x Nvidia GeForce Go 7950GTX video cards with 512MB DDR3) | 6,871 PCMarks |
| Alienware m5790 (1.83 GHz Core 2 Duo, ATI X1800 256MB) | 4,411 PCMarks |
| Fujtisu Siemens Amilo Xi1554 Review (2.0GHz Core 2 Duo, ATI X1900, Windows XP) | 5,066 PCMarks |
| Fujitsu LifeBook N6420 (2.00GHz Core 2 Duo, ATI X1600) | 4,621 PCMarks |
| Fujitsu LifeBook N6410 (1.66GHz Core Duo, ATI X1400) | 3,487 PCMarks |
| Sony VAIO SZ-110B in Speed Mode (Using Nvidia GeForce Go 7400) | 3,637 PCMarks |
| Asus Z84Jp (2.16GHz Core 2 Duo, Nvidia Go 7600) | 4,739 PCMarks |
| Asus V6J (1.86GHz Core Duo T2400) | 3,646 PCMarks |
| Alienware M7700 (AMD Athlon FX-60, Nvidia Go 7800GTX) | 5,597 PCMarks |
| 2.8GHz  | 3.4GHz  |
Performance Difference (PCMark score): 9.15%
The overclocked processor improved overall performance by about 9%. The M1730 is already blazing fast so it is doubtful whether the extra increase in CPU power will be noticed while running everyday applications. For more processor intensive applications, a nearly 10% gain is significant.
Processor Performance Benchmarks
wPrime
wPrime is a program that forces the processor to do recursive mathematical calculations, the advantage of this program is that it is multi-threaded and can use both processor cores at once, thereby giving more accurate benchmarking measurements than Super Pi.
| Notebook / CPU | wPrime 32M time |
| Dell XPS M1730 (Core 2 Extreme X7900 2.8GHz) | 29.477s |
| Dell XPS M1730 (Core 2 Extreme X7900 3.2GHz) | 29.251s |
| Sony VAIO NR (Core 2 Duo T5250 @ 1.5GHz) | 58.233s |
| Toshiba Tecra A9 (Core 2 Duo T7500 @ 2.2GHz) | 38.343s |
| Toshiba Tecra M9 (Core 2 Duo T7500 @ 2.2GHz) | 37.299s |
| HP Compaq 6910p (Core 2 Duo T7300 @ 2GHz) | 40.965s |
| Sony VAIO TZ (Core 2 Duo U7600 @ 1.20GHz) | 76.240s |
| Zepto 6024W (Core 2 Duo T7300 @ 2GHz) | 42.385s |
| Lenovo T61 (Core 2 Duo T7500 @ 2.2GHz) | 37.705s |
| Alienware M5750 (Core 2 Duo T7600 @ 2.33GHz) | 38.327s |
| Samsung Q70 (Core 2 Duo T7300 @ 2.0GHz) | 42.218s |
| Acer Travelmate 8204WLMi (Core Duo T2500 @ 2.0GHz) | 42.947s |
| Samsung R20 (Core Duo T2250 @ 1.73GHz) | 47.563s |
| Dell Inspiron 2650 (Pentium 4 Mobile 1.6GHz) | 231.714s |
| 2.8GHz  | 3.4GHz  |
Performance Difference (32M): 00.77%
Performance Difference (1024M): 10.44%
The effect of the higher-clocked processor is slightly more pronounced in wPrime, which is almost entirely CPU-dependent.
Hard Drive Performance
HDTune 2.53

The HDTune result for the dual 200GB 7200RPM Hitach hard drives in RAID 0 is stellar. This is the fastest hard disk performance we have seen in a notebook to date. Both drives were inaudible, even while reading and writing.
Synthetic Gaming Performance Benchmarks
3DMark06
3DMark06 Results and Comparison:
| Notebook | 3D Mark 06 Results |
| Dell XPS M1730 (Core 2 Extreme X7900 2.8GHz, Dual Nvidia GeForce 8700M-GT 512MB) | 8,536 |
| Dell XPS M1730 (Core 2 Extreme X7900 overclocked to 3.2GHz, Dual Nvidia GeForce 8700M-GT 512MB) | 8,572 |
| Sager NP5791 (Intel Core 2 Duo T7500 2.20GHz, Nvidia GeForce 8700M-GT 512MB) | 4,941 |
| Alienware M9750 (Intel Core 2 Duo T7600 2.33GHz, Nvidia go 7950 GTX video cards with 512MB) | 7,308 |
| Sager NP9260 (2.66GHz Core 2 Duo E6700, 2x Nvidia GeForce Go 7950GTX video cards with 512MB DDR3) | 9,097 |
| Alienware m5790 (1.83 GHz Core 2 Duo, ATI X1800 256MB) | 2,625 |
| WidowPC Sting D517D (Core 2 Duo 2.33GHz, Nvidia 7900GTX 512MB) | 4,833 |
| Apple MacBook Pro (2.00GHz Core Duo, ATI Mobility Radeon X1600 128MB) | 1,528 |
| Alienware Area 51 m5550 (2.33GHz Core 2 Duo, nVidia GeForce Go 7600 256MB) | 2,183 |
| ASUS A8Ja (1.66GHz Core Duo, ATI Mobility Radeon X1600 512MB) | 1,973 |
| Dell XPS M1710 (2.16GHz Core Duo, nVidia GeForce Go 7900 GTX 512MB) | 4,744 |
| 2.8GHz  | 3.4GHz  |
Performance Difference (3DMark score): 2.52%
Performance Difference (CPU score): 13.22%
While the overclocked processor did not make a large impact on the 3DMark score as a whole, it did increase the CPU score by over 13%.
Ageia PhysX RealityMark
The M1730 is unique in that it has an Ageia PhysX physics processor card. Ageia's RealityMark is a physics performance measurement tool which, according to Ageia, can be used to gauge overall gaming performance during a game with a high amount of physics calculations. It is based on Artificial Studio's CellFactor: Combat Training game and its Reality Engine.
| 2.8GHz  | 3.4GHz  |
Performance Difference (PhysX Hardware): 8.80%
There is a huge performance difference between software rendering and hardware rendering as the benchmark shows.