A few days ago, the company purchased an electric power detector, which made a health check for its own notebook. Let's take a look at what factors in the notebook affect the overall power consumption.
Paste the sub-Configuration:
Notebook Model |
Dell XPS m1210 |
CPU type |
Mobile dualcore Intel Core Duo, 1866 MHz (14x133) |
Motherboard chipset |
Mobile intel Calistoga-G i945gm |
System memory |
2048 MB (DDR2-667 DDR2 SDRAM) |
Display Card |
Mobile Intel (r) 945 express chipset family (256 MB) |
Display |
12 "Broadband LCD |
Sound Card |
Sigmatel stac9220 A1-high definition Audio Controller |
Hard Disk |
Hitachi hts541612j9sa00 ata device (120 GB, 5400 rpm, SATA) |
Optical Drive |
Tsstcorp DVD +-RW TS-L632D ata Device |
Wired Nic |
Broadcom 440x10/100 integrated controller (192.30.0.74) |
Wireless Network Card |
Intel (r) Pro/wireless 3945abg network connection (192.30.2.21) |
Shutdown and standby
First test the power consumption in the shutdown state: Shut down, plug in the power adapter, the dial shows 0.7 W, it seems that the power consumption of the adapter is not low, if the long-term shutdown, it also consumes a lot of power in the plug-in (some people are too lazy to pay attention ). Test the energy consumption in the standby mode: 1.2 W; skip the test when the sleep mode is switched off.
Energy Consumption of wireless NICs
Start the system normally and run only a few frequently used software in the memory. The energy consumption fluctuates between 18 and 20 W, and the wireless network adapter is closed and immediately dropped to the 16-18 W range. It can be inferred that, the energy consumption of wireless NICS is about.
LCD energy consumption
I always thought that the LCD was a big energy consumer. This test also proved my idea that the energy consumption changed from the highest brightness to the lowest brightness, which was nearly 4 W. However, after that, I completely disabled the LCD, the energy consumption has not changed much. It can be seen that the LCD is completely disabled and the lowest brightness, and the energy consumption is basically the same, which is unexpected.
CPU consumption
CPU, mainly want to see how powerful the CPU speedstep technology intel has been advocating is. Designed several scenarios: 1. run the minimum number of applications, and the CPU usage is around 1%. Use swichspeed to adjust the CPU frequency and capture several groups of data. run super Pi, the CPU usage reaches 100%, capture a set of data. The test results show that the CPU power consumption and frequency are not closely related, and are proportional to the CPU load (workload). The larger the load, the higher the energy consumption. The test result is as high as 40 W. It can be seen that the CPU is indeed called an electric tiger, and the actual effect of the speedstep technology is far less than that of the promotion.
Hard Disk energy consumption
Next we will look at the hard disk. The test tool is HD tune pro 4.01.
In the case of frequent file reads, the energy consumption increases by 5 W; in the case of random sector reads and writes, the energy consumption increases by 7 W. It can be seen that in the random mode, the hard disk's magnetic arm is frequently moved, causing more power consumption.
Conclusion
In this test, apart from the integrated graphics card, you cannot test the power consumption of several major components separately. Summarize the test results:
Shutdown and sleep |
0.7 million |
Standby |
1.2 million |
Minimum Energy Consumption Under Windows Load |
18 W |
Maximum CPU consumption under load |
38 W |
Wireless Network Card energy consumption |
+ 2 W |
LCD energy consumption |
+ 4.5 million |
Minimum LCD energy consumption |
+ 0.5 million |
Hard drive energy consumption for frequent file reads |
+ 5 W |
Random sector read/write hard drive energy consumption |
+ 7 W |
In addition, there will be the optical drive part, and will be added later.
How to extend battery life
After so many tests, the goal is nothing more than one: how can we maximize the battery life, while at the same time contributing to the green environment.
The following are my suggestions:
1. Try to adjust the LCD to the lowest brightness;
2. the backend should run as few applications as possible. The test shows that this has the greatest impact on energy consumption and cannot be ignored;
3. Disable it immediately without using a wireless Nic;
4. Reduce the reading and writing of a large number of small fragment files. For example, when accessing the Internet through a browser, you can transfer the cache to the memory disk to avoid hard disk reading and writing.