Linked by Thom Holwerda on Mon 8th Feb 2010 23:55 UTC
Thread beginning with comment 408474
To view parent comment, click here.
To read all comments associated with this story, please click here.
To view parent comment, click here.
To read all comments associated with this story, please click here.
Also, just because a system has ACPI bios doesn't mean you're machine will adjust itself (scale back the CPU and/or FSB when not needed, etc). This, too, typically requires a driver or utility as the os itself doesn't often handle it.
Fair enough... this was one major lacking feature in XP (likely present in W7) - I do recall installing "Intel Speedstep" on several of my laptops in order to reduce CPU performance.




Member since:
2008-07-15
ACPI is standardized in theory, but hard-coded tables are only one of the issues. Some ACPI bioses behave badly, either not responding to certain functions or returning bad information. Typically these machines come with Windows drivers that compensate for it. I'm not talking about the crapware power management utilities but the driver component itself. Also, just because a system has ACPI bios doesn't mean you're machine will adjust itself (scale back the CPU and/or FSB when not needed, etc). This, too, typically requires a driver or utility as the os itself doesn't often handle it. In the case of Windows this is taken care of by installing drivers, in the case of Linux and other *NIX oses you need to configure the CPU frequency scaling and FSB scaling yourself if there's no utility to do it for your machine. ACPI is a standard at least on paper, but it merely provides the os a way to communicate with the ACPI bios and through there control the system's power management. It doesn't do much on its own. This is in contrast to APM which was used before ACPI became commonplace, which handled a lot in the bios itself and much less through the os. Having dealt with both of them, I'd say I prefer ACPI (even buggy ACPI) over APM.