I am always amused by "benchmarks" that measure how long it takes to shut down or start up a PC. Think about it: how do you accurately measure the time it takes to shut down a PC? With a stopwatch? You certainly can't do it with software, since the software won't be running when the shutdown is complete.
Similarly, how do you know when a PC has finished starting up? How do you know what time it was switched on? As far as I know there is no counter in the processor that says how long the CPU has been running. What we do have is an accurate clock, which can provide the number of seconds and milliseconds since midnight.Enter the Reboot Timer: a free utility that notes the current time, stores it in the registry, and then reboots the machine. Because there is a shortcut to the program in the Startup menu, once the PC restarts the program can compare the current time with the stored one, and calculate a time difference. The screen shot shows that it took my laptop 2 minutes and 32.09 seconds to reboot.Now I can use this to determine the time saving achieved by defrag programs or other software that claims to "improve" a PC's startup time. I can also use it to measure how long a boot time defrag takes.Update Wed 10 June: PerfectDisk 10 Professional on my Vista 64-bit machine improved the average boot time of the machine by 7.3%. I have nothing to compare that to, but it is an improvement of 6 seconds. More results will follow as they become available.
3 comments:
The only problem Donn is that bootup will be slowed down slightly by this app running at startup.
@DooGie: Good point, but this isn't a "speed up" program. The app is very lightweight: only 40kb for the EXE, excluding the VB runtime files. But the slowdown can't be measured, because the app itself is doing the measurements. %-) Think about it.
The boot time measured is only useful for comparison purposes with other reboots on the same machine. i.e. has my machine slowed down after installing Vista Service Pack 2, or after doing a full defrag?
So I'm using it in the benchmarking tests to do "before" and "after" comparisons.
You can't really use it to compare two different machines, except at the very general level (i.e. minutes, not seconds) and you have to have an automatic logon. Otherwise you're essentially measuring how long it took to type in the user password.
My laptop takes 152 seconds to boot, but FRAG averages at 86 seconds. So my machine is "slower" because I have to type in a hard drive password and a log on password. I don't know how much of the 66 seconds is used up in this way, but it doesn't matter because the one is a laptop and the other is a desktop and I would be worried if there wasn't a difference. ;-)
what about using event viewer instead, its a part of vista just click on application and services - microsoft - windows - diagnostic performance. Just suggestion maybe yours way is better...
Post a Comment