New
#101
Or, it could be that a couple of restarts are required before you get a settled down system - remember, things like Super Fetch, prefetch, Readyboost (if you have it) and the like are all working to help speed up the optimization of your booting process - a single restart after messing around with your system is not going to be an accurate reflection on your boot time, because in that single restart, things like locked files being removed during reboot, fetch algorithms being updated due to removed programs and services, etc. are going to slow your computer down.
If I had the time, I'd love to rewrite the script to allow for up to 9 different restarts (selectable by the user so if a user wanted only 3 iterations they could enter 3) to get a more accurate representation of the restart time, that would allow for things like statistical analysis and standard deviation parameters being taken into account....
For example, if I ran the test 7 times, and 6 times I got times in the high 60s and low 70s, but one time I got a 99, it would statistically determine that the 99 was off in left field and discard that value, and give a better average that showed a more true representation of the restart time.
I know that eventually I'll be able to do something like this with F# (or a combination of VB and F#) but for now that is simply taking on too much to do with this little script.
I suppose I can look at the script to make it user interactive and run the test on a set of iterations, though....
What statistic?
As John mentioned, you usually need a couple of restarts for Windows to settle down after changes.
I usually do a few reboots to find work out my average and I noticed that enabling all four cores gave me a quicker average reboot time overall.
I just save the best speeds to post