C.Joseph Drayton said:
Blake,
You could try some 'real' world test. Take a drive that has had a
large number of installs/uninstalls and test if for speed where a
large number of small DLLs are loaded and unloaded as needed. Take
that drive and run a registry cleaner on it properly and do the same
test, you will see that their is an increase in speed.
As to risk, one should never allow a registry cleaner to automatically
remove items. A person should look through the list to confirm that
the items the cleaner has flagged as no longer necessary are in fact
no longer necessary.
I contend and will always favor that users should learn how to
properly maintain there computer . . . which means learning what the
registry does and how it is being used by applications. There is risk
in anything but the risk diminishes when one equips themselves with
knowledge.
Sincerely,
C.Joseph Drayton, Ph.D. AS&T
CSD Computer Services
Web site:
http://csdcs.site90.net/
E-mail: (e-mail address removed)90.net
That's true CJ, but it's going to fall on deaf ears or minimum ignorant
ones with questionable intentions.
In addition, the better ones, such as both of the ones I use right now,
also give you choices such as 1.Let me show it where the missing data is
that it couldn't find, 2., Remove the entry because I don't use that app
any more, 3., Try to fix it manually, meaning, jump to the registry
entry to look around in it, 4, Don't do anything, and finally, 5, Ignore
& don't report this issue in future scans. Seems like I missed
something, but that's at least most of it.
They both define what will be edited, removed, added or otherwise
manipulated based on its own search of the disk drives and possible
solutions it found. Oh yeah, it shows a severity level too. These are
the specific things that make some of them great tools even for newbies,
because they use language that most are goign to understand.
I do admit that I don't delete registry change stores until a few
weeks after the changes just in case, but it hasn't disappointed me yet.
They take so little space it's usually months before I actually delete
them and they always end up in my archives the next day anyway, so ... .
Although the scenario you gave is a good one, it's fairly possible
for the time differences to not be very substantia, depending on a lot
of variables. In most cases it won't be the increase in speed that will
be as noticeable, IME anyway, as the inprovement in load and cpu
intensive applications. I don't mean there will be NO difference, just
that, amongst all the other things in a poorly maintained machine, it
may not be very noticeable in the overall scheme of things.
Just for grins, on a machine I was about to rebuild, I once used
Regedit to export the entire registry and then ran that resulting .REG,
twice, pulling back in all of those same entries twice. It begs the
question why it takes a registry cleaner to notice multiple duplicate
entries, especially since the registry is a database, but ... that's how
it is.
That was on early XP, SP1 I think. The results were not only
noticeable, but boot times more than doubled, shut downs took forever,
and the odd long pauses cropped up here and there on the machine. I ran
out of time and didn't get to verify the "why"s of it all. I have to
assume the besides all the extraneous entries I created, that it also
sort of baffled the OS on what improprieties were also present in the
registry. So if one registry search "miss" had to go thru the 20S
timeout, now it went thru it twice or thrice, and so on.
Computers are stupid: They can only do what they're told to do, not
what we meant to do<g>.
Regards,
Twayne
. ut