registry cleaner

[Curt Christianson]

Discussion of Registry "Cleaners"
http://aumha.net/viewtopic.php?t=28099

--
HTH,
Curt

Windows Support Center
www.aumha.org
Practically Nerded,...
http://dundats.mvps.org/Index.htm

| what if the registry I want to delete is from a program that I deleted
using
| the Add/Remove option??
|
| "Pegasus (MVP)" wrote:
|
| >
| > | > > hello i need help to solve this problem in windows live onecare scaner
was
| > > been found 2 invalid registry ,and 1 i use at long time oneclik that
one
| > > found 166 invalid registry ,i download uniblue registryboost and we
found
| > > 491
| > > invalid registry ,this all about becouse my computer crash , programes
| > > stop
| > > midle of tasks,is any thing i can change for improve my computer
| > > performance
| > > ,many thanks.
| >
| > I suspect that your computer's unstable performance is caused
| > by you using a registry "cleaner". Do not use them. At best they
| > do not deliver any benefit; at worst they cripple your machine.
| > In your case you probably have to reload Windows and re-install
| > all applications. Remember to save your important files and your
| > EMail to an independent medium first.
| >
| >
| >

 

[Pegasus \(MVP\)]

Stop watch. Dozens and dozens of machines. I just cleaned up a computer
that took TEN MINUTES to boot up. Now it boots up in 46 seconds. Enough of
a benchmark for you? I removed almost a THOUSAND errant registry entries.

Unless you supply full details, we have to take your word for it
since we cannot repeat your tests. It therefore becomes an
issue of belief rather than scientific proof.

 

[Alias]

Unless you supply full details, we have to take your word for it
since we cannot repeat your tests. It therefore becomes an
issue of belief rather than scientific proof.

You know it's impossible for me to provide you with all the registry
entries I removed. The computer isn't mine but a client's who lives
cross town. Whether you believe me or not isn't material to me. I know
both registry fixers work so I use them on every machine I encounter and
have been doing this for years with no problems. If that isn't good
enough for you, well, so be it.

I will, however, challenge your FUD regarding registry cleaners every
chance I get.

 

[Stephen]

Unless you supply full details, we have to take your word for it
since we cannot repeat your tests. It therefore becomes an
issue of belief rather than scientific proof.

This is getting silly. You may well be correct in your assertion that
registry cleaning is unnecessary. However, nobody could supply full details
of such a test so that it could be repeated (in a 'scientific' fashion), as
this would entail knowing exactly what state the computer was in before
cleaning.

We do not need to descend into a 'religion versus science' argument. I would
rather ask "Alias" (on a non-confrontational basis) if he has any idea
whether it is the number of errant entries that makes such a difference, or
some particular entry or type of entry, or what. If we could make progress
in narrowing down the source of the reported performance improvement,
perhaps we could learn something useful.

As one who doesn't know that much about registry entries, I just wonder if
one could be trying to access a (now non-existent) mapped drive, or
something similar, and so waiting for a timeout. If this kind of behaviour
is possible, presumably removing this particular registry entry could have a
significant effect on performance, especially boot-up time. I would welcome
advice on this.

Stephen

 

[Alias]

This is getting silly. You may well be correct in your assertion that
registry cleaning is unnecessary. However, nobody could supply full details
of such a test so that it could be repeated (in a 'scientific' fashion), as
this would entail knowing exactly what state the computer was in before
cleaning.

We do not need to descend into a 'religion versus science' argument. I would
rather ask "Alias" (on a non-confrontational basis) if he has any idea
whether it is the number of errant entries that makes such a difference, or
some particular entry or type of entry, or what. If we could make progress
in narrowing down the source of the reported performance improvement,
perhaps we could learn something useful.

As one who doesn't know that much about registry entries, I just wonder if
one could be trying to access a (now non-existent) mapped drive, or
something similar, and so waiting for a timeout. If this kind of behaviour
is possible, presumably removing this particular registry entry could have a
significant effect on performance, especially boot-up time. I would welcome
advice on this.

Stephen

I don't know if it was a particular entry or combination of entries. I
do know that it booted up much quicker after removing all those entries.
Removing 50 viruses didn't improve the boot up time as much as removing
the registry entries, though. Over 4000 fragmented files being
defragmented helped boot up and performance as well but I did that last.

 

[Pegasus \(MVP\)]

This is getting silly. You may well be correct in your assertion that
registry cleaning is unnecessary. However, nobody could supply full
details of such a test so that it could be repeated (in a 'scientific'
fashion), as this would entail knowing exactly what state the computer was
in before cleaning.

We do not need to descend into a 'religion versus science' argument. I
would rather ask "Alias" (on a non-confrontational basis) if he has any
idea whether it is the number of errant entries that makes such a
difference, or some particular entry or type of entry, or what. If we
could make progress in narrowing down the source of the reported
performance improvement, perhaps we could learn something useful.

As one who doesn't know that much about registry entries, I just wonder if
one could be trying to access a (now non-existent) mapped drive, or
something similar, and so waiting for a timeout. If this kind of behaviour
is possible, presumably removing this particular registry entry could have
a significant effect on performance, especially boot-up time. I would
welcome advice on this.

Stephen

Let's introduce some clear thinking into this thread.
If Alias had a systematic approach then he would
report something like this:
1. I did a clean installation of WinXP.
2. I did a clean installation of Office XP.
3. I arranged it so that Word would auto-start.
4. I measured the time from when I turned on the
PC until Word would accept my typing.
5. I applied registry cleaner "xyz".
6. I repeated test #4. The time reduced from
85 to 65 seconds.
7. I observed similar reductions on a number of
other machines.
So far Alias has reported no such thing. It is therefore
impossible for me or anyone else to repeat his tests
or confirm the results. His reports are purely anecdotal
and therefore a matter of faith. Is there room for faith
in a PC?

 

[Gerry]

Stephen

As one who doesn't know that much about registry entries, I just
wonder if one could be trying to access a (now non-existent) mapped
drive, or something similar, and so waiting for a timeout. If this
kind of behaviour is possible, presumably removing this particular
registry entry could have a significant effect on performance,
especially boot-up time. I would welcome advice on this.

Problems like this appear in Event Viewer so that you can pick and
correct what you know to be a problem without resorting to a Registry
Cleaner.


--



Hope this helps.

Gerry
~~~~
FCA
Stourport, England
Enquire, plan and execute
~~~~~~~~~~~~~~~~~~~

 

[Gerry]

Alias

The only way you can demonstrate your point is to test performance after
each stage of housekeeping. Unless a registry entry causes a timeout any
superfluous entry will have so little affect than you would not notice
any difference in performance. You mention removing 1,000 registry
entries but how many of those entries would have been accessed when you
restart the computer? Very few I suspect. If they are accessed to
provide a false start-up the best solution is to remove them manually
using Autoruns, having first confirmed what affect they have by
disabling rather than removing. The processor handles so many
transactions continuously that a few extra do not make a significant
difference. Event Viewer Reports generate errors / warnings if there is
a problem and it is logical to follow up from there.

Based on what you say you have been doing I do not see that you can
distinguish between the effect on performance of removing unnecessary
registry entries and defragmenting 4,000 files. Defragmenting in the
situation you describe will have a noticeable impact. Your approach may
not pose a significant risk in your hands but if the risk converts to
reality you can have a major problem which may be very difficult to
recover from.


--
Regards.

Gerry
~~~~
FCA
Stourport, England
Enquire, plan and execute
~~~~~~~~~~~~~~~~~~~

 

[Alias]

Alias

The only way you can demonstrate your point is to test performance after
each stage of housekeeping. Unless a registry entry causes a timeout any
superfluous entry will have so little affect than you would not notice
any difference in performance. You mention removing 1,000 registry
entries but how many of those entries would have been accessed when you
restart the computer? Very few I suspect. If they are accessed to
provide a false start-up the best solution is to remove them manually
using Autoruns, having first confirmed what affect they have by
disabling rather than removing. The processor handles so many
transactions continuously that a few extra do not make a significant
difference. Event Viewer Reports generate errors / warnings if there is
a problem and it is logical to follow up from there.

Based on what you say you have been doing I do not see that you can
distinguish between the effect on performance of removing unnecessary
registry entries and defragmenting 4,000 files. Defragmenting in the
situation you describe will have a noticeable impact. Your approach may
not pose a significant risk in your hands but if the risk converts to
reality you can have a major problem which may be very difficult to
recover from.

Hasn't happened in over five years and hundreds of computers.

 

[Gerry]

You're only commenting on the "risk" and ignoring the other points I
made.


--
Regards.

Gerry
~~~~
FCA
Stourport, England
Enquire, plan and execute
~~~~~~~~~~~~~~~~~~~

 

[Doug W.]

Yeah, there is. Get Crap Cleaner from www.ccleaner.com and use
the Registry function. Be sure to also use the back up feature
before removing registry entries. If you've installed and
uninstalled a slew of programs. Crap Cleaner will find
hundreds of entries that should be removed, regardless of what
*anyone* says. Once you've done that, you will see a marked
performance improvement.

-
BS...should notice little if any improvement in performance.

 

[Doug W.]

Hasn't happened in over five years and hundreds of computers.

-
I have used a good many so-called registry cleaners over the
last twenty years. I have found thousands of entries that I
removed. This did not improve performance one iota...but the
registry is sparkling clean and free of dross. All it did was
houseclean and made the various registry sections a bit smaller
so they took up less space on the harddrive.

Alias...no offense...but you are so full of BEANS.
-
Doug W.

 

[Zilbandy]

You know it's impossible for me to provide you with all the registry
entries I removed. The computer isn't mine but a client's who lives
cross town. Whether you believe me or not isn't material to me. I know
both registry fixers work so I use them on every machine I encounter and
have been doing this for years with no problems. If that isn't good
enough for you, well, so be it.

Likely running some anti spyware programs would have fixed it. I don't
like to see clutter in the registry either, but I've never noticed any
change in performance after cleaning my registry, and yes, I
periodically 'clean' my registry. It's sort of like cleaning the
backseat of my car. It doesn't make it run better, but it makes me
feel better. FYI, I'm using an older version of Registry First Aid
3.04 and I do backup registry with Erunt prior to using it.

 

[Ken Blake, MVP]

I don't
like to see clutter in the registry either, but I've never noticed any
change in performance after cleaning my registry, and yes, I
periodically 'clean' my registry. It's sort of like cleaning the
backseat of my car. It doesn't make it run better, but it makes me
feel better.

There one *very big* difference between cleaning the registry and
cleaning the back seat of your car. Cleaning the back seat of your car
can't result in your car not working.

FYI, I'm using an older version of Registry First Aid
3.04 and I do backup registry with Erunt prior to using it.

Glad to hear you do backups first (and Erunt is a good tool for that),
but you need to be aware that if the registry cleaner screws up badly
enough (and that's always possible) you can be left with an unbootable
computer.

You are taking a risk that as far as I'm concerned is unacceptable,
since you are taking it for what you yourself admit provides no real
benefit. But it's your computer and your choice, of course.

 

[Zilbandy]

Glad to hear you do backups first (and Erunt is a good tool for that),
but you need to be aware that if the registry cleaner screws up badly
enough (and that's always possible) you can be left with an unbootable
computer.

If I can't boot it in safe mode, then it's time to use the Acronis
True image boot disk and restore my last image. If I can't boot from
that, then I've more problems than just a screwed registry. Do you
ever wash your car? Why? Ohhh, you like it to look good! Why? Get my
point?

 

[Ken Blake, MVP]

If I can't boot it in safe mode, then it's time to use the Acronis
True image boot disk and restore my last image. If I can't boot from
that, then I've more problems than just a screwed registry. Do you
ever wash your car? Why? Ohhh, you like it to look good! Why? Get my
point?

As a matter of fact, no, I don't.

 

[Edward W. Thompson]

Stop watch. Dozens and dozens of machines. I just cleaned up a computer
that took TEN MINUTES to boot up. Now it boots up in 46 seconds. Enough of
a benchmark for you? I removed almost a THOUSAND errant registry entries.

Your assertions are patently false (in crude terms 'lies') and unfortunately
I think you know it. This thread is nothing but SPAM. I am suprised that
anyone has given you any credence.

 

[Stephen]

Let's introduce some clear thinking into this thread.
If Alias had a systematic approach then he would
report something like this:
1. I did a clean installation of WinXP.
2. I did a clean installation of Office XP.
3. I arranged it so that Word would auto-start.
4. I measured the time from when I turned on the
PC until Word would accept my typing.
5. I applied registry cleaner "xyz".
6. I repeated test #4. The time reduced from
85 to 65 seconds.
7. I observed similar reductions on a number of
other machines.
So far Alias has reported no such thing. It is therefore
impossible for me or anyone else to repeat his tests
or confirm the results. His reports are purely anecdotal
and therefore a matter of faith. Is there room for faith
in a PC?

OK. Maybe you're right - and I think that's likely. (Although, of course, a
real scenario would be hugely more complicated - clean installations are not
likely to be improved by cleaning!)

Let me ask you a similar question, though. Can you provide full details of a
registry key that is removed by a (specified) registry cleaner that causes
the PC not to boot? That's what has been alleged, but without details (which
may well be available, but I've never seen posted), such an allegation is
"purely anecdotal and therefore a matter of faith".

Stephen

 

[Pegasus \(MVP\)]

OK. Maybe you're right - and I think that's likely. (Although, of course,
a real scenario would be hugely more complicated - clean installations are
not likely to be improved by cleaning!)

Let me ask you a similar question, though. Can you provide full details of
a registry key that is removed by a (specified) registry cleaner that
causes the PC not to boot? That's what has been alleged, but without
details (which may well be available, but I've never seen posted), such an
allegation is "purely anecdotal and therefore a matter of faith".

Stephen

Sorry, I can't. Since I don't believe in registry cleaners I do not use
them, and my machines all boot the way they should. When a
registry-cleaner-prophet such as Alias will produce some hard and
reproducible evidence that registry cleaners do more good than
harm then I will review the issue. Up to now the message has
always been "Believe me, it did miracles on x hundred machines".
Strong faith, weak evidence.

 

[Stephen]

Sorry, I can't. Since I don't believe in registry cleaners I do not use
them, and my machines all boot the way they should. When a
registry-cleaner-prophet such as Alias will produce some hard and
reproducible evidence that registry cleaners do more good than
harm then I will review the issue. Up to now the message has
always been "Believe me, it did miracles on x hundred machines".
Strong faith, weak evidence.

Fine. My point is that your "belief" in how bad registry cleaners are is no
better founded than Alias's "belief" in their efficacy. There's nothing
wrong with "belief". Much of what we do in life isn't backed up by firm,
scientifically-repeatable proof, but is based on "belief", often founded on
some kind of evidence that falls short of proof.

As a matter of interest, I don't use registry cleaners either. I have done
so in the past and never experienced any problems. However, neither did I
see any improvement. I just think it would be better if BOTH sides of this
debate would put forward sensible evidence (not necessarily proof) for their
beliefs, rather than wild (and probably unsupportable) claims.

Stephen