[hewee]

Plus with each defragger having its own way of laying out the files the 1st time you use it may take a very long time because it may move almost everything on your computer around.

I can defrag at boot up the system files and so on that are locked when windows run.
You can really tell how it speeds thing up.

But how often should you defag at boot?


I use PerfectDisk 10 Pro.

[DorianVonRichter]

Originally Posted by Elvandil

You'l probably find people who disagree (especially all those fake articles about defragging posted all over the net by the good people at Diskeeper), but I'd like them to prove it.

Have a good one.

Registry cleaner? You may not see the bad effects until it is too late. The reason there are so many of them is that they are easy to write. All one need is a set of templates to compare the keys to. If they don't match, they are an "error". Too bad the templates can't possibly cover all the possibilities. The real problem with cleaners is that they are a risk and do absolutely no good whatsoever.

http://forums.techguy.org/6657814-post1.html
http://forums.techguy.org/all-other-...-computer.html
http://www.edbott.com/weblog/?p=643
http://forums.techguy.org/all-other-...gcure-any.html
http://forums.techguy.org/all-other-...y-cleaner.html
http://forums.techguy.org/all-other-...g-cleaner.html
http://forums.techguy.org/windows-vi...file-lost.html

Even if registry cleaners actually worked and did what they are supposed to safely, what would be the net gain?

I have over 800,000 entries in my registry. So let's say 1 million as a good approximation. So if we also assume that the "speed" of the registry is an inverse relationship, in other words, if it were half as large, it would be twice as fast, how much would this cleaner actually speed things up?

If it found 5,000 "errors" (they aren't really errors, but let's assume they are since the cleaners do), then the increase in the speed of the registry would be 5000/1000000, or 0.005, or 1/2%. So in other words, by running a registry cleaner and remocving 5000 entries (which is more than most find) and risking serious damage to your installation and programs, you have succeeded in increasing the speed by 1/2 of 1% (if it were even true in the first place that registry access speed is dependent on size in this way, which it isn't).

Does that seem worth it to you?

Registry "defraggers" ("compactors") may actually improve your registry access speed, however. These tools do not hack out possibly needed entries like "cleaners" do. They simply rebuild the registry, leaving out blank space and reducing the size. Auslogics may be one of the best since it does the compaction offline on the next machine boot. But it is no longer free.

Free registry defragmenters (compressors):

NTREGOPT - http://www.larshederer.homepage.t-online.de/erunt/
RegCompact.NET - http://www.aplusfreeware.com/categor...egCompact.html
WinASO RegDefrag - http://www.winaso.com/
Free Registry Defrag - http://www.registry-clean.net/free-registry-defrag.htm

You can benchmark your registry access speed to compare it at different times, or after a "defrag" ("compaction") to see if any real improvement in access speed occurred (use identical conditions). The output of regbench HKLM -auto looks like this:

Benchmarking registry root:
HKEY_LOCAL_MACHINE [HKLM] hKey: 0x80000002
/
== Keys in hive : 359169
== Enumeration time : 135625 ms (135.63 secs)
== Total accesses : 100000
== Total access time : 8609
== Time per access : 0.086090 ms
== Keys per second : 11615.75
== Total bytes read : 10000046
== Time to read all : 10157 ms
== Time per byte : 0.001010 ms per byte
== Bytes per second : 990099.00

Alright, no more registry cleaner in CCleaner. I now have two questions for you. First, would YOU recommend a registry defragger? Second, if so, which free one would YOU recommend?

[Kenny94]

Elvandil, you left this one out:

http://miekiemoes.blogspot.com/2008/...eaking_13.html