DIY

Registry Cleaners: Are they a good thing or possibly bad?

Many kinds of malware keep rearing their ugly heads because of some embedded trigger planted in the Windows Registry. Can registry cleaners that are designed to remove these things actually do more harm than good?

Many kinds of malware keep rearing their ugly heads because of some embedded trigger planted in the Windows Registry. Can registry cleaners that are designed to remove these things actually do more harm than good?

-------------------------------------------------------------------------------------------------------------------

I've seen it happen on more occasions than I can count. A corporate or home user runs a registry cleaner in an attempt to remove some malware, only to find that Windows behaved worse after the registry was cleaned than it did before. Or perhaps it was run in an attempt to simply clean up some junk files so Windows might perform better, only to discover it ran worse -- or possibly not at all.

Although there are a lot of good registry cleaners out there, there are just as many -- or maybe even more -- that simply aren't worth the time (or money). In fact, some of them are actually designed to plant malware, not clean it out. Some registry cleaners are offered as free downloads, while others have a price tag attached to them. Suffice it to say, not all registry cleaners are created equal.

I will admit that I've never been a huge fan of registry cleaners, even the ones we might consider good. They've certainly gotten better over the years, but in the earlier days of Windows it was an exception to find a good one. Even the ones that were reviewed to be good would often create some unintentional consequence that actually made things worse in the long run.

I've always been a believer of backing up the registry after the computer was initially configured pretty much the way it should be, and that replacing a corrupted registry with an older iteration was a better solution than attempting to clean it up. And an even better solution would be to reinstall Windows from scratch, the logic being that when the registry became corrupted, the time it takes to reinstall Windows is better spent than the time it might take to repair, replace, and reconfigure, especially considering a clean install would guarantee a smooth running system. But, like I said, registry cleaners have evolved like everything else, and there are probably some great products out there.

Sorry, but I don't have a recommendation for a good registry cleaner. What prompted this blog piece, however, was that very question someone recently asked me. I didn't have an immediate answer, except to recommend a search for product reviews and comments from people who've actually used the products. That's something I'll do myself, but since I also have the option of throwing the question out to my TR peers, that's exactly what I'm doing.

So please share your experiences -- and even your recommendations -- about the many registry cleaners that are available. Share the good, the bad, and the ugly, and perhaps by the end of the discussion a bright light will shine on the best registry cleaner out there.

129 comments
sc3reface
sc3reface

Recommended tool with many features been using this for quiet a while now

Mark A Demers
Mark A Demers

This is a great program to run once a day or more - it won`t harm your registry but the drawback is it won`t do as good a job as a paid one - i mainly use CCleaner to clean my system of cookies and some viruses that accompany them . Two thumbs up for CCleaner .

Mark A Demers
Mark A Demers

I`ve been using Registry Cleaners since i finally got online about 6 years ago and when i let my registry cleaner subscription laps my pc crashed within 2 weeks . I don`t know about you but as for me i can`t be without one. A very good one i`ve been using even has Microsoft and Adobe updates and more included in the scan to save time doing them separately. Registry Cleaners to me is are a vital part of keeping my pc up and running smooth. Most of all running and not crashing. Personally i like Error Fix because of the nice job it does and the extra features it has - when i ranked these tools i placed a value on customer service too and that is why i ranked Error fix # 2 - if they were a bit more reachable they would have been # 1 --like i said because of the extra features they were my personal choice. My top rated is RegTOOL because it has practically the same features , scanning ability and cleaning ability but has a better customer service. So take your pick --More features (Error Fix) or better customer service (RegTOOL) Those are my two top choices for now - things may change as i review more of them. Have a Great Day !

repairregistry
repairregistry

Many more advantages we know but the disadvantages is not possible to find a key which is removed by the program. Windows is not open source; therefore it is impossible for the registry cleaners to conform whether a key is unnecessary. It may sometime remove the key that is necessary and also it cannot resolve the problems of removing the required key. They may also cause instability in the system. http://www.repairregistrycleaners.com/

repairregistry
repairregistry

It's always a good idea to read some reviews of registry cleaners to find the best software for your computer. Because the registry is such an important and sensitive part of your computer, we recommend using only quality software when cleaning it. Repair Registry

phlashman
phlashman

I've used CCleaner for years, not only on my computers at home but on my customers as well. Throughout all that I've "never" had an issue with it...not even once! Its handy and easy to use. As far as "pay for" cleaners, "Tuneup Utilities" is an excellent program as well. Its more robust and can be used to correct a lot of other problems.

jeremial-21966916363912016372987921703527
jeremial-21966916363912016372987921703527

While I agree with the consensus about CCCleaner, I would like to add my lament regarding the fact that we need such tools. One of my duties at work is software packaging and delivery to end users. In doing so, I am appalled at the amount of crap that "legitimate" applications leave behind after an uninstall (directory files, shortcuts, registry entries, etc. etc. etc.). I'm not sure if it is ignorance or laziness on the part of the programmer, but I find myself writing scripts and embedding them into the MSI, just to get rid of the things they leave behind. Why is it so hard, when creating an application for market, to simply uninstall everything you put on?! It makes no sense to me.

jouni.flemming
jouni.flemming

It's now official: A new study shows that most registry cleaner products do more harm than good. The study shows many concerning trends, such as that 9 products out of 31 produced false positives making these products more potent in creating Windows errors than actually fixing them. And speaking of fixing of errors, the results of the first ever test of different products' ability to detect and fix actual, serious registry errors are even more alarming: only 1 product out of 31 was able to detect and fix a serious registry error. Another interesting and equally alarming result is that a whopping 18 out of 31 products actually left some parts of themselves behind after using the products' very own uninstallation option. Such badly working uninstallers are the root cause for many types of registry errors and the cumulation of left-over junk data. In other words, many of the tested products created the very same problems they were said to solve! These findings can lead to only one conclusion: Most Registry Cleaner products do more harm than good. And there is very little real scientific information available for consumers to distinguish the cold facts out of the marketing hype and buzz words. You can read the study here: http://www.macecraft.com/registry_cleaner_comparison2/

deepsand
deepsand

"[i]... replacing a corrupted registry with an older iteration was a better solution than attempting to clean it up. And an even better solution would be to reinstall Windows from scratch ...[/i]" Given the exceedingly dynamic nature of the Reg. Hive, both of these recommendations must, if followed, result in the loss of valuable information at best; at worst, critical & unrecoverable data will be lost. While I must allow that not all Reg. cleaners are worthy of consideration, the fact remains that maintenance of the Hive will eventually be required.

Sirgwain
Sirgwain

For once I would like to see one of the companies that produce this software to make one that works! I have tried several and none of them does the job and all find problems that others were supposed to "fix". I have even lost data because the cleaner deleted the wrong key(s) from the registry and I had to install the backup. A tip to the wise: backup, backup, backup before using one of these. Last, I would like to be able to go to a web site where and INDEPENDENT analysis of registry cleaners is done to get unbiased assessments, not sales pitches!

remanabat
remanabat

I've used different kinds of reg cleaners and as you've said, they don't seem to function at all. After using one reg cleaner, the other reg cleaner also find the same issues which means to say that the issues were not removed. Seems to be just a hype..

rdorris4
rdorris4

I've been using Registry Easy since Oct. 2008. I paid $44.90 for the program and a 3 year automatic upgrade license. I run it roughly every two weeks or so, and so far have not had any problems with it at all. In fact, I actually like it. It takes a lot of dead files out of my hard drive and, I think, actually lets my computer run a little faster. It's doing what it's supposed to do without any problems. I'm happy with it. rdorris4

mikifinaz1
mikifinaz1

Most people use tools as band-aids. Wrong. Prep is the heart and soul of keeping your system running. 1. Image OEM system to DVD-CD. 2. Modify to customized preference then save image. 3. Do backups, backup registry etc. etc. 4. Review any major changes as they happen before committing to them and plan a monthly review otherwise. Once you get into a problem, with this system you usually have a fall back position.

chaz15
chaz15

Use any reputable registry cleaner to back up the registry first, just in case. Then run reg cleaner. CCleaner seems absolutely fine,at least with XP.

kevin
kevin

I have been in systems since 1984 and I have learned along the way that there is no magic piece of software to fix a buggered system. The majority of messed up systems are caused by the end-user and they will always say "I didn't do anything". The remaining problems that foul a registry relate to software that fails to cleanup after itself during uninstall. Even Microsoft admits that system restore may not correctly rewrite the registry and can cause both program and system malfunction. While I agree that wiping and reinstalling will most likely take less time than trying to re-tune the computer to run correctly, I have found a much more effective method. Rather than invest money (hands down reject free-ware) in a reg cleaner, Buy imaging software such as Macrium Reflect. Set the system up as it should be for the end-user's purpose. Take a snap-shot of the system and if the machine gets fouled up...re-image it. It will take about 20 minutes to bring it back to a known good condition. Not only does it save time, especially for a large IT support group it also establishes a standard that can be more reliable. Oh, and some peace of mind for IT and the users. Kevin McDonald

melekali
melekali

...ERUNT (Emergency Recovery Utility which does an auto backup) & NTREGOPT (Registry Optimization).

melekali
melekali

...the only registry cleaner I trust that I have personally seen speed up Windows is Microsoft's free RegClean. Other than that, I clean it manually.

jimdrvr99
jimdrvr99

I have 4 computers in my home and they all have CCleaner installed, this is a free product and it not cleaners the registry but it will also clean and optimize old files. You can configure is to your spec's. FCleaner is another good one. I use it along with ccleaner on my personal computer and they both work great. I would recomemend both. Good luck in your search

msainsworth
msainsworth

I've used CC Cleaner for a long time and have found it the best around.

TheOlderEd
TheOlderEd

Actually, I use a couple of them. CCleaner has been named, and I find it a bit too aggressive, but still use it on occasion (such as when I have to remove Norton from a computer). But, I have also found that the registry fixer in System Suite is also pretty good at cleaning out "dead" links. Naturally, System Suite is not free. But, it has lots of tools built in that I also use from time to time.

Taoufik.bhaja
Taoufik.bhaja

Hello, Certainly,registry cleaners do affect windows stability.Some of them are dangerous to use for a neophyte.Do save your registry before any cleaning;it's safer if a problem appears later. "Tune up utilities" is a good optimization software that do clean registry in 1 click.I'm using it ,and it has never caused me a problem.But ,i think,the 2007 version(which i'm still using) is the best one, as it's simple in use and does'nt go TOO DEEP in optimisation and cleaning. Also, there's "JV16 Power tools",which is very performant,but dangerous to a neophyte .Take care if you use it.I've tried it ,and it's a very powerful tool.

pjbruce
pjbruce

Glary Registry Repair - The first and only cleaner I've ever found that actually improves the system and solves problems with outlook express etc.

deepsand
deepsand

1) The developers believe that the detritus left behind is benign in nature; and, it is therefore economically unproductive to invest resources in producing a product that removes such. 2) The developer believes that the application cannot completely & accurately determine all Reg. Hive entries which can be removed or altered without adversely affecting the OS and/or other applications. 3) The developer and/or his superior(s) do not attach a great import to this matter, in that a) with their product uninstalled, it's no longer of concern to them; and/or b) 3rd party applications, such as those here under discussion, are equal to the task & will be used.

pdx-man
pdx-man

This "study" is performed by Macecraft.com - which (SURPRISE!) makes the jv16 powertools software that WINS the comparison study!!! IF macecraft's software is so great, then they should reference an independent study, not their own comparison test. The fact that they win their own test study calls that study's validity into serious question.

deepsand
deepsand

The better question is how well suited are they for one so skilled.

MeadowsPV
MeadowsPV

Although ERUNT and NTREGOPT are not part of the registry cleaners group; I am glad to see that I am not the only one that [still] uses ERUNT and NTREGOPT. I have been tempted, often to remove it from the startup. Often, if I did, I learned thet it was a huge mistake, since it does a full bakup of the registry, by date, into a known location. As for any "Tool" registry cleaners come in many shapes and capabilities. I do Like CCleaner and recommend it to those that are "Administrators [with all noted chagrin]" of thier own machine[Big Exception - The work Place - let the IT department do thier job AND not make it harder!!!]. Like so many other tools, there are ones for every level of expertise. Several of the agressive cleaners and optimisers should be left in the hands of a person that can repair/restore what damage is/can be done with these agressive tools. The analogy might go like this: CCleaner is a tap hammer and good for use around the house resident. Other 'Tools' are like either a six pound sledge hammer, or a scalpel for surgery. What house resident would perfom demolition or self surgery, without the requisite knowledge, experience AND support systems to 'undo' what is/was done. As always, "The Right Tool For The Right Job [ by the right person]".

dwood
dwood

After cleaning the registry, is it also necessary to defrag the registry? I think that when one performs a new format and reinstall, it also of course had a defragged registry, which may be one reason reinstalls helps the most over registry cleaners.

plutchok
plutchok

I have used CCleaner a number of times without problems, and that's the main criterion for a registry cleaner. However, CCleaner seems a bit timid. (Advanced System Care also has been fairly reliable and unaggressive.) For a more aggressive cleaning I have generally been happy with Glary Utilities. However, the last time I used Glary, a few weeks ago (first time on with Vista), it also, surprisingly, wiped out Comodo Firewall's rule base, so now I have had to retrain Comodo Firewall to recognize all the trusted applications. Next time I'll try Comodo's new registry cleaner, and expect better results.

OnTheRopes
OnTheRopes

You are so right, dangerous to a neophyte. Even an experienced user can have problems unless they're very careful. JV16 is powerful and not to be taken lightly.

jouni.flemming
jouni.flemming

The fact that the authors of the study are also affiliated with one of the tested products is clearly stated in the study. That is also one of the reasons why the study only addresses objective, factual properties of the tested software, not for example ease of use or GUI etc. which can be more matter of taste. We (as in Macecraft Software) do not refer independent studies because no such study exists. All the so called "studies" online are simply marketing tactics of certain companies, not a single scientific study on this field existed before we started to do them (as far as I can tell, of course). In fact, in the very first study we made we urged others to do similar tests. No one did. Most likely because very few people have the required skills to perform these tests, and most of these people are already affiliated with some of these products. I, for example, am more than willing to offer my help free of charge to anyone interested in doing this kind of research.

deepsand
deepsand

For example, Registry First Aid does not simply flag entries as "bad," as implied by the published results, but, within each category of entry types, sub-groups its findings according to a "Green/Yellow/Red" safety scale. Furthermore, in addition to giving the user its recommended action, it also provides a drop-down list, for each flagged entry, of suitable alternative options. Lastly, the user has the ability to allow or disallow action at the individual entry level. Even if the user simply accepts the recommended actions, the app will not effect any changes other than those deemed safe, the "Green" ones, unless the user expressly forces such action. The published results of this study imply that RFA actually acts on all those items deemed to be "false positives. No doubt other apps have been likewise incorrectly portrayed.

allanrockwell
allanrockwell

This time around i think i have to write back to say that an uninstaller will help a lot here. I use Revo - and there are other good ones too - and i haven't found left over entries when i checked with a registry search. Some uninstaller will work at their best if they've "seen", monitored the installation. If not, you can reinstall, just for that purpose of uninstallation. And also the miracle cure for speeding up my old laptop was cleaning old Java versions with JavaRa. I can't remember for sure if i had used Revo first.

seanferd
seanferd

I've used that since it was Reg Cleaner. It was nice back in the day when it was free. It is powerful, but I don't think you can go too far wrong if you stick to the "Safe" list, and don't randomly mess around with the advanced functions.

seanferd
seanferd

Too much time, rather. And the default registry editor and its search function don't help, either. You end up running across the same entries repeatedly due to having chunks of one hive displayed within another. I agree, a lot of the organization of the structure makes little sense to me. That's a fairly comprehensive Wikipedia entry. The "advantages" listed at the end are pretty much what I've seen in various blogs or articles, so no point in looking for what I've read in the past. Frankly, I thing the spread out reg entries per app are harder to track than any INI file. All that was needed was an updated INI file capability. It wouldn't be so bad if, as you have stated in the past, the registry was for the OS only.

seanferd
seanferd

I've read a couple of articles by MS developers in the past. I'll see if I can find links. But basically, I don't think they have a really good reason, myself.

deepsand
deepsand

What, if any, was Microsoft's given reason(s) for introducing the Registry Hive, while continuing to not only support [i]ini[/i] files, but their using both [i]reg[/]i & [i]ini[/i].

seanferd
seanferd

I see ini files even in installation files. And yes, XP is full of them. You'll find them liberally sprinkled throughout Windows 7 as well.

deepsand
deepsand

Odd thing is, not only does XP still support .ini files, MS still uses them! Do a system wide search for "*.ini" and note the great number of results. On the machine I'm on at the moment, the count is 283 files! Noticed that Adobe had some sprinkled in there as well!

seanferd
seanferd

I've only ever come across a fistfull of apps that do this, or offer it. ini v. reg, the battle continues, now with more xml.

deepsand
deepsand

Don't let the Reg. Hive be used by anyone other than the OS; require apps. to maintain all stateful data in their own store, such that uninstalling entails no more than the deletion of the app's files.

seanferd
seanferd

I appreciate you taking the time to notice the question and reply. And thanks for writing wicked software as well. :-bd

jouni.flemming
jouni.flemming

There is just one Jouni. I changed my last name from the overly difficult "Vuorio" to the last name of my great-grandfather "Flemming" in last year. English sounding last name makes my life a lot easier :)

seanferd
seanferd

JV16? Or are you the same guy using different surnames? :D Personally, I've always liked the various JV16 tools the best. What an app suite that is.

jouni.flemming
jouni.flemming

>To repeat, in order to unequivocally state that something is a "false postive," you must first be able to assert that : 1) every entry must be either absolutely necessary & correct or not; and, 2) that such determination can be made with absolute certainty all of the time. To repeat, both conditions were satisfied. The comparison was only made based on 2000 items of synthethic, manually added data of the registry of which 50% were manually verified to be registry errors, and 50% were manually verified not to be registry errors. As explained in the study as well. And as a tester I first created the errors using the general rules and my knowledge, I did not cherry pick anything to favor PowerTools, but for that, you of course only have my word. But these tests can be easily repeated, simply create you own set of test data and try it out yourself. And you will see the same results, many programs produce false positives. >And, were such assertion true, we would not have a plethora of registry tools, but only a single perfect one. Well, it's sad but if you haven't really noticed, most registry cleaner products are not very high quality and their developers are only interested of making some quick money than really develop the cleaning engine.

deepsand
deepsand

To repeat, in order to unequivocally state that something is a "false postive," you must first be able to assert that : 1) every entry must be either absolutely necessary & correct or not; and, 2) that such determination can be made with absolute certainty all of the time. I.e., there must first be a set of absolutely complete & accurate rules that can be used to judge each and every entry at all times. And, were such assertion true, we would not have a plethora of registry tools, but only a single perfect one. To create synthetic entries that you deem to be "good" but were "possible false positives" does not perforce mean that an at arms length observer possesses sufficient information to discern such from real errors. That jv16 performed so well owes to the fact that the tester and the tested had access to the same definitions of "true errors" and "false positives." I.e., the tested had advance knowledge of both the questions & the correct answers.

jouni.flemming
jouni.flemming

>Unless, that is, you assert that every entry must be either absolutely necessary & correct or not, and that such determination can be made with absolute certainty all of the time. The results of each tested registry cleaner were verified using synthetic data added to the registry. Set one of the data contained registry errors, set two contained data that ver possible false positives (i.e. data that might look invalid but is in fact valid). RFA listed numerous items of set two, meaning its registry cleaner engine is flawed. All this was also explained in the study itself.

deepsand
deepsand

Unless, that is, you assert that every entry must be either absolutely necessary & correct or not, and that such determination can be made with absolute certainty all of the time. In fact, were such assertion true, we would not have a plethora of registry tools, but only a single perfect one.

jouni.flemming
jouni.flemming

>The published results of this study imply that RFA actually acts on all those items deemed to be "false positives. The fact of the matter is that RFA listed false positives in its results. It is not relevant whether it marked these items Green, Yellow or Red, as the test was about the quality of the actual registry cleaner engine. The product's engine listed numerous false positives as errors, a false positive is a false positive no matter what color code you would use with it.