Jump to content
Jonesy200916

Registry cleaners

Recommended Posts

this this stoped it from booting ?

Completely. I don’t know what it did to achieve that, but it fucked me right off too - it happened late at night and it was my only form of entertainment. To top that off, my DVD drive was poked so I couldn't reinstall/repair.

Share this post


Link to post
Share on other sites

I use CCleaner's registry cleaner, but I by no means consider it thorough, I've had Auslogics Registry Defrag completely stuff Windows 7 quite a few times.......use it at your own peril. I used to use JV16 registry cleaner with XP for years, with great results, but for Win7, I've found Wise Registry Cleaner to be excellent there's both a freebie and a paid-for version, which is only a trial.I've had no problems with it at all - there's also a Wise Disk Cleaner which is very good (again a free, and a trial/paid for version) I use both of these (the free versions) in conjunction with CCleaner

then I defrag with Auslogics Defrag

Share this post


Link to post
Share on other sites

I use CCleaner's registry cleaner, but I by no means consider it thorough, I've had Auslogics Registry Defrag completely stuff Windows 7 quite a few times.......use it at your own peril. I used to use JV16 registry cleaner with XP for years, with great results, but for Win7, I've found Wise Registry Cleaner to be excellent there's both a freebie and a paid-for version, which is only a trial.I've had no problems with it at all - there's also a Wise Disk Cleaner which is very good (again a free, and a trial/paid for version) I use both of these (the free versions) in conjunction with CCleaner

then I defrag with Auslogics Defrag

Interesting. We've been using Auslogics on 3 systems here every 2 weeks or so since Win 7 Beta came out. No problems yet.

 

I knew there was a reason I also updated those backup images regularly. :-)

Share this post


Link to post
Share on other sites

*EDIT* Now I've answered your questions, would you mind answering mine? Or are you contesting that it's just impossible to test registry cleaners, so we should just assume they work?

It is you who is setting up a test designed to get the result you want, not me.

 

The question is, does your environment duplicate what the average user (if there is such a thing), does to their PC over time?

 

Your answer is to list the things that have been installed over the past three months. Again, that is NOT starting with a known environment, and in no way attempts to duplicate what happens in the real world. I look at my kid's PC and they may go for three months not have added or removing ANYTHING, so what you are doing this step that you list provides no value.

 

You then say you are going to ask users what they have installed and removed.

 

You're joking right?

 

If I ask the families of the PC's I've had to fix what they had installed over the past year, there is no way they can remember half of it.

 

You asked me to be constructive, I was. I said grab 100 different programs, a combination of commercially available products and shareware/freeware, and install them, remove them upgrade them.

 

This is what people do to their PC's over time.

 

Now I picked the number 100 out of thin air. I don't know how many installs/upgrades/removals a typical PC goes through in a year, but 100 for a family PC wouldn't surprise me.

 

 

The bottom line is the PC's that I've seen that have benefited from a registry cleaner are owend by average families, where multiple users add and remove stuff, BUT nobody really "maintins" the PC. This would be very common I'd think.

 

So that's what needs to be duplicated. Instead, you seguest a benchmark where the number of application installs/updates/removals is unknown (apart from the past 3 months).

 

If you were seriously trying to put together a realistic benchmark, you wouldn't have such unknown variables. Instead it appears you design a benchmark to get the result you want.

 

I honestly think benchmarking it is a great idea, but you have to be serious about it first.

 

Do you know how many applications have been installed and removed from your client's PCs?

No, of course not.

The above is the crux of the problem - you are trying to benchmark the impact of a utility on a highly variable UNKNOWN environment. You're not serious. You need to define the variables so that their impact on the results can be observed. These include :

 

* Hardware configuration

* OS version and configuration/modifications

* Applications installed

* Applications removed

* Applications updated

* Failed updates/removals/installations

 

I think you would agree that ALL of the above can have an impact, but your benchmark scenario doesn't measure them. How can you interpret the results?

Edited by Mac Dude

Share this post


Link to post
Share on other sites

The above is the crux of the problem - you are trying to benchmark the impact of a utility on a highly variable UNKNOWN environment. You're not serious. You need to define the variables so that their impact on the results can be observed.

OK. So basically you want to create a synthetic environment that you hope might be like actual peoples PCs. The only way to do it properly would, of course, be to set up a bunch of PCs and monitor them fully for, let's say, five years. The synthetic install and uninstall programs method would be set to an arbitrary level (you say 100 programs on and off, which is entirely arbitrary) and would only reflect that synthetic benchmark.

 

I consider that as an effort to simply bypass reality and look for an unrealistic result - which, to put it your way, cannot be serious.

 

The first sentence I quoted described exactly the kind of people's PCs I'll be testing. Normal people, with normal computers, usually having been running for many years. I won't be testing one PC - I'll be testing as many as I can. I'll be getting all the information I can get. And most importantly they'll be multiple real, highly variable, unknown environments. I'm planning on testing in a real world environment. See the impact on actual PCs that have been run for years by real live people.

 

We obviously have different criteria. I care about the real world. You want to guess what reality will be like and attempt to synthesize that to a self-mandated level. I prefer my method. You, apparently, don't think the computers of the people you described this would benefit the most are suitable for testing. Which is confusing.

 

Tell ya what, though. You test your theory - get yourself an old computer and install and uninstall a bunch of stuff. Document it for us. Run your reg cleaner and defragger, and no other products (like actual disk defragger, clean out msconfig/HJT etc.). And do some measurements. I'll stick with real situations.

 

*EDIT* And for the record, my method does cover:

 

* Hardware configuration

* OS version and configuration/modifications

* Applications installed (current)

Edited by tantryl

Share this post


Link to post
Share on other sites

The above is the crux of the problem - you are trying to benchmark the impact of a utility on a highly variable UNKNOWN environment. You're not serious. You need to define the variables so that their impact on the results can be observed.

OK. So basically you want to create a synthetic environment that you hope might be like actual peoples PCs. The only way to do it properly would, of course, be to set up a bunch of PCs and monitor them fully for, let's say, five years. The synthetic install and uninstall programs method would be set to an arbitrary level (you say 100 programs on and off, which is entirely arbitrary) and would only reflect that synthetic benchmark.

 

I consider that as an effort to simply bypass reality and look for an unrealistic result - which, to put it your way, cannot be serious.

 

 

I never said to do the above, so saying I did is just setting up a false faulty argument you can pull down. Fine.

The first sentence I quoted described exactly the kind of people's PCs I'll be testing. Normal people, with normal computers, usually having been running for many years. I won't be testing one PC - I'll be testing as many as I can. I'll be getting all the information I can get. And most importantly they'll be multiple real, highly variable, unknown environments. I'm planning on testing in a real world environment. See the impact on actual PCs that have been run for years by real live people.

You haven't stated how many PC's, you haven't stated what environments they are coming from, just that they are your clients and therefore represent the average user. You expect me to just take that on faith. What if you test 2 PC's because that's "as many as you can". It meets your criteria, but not mine because it's not realistic.

 

The state of the PCs you start with is unknown, the number of PCs is unknown, yet you expect people to just accept that.

We obviously have different criteria. I care about the real world. You want to guess what reality will be like and attempt to synthesize that to a self-mandated level. I prefer my method. You, apparently, don't think the computers of the people you described this would benefit the most are suitable for testing. Which is confusing.

As I said, you are saying that the PCs you will be testing represent the real world yet you have no idea how they have been created modified or used. I spent about 7 years trained and performing server benchmarks and computer modelling, and there is no way I'd place any weight on a benchmark from an unknown environment.

 

You are assuming the individual PCs you will be testing represent the real world, yet you cannot define how they have come to be in the state they are in, and you cannot say how many PCs you will test.

 

Is it going to be a representative sample? It's a question you cannot answer, yet you accuse me of assuming things.

 

I'll say it a last time, I think your idea is a good one, but your methodology is based on assumptions which may or may not be flawed - you cannot know, yet you accuse me of making guesses. I guess your guesses are ok but mine aren't, I guess...

 

But please, post the results when you're done.

Edited by Mac Dude

Share this post


Link to post
Share on other sites

I'll say it a last time, I think your idea is a good one, but your methodology is based on assumptions which may or may not be flawed - you cannot know, yet you accuse me of making guesses. I guess your guesses are ok but mine aren't, I guess...

See, this is where I'm confused.

 

What am I guessing about?

Share this post


Link to post
Share on other sites

I found the best way to clean the registry was to buy a Mac...

 

...that cleaned it, totally!

 

;p~~~

Newb, you now need /etcCleaner.

Share this post


Link to post
Share on other sites

Mac Dude,

I wouldn't waste much time trying to prove anything to non-believers. Even when you post results, they try to twist the statistics around in favour of their argument. So this is just seeing what they want to see as far as I'm concerned.

 

You've seen the difference. I've seen the difference. More importantly, our customers have seen the difference and are happy with the results.

 

Let them experiment as they see fit and come to their own conclusions. At least that's better than blanket statements about how it doesn't affect anything when they have no experience to back it up. Until they actually run into a system where it makes a significant difference they're not likely to believe. That may or may not happen, and won't change what we've seen or our opinions about it either. :-)

Share this post


Link to post
Share on other sites

Mac Dude,

I wouldn't waste much time trying to prove anything to non-believers. Even when you post results, they try to twist the statistics around in favour of their argument. So this is just seeing what they want to see as far as I'm concerned.

 

You've seen the difference. I've seen the difference. More importantly, our customers have seen the difference and are happy with the results.

 

Let them experiment as they see fit and come to their own conclusions. At least that's better than blanket statements about how it doesn't affect anything when they have no experience to back it up. Until they actually run into a system where it makes a significant difference they're not likely to believe. That may or may not happen, and won't change what we've seen or our opinions about it either. :-)

My goodness, someone convinced a product they don't really understand how works, convinced it does work, and not happy that someone wants to test things. What a shock.

Share this post


Link to post
Share on other sites

This topic was an interesting read.

 

My experience is basically almost the last 10 years, doing pc repair, and fixing *a lot* of home users laptops. Many users with no clue how to maintain or look after a PC, or often even apply updates. Personally, i have never noticed an improvement with registry cleaners, except in very rare cases, such as the example master_scythe gave of overly large program lists. Even then, these issues are kind of irrelivant, I would think.

 

For actual performance, it has been(in my experience), more to do with lame OEM installed programs, malware, toolbars etc., and never actually the size of the registry.

 

I am not completely sure, but I would think that due to the way the registry works, the size would not be a factor for performance. The applications will request the keys that they need, which is handled by windows, rather than actually having to seek through the actual file.

Edited by willm

Share this post


Link to post
Share on other sites

This topic was an interesting read.

 

My experience is basically almost the last 10 years, doing pc repair, and fixing *a lot* of home users laptops. Many users with no clue how to maintain or look after a PC, or often even apply updates. Personally, i have never noticed an improvement with registry cleaners, except in very rare cases, such as the example master_scythe gave of overly large program lists. Even then, these issues are kind of irrelivant, I would think.

 

For actual performance, it has been(in my experience), more to do with lame OEM installed programs, malware, toolbars etc., and never actually the size of the registry.

 

I am not completely sure, but I would think that due to the way the registry works, the size would not be a factor for performance. The applications will request the keys that they need, which is handled by windows, rather than actually having to seek through the actual file.

 

Yes i do agree with this it is very rare to see a speed increase from cleaning the reg but it does have other uses.

 

1 example is for a program that doesn't uninstall properly and u need to reinstall it a reg cleaner can be very help full

 

Ccleaner best effect is the flushing of temp directory's before scanning for virus's

Edited by built by noob

Share this post


Link to post
Share on other sites

Mac Dude,

I wouldn't waste much time trying to prove anything to non-believers. Even when you post results, they try to twist the statistics around in favour of their argument. So this is just seeing what they want to see as far as I'm concerned.

 

You've seen the difference. I've seen the difference. More importantly, our customers have seen the difference and are happy with the results.

 

Let them experiment as they see fit and come to their own conclusions. At least that's better than blanket statements about how it doesn't affect anything when they have no experience to back it up. Until they actually run into a system where it makes a significant difference they're not likely to believe. That may or may not happen, and won't change what we've seen or our opinions about it either. :-)

My goodness, someone convinced a product they don't really understand how works, convinced it does work, and not happy that someone wants to test things. What a shock.

 

That's the stupidest thing you've ever posted. I don't understand how gravity works either, but I sure as hell count on the fact that it does because I see it work every day. Obviously you understand all the intricacies of gravity, or don't believe that works either. I also said I'm happy that you are finally testing it rather than blindly forming opinions, but with reading comprehension such as yours its no surprise that this went over your head too.

 

I posted actual results where a system sped up, with the customer sitting right here and laughing at your silly reponses. You twisted the results to see what you wanted to see. Oh, 0-4 seconds difference... so it can be zero. Doh!! If I wanted to twist those same statistics further to prove my case, I would have said it made 3-7 seconds difference, which would also be accurate from a biased point of view, but I counted on you misinterpretting the data and you obliged with flying colors. Doh!

 

We shared a good laugh at your expense and she went home quite satisfied with a faster system.

 

That's what I love most about trolls, they're completely predictable.

Share this post


Link to post
Share on other sites

If you'd been following the thread, that was comparing you to people who write testimonials for fuel additives and the like.

 

You clearly didn't measure anything, and were relying on your feeling of things. Otherwise you wouldn't have given such large ranges. You also sound like you didn't even restart the computer, or ensure circumstances were identical. Which means due to the way RAM loading works, subsequent re-use of programs that had already been used would be faster. You also sound like you ran the entire Auslogic "One Touch" suite, not just the reg defrag and cleaner, which includes many other tweaks.

 

I, of course, HAVE to assume these things because you haven't put any detail into them. Just like every other testimonial.

Share this post


Link to post
Share on other sites

from what i gathered both sides are neither wrong, i've been using system cleaners since back in the day and use them almost all the time particularly after a browsing session or maintenance runs. i only use ccleaner these days mainly cos its easy to use and never had issues with it, i've tried other fancier cleaners but i find the more options they have the more dangerous they tend to be especially if you don't know what your doing and it only takes 1 deletion of a registry key or file to cripple your o/s. in terms performance tho, i'd have to agree with tantryl's original statement, the registry is just a database that holds the info for all the locations of the o/s's files, there is a false sense about registry cleaners here and this is where people go wrong. a reg cleaner is part of a system cleaner, take ccleaner for example it has both tools 1 for general cleaning of temps etc. and 1 for the registry, i use the main one most of the time and the reg cleaner only when i've uninstalled a program but there is no impact on performance by cleaning the registry in my experience. You'll likely find that Performance degradation is due to other issues from the hardware down to malware/viruses or fragmentation of the disk etc.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×