AV Tests

This is his own test done in his free time. If you don't want to believe his results, than don't. He didn't claim to be doing any sort of professional work, but many still like his tests.

 

That's not a right conclusion, because the number of infections MBAM finds depends on which malware got throught. Some malware make more registry changes than others.

Avast let throught less malware, but one of them (or some of them) made more changes on the registry than the ones Norton let throught.

 

Some interesting discussion going on in here.

For those that are ticked off about the Norton results and don't like the fact I was using Opera because "it's not supported", well, that's the browser I was using. When I installed the Norton trial, it didn't say "Oh, by the way, for the most effective results and for us to protect you best, you MUST not use your default system browser."

As for the accuracy of the test, or this product would have done better if you had tweaked the settings or set this or that to max, well, that's simply not the way I chose to conduct the tests. If an AV product is very strong with certain settings, then the demo should be installed that way by default. Most users (those of us not on this forum and most of my clients) have no idea what any of these products do or what any of those settings are and would simply not touch the settings other than leave the products set the way they were when they were installed.

 

Interesting tests, thanks Igster. I am amazed that Panda Cloud didn't detect anything.

 

Amen to that brother! For the rest of you whose product didnt cut the muster, buy another.

I totally agree that out of the box settings should provide max results.

 

So if you want to use Norton and have full protection you have to use IE or Firefox?seems a bit of omission on the part of Symantec don't you think?

 

Hi TheIgster,

In your next round of tests, can you include Comodo also.

Thanks

 

Also I fond that Avast,s web AV not supports Opera. Seems to support IE only.

It,s not so good. They must support atleast major browsers like IE, FF, Opera and Chrome.

 

webshield is scanning my firefox, so it's not just IE.

 

In my tests, the web shield blocked most of the links I tried to access when Avast stopped the incoming malware. All tests were done using Opera, so it supports Opera as well.

 

Anyone know if the web shield also supports Google Chrome?

Also, doesn't Opera have some built in web page reputation or safety program? If so, did that block or alert to any sites tested?

 

Yes, it does support chrome.

 

Just a curious question.. Why is there so much hubbub over this test ? I just don't get something as random of as this test was could really show any value. Example if you ran the same AV's again over new malware you will get a totally different result. If you take a step back and look, You will see this test really has no value nor could it give any value to people.

Maybe I'm missing something ?

 

Because it stepped on somone's favorite AV's toes

 

Avast's web Shield works with IE, FF and Chrome. I tested each browser with it. I did not try Opera.

Ice

 

The OP has put some solid work into the tests. But you highlight a key and important word with all smaller tests, 'random'.

 

I use Opera all the time and I've never seen it block anything itself. No, it did not block anything in any of the tests.

 

Something interesting to talk about? Random, sure, but each AV was given the same "random" links during each round of testing, so they all had a fair shake.

Again, not scientific or huge sample sizes, but I think interesting none the less and it seems others found it interesting as well.

 

Yes, and 'random' is exactly the way our computers are operated, and exactly the way we need our security applications to perform. None of us are waking up in a lab each morning, ready for another round of clinical testing.

The more I think about it, the more value I think these tests have. And it is their randomness that make them valuable.

 

Good point.

But there are a lot of questions you can ask.

eg. some vendors might not be aware of all these malicious link sites. Are these tests proving actual detection rates, or only detection rates of malicious links on malware list sites?

Is the malware/links being circulated as readily as other malware? For example, XYZ2020.exe might destroy all AVs except for one, but is this .exe going to even affect users (except for those accessing the download site) if the file isn't available on popular sites.

I too also frequent the malicious link sites for testing. But, if some vendors are adding the links from these test sites faster than others, are our personal/youtube type tests a true representation of a product's detection ability, or a representation of a product's detection ability of links provided on malware list sites?

Do I have the answer? Nope!

 

Saraceno, you ask good questions and make a good point as well.

But here is what is hard to get away from, or ignore, or make excuses for...

each and every application had an equal shot at blocking the malware.

And, really, to my (sometimes admittedly limited!) thinking, that is what levels the playing field right there.

 

No problem, respect your thoughts.

I think what it comes down to, most common threats affecting users, the majority of AVs perform well. The PC Security Labs report demonstrated this.

But the more obscure files and downloads, that's where you'll notice a difference in programs. Regular users should be able to get by with your standard AV, while those engaging in more risky browsing/downloading, would benefit from a program providing better 'zero-day' or close-to protection.

 

This is so true TheIgster, especially the part where you tell that some users don't bother to change the settings at all. Aside from detection, removal and prevention, [size=+2]ease of use[/size] is important as well. AVs are still consumer products (with service support) after all.

 

I agree page that computers are operated just in that way random. But as for seeing what a product can do thur these kind of tests, I feel is just pointless. The product that won the first time could totally fail the next time and the other way around. In the end it really gives you a false sense of what a product can do and a flawed view of whether a product is protecting you or not. But then again this could just be my view.

 

Hey Fajo
Pointless perhaps in terms of identifying a clear and consistent winner, but not so much if the lesson learned is what you just said... that the product that did best the first time could fail the next time. There is value even in that lesson, although that extreme hasn't been the case so far, I don't think.

In the end, the tests accomplish what the OP set out to achieve, in my opinion. Entertainment and a topic for discussion for many of us... plus another opportunity to learn a little bit more about the protections we all use.