Is G Data AVK really superior?

Well I am a Kaspersky man myself and in no way a fan of multiple engines, I was looking at G data Anti Virus Kit, it used to use engines from BitDefender and Kaspersky and now days BitDefender and Avast, adding more engines didn't prove to add extra detection. When you take a look at the testing results from http://www.av-comparatives.org/, http://www.virusbtn.com/news/2008/09_02 or http://malwareresearchgroup.com/?page_id=2 you can see that there is not "much" difference between AVK and the other Antivirus programs (singe engine ones), so what is the point then?

 

Hey,
i agree a decent single engine is very close is the dual engine solutions. dual engines could potentially use double the resourses for what less that 1percent detection? is someone really that paranoid to warrent the extra resourses needed?

 

Well I guess some people are, otherwise they wouldn't make these products. In my opinion multiple engines are not worth the sacrifices in terms of system resources.

 

G DATA seems to be a German phenomenon.

True to the motto - I want to feel my security software, the more the better.
AVKWCtl.exe about 100 MB in memory, AVKProxy.exe about 160 MB, that's ridiculous, even with a double-scan-engine.

But... G DATA does a lot of magazine advertising in Germany and wins nearly all these magazine tests and all tests are done by German AV-Test.org.

Even a dedicated German gamer magazine presents G DATA as their test winner and as the best solution for a gaming PC.

Of course, their detection rate is great, but unless you have a dual core and a lot of memory (to waste), it makes your PC completely unusable and even if you have, it's far beyond reason to use such a performance/memory hog.

However, their international recognition can't be that great, their localized websites contain not even half of informations and links their German site offers.

As said, it seems to be a German phenomenon, pushed by local magazines and testers.

Cheers

 

Well I have been using it for about 8 months now. My computer is far from top of the line, its a P4 630 with 3gigs of ddr2 ram. It does use a fair amount of ram when surfing, less when the browser is closed. I feel no system slowdowns with this a/v. The web scanner does slow it up some (like most others do) but it is not that bad. If I disable the web scanner I have zero slowdown. System responsiveness is good & it feels no heavier that any other a/v.

I have tried many other a/v's & this one runs better than most on my system. The key seems to be the amount of ram. The more ram, the better. If you have the amount of ram I do, you will not notice any slowdowns.

 

I don't think so.

When GDATA used Kaspersky engine it was very good, but actually I really don't believe that Avast + BitDefender can get good results, despite some tests. What I see when I use the new GDATA for testing is very different.

And a lot of resource usage too...

 

I'n my personal oppinion adding bitdefender was a huge mistake, a very huge mistake.

 

Well to be honest I visited a German website where they were talking how G Data is the best one, they were showing the test results and so on, that is why I wanted to see your opinion because, detection rate is one thing, but stability and recourse usage is also very important. G Data may be a little more then the average users really need.

 

G DATA's main territory is indeed Germany, well-organised advertisment campaigns and the friendly help provided by e.g. Germans largest computermagazine ComputerBild deliver G DATA numerous German customers.

 

This is true, I visit Greman security sites and forums very often, on most of them you can't find a single bad word about this product and the marketing is phenomenal.

 

At first, the combination of BD-engine and AVAST-engine is more effective than the old combination.

Of course memory consumption can be optimized (and surely it will ), but when your systems RAM >= 1GB you won´t have to suffer

At least the great detection rate isn´t a (german) phenomenon - it´s reall.
Looking at absolute numbers you don´t have to be paranoid to be exalted.

Use the trial-version and you will see.

Best regards
Ford

 

Don't misunderstand me, AVK is ,when you look at the numbers, a great product and nobody can doubt that. The point that I'm trying to make here is that AVK even with 2 engines is not all that much better then it's rivals (Avira, Kaspersky, Avast...), when you look at the tests there is very little difference between Avast and AVK, so I hope that you get my point here

 

I think I got your point, but here´s my point:
av x detection rate: 97,6%
av y detection rate: 99,7%
av z detection rate: 99,9%
Little differences?

Probability that your system is infected when it was attacked by 100 independent "bad guys":
protected by x: 91,19%
protected by y: 29,95%
protected by z: 09,52%

Still little differences? Please correct me if my calculation is wrong

 

there is alot more to an AV product that detection rates,

ie. how are Gdatas drivers?

 

You are right, but detection rate is this threads topic.

I didn´t hear about vulnerabilities in G DATA´s current drivers.
Especially the wfp drivers "feel" fine.

 

Link to Eng. trial please?

 

v.2009

 

I wasn't talking about AV x,y or z, I was talking about Avira, Kaspersky and Avast. Your calculations are OK, but once again if we take the three products that I mentioned, there is no big difference. We are talking about less then 1% , it would be surprising is AVK couldn't secure a lead. Keep in mind that AVK didn't win in all tests

 

I´m sure there exists a comparative with results, so that you can substitute x by kaspersky, y by avira and z by G DATA
(e.g. take a look at http://www.rokop-security.de/index.php?s=&showtopic=16855&view=findpost&p=260911 - latest results for avira and kaspersky are not so good as in my example)

But even if we don´t talk about concrete vendors - once again my point:
"little" differences (minor than 1%) in detection rate can have big consequences.

cu,
Ford

 

*agree with Ford Prefect*

 

Yes and I agree, but in real life that doesn't mean much, you can't expect somebody to be infected with hundreds of thousands of viruses, so the extra guns are not really necessary. I agree that AVK has better detection rate then the others, but like I said before with 2 engines that is not really a surprise.
Maybe they should test the programs with multiple engines (and signatures) separately, it is not fair to test 2 programs against one, but who's complaining

 

An AV having 2 engines is not justification for treating it differently from single-engine AVs. If it detects more, good, its users will be better protected. Thats what most people want to see with tests/comparisons, how effective their AVs are at detecting in comparison to other AVs.

Having 2 engines doesnt mean its in a different class or league, its still an AV and the tests are simply showing how well AVs detect in comparison to others - what most users and companies want to know.

Cant shoot it because it incorporates different technologies or has merged other companies solutions, if it detects more, good for it, regardless of what technology its using.

 

That is what I meant when I said "who's complaining"

 

Your calculation is fine, although there's a minor typo (protected by y should read 25,95%), but I would take issue with how realistic is a 100 challenges in the real world. That is a lot of malware encountered in casual usage.

I'm sure one can surf with that type of result, but my own impression is that you really have to work at it.

In my world, genuine challenges number 1 or 2 every couple of years. This is really probably more typical, and in that case the differences on a yearly basis are quite minor.

The calculation is useful for those who court exposure via their surfing style, and that could form the basis for selection of a top tier detection engine, although I'd also tend to point to alternative technologies that are useful approaches as well.

Blue

 

Wow - nice that you ´ve found my typo!
I agree - probably most users are not confronted with so many attacks.

However - detection rate is one of the most important criteria to me.

Regards,
Ford