AV-Test Certifications for the 1st Quarter 2011 (Windows 7)

Yes I did because I thought it would make it faster but I was wrong, I saw no difference.

Thanks.

 

Scores Drop in AV-Test Lab Results

 

This test is a shame, why a product with a 100% in the dynamic test is not certified? with a 100% you don't need any repair who care about how good is the repair?

Many products that allow your computer being infected are certified and 1 product that protects against the 100% of the malware in a dynamic test (real test type) is not...


By the way the detection rates of some products against 0day malware smells bad, probably they were not using 0 day malware at all.

 

Repair is important when you want to restore an already infected system.

 

Which av is gonna clean your system if it lets the malware infect your system? none. You are always going to use other apps to clean it or call to the support of the av to do it. The relevance shouldn't be the same. I think this test is totally bad balanced

Why mix some obsolete tests like detection test with real test like the dynamic ones?

it's like an AV that scores a 100% in the dynamic test by avcomparatives here is not certified, because the interface or the repair abilities.

Anyway, this is just my opinion, with this scoring system you can make look good to any AV.

 

was there a case of 100% detection? I could not find it

 

I agree but how do you, really know, it repaired it correctly and didnt leave remnants.

 

Comodo if I remember well

 

That is why repair needs improvements every year and that is why the are testing the products in this way. The optimal situation would be if i dont need to download a bunch of other tools to remove viruses.

 

KAV has done it twice for me (with malware that infected the system before I installed KAV), although I've always verified manually that no serious traces were left. I agree that they should've given much more importance for the dynamic testing part though, but detection, repairing and usability is important too.

 

So KAV didn't do it because you haven't it installed when your pc got infected.

 

I would assume that in future Avira's Protection and Repair scores will go up with the release of SP2.

 

It is tested but no 100% detection... 4.0 points out of 6. So which one?

 

Ah, you meant it like that. Well, I think it's possible that if an AV lets malware infect the computer, that malware will be detected after future signature updates and then comes the repairing part.

 

Again, Comodo has 100% in the dynamic testing.
In the real world test if there is any file left by the sandbox they count it like a bypass (if I remember well) although there is no malware active in the computer during the test or after restart it, so for me this does not count, especially when this useless malware can be detected and deleted in any on-demand scan some days later.
A bypass is when your computer is infected not when there are some deactivated and useless rests of a malware.
Would be nice to have a document with the methodology explained, I can't find it on the web, but probably does not exists.

All this is just my point of view, you can agree or not xD

 

Ok.. I see.. you don't have access to the methodology used for scoring so you assume that the failure on protecting from 0day threats is likely not a failure.

 

No, the last time I asked them about the methodology they told me that the traces left by any malware although not active were count like a fail.
But I shouldn't need to ask them because this kind of info should be available.

 

Sorry, nothing personal but the methodology seems not public (at least I could not find it) so still please allow me the benefit of the doubt on this 'left overs' consideration and the 100%. And even if this was the case, the criteria is applied to all products tested, so they are all on the same boat.

 

They plan to publish the methodology eventually. guest is correct, leftovers count as a fail.

 

So, looking forward to the methodology. Establishing a certification system, testing commercial software and 'market' results without giving open access to the method is not impressive to say the least.

 

They also told me that on the 1st or 2nd quarter of 2010