New german PC-Welt Test by av-test.org

Can you just explane to me how it is possible to collect 300000 trojans for 1 year?

Do you know how many different trojans it is every 10 minutes? (Keep in mind that this is not antivirus vendor)

I am sure that even the best antivirus laboratories does not collect this amount of different trojans per year. I would like if Mr. Inspector or Mr. Stefan can explane it to me. (or I did not properly translate that German language)

The other less suspicius thing for me is that Symantec has score better than Kaspersky and AVG better than McAfee.

 

Hello Durad

I can't explain, this is above my knowloge. What I do know is Wilders is a special place where true experts mingle with us common folks.

Many occasions Inspector Clouseau, Stefan Kurtzhals, vlk, Marcos, IBK and others have jumped into subjects, set the record straight or added words of caution.

For this subject maybe they can't comment ? For the concerns you raise maybe they will ?

 

Re: F-Prot 6 or Antivir premium?

It is not a problem to have so many trojans. Especially if you take into account that a trojan (for example a backdoor) can have many parts such as Server, Server Editor, Client, Plugins etc. This multiplicates the counter for some versions already with at least 3 for some backdoors. Then you have several types which are having 100's or even 1000's of variants. (RBot / SDBot / Methodbot / IrcBot) etc. So don't worry about the number of trojans/backdoors - you can easily reach this.

Personally i dont like tests aka "I have a lot of malware, scan it". For several reasons: As more malware you have, as less is your chance to have a "clean" testbed without trash/garbage. There's no way in hell that you can make sure that all files are really malicious without relying on other companies "results" eg. that at least 2 or 3 scanner report it. And what if 3 companies report it wrong? Whom to blame them? Normally it would be the duty of the tester to make sure that the files are malicious. But there we already reach the time limit. It would take years to verify this. And because of this there is no (and will be no) AV Test with that many files overall scan which is 100% flawless.

The most interesting tests - i mean the real world scenario tests - are not performed (or very bad) because it takes a huge amount of time and knowledge to prepare such tests. Tests like "How good is a av solution at curing / cleaning ?" That would give a lot of "unexpected" results FOR SURE.
Aka "Detects it, but is unable to clean it". What to do then? Next part you should check in this test would be "is the support able to help you with this issue". Because otherwise it drives the user nuts if he knows he has malware on his system and cannot get rid of this. It usually results in a huge amount of personal research, such as trying other av solutions if they can clean it or asking in forums. As i said before such tests taking a huge amount of time because you have to install the malware, scan/clean it and recheck. That includes that the tester MUST KNOW what the malware is doing in order to verify that it cleaned it properly. (Eg also deleting dropped components and not only the main dropper)

 

Re: F-Prot 6 or Antivir premium?

Wrong thread I'm still in the bed and didn't have coffee yet
Can somebody move the last post to AV-Test Thread pls?

 

Re: F-Prot 6 or Antivir premium?

For a minute I thought my short term memory was playing tricks on me

 

Done.

Blue

 

Inspector Clouseau said "Personally i dont like tests"

So "if" in some future time F-Prot got a good rating in an av test, i wonder if he would say the same then, or be pleased?

We will have to wait to see, if and when that day arrives!


StevieO

 

Spanner, first at all you should quote the full sentence and not a part of it. I said that i do not like THIS type of tests and not that i do not like ALL tests. That's a difference. Next thing is that it has NOTHING to do with the results, it's a matter of HOW a test was performed. So please stop assuming here things which i didn't even say.