avast! earned ADVANCED rating in AV-Comparatives test.

no.

NOD32 prowes that but you certanly tolerate more FPs easily if AV justifies them with extreme detection rate of actual malware (opposed to AV with awful detection and lots of false positives).

@Andreas
Can you elaborate a little on how Avira`s score deteriorated from what looked great to standard rating

its due false positives.
getting high scores (also in the tests of august) with a noisy heuristic is easy. as FP’s are not something tolerable (and passing our FP test “should” be easy) it does not deserve the A+ or A rating.
e.g. in virusbulletin tests only 1 FP is needed to do not get any award.

This just means heuristics are indeed strong, but not fine tuned enough. NOD32 is a live proof of that, they offer excellent heuristics with very low false positive rates (if we look globally, not just this test). Thats why avast! scored that well too. They have detection and false positives nicely balanced.

But, the thing is if you look at the rate of improvement for AVIRA, when its detection rate was something like 41% it received Advanced++ standing, so it increased its number of false positives from about 6 to 16 but its detection increased from 41% to 81% and now its just standard. The AV comparative seems very made up.

A++ does not exist. A+ is the highest. FP’s are sent to the vendors after the tests, so if FP’s still raise in our FP tests it means that a product has a very noisy heuristic / weak signatures and this makes it easy to get high scores when scanning malware. to get A+ it is only needed to reach 40% and have only few FP’s. the categorization is unchanged since years; changing the rating system to make avira look better is unfair. FP’s are an issue and should not be talked down.

Statistics:
Depending on how they are put together, can make 1+1=3
The best way to judge a product is by how it performs for you on your system. :slight_smile:

Well, i totally agree with IBK and his rules. Most of AVs even show they can work just fine with those rules (NOD32 and KAV7 specifically, you can also include avast! in there). Avira has to iron out false positives in their heuristics engine. As simple as that.

Aggressive heuristic could sell an image that the product is detecting MORE than the others. It could be used as an advertisement. I agree with IBK and RejZor. Also with Bob, as no product will be perfect and we do like layered defense 8)

When will we get the “On-demand comparative February 2008” ? :wink:

They will be released on 1st March :wink:

Thank you :smiley:

i don’t think tha AV Comparatives are too important …

to understood what i mean read
http://www.wilderssecurity.com/showthread.php?t=199292
especially since http://www.wilderssecurity.com/showthread.php?t=199292&page=7

it get’s interesting where company experts explain why security companies bail out of AV comparatives due testset and methology mistakes etc …

quite shocking (tho i knew some of it - i never paid closer attention to that mess around) …

and in thread are mentioned some interesting numbers from russian virus tests and other infos …

the opinion of bontchev about tests with large test-sets is known. he does not represent the position of f-prot as company. if you read the whole thread you will better understand who is just trying to discredit a test/tester and why.

suddenly IBK's testbed is full of garbage files. Until recently, AV-C tests were considered to be reliable by the industry, with garbage files kept to a minimum. im sure the top-scorers just recieve the dvds from IBK and just add them all to their database. 20 minutes work, simple and scores high in the test.

I can’t believe it ::slight_smile:

Heh, why ppl always jump so fast on “garbage” files as soon as they fall behind competition. It’s not like you can “desinfect” trojans and break them like it used to be in the file infectors era, DSL and cable connections are now almost 100% error free compared to crappy 56k full of errors and packet loss. Where the hell all these “garbage” comes from? If you ask me it’s just an lame excuse for AVs that suck and just can’t get over that fact. Why can avast! score good, yet ALWIL guys still sometimes have slow samples adding and similar administrative problems, Dr Web which has oh and so heuristics, hourly updates and god knows what is still worse in lots of points?
Same for F-Prot. Long history and developement, recently a very good dev joined their team, yet they still just can’t make any visible progress. Then there is AVIRA on the other side, lets face it, they were pretty lame like year and a half ago, AntiVir 6 was like ancient software with it’s non incrimental updates and quiet crappy detection. But look it now, probably the best heuristics, they offer hourly updates and they’ve beaten pretty much all vendors hands down in pretty much all tests done to date. Why can AVIRA actually show the progress and improvements and others can’t? Garbage files, right…
Same goes for avast!. It’s not THE best but constantly achieving ADVANCED rating and also improving in proactive field without any heuristics does show that devs can keep up a certain level of quality. Garbage files pheh…

avast! earned ADVANCED+ rating behind AVG and Avira :o

What can i say, excellent job ALWIL!

Great job ALWIL team :wink: