Jotti's A Good Indicator?

[JerryM]

I have been looking at the results of Jotti's malware detections the last couple of days.

Do you think it is a good indicator of the effectiveness of an AV?

Thanks,
Jerry

[DavidR]

I would say it is a god indicator if a suspect file is in fact a virus/malware because of the multi engine scanning.

However, since it contains all suspect files which could be adware/spyware which avast and some other AVs don't specialise in, so it would be hard to compare apples with apples and many of those apples are trojan detections which are mostly spyware.

So I would say a multi-level/application approach is going to be best rather than rely on one single means of detection. avast! with Ewido (or a-squared) would give good all round protection and neither Ewido or a-squared feature in the tests. Include AdAware, Spybot S&D, SpywareBlaster, etc. and you improve that protection further.

[Lisandro]

For generic detection, all people around the world, it could be useful.
For YOUR particular use (that asks for another things like your system performance, your surf habits, your other security programs to get layered defense, etc.) it could not be that useful. You must be protected by the infections that could *in fact* reach your system.
Just my 0.02  

[JerryM]

Thanks for the replies. I completely agree that layering is important, and that is why I consider Ewido so important on both my systems. I think it takes Avast out of the average and makes it first class.

My question was prompted due to the fact that I have sometimes seen Jotti's referenced when wanting to prove X was better than Y.

For the last few days, off and on, I have kept track of some of Jotti's "tests."
I was a little surprised at what I found. Not in all cases, but in some.

I took 20 tests, and 7 AVs, and noted how many malware detections were shown.
Here are the results. I realize another 20 might show different results.
Bit Defender - 6 detections out of the 20.
Avast - 4
KAV - 12
NOD - 9
Antivir - 8
Dr Web - 11
AVG - 7.

I was surprised that Avast did so poorly, and Dr Web so well. I also expected BD to do much better.
Comments?

Jerry

[TAP]

I have been looking at the results of Jotti's malware detections the last couple of days.

Do you think it is a good indicator of the effectiveness of an AV?

Thanks,
Jerry

No, in my opinion such results are some kind of misleading by many factors such as corrupted smaple, false positive, non-malware sample, ect.

And a problem with scanner itself, http://forum.avast.com/index.php?topic=18923.0

[RejZoR]

Jotti is the least reliable way to test because it's using Linux scanners.
ANd even VirusTotal with Windows scanners often fails to detect even though local scanner detects the stuff just fine.