PC World AV Chart

Excellent heuristics WITHOUT TruPrevent and so few False Positives. Maybe the tested clean file set of 20.000 samples didn't include any packed samples!

Best regards,
Firefighter!

 

No, I think runtime compressors are always there. Panda has improved a lot in recent days, but there is still some way to go. Their unpack engine is not so bad at all IMO.

 

PC World June 2007 Top Antivirus Picks

Using AV-Test.org's malware sample the magazine listed Kaspersky AV 6 as there best buy from those tested. These were NOT suites but stand alone tools still possible as a base for a mix and match approach. They excluded free tools like AntiVir which is an error in my view, but they are a magazine of course!

Rankings were based on Design and ease of use, Detection rate, Proactive Detection of Unknown threats, False positives, System Slowdown, Outbreak response time.

IMHO the least important of these variables are False Positives and system slowdown since my Product BD 10 did the worst on those

1 KAV AV 6
2 Norton AV2007
3 BD AV 10
4 Eset NOD 32
5 Panda AV 2007
6 Avast 4 AV
7 AVG 7.5 AV
8 Trend Micro

1 had 85 points, 2,3 4 all were tied at 84 points. The rest of the field was all scored below the 80 point level.

Based only this list and my bias toward those with the best heuristics I would pick either BD 10 AV again since the slowdown is a known bug and will be fixed or NOD 32.

KAV was ranked 1 in spite of it's less than top of the line heuristics.

 

Re: PC World June 2007 Top Antivirus Picks

I want to clarify that Avira was considered for inclusion but PC World was too far ahead with the test when Avira released the Vista-compatible version and hence Avira was not included.

Other than that...no comment

 

If Avira Antvir had been tested, it would become more famous.

I notice NOD32 is declining, especially the detection of keyloggers/trojans/backdoors-like malware, well before the time that many people praise it is an unbeatable product.

The public is usually slow to accept new information. It needs time for them to absorb to new information. There is "a lag" in their mind. They will refuse to accept the test and call it unreliable or anything similar if the test results don't fit their preferences, instead of judging based on sample sizes/methodology and so on.

Based on history, very few antiviruses manage to maintain at top-rated every year. The exception is Kaspersky which is worth praising. However I won't take it for granted it will keep on forever, nor assume it *must* come first in every test result.

 

If you read the report, the sample size of polymorphic virus detection is very small anyway. The PDF report in AV-comparatives list all the polymorphic virus details.

I'm not sure how many samples AV-Test use to test the polymorphic detection.

 

Nod went from best to worst in only four months. Imagine that.

 

AVG reaches in AV-Comparatives 61% (67% if you also count the other 2*2000 samples), depending on how you count the detections. 12*2000=24000. If some more samples that are detected reliably by all scanners would get added, the score would be higher.

 

Hard to believe, isn't it? Also, if you happen to visit the support fora for Avira and Kaspersky s/w, it's quite easy to see exactly why they're on top; absolutely no issues with those products and their level of protection is absolutely outstanding. Amazing!

 

AVIRA and Kaspersky are not without their own problems, but in the end they are much more responsive and helpful than some others, which is why they stay on top.

 

Kaspersky has a nasty bug in Vista when you copy 16K+ files you get "out of memory errors" which they are still addressing w/ MS.

http://forum.kaspersky.com/index.php?showtopic=33258

 

Worst? There are much more antivirus which are worse than NOD32.
Even if you take the more reputable ones only into account, it is still not the worst. At least Dr.Web sits firmly in the last row at av-comparatives.org

How do you know it is only four months exactly? Just by the publication dates of some sites like AV-comparatives? I observe it actually starts to decline about 1 year ago. The recent test results proves my observation.

 

NOD32 scored four Advanced+ ratings in AV-Comparatives during 2006. What results are you talking about?

 

Actually I can only find 2 reliable and respresentative organizations for AV tests, ie AV-Comparatives & AV-Test.

There are 3 AV-Test published in the magainzes:
- PC-WELT(a German IT Magazine) on September 2006
- PC-WELT(a German IT Magazine) on November 2006
- PC WORLD (an USA IT Magazine) on April 2007 (this one here!)

Together with AV-Comparatives:
- On-demand comparative on February 2007

I also do some small tests on antivirus programs for my own interest. I collect the samples and do the tests.

Always read the full reports (not just the badge/certificate) if available. You will see and know more.

 

You stated their results began declining about one year ago. Where are the results to support this claim?

 

I feel it starts to decline about 1 year ago. "About" only!

I notice that due to my observation, not pure guesswork.

The results stated above prove my observations, from Sep 2006 - May 2007.

The results are from AV-Test (3) and AV-Comparatives (1).

You can see the decline too from both AV-Test & AV-Comparatives. Read and find out yourself. I'm not going to list all of them.

Some evidence:
Kaspersky vs NOD32
File viruses 97% vs 79%
Trojan horses 97% vs 89%

Is that clearer now?

PS: I have published my own small test here on about Nov-Dec 2006(?). NOD32 ~30% vs KAV ~60%. Small samples only.

 

Please provide links to the tests that support your claim. Additionally, are you a professional virus tester?

 

I have updated the previous post. Read again.

Sorry, I don't have enough time to do the work for you.
I have provided enough info about the source. They are known to others too.
Google is your firend, or ask other posters to provide links for you.

Note: If you don't have an open mind to digest the info, then just forget what I say. Don't take it personally though. Nothing against you. I realise many people generally has strong attachment to particular products. It is normal to them to accept new info easily. Maybe NOD32 will recover late this year. Who knows?

No. Just amateur. Similar to what aigle is doing here. Test security programs for one's own interests only.

About time to go out.

 

You made the claim. It is not unreasonable to ask for proof. I still don't see any.

 

If you look at a developers comments, it looks like something in vista is wrong, as other drivers with the same functionality will also display such behaviours.

 

Why would it have to be "unpacked"? Shouldn't a good AV detect malware in any form, packed or not?

Trev

 

The point is to test actually working files. Since Advanced heuristics performs deep code emulation, it doesn't flag files that are non-functional. Of course, that does not mean that all functional malicious files will be flagged by AH, however, it's still being evolved to detect more and more threats proactively.

 

Perhaps because you're sitting tight and refusing to look?

Like Wai_Wai said, Google is your friend. If you're genuinely interested, the facts are always easily accessible to you. If not, well, why bother? Forcing other people to do the work for you is a great debating tactic to get them to back off, but in the end it's nothing but that; a debating tactic.

I'll just chime in that NOD32 has always performed against with Chinese malware, as far as I've known. Again, no hard, scientific statistics, just personal experience from delving through VX forums.

 

All I know is that NOD32 was rated the number AV for 2006 by AV-Comparatives.
If someone has facts, NOT INNUENDO, let's see them. I did not claim that NOD has declined. Post the links. People on this forum can make up their own minds if they have data to look at.

 

The recent PC-Welt test regarding the detection rates of the ANI exploit, the PC World test where NOD32 scored lower detection rates than avast! And AVG, and of course, the most recent AV-C test are three I can recall off the top of my head right now. NOD32 missed a VB100% award in April as well, I believe, but since it was due to an FP rather than a missed sample IIRC, I won’t hold that one against NOD32.

Of course, since large-scale, reliable tests aren’t carried out every other weekend, it’s kind of hard to think of a lot of tests which point out NOD32’s shortcomings lately. But the proof is there, and again, as I’ve said, Google is your friend. The big tests, carried out at infrequent intervals, simply discreetly point out a trend that’s been going on for a while now. As opposed to spreading malicious innuendo as you seem to believe, I’m simply telling you this in case you’re genuinely interested in facts rather than fancy talk, but in the case that you’re not and are simply content to pretend that what you refuse to verify for yourself is not true, it really doesn’t matter to me either way, to be honest.