maybe special test in june

EDIT: ClamAV/ClamWin retired the permission, so it will be not included. Sorry

So now it looks like this:
UNA and Rising are willing to participate.
eTrust(VET), Kingsoft, Jiangmin and VirusBuster: waiting for final reply (or any reply).

 

IBK :

I am a Chinese living in Shanghai and I am very glad to see that almost all famous AV software in China may be in the list.

If you can not find the AV software of Kingsoft, I can offer the download address of Kingsoft Internet Security 2006 on Kingsoft Official Chinese site.
This exe file also offers English GUI.
The address is
http://telecom.event.www.kingsoft.com/db2006.php?id=1
or
ftp://dl3.kingsoft.com/duba/download/install/DB06_SP_060419/DUBA06IS.10.EXE

It is a trial version. You can only update for 7 days after you register a "passport" of Kingsoft in this trial version. If you need the passport,I can IM to you.


You can also download Kingsoft Internet Security 2006 from its official site in Japan:
The address is:
http://210.188.221.7/soft_plus/kingsoft_000/kis2006.exe
This can be used for 6 months for free.
But it seems that this is not a trial version, i don't know if it is legal to download it outside of Japan.

BTW, can you consider Guanghua, another famous AV in China (I remeber you said Guanghua and Jiangmin are ..., since you test Jiangmin, then why not test Guanghua)?

If you need any trial version of AV in China, I can give you the download link. It's easy to get for me.

 

1. I test/publish only if I get a permission.
2. Jiangmins detection rate is lower than Rising, so it currently does not appear to use another engine.
GuangHua I will not test, because it defintly use various material from other companies, the rates may look good, but most probably the reliability and disinfection is not good, as they do not even have the malware that they detect. Better you get instead one of the original products.

 

Thank you. I understand.

 

I guess "Open Source" isn't very open.......

Time to ditch the clam

 

Currently:
Rising, UNA & VirusBuster will participate.
Waiting for answer from: eTrust(VET), Kingsoft and Jiangmin. Should they not answer within Friday night, I will consider them as not willing to participate .

In any case - whatever the results will be - I find it very nice from the participating companies that they allow their users to see how they score, in order that the users will see how their products improve over the time. It is a big plus for companies to do not hide the scores of their products, as I consider it as a sign that they have nothing to hide and are convident in their products.

 

Would it be possible also to add databaseless anti-malware, like i.e. Viguard?
Or am I talking nonsense?

Gerard

 

IBK, so what about KAV6 Proactive Defense module? Will it be tested?

 

as it will be an on-demand test, Viguard can not be added. would also be too much work considering that is a special test done in "free" time (and for free), while preparing the other things for the regular tests.

 

It was already tested. The press release about its results should be released maybe already during this week or next week and posted on my website and weblog.

 

Excellent

 

nice. ... I can't wait to see it.

 

here's a link to Kaspersky's response to the proactive test - hopefully we will get to see the actual report too.

http://www.kaspersky.com/news?id=188369257

 

No, the press release is all what you are going to see I think. http://www.av-comparatives.org/seiten/ergebnisse/KAV6_PDM_test.PDF
the results by executing the malware was 99,4%. Kaspersky will soon update some rules to cover some of the not-blocked malware, after that it will have 99,5% (of the ~6329 samples). Note that this test was done on-execution, and not on-demand, as the PDM of KAV v6 works only during execution of files.

 

sorry if this is a stupid question, i dont pretend to know much about proactive detection...but do false positives come into the equation in this type of test?

 

No, because executing millions of clean files would take too much time. this single test only shows how much the PDM (behavior blocker) of KAV6 is able to block.

 

Andreas, what were the settings that you used in the PDM?

 

the highest available settings.

 

You used just PDM right? No File Anti-virus module i assume.

Btw, PDF document link is dead.

 

~Yes, using only the PDM was a little bit tricky, by replacing the new databases in order to test only the PDM.
I fixed now the URL, it was case sensitive.

 

So bottom line, KAV6 PDM module is extremely effective? I can only see such result if it was able to detect 99% of everything that you have in your test bed.

 

The PDM (behaviour blocker) blocked over 99,4% of the used test-set (6329 samples) - new samples that are not recognized by the signatures. With signatures, KAV already detects over 99% in the on-demand test of February, so if you take those 2 results together, KAV 6 will be quite effective I think.

 

It's good to finally know how effective is KAV6 PDM and KAV6 overall considering both reactive and proactive parts together.

 

WOW!!! KAV 6 is truly one of the best (I'd personally say it's the best) antivirus software.

 

If I understand, the test was conducted using the samples that were not in the signature data base. Only the PDM was turned on, and instead of on-demand scanning the samples attempted to infect the system. PDM intercepted 99.4% of those samples.

Instead of using heuristics to find malware already on the computer, this test demonstrated the capability of PDM to prevent infections.

Is this correct?

Thanks,
Jerry