Update - AV-comparatives dynamic test: September 2012

Some surprising changes....

http://www.av-comparatives.org/comparativesreviews/dynamic-tests

 

Yes, TrendMicro looks excellent. Gdata, F-Secure and Bitdefender are also great. Webroot is to much user dependent for my liking.

 

How exactly is Webroot user dependent?
What choices does a user of Webroot have to make?

 

what does user dependent mean in these tests guys?

 

Yes, not sure. On defaults Webroot has no user input.
May be its the pop-up asking for scanning the system when something is detected that can be canceled?
Or the pop-up box when a malicious website is identified

 

Block and Allow

 

you can also choose to have WSA to block automatically so the user intervention would be taking out of the equation. In default mode it isnt enabled though.

 

based on that, WSA would have come in around 99 percent detection. Pretty damn good if you ask me.

 

I wonder if this is something that can be fixed by Webroot or it will lead to more FPs.

 

But normally the pop-up is quite clear about the malicious content, if I remember well.

 

Hopefully Joe and his team can take a look at these results and get back to us.

 

Is TrendMicro surging after this result and the September on-demand test? Are forum members using this AV?
Eset with its typical 95-97% result.

 

Wow!!! Trend Micro 100%!!! Unbelievable!!!

 

These prompts will occur on default settings (as far as I'm aware, AV-C tests all products "out of the box" in this test). WSA isn't always completely silent but it does try to be - it uses the cloud's view of files to reduce FPs so the average user will likely never see one.

The popup should be pretty straightforward - it defaults to Block and has a clear message as to why it's being blocked. We've been phasing in this protection over the last two months and have seen no uptick in false positives so you shouldn't experience any problems with it.

I can't quite make out our exact percentage in the test (the bar is pretty small) but I'll ask around internally to see if we have an exact number from the results.

 

Bitdefender and F-Secure seem to be very consistent test after test.Congrats to Trend, they stepped up big time.

 

Looking at the month of Sep in the chart:
http://chart.av-comparatives.org/chart2.php

Stats for Webroot appear to be:

Blocked 88.2%
User dependent 11.2%
Compromised 0.6%

With a savvy user, yes a good score with compromised only 0.6%. With an average Jo user who likes to get where they are going and clicks 'Allow' when given the choice (quite frequently according to this test), not so good. Could default setting be changed to automatically block?

 

These types of detections in the course of a normal user actually using their PC are extremely rare and show a modal message encouraging them to block the file. While they can still click Allow, it's a pretty firm message to block by default. I do think it may be worth just blocking these, but I haven't seen any case of a user answering Allow and then being infected. I'll see what else we've analyzed internally and assess if there are changes we can make to improve this.

I would be curious to hear if any user here has seen these prompts when using the default settings (feel free to PM me or write over the Prevx section).

Thanks!

 

But PrevxHelp said an average user is unlikely to see these prompts:

 

Webroot are being a little mean with its trial period of 15 days.
Most companies offer 30 days minimum.
Just a Thought.

 

I have been using WSA on one of my machines for the past 3-4 weeks. No pop-ups. The only time I have seen pop-ups, is when I am testing it against malware.

99.4% in September! Congrats on the improvement Webroot.

 

I'm glad to see Panda came back up a bit. I've been considering trying them out after being an Avast user for many years. There's not one single thing wrong with Avast on my end, I just like to change things now and again for "kicks".

While I do appreciate and pay attention to these monthly tests, I always advise others using it to base their choice of security on to be careful. These tests are far from useless, but considering results change almost every test they aren't the end all-be all. I am a bit entertained admittedly when these come out and suddenly people who were supporting a product last month at 98% decide to shun it the next month when the product is at 93%. Sometimes I wonder if these people think of the Internet as some Stuxnet-peppered minefield and that anything below 100% (which in real world use is impossible) should just be thrown away.

If a product consistently stays above 90%, it's worthy in my mind.

 

Wow, 100% from Trend Micro

Avast! would be much higher on the chart without the user dependence, but they are still great imo.

 

Lol....................WSA got spanked. All WSA users look in the mirror...........Do you see this?

 

You really hate WSA don't you?!?!