Matousec Firewall Challenge = new Test

ZoneAlarm is like swiss cheese Huge holes.

 

Based on what?
Mrk

 

One thing you may not know, it's not every Swiss cheese that has holes. It depends if you speak about Gruyere or Emmenthaler.

 

Loool, what about Gouda or Tilsiter?

 

Based on what?

I was never very impressed by the free ZoneAlarm, but ZoneAlarm AntiVirus, with Program Control works well.

 

Can anyone test dnstest (not dnstester) with the latest Comodo ? What popups and in what order do appear in case you allow everything step by step ?

 

This new test is absolutelly irrelevant, mainly because "level up" test procedure. A list of passed and failed test,like the old method would have been much better and relevant to full capabilities of a "firewall" (most of the security products are not firewalls, but HIPS+firewall, that's why I use quotes for the term firewall).

 

I disagree. If your firewall cannot pass 5 of 9 basic level1 tests, does it matter if it can pass level 7 tests?

 

Still something wrong there.
For example, OA passed level 10 long ago, a day or two after BSODhook was released, but failing level 9 had no chance to have 100% level 10 score.

 

Hey, we should pass level 9 now with the 119 update we just released!

 

Do you ever rest ?

 

Matousec says this in the introduction:

 

It's like I said. Matousec is neither perfect nor good enough, but keep smiling.

 

Hi Mike,

Will build 119 meet level 10?

 

Leve 10 was not a problem. But according to Matousec methology a product cannot skip failed level, so OA was not tested on level 10.

"Those products that reach the score limit of level 1 and thus pass this level will be tested in level 2 and so on until they reach the highest level or until they fail a limit of some level."

 

What's your point?
I consider Matousec tests' completely irrelevant, but you need to read the whole thing before making conclusions. Choosing a firewall because it scores high on someone's test is as wrong as bashing the test/tester for the wrong reasons.

 

Yes, I did know why the level 10 test was not carried out.
I just wanted to confirm that OA would pass the test if it was carried out.
You have done that for me - thanks.

 

Matousec is much better than those who do nothing except talking. He does what he does. You may hear to him or not. But did you do something better than he to criticize him ? His view and methodology may be questionable, but his tests are accurate. At least nobody complainded that his results are not correct. In any case all the tests are available with the sources now. So while your words are against his actions you are a loser.

 

Choosing a firewall based on this (Matousec) test, would be the same as to judge the quality of a firewall in (example) BitDefender suite based on its AV detection rates.
There is nothing wrong with Matousec test, it is the common misinterpretation that makes the confusion here.
Well, you cannot drive a car without a driver's license, right? The same should be applied for reading the test results imo.

 

That's my point

 

Sorry guys, I know not everybody here respects or values Matousec tests, but I think this info is worth posting to this thread.

So, as it was promised by Mike, build 119 scored 100% again

http://www.matousec.com/projects/firewall-challenge/results.php

 

OK, I have tested it myself.

And I should say OA 119 passes this test twice better, than Comodo.

The both show popup on dnstest.exe trying to start svchost.exe, but

if you allow this with Comodo, then test fails
if you allow this with OA, OA warns that dnstest.exe tries to change memory protection of svchost.exe.

The same story with awft1.

I'll continue comparison, there are some actually interesting tests in the suite.

 

OA execution alerts do not count. So its still one. And your the one talking about comodos site being biased

 

I did something way better than Matousec: I tested all the firewalls I wanted with the configuration and the rules I wanted. And I recomend that to anybody. Matousec tests are completely irrelevant for a particular configuration. I can score 90% with Kerio 2.1.5 correctly configured or 10% with latest Comodo if I add a badly wrote rule. It's the same for packet filters as it is for HIPS.

 

If you mean you can score 90% against matousec's tests on Kerio 2.1.5, then that is literally impossible, at least without the aid of HIPS or similar security apps. The quote from Matousec's site in lucas' #62 is all that really matters with regards to the importance of these tests. If your concept of a personal fw mirrors that of Matusec's, the tests might matter, otherwise if you view- as many do - a fw as only needing to provide inbound protection, then the tests don't matter.