Tested Firewalls & Leaktests

Just released today, don't know if it's been already posted. Very interesting.

http://www.matousec.com/projects/windows-personal-firewall-analysis/leak-tests-results.php

 

Good info. Thanks.

I recently replaced Safety.Net with Comodo. Looks like I did the right thing.

It dosn't seem so long ago that Sygate and Outpost were the highest rated firewalls. Times change.

 

Also read the statement about Outpost, it says something about cheating...

 

not cheating, just protecting in an inadequate way. According to them, user-level(ring3) hooking is weak against real malware
Thumbs up for Jetico and Comodo although leaktest performance shouldn´t be what decides a firewall choice

 

Heh:
Matousec is creating a bit of a stir; good; there have been previous threads here about the "business model".

The firewall leak tests are impressive work.
The ratings are interesting

There will be the usual: "leak tests not real world" blather: and there may be some truth in that.

What piques my interest more is the faithful recording of how various vendors responded. That is VERY interesting. It is only proper that vendors respond in a way that reinforces confidence (or otherwise heh) in their product.

What is good to know is that the tried and true are all at the top of the pack.

Shame that Avira was not around.
Be interesting as to how Norton 2007 would go.( maybe no real changes in coding?)
I see Outpost is at current version !

Not sure why CA did so poorly if Tiny is now part of their utility?

 

It was already posted
https://www.wilderssecurity.com/showthread.php?t=155955
and this post is related
https://www.wilderssecurity.com/showpost.php?p=891805&postcount=41

Comparing to the March leaktests from www.firewallleaktester.com it's notable the improvement that Comodo made during the 7 months.
Jetico 2.0 is improving too, it failed 2 tests while the 1.0 version would failed at least 4. Also v1.0 doesn't have self-protection.

Anyway the best outbound protection available is completely useless if the user always click allow without thinking twice.

 

What surprised me the most was the excellent showing of/by Trend Micro PC-cillin Internet Security 2007. Who would have thunk it

 

Hello,
I'm scared now. I need a firewall that passes all leaktests. I need something to protect me from myself.
Mrk

 

They got you covered..

http://forums.comodo.com/index.php?&topic=4163.0

 

Hello,

LOL

 

Well, just wanted to say that in fact KIS passes PCFlank and Surfer leaktests, not as noted on that review, based on my personal experience using KIS.

 

Hello,

Here's a little story:

John got this great game from a friend Oliver - of course, it's cracked.
Oliver tells John that he needs to mount the .iso using deamon tools to avoid the copy protection.
Then Oliver tells John he needs to run the keygen.exe to create a key for the game.
John begins with deamon tools. A bit of conflict with his AV, so he shuts it down.
He reaches the installation stage of the game and is asked for the key.
John runs a great firewall, which pops a question, asking John if he wants to allow keygen.exe to access the network. Of course he does.
All is well, the key is successfully created.
John installs the game.
Before the game launchers, the firewall asks him if he wants to allow the game to connect to the Internet. He heard from Oliver that you can download extra weapons from the game's site, so he allows it.
But now, the firewall asks him if he wants to allow the game to launch some dll (warns him of application hijack, dll injection, crc check). John is confused so he calls Oliver. Oliver tells him it's ok, that "hijack" option allows you to use ingame chat, that's why you need to launch the browser. Oliver knows his ****, so John does not doubt the doubtful explanation.
John allows the cracked game to connect.
Game over.
Mrk

 

lol, , SOLUTION: Next time John should buy the game.

 

So, John. What firewall did you decide to protect you from yourself?

(That was a biography?)

 

Hello,
No, I happens to buy me games - honest.
That was a hypothetical situation that probably applies to 90% of world's populace - the part that has computers that is.
I use no firewall to protect myself from myself. I just use a simple firewall to filter out the Internet noise. I have no intention of trying to nuke my computer to see how well it could survive me own follies.
BTW, no firewall can survive a format. Or help you if you decide to delete a few random system files here and there.
Mrk

 

Comodo PF are doing an excellent job on this area.

It's a program very easy to use, and also easy on resources...

Hope that they not add features that not belong to a firewall in the future.

 

Echo that

 

Well, since firewalls pass leak tests thanks to those features (HIPS, sandbox), it is difficult.
People want from firewall to block everything, but in order do so, it has to control everything.

By the way. the latest CPF beta passes the Coat test and in the list is not mentioned any security update. I note this, because there was a thread about Outpost firewall 4 vs old one and if people should use newest versions, when the old versions work just fine. Judge yourself.

 

I wouldn't trust anything that comes out from that place except only what is independently verified by a third party like Secunia, iDefense, etc.

It is nothing more than a store that has a product to sell: Firewall Bugs for Sale.
The bug descriptions are puffed up to look like reviews so that we get all stirred up over the product he is force selling to the firewall vendors.
Why settle for product descriptions when you can read real reviews?

Extensively covered here.

 

Hello,
I think the only real review is a self-test. Install it & try it.
Mrk

 

Hi Mrk,
Very true!
The review will have no bias but your own.

 

Well, I've done all the tests myself, including Coat, and I pass them all...
Using the latest beta of Comodo Firewall.

 

Hello,
Well, I pass them all, too. And I'm not using Comodo, even though it is a nice firewall.
Mrk

 

I dont know about others but i think the emphasis on firewalls passing leaktests is getting really monotonous.And just what is the point in matousec rating these firewalls leaktest results when he classes most if not all of the firewalls that have so far been tested for bugs as not worth a jot anyway?
I get the sense that maybe matousec thinks leak test scoring will somehow stimulate interest from the masses more than the bug analysis did.
ellison

 

So what firewall are you using?