Was that to me?
I was answering you. I should have used quotes to clarify that.
The way I see it. Comodo AV is relatively young. The protection that CIS offers exceeds what any AV can offer. Just give the detection rate time to mature some more before it hits the limelight of detection rates based tests and matching machismo (my ~snip~ is bigger than yours … 
).
I started using the AV with CIS v3.5 because the suite was light; even lighter then CFP with Avira (I was running Vista on older hardware; so every bit of system load mattered). Even though the detection rate was below average back then I never got infected. That experience shows me prevention is more important than detection.
Relatively young? bad excuse, another?
You are comparing apples and oranges… why do you avoid to compare the AV’s? relatively young? xD
So what?
There are many Prevention/Protection tests, and CIS is usually bypassed by malware with false certificates, the actuall 5.5 CIS can also be bypassed due to some bugs, and there always will be bugs.
https://forums.comodo.com/av-false-positivenegative-detection-reporting/report-trusted-and-whitelisted-malwares-here-dont-attach-live-malware-t67172.0.html
Great that Comodo AV did good in this test. I have done test with the AV module myself and have always considered the detection to be very high 
@ lordraiden - How do you even know that there are no changes/improvements being made. In the end you are just a member on this forum. You can’t know what is going on behind the scene’s. Not all small details/improvements are being noted in the change-log. Also Melih already stated that the AV engine is fine, and does not need major changes. Because of that most improvements are being made in creating generic sigs, introducing services like valkerie, and decreasing the size of the AV database.
Keep up the good work Comodo :-TU
- 100
I have said it many times before in this forum.
CIS is built to protect. There are new ways to protect the users. These testing organisations don’t have the ability to test that (they readily admit it and say it consumes too much of their time to test something like CIS).
so, what difference extra 1-2% it detects more or less compared to others? how will that effect the user? Well It won’t, because of “patent pending” “Automatic Sandboxing”, users will be protected from unknown malware because it will automatically be sandboxed. Thats a new kind of protection, we don’t need to detect to protect.
The only value of these old method of testing is for marketing purposes only. Marketing aspects are the only value to Comodo to be honest.
And there are quite few tests done by respectable organisations and you can find the results in this forum (there is a topic about that), you can also find real world tests, by real people done against real malware which is the real deal and most respectable one imo.
Melih
Please explain this to me: Norton got 95% detection (which is high) in this test
Yet it failed miserably in this real world test
How many proofs does anyone need to understand “real world of malware” is different than “test environments” that AV vendors PAY for?
Melih
I would say the more malware you use, you will get a more reliable test.
and 400k malwares is a lot more then languy used.
but the other way around i guess those languy use is newer? as those av-comparetives use, are up to 6 month old.
btw weren’t we once told that CIS av would get tested in the next av-comparetives?
whether old or new…the question the user want answered is : Will it protect me?
And how would 99,9% of detection help you when your PC has just been infected?
I tested some AVs on a virtual machine. Avira has mostly 99% of detection
in AV-tests and in my test Avira could barely detect my malware samples so the system got infected.
Like Melih said, there is no 100% protection by blacklisting. When an AV-test is released, there is already new malware that might cripple the winner of such an AV-test. So it’s not enough for an AV to have a huge database of signatures, it needs also methods to block malware which doesn’t have a signature yet; that is much more important!
At first the malware must be blocked, the signature can (or better said: will) be created afterwards and I think Comodo does a great job here.
Fair enough answer, and one that I appreciate. Thank you!
My understanding is that you Comodo guys are taking extra steps to ensure 5.8 will be a great version. Thanks for that and for releasing a great HIPS firewall to the public for free. And you have already submitted your product to 2 serious testing organizations. These are good steps to the right direction.
But I don’t think that such threads like this one, especially when commented upon by Comodo staff, are steps to the right direction, quite the opposite in fact: “According to faravirusi.com (blog of a Comodo Malware Research Team volunteer) […] 98.33%”. And why would a Comodo volunteer not pick the right samples to achieve such a result ? Taking such figures seriously (the same is true for many video reviews of Comodo) is like taking Bob (avast’s fan) seriously when he derides Comodo.
I understand this is a free forum, everyone is allowed to post their opinion (and as a testament to that, Melih has been shown to be extremely tolerant to unfair insults directed to him), but, Comodo staff, please refrain from making comments on test results from private users. It’s a form of approval and you are IRRITATING experienced users.
I do not avoid anything. I am not a Comodo employee and you have been around long enough to know that mods are end user volunteers. I don’t call any shots within Comodo.
Speaking for Apples and Oranges. Detection testing is the Apple and HIPS/sandbox based solutions like CIS are the Oranges.
Why test an orange like it were an apple? I know that getting your name out there in the Apples world may drive people to the world of Oranges. But, in due time…
So what? There are many Prevention/Protection tests, and CIS is usually bypassed by malware with false certificates, the actuall 5.5 CIS can also be bypassed due to some bugs, and there always will be bugs. https://forums.comodo.com/av-false-positivenegative-detection-reporting/report-trusted-and-whitelisted-malwares-here-dont-attach-live-malware-t67172.0.htmlI did never say that CIS is flawless. Please refrain from putting words in my mouth just for the sake of trying to start an argument. You are still on the mods radar.
To comment on the above. Digitally signed malware is a problem. I take Umesh word for it that it is usually “just adware”. Kaspersky Internet Security will by default trust everything that is digitally signed. At least with Comodo there is the intermediate Trusted Software Vendor list; not all digitally signed applications are automatically trusted and vendors can be taken of that list. And yes I am proponent of making the TSV list more configurable and closer to how it was in older versions.
WTF are you on? i haven’t said anything about 99% detection.
And am not really sure why you give me that answer, to my answer, that you can’t really disagree with unless i am wrong about how old those virusses languy use.
It would be like if you compared the difference between 2 different kind of car crash test, and then i suddenly started telling you that nobody is 100% safe in a car crash, but you have to use the newest technology to avoid them instead.
2nd There have recently been found 2 malwares that bypass CIS easily, so you cant say CIS is 100% secure either, and perhaps some have got infected while using CIS because of that, but would not have been if using avira.
3rd some people can’t figure out how to use CIS, and have to give up, they have to rely on normal AV.
It will, because CIS doesn’t put all the eggs in one basket.
Normal AV’s all put the eggs in one basket, a basket called detection.
I’m glad I was able to test Comodo 5.8 RC and have this results. Comodo is on a high way and I want to congratulate this team. :-TU