Free Comodo AV is out!

Firecat,

You are a gentleman! Thanks for kind words. Comodo has been in this business since 1998 and we are the second largest Digital Certificate provider in the world after Verisign. First we launched the personal firewall and kept immproving it and we will continue to improve it, now we are going to do the same thing for the Comodo AV and Comodo Anti spyware (soon). So please continue to support us by using our products and providing us with your valuable feedback about how we can improve our products.

thanks
Melih
Comodo

 

Fellow Creatures,
I will be looking to replace CA AV in a few months if interested in the "why" see my posts in this link:

https://www.wilderssecurity.com/showthread.php?t=123623

I need good detections with low resource useage for my older XP compaq machine running 800MHz Celeron 256 Ram. It's older but very solid and reliable but about all tapped out for expansion upgrades. I also do not care for suites as a genral rule.

I will keep a sharp eye out on this new AV as it is attempting to address my needs it seems. I like Firecat am cautious. .

I have done some research seems like this company is on track and there business model it makes since. Their site is informative certainly the company is approaching 10 years of life with a product line. Would like to see more tests myself.

I will watch this product with keen interest and hope to see improvements in the areas disscussed in this thread so that my comfort level is increased to the level I can give them a try.

Good luck I wish you well Comodo!

 

Thanks Mercurie,

You can even help us by telling us the features you would like to see in this Comodo AV. You can, if you like, install it for a short period so that you can play with it enough to give us your "order list" of your "wishes". This way you will have an AV that you effectively ordered . We are already well down the track with addressing mem usage, increasing detection etc, we want more feedback about how "you" would like to see AV operate. So any feedback will be greatly appreciated.

Thanks
Melih

 

This would be nice when it also would work on Windows 98.

 

Time to upgrade man, it really is

 

Just reinstalled it and it works well
So if Comodo offers nice freeware i would like to try it.

 

Show him the money, man....really

 

Melih, can you name just few packers that are already supported by Comodo AV ? Also do you already detect Polip/Polipos? Last time i checked it was still not detected. If not, let me know and i'll send 2 samples.

 

RejZoR

We have a support for 4 main ones at the moment (i don't know which ones top of my head but can find out if you like (i know one is UPX ). we have a team of people working on writing around 27 different unpackers but will take time.
Please do send me the samples so that we can analyse it. Much appreciate it.

We are also extending our Heruistic engine to catch viruses without signatures, any help is appreciated.

thanks
Melih

 

If Comodo permits me to post here the results of their AV solution against the av-comparatives test-sets), I will do that.
P.S.: I do not tell results in advance or privatly to peoples.

 

I'm certanly interested in results. I hope Comodo guys will grant it.
Even if they don't score that well, thats ok. It's new program so there's still lots of time and room for improvements.

 

Please go ahead. We know we have to improve detection rates which we do on daily basis, especially with the next version. Our aim is to be have the best detection rate in the market and it will take us few months to get there ;-) . In order to save you some time, if you wait until the next version (by end of this month) then publish it, it will give us time to improve it further. However, please feel free to publish the current ones if you wish.

Also, you can help us by submitting us virus samples you might find/have, or even help us write some unpackers ( we have 4 in the next release and our guys are busy writing another 27 ) or even have ideas about how to improve heuristic engine. All help welcome Comodo AV is your AV product! You decide what goes in, you decide how it should function, literally!

thanks
Melih

 

Great! IBK, keep us posted

 

Thx for letting me post it.
Test-Set of February used. The scanner is a bit slow. It also crashed sometimes, maybe I will try to find out why when I have time for that.

OtherOS .......................9%
othermalware................19%
script...........................8%
worms.........................35%
windows......................34%
macro.........................12%
backdoors....................25%
trojans........................26%
dos.............................29%

TOTAL.........................27%
TOTAL without DOS.......24%

It is highly unprobably that ComodoAV will ever get the "best detection rate in the market". Currently even the little tool that Inspector C. wrote in 2 weeks is able to find more.
For good reasons, my rules do not allow to help writing your AV product or to send you samples, sorry. But I will from time to time look how your product improves .

 

Oh the harsh differences between possibilities and wishful thinking...

 

Isn't possible in a few months, but maybe in a few years...
But don't forget that other AV companies want the same...

 

Lack of good packers support is cetanly the problem here i'd say. I mean without unpackers you're basically toasted in these days. Let alone other goodies like emulators, generic unpackers, decrypters, heuristics and stuff like that (not to mention lots of signatures!).
But i'm certanly interesting to see how it will improve.
If it manages to get up to the Standard ranking on AV-Comparatives test, that would be a very good sign.

 

It is more than this. Most important thing which comes before all this are fileformats. It starts already with this - exploring fileformats ( such as Nullsoft Installer, Inno, Wise, Installshield etc ) takes a lot of time. Writing an emulator which is able to support an AV Engine in a reasonable way takes for a good team around 2 years and up. All other timeframes and "futuristic" thinking is unrealistic. Including highly complex malware detection took respected av experts sometimes months for only a single virus! Example: Zmist-based stuff. Zmist.D you can easily find - the fun starts with all other variants of this virus. Or ACG dos virus family - without detecting this you will have very bad results at Virus Bulletin

 

@Melih-Comodo: please correct the table on http://www.antivirus.comodo.com/comparison.html?currency=EUR&region=Europe&country=IS , it contains absolutly wrong information in various points.

 

Lets wait until the end of the month and test it again please with the new version. We have identified few bugs that caused unpackers (there are 4 at the moment) to give wrong results and some problem with heuristic engine also.

Lets just watch and see (and also do help pls ) how we can continually improve this.

(thanks for the help guys)

Melih

 

Aaaahhh, you work for a competitor company who charges for AV (?) and helping us to make AV for free will cause you not to have a job.(?)
You can always come and work with us, all good people are welcome at Comodo

Melih

 

IBK, thanks for testing the product, one question i have though, on what basis do you think its highly unprobable? Do you know how how many people are working on this product or what experience they have, or even what AV product they have written before or what company they worked previously?

Again, thanks for testing and look forward to your continued support in informing everyone at every release so that they can see possible improvements.

thanks
Melih

 

Does not matter. Even e.g. Symantec which has many employes and very good av experts and many sources where they get new samples have to work hard to stay under the tops regarding detection rate, a newcomer has a backlog of work to do to reach such a level of quality and as every day new work to do comes, it is probably that it will always stay behind the others. I do not think ComodoAV will ever beat e.g. KAV regarding e.g. detection rate. Its just not realistic imo.

 

Ok, thank you for your opinion.


Melih

 

Contrary to IBK (Andreas Clementi) I am not one of the most well-known AV testers. Nevertheless, here are my 2 cents:

1.
I agree with IBK that it will be very hard (if not impossible) for a newcomer to beat Kaspersky's "detection rate".

2.
I doubt, however, that the "detection rate" (as determined by standard AV tests) is of utmost importance:

As regards replicating malware (e.g., worms) it's important that you detect the samples that are currently "in the wild". This goal can be reached (also by a newcomer). It's less important that you detect a vast number of zoo nasties.

As regards non-replicating malware the "detection rate" does not really matter. This is because non-replicating malware is usually compressed or otherwise modified. In other words, it's important that you can detect a slightly modified Bifrost trojan and it's quite irrelevant whether you can detect a rare alpha-version of an unstable Chinese keylogger.

Consequently, the size of the signature database should be considered less important than the quality of the scan engine/unpacking engine.

If you test the quality of a scanner (and not the quantity of its signature database) you may easily come to the conclusion that, for instance, the technology used by Kaspersky is far from perfect. This also applies to the static unpacking engine.

3.
Unfortunately, major AV testers like IBK or Andreas Marx do not comment on this issue. Therefore, most people solely look at the "detection rate".

I also believe that major AV developers are quite happy with this situation. Because they have already collected a huge malware archive they are in a comfortable position to compete with newcomers (i.e., it would be foolish to admit that size (of the database) does not always matter).