Very surprising Anti Virus results

  Totally-braindead 13:34 15 Feb 2007

I just bought the Computer Shopper magazine April 07 mainly because of the anti virus tests advertised on the front. The results are surprising, very surprising.
They tested for the ability to find and destroy viruses and malware and heres what they found.
12 were tested.
Kaspersky Anti Virus 6 72.6%
Steganos Anti Virus 2007 71.2%
Eset NOD32 2.7 63.6%
F-Secure Anti Virus 2007 63.1%
Grisoft AVG Anti Virus 7.5 Free 59.2%
Panda Antivirus 2007 56.2%
Bullguard 7 55.3%
Symantec Norton Anti Virus 2007 54.8%
Softwin BitDefender Antivirus 10 53.7%
McAfee Virusscan Plus 2007 50.5%
Trend PC-cillin Internet Security 2006 44.5%
ALwil AVAST 4.7 Free 44%

Now they make a big thing about the AVG free beating a lot of paid for anti viruses. Which I must admit does surprise me as well but the other things that I notice looking at what they say the AVAST doesn't do as well and although I don't use it I always thought AVAST and AVG were about equal and this does not appear to be the case. I am also very surprised that NOD32 didn't win.
Also looking at the results even the best could only manage 72.6% of the viruses and malware, leaving 27.4% undetected.

Now I have to point out that reading the way they have done the tests I do think the results are slightly skewed and this is why I think so.
The results I've posted are a combination of 2 tests for both viruses and malware, its an average. The best at detecting email viruses was Kaspersky and the best at detecting malware was NOD32. Now its all very well to run these sort of tests but reading the final results you think some are better than others and they aren't in my opinion at least. I was always under the impression that Malware and viruses were different and therefore looking at the anti virus aspect alone is in my mind a fairer way of looking at them as they are anti virus programs and not malware programs. If you look at the antivirus alone and ignore the malware you get
Kaspersky 90.8%
Steganos 90.4%
Panda 89.2%
F-Secure 78.1%
NOD32 76.5%
McAfee 75.3%
Norton 74.5%
Bullguard 72.9%
BitDefender 68.1%
AVG free 66.9%
PC-cillin 64.9%
AVAST free 59.8%
What does everyone think about this. Is the ability to detect and destroy malware as important as the anti virus aspect or is the ability to destroy viruses more important?
After all we have a load of free tools to deal with malware already so to my mind the anti virus aspect is the most important, they are anti virus programs after all.

  Totally-braindead 13:37 15 Feb 2007

Should have said, their recommendation was for Steganos as it was only £20 and is the same software as Kaspersky which is £8 dearer.

  anskyber 14:03 15 Feb 2007

In short I do not know. I imagine you like others including me keep a weather eye open for the reviews of different AV or security suite software.

The only consistent think I have found is a remarkable lack of agreement (generally) between reviews. I understand there is a standard group of 1000 nasties which are used to benchmark AVs and it is suggested some AV companies fine tune their software to perform well against the 1000. A result which could compromise performance elsewhere.

The truth is, given the nature of nasties, they will always appear before the solution to block them. Even AVs with "guessing" features can do only that and never be perfect.

Gandalf has said so many times here that the biggest single contributer to keeping nasties out is free, common sense.

I agree with him but common sense with suitable Avs etc is the ultimate policy. Otherwise read around the reviews, accept none on face value (as some, but not you, have here) and take a view on a suitable product.

  robgf 14:05 15 Feb 2007

I would have said that malware is adware, spyware etc, which would be very difficult to rate, as one persons adware, is another persons useful program.
So some scanners could be checking for fairly benign malware, that other scanners don't even consider to be malware.

No test results can give a truly accurate result, as it depends on the viruses that are used for testing, some antivirus programs have enormous definition lists, that still protect against very old viruses, that you are never likely to encounter.

My only definite trojan infection, was a drive by attack when using IE.
But RegProt notified me of an attempt to alter the startup registry and I was able to block it. However the trojan still managed to gather together my passwords in a text file in its folder.(RegProt had given me the program path)
During the attack AVG didn't make a peep and even when pointed at the correct folder didn't detect it.
Out of curiosity I then ran several online scans (Panda, Bitdefender) neither detected it.
Spybot and Adaware similarly failed to detect it, but A2 scanner detected and removed it (hurrah we have a winner).
Which illustrates how unreliable antivirus programs are, despite the makers claims to the contrary.

Prevention by good practice is better than cure, or antivirus software.

  Totally-braindead 14:57 15 Feb 2007

It is certainly right that a bit of common sense and an AV program is the right way about it.
I am somewhat concerned at the attitude they had about the results.
They mention Kaspersky, Steganos, F-Secure and NOD32 as being the best then go on about AVG deserving a mention because although its not as accurate as the aforementioned ones it does a better job than most of its rivals and is free.
I've no problem with that it the next bit I'm not keen on and I'll quote directly.
"It is a sad situation when the home anti-virus products from the three world leaders, Symantec, McAfee and Trend Micro, fail to impress. Norton AntiVirus and McAfee VirusScan are distinctly average and PC-cillin's virus detection abilities are poor. If you are using any of these, we recommend that you replace them as soon as possible with an alternative from the above list."
Its the recommendation I have a problem with.
I use AVG and its kept me virus free for about 3 years now but when you look at the actual virus scan results McAfee and Norton are smack in the middle. Its only when you add in their poor results at malware that they drop down to nearer the bottom.
And to me at least as I said before I think anti virus is for viruses not for malware.
I'm happy with AVG and am not going to change its just, I'm unconvinced with their arguements is what I mean.

  Watchful 15:36 15 Feb 2007

Well from those figures there's not a lot of difference between the paid-for one I had (PC-cillin) and the free one I now use (Avast) and as I have not had any major problems for four years and do surf sensibly, I'm sticking with what I have. In the end you can only go off your own experience with an application whatever the figures show.

  GANDALF <|:-)> 16:07 15 Feb 2007

Go to 3 different virus testing companies and you will get 3 different results which is why I never take any notice of what they publish. I do suspect that they may use files that would never be seen in the wild and I prefer to base my opinions on experience in real life.

  Totally-braindead 16:18 15 Feb 2007

Yes I see what you mean GANDALF <|:-)>, this one was done by a magazine but they mention using 251 real life viruses that were spreading across the internet at the time the tests were done and each file had been detected in the wild for at least 2 weeks, which I presume means that the AV companies should have had time to write a virus definition. They do mention they got the viruses from MessageLabs who might be affiliated with some AV company I don't know.
I always read the tests but do not necessarily completely believe them. I too believe in what I and others I know actually experience rather than what the test centres say but it is interesting.

  SB23 16:22 15 Feb 2007

I agree with GANDALF <|:-)>, I'm one of these people that would rather find out about how good, or how bad a program is for myself.
I tend to find programs that I wouldn't give a second thought to, do well, and programs that I thought were good score badly.
As for Avast, my program of choice, coming bottom, I don't care. It has served me well since I ditched Norton, and McAfee, and I would never change that.

  Z1100 16:26 15 Feb 2007

I would believe that the results are fair. Unless some clever bod comes up with a test that can recognise as yet un-known viruses, and by that I mean at least flag a file as suspect, then the tests will be skewed.

No good looking for a car in a car park if you already know one is there. And I would have thought most users would have defence against Malware Adware and all the other threats out there since they are all free in magazines.

I am now off to look at Kaspersky...


  mammak 16:57 15 Feb 2007

Thanks for the information and like you I am shocked by the results,

but like SB23 and Watchful Avast has served me well and I am sticking with it also.

This thread is now locked and can not be replied to.

Elsewhere on IDG sites

Dell XPS 13 9370 (2018) review

The art of 'British' pulp fiction

Best password managers for Mac

TV & streaming : comment regarder le Tournoi des Six Nations 2018 ?