I am really willing to point out a statistical analysis on the detection rate of antivirus programs via some results from some Virus Total Results. Right now I have only gathered five VirusTotal results tested on a Javascript, a Backdoor virus, a Trojan Dropper, a Trojan and a PDF Exploit. The results are compiled into a single MS-Excel file which you can download from the link below:
I will gather 100 VirusTotal and NoVirusThanks results and will then anlalyse them. In such a way I can find which antivirus is the best in terms of detection rate. For more info visit:
in the general topic section…but it’s too late. And mods here don’t move threads when needed…anyway, as far as I’m concerned, that’s advertising a web site with no serious goal in mind, concerning security matters. Your site basically prompts people to register to see anything. And statistics elaborated on the basis of a few observations with VirusTotal don’t mean much.
The most serious goal in mind is verifying the detction rates of the 42 concerned antivirus products and check their accuracy.
My site does not require registering, you can view threads and download attachments anytime you want.
AV Comparatives only check the antivirus solutions against 6 moth old samples, so they are not really out in the wild.
The VirusTotal Results have been gathered from many support forums where the posters have are talking about a new sample, not 6 moths old ones. Though I have been at the receiving end from Kaspersky who say that as VT uses a very old version of Kaspersky’s antivirus products, the detection might not be accurate. I agree with them, but that’s the best way we have got. I am in short collecting the VT results for all new threats received and analyzed from malicious urls and exploits that have been submitted in the forums. I lay no faith in detection of old samples as they are rarely out in the wild.
I mean people like the OP here make a big mistake. Same mistake is observed at professional level as well as at amateur level attempting to advertise themselves behind the pretext of helping security software companies with their tests and reports ;D One thing is clear, security software companies improve their products based on users’ feedback, people actually running it permanently, posting and reporting on forums. Serious people don’t give a crap for external testing and comparatives of security software (unless they’re called Comodo ;D :-X …yeah, I said serious people ).
Uptil now, 9 results have been analyzed. I shall be producing an early summary(not final) after 20 results have been analyzed. Some newly added categories are Bots and keyloggers. It seems that till now Avira, Symantec and Microsoft Security essentials can go far, though I am surprised with Avast’s (bad) result. You can download Update 1 from the link below:
My testing methods are very very simple. I gather VirusTotal results from various security forums and bring them undera statistical analysis to find out which of the 42 antivirus engines have caught the most number of samples. The test is depended totally on VirusTotal. The tests have been conducted on various virus categories. After 20 results have been analyzed, then a mini summary will be made. The final summary will only be made after all 100 results have been gathered and analyzed. Right now, I have gathered and analyzed only 14, the next Update 2 will be released with 15 results in all total. And please, I am not advertising my website in any way.
Update 2 has seen Avast catching 4 of the 5 new samples but Avira almost everyone of them. MSE and AVG are still very balanced in their detection rates.
I have released the Online Scanner Detection Rate Final Update. You can download the Summary or the Full Project Info(Along with which virus sample was tested and full review of the 20 tested samples etc.) I will not give any sneak peak about the results but the highest detection rate obtained is 70%(14 out 20). Download from the links below:
