PC Magazine: Which Antivirus Offers Best Real-World Protection?

Rather than test just one component of antivirus products, AV-Comparatives gives every component of each product a chance to do its malware-blasting job.

By Neil J. Rubenking
July 15, 2015

Some lab tests analyze how well antivirus and security suite products recognize and eliminate static malware samples. Others evaluate behavioral detection techniques, and still others measure each product's ability to detect malware-hosting URLs. But to the user, just what component blocked an attack doesn't matter, as long as the malware got whacked. The Whole Product Dynamic Protection test just released by AV-Comparatives lets every component of each security product have a chance to shine, and gives credit for successfully preventing attacks no matter how that success was accomplished.

The test is grueling, requiring researchers to challenge 20 or more antivirus products with brand new malicious URLs every day for months, making sure to keep them up to date. Over the course of the test, they hit each product with nearly 2,000 real-world samples. A product gets full credit if it either prevents the malware attack completely or removes all significant traces. If the malware compromises system security, naturally the product gets no credit.

What about cases where the product displays a notification asking the user what to do? In those cases, the researchers always choose Allow, OK, or the equivalent wrong choice. A product that prevents infestation despite the user's error gets full credit. One that allows system compromise if the user makes a mistake gets half credit.

Click the URL above to read the full article