A lot of YouTube testers try to achieve this. But it causes problems. A lot of vendors components work together. For example when real-time Protection is disabled in ESET the HIPS component is not fully functional. So it fails to detect certain aspects of malware it would normally be able to detect.
I was just giving an example of what information one can gather from such testing. The good products tend to perform consistently , including Bitdefender (which I really like the 2020 version). In my example if a product were to consistently or regularly perform below the average then it could provide information. My point is only that the testing is a valuable piece in a larger data set for one to make a decision.
This is true, but the point was that most AVs don’t allow you to disable signatures only while keeping all other components intact. In that example I believe ESETs BB is also disabled, though it is not well understood or known for its performance.
Maybe 5 years ago. Nowadays every test is essentially the same. Not a single product has pushed past 5% compromised over a year-long period since 2015. Hell, even on a month-by-month basis nothing's gone above 3% since 2018.
Yeah, but if a product is compromised more than 2-3% when all other vendors have no issues that’s kind of a glaring negative mark.
Even then it's only evidence of a downturn against the specific samples used by the testing lab. They have no relevance to how someone's antivirus is going to react to any malware they encounter in the wild.
My apologies, I thought you meant without any signatures as well as offline. I believe AV Comparatives offline detection does actually give some hints. Windows Defender is terrible offline. But it can’t separate the companies with good signatures that are updates regularly and good BB. I think the best test for good BBs is zero day malware, as we see in the hub, and ransomware variants that aren’t prevalent.
This is very true. Which is why I think it should only be used as one data point of many.
And looks like we have a new FP champ.
Have to agree. Very happy with Eset's performance. Looks like it's a keeper.Seems ESET at default is just fine.
So they cannot be average.
avlab.pl
checklab.pl
This is normal due to statistical errors. There are 3 new malware threats created per second. So, daily you have about 260 000 new samples. For 2 month test, we have 15 mln samples. Even if 99.9% of samples are polymorphic or similar, then you have 15 000 sufficiently different samples (it is an example, the real number is not known).
