Discussion in 'other anti-virus software' started by anon, Jul 17, 2014.
How does Panda score against PUPs today? I'm asking this cause during my own tests of v2.x it let many through.
FP picking up in Trend Micro and McAfee.
Yes even F-Secure got downgraded because of FPs. At my work we get a lot of calls from our customers because Avast quarantined our software or some program component (usually dll). I'm quite surprised it got 0 FPs in this test.
congrats panda, another good result
So, the essence is that the test set is too small to justify the average performance of any antimalware. So, all of us should take such results with a grain of salt.
So, that is the worlds largest Real-World Test of Internet Security Products
I wouldn't say that. At my work I'm in specific situation - I support users of our proprietary software. It is not widely used software so some AVs sometimes detect it as FP.
I agree that AVC uses the largest test set but that is too small compared to the actual number of new malware samples evolving everyday and the number of existing clean executable files (huge).
Ref: An old inforgraphic report from McAfee reports that new malware samples increased 44x in 2012 from 2011 and also it increased 35% from Q3 (10M) to Q4 (12M) of 2012. Now, its an old data back from 2012.
..... but that's how the statistics works.
I know but this is why the perception about a product varies from user to user.
Compare the AV-Comp results with these:
And see what money can do for an AV company.
I see you on the boards over there - if it's rubbish, why?
Have YOU done any testing to back up your assertion? Why not? See for yourself, don't be spoon fed.
Have you done any testing as to the prevalence and behaviour of the malware?
My tests are over there - I know you troll MapwareTips, you as a member of MalwareTips should know that. Where's all your tests and reviews? Bty - I just posted your comment. Expect a warm welcome.
If a file doesn't exist in the real world, then how do I get it?
Btw, why do you even care about what he said? He knows nothing about testing or stuff like that.
I am not the one accusing AV-Comparatives and the participants of their tests of criminal behaviour without proof. And no, I don't consider a bunch of people and their premature playful relationship with security software as a reliable source of information.
To be fair, despite my high opinion of AVC, McAfee got better results there than what a lot of user complaints about poor detection would suggest at some point (old heavy Norton doesn't count because it did poorly at heuristics), meaning even they badly screw up from time to time. Not to mention even amateur tests are better than the likes of AVT.
Some off topic posts removed.
Let's keep this thread on topic and civil. Avoid attacking each other for any reason.
Yes somehow this bytetrash exists but it doesn't mean that this file is spread. There are enough sites and malware play-samples but only for internal testing, files that are never itw or affect real customers.
It was shown often enough that those everyday test packs like you see for example on malwaretips contain some trash, some harmless files (even installers of other AVs ), files that itw don't exist in that form. Beside that often the only way for the self-called testers to identify a file as malware is if some AVs (most HMP or MBAM) detects it. Yes, very valid...come on.
Those tests are good to see some popups and get an impression of how products work, but they can not be used to compare detection rates or to compare with professional tests that use reliable prevalence data.
I don't put installers of other AVs, are you sure it was my pack? I can recognize an AV installer from a malicious file, I don't say a harmless toolbar is a rootkit... I can see how many files are really malicious and which are clean based on replies from AV vendors.
Here's Avira: https://analysis.avira.com/en/overview?uniqueid=F74zKbvYHp491yPH50LsEPK3H9rpGD2g
The files used for testing by AV-Comparatives are delivered by AV vendors, so they can add everything. Note that I don't upload malware packs for testing purpose, but to submit undetected files to security companies. Although these YouTube tests aren't the best vector to check if an AV is good, they're still much better than the tests done by AV-Comparatives.
I still don't understand how I upload files that never affect real customers...
Where did you get that "info" from?
Because random files from random sources on malc0de etc. didn't necessarily propagate and actually end up on end-user machines, users other than some people who reverse engineer them for a hobby. (or folks like you who input them in your browser url)
Not sure, most likely from AV-Comparatives' site. This is known for a long time.
I don't use malc0de or any malicious URL lists.
Separate names with a comma.