PC & Tech Authority - - CONTENTS -

Cy­ber threats are on the in­crease, but is your se­cu­rity suite re­ally pro­tect­ing you? We pit 10 of the best prod­ucts against each other to nd out

Ev­ery year the risk of at­tack goes up, with cy­ber crim­i­nals find­ing newer, smarter ways of at­tack­ing you. While Win­dows 10 has boosted base se­cu­rity lev­els, you’re still not com­pletely safe.

Noth­ing shows this as much as the re­sults of our test­ing. The last time we tested se­cu­rity suite, the low­est-rated prod­uct man­aged a pro­tec­tion rat­ing of 88.1%; this year, the worst prod­uct man­aged just 48%. That’s quite ter­ri­fy­ing.

For­tu­nately, there are bet­ter prod­ucts out there that can dra­mat­i­cally im­prove your pro­tec­tion and stop you get­ting hacked or your pri­vate de­tails be­ing stolen.


Test­ing se­cu­rity soft­ware means turn­ing to the ex­perts. This year all the tests were per­formed by SE Labs, run by Si­mon Ed­wards.

SE Labs uses the Anti-Mal­ware Test­ing Stan­dards Or­ga­ni­za­tion (AMTSO) stan­dard for test­ing. What this means is that all of the prod­ucts have been tested thor­oughly in a way that the re­sults demon­strate re­al­world per­for­mance.

All prod­ucts are tested us­ing a dou­ble form of at­tack. First, there are threats col­lected in the wild, which means the kinds of threats that your com­puter is ex­posed to. These should be rel­a­tively straight­for­ward for a prod­uct to deal with. Sec­ond, prod­ucts are ex­posed to tar­geted at­tacks. These are cre­ated us­ing pub­licly avail­able free hack­ing tools, so no unique mal­ware was writ­ten and there’s no tech­ni­cal rea­son why any soft­ware should do poorly here.

When it comes to block­ing a threat, there are de­grees of suc­cess. Ideally, a prod­uct should block mal­ware be­fore it runs, but some prod­ucts let an ex­ploit at­tack, stop­ping it af­ter it’s de­tected. To show this, prod­ucts that block a threat com­pletely are rated higher than those that don’t. Prod­ucts that al­low mal­ware through are heav­ily pe­nalised. The scor­ing sys­tem works like this:


If the prod­uct de­tects the threat with any de­gree of use­ful in­for­ma­tion, it is awarded one point.


Threats that are dis­al­lowed from even start­ing their ma­li­cious ac­tiv­i­ties are blocked. Block­ing prod­ucts score two points.

NEU­TRALISED (+1) Prod­ucts that kill all run­ning ma­li­cious pro­cesses ‘neu­tralise’ the threat and win one point.


If, in ad­di­tion to neu­tral­is­ing a threat, the prod­uct re­moves all sig­nif­i­cant traces of the at­tack, it gains an ad­di­tional one point.


If the threat com­pro­mises the sys­tem, the prod­uct loses five points. This loss may be re­duced to four points if it man­ages to de­tect the threat (see De­tected, above), as this at least alerts the

user, who may now take steps to se­cure the sys­tem.

Pro­tec­tion rat­ings are then weighted, scor­ing prod­ucts that pro­tect against threats far higher than those that were com­pro­mised. This in­for­ma­tion is then scored as a per­cent­age.

It’s im­por­tant that any se­cu­rity prod­uct lets le­git­i­mate pro­grams carry on with­out be­ing blocked. Af­ter all, any AV soft­ware that proves too an­noy­ing to use will likely be turned off. Prod­ucts were also tested to see how they dealt with false pos­i­tives.

Again, there are de­grees of suc­cess here. A prod­uct that blocks the lat­est ver­sion of Word was given a lower score than a prod­uct that blocked an ob­scure util­ity. How the in­for­ma­tion is pre­sented to the user is also taken into ac­count. It’s rare for se­cu­rity soft­ware to sim­ply block soft­ware, and it will usu­ally

ag up a warn­ing with ei­ther a rec­om­men­da­tion to block or con­tinue.

All of this in­for­ma­tion is put to­gether as a to­tal ac­cu­racy rat­ing, scored as a per­cent­age, demon­strat­ing the over­all ef­fec­tive­ness of a prod­uct. We’ve graphed the to­tal ac­cu­racy rat­ing. We’ve also graphed each suite’s ac­cu­racy in iden­ti­fy­ing le­git­i­mate soft­ware, so you can see which are the best for avoid­ing an­noy­ing false pos­i­tives.


SE Labs tested the cur­rent ver­sions of prod­ucts avail­able, per­form­ing tests for a three month pe­riod. Typ­i­cally, ver­sion up­grades don’t af­fect the per­for­mance of a prod­uct, but new fea­tures are added. In­stead, most se­cu­rity soft­ware is up­dated at least daily to im­prove its threat de­tec­tion and block­ing.

For our re­views, we’ve used the best-value ver­sion of a prod­uct avail­able. Typ­i­cally, higher-cost suites just pile on the fea­tures and num­ber of sup­ported de­vices; lower-cost pack­ages have the same ba­sic pro­tec­tion.


Se­cu­rity soft­ware isn’t just about mal­ware, and man­u­fac­tur­ers have added lots of fea­tures over the years to cover all ar­eas. Some add-ons are a waste of time; oth­ers are sur­pris­ingly use­ful. You’ll of­ten get cre­den­tial/pass­word man­agers that in­te­grate into your web browser. These can do some use­ful things, in­clud­ing gen­er­at­ing im­pos­si­bleto-guess pass­words, auto lling pass­word forms and auto lling other web forms, such as store check­outs. These are all pro­tected by a master pass­word known only by you.

Parental con­trols vary slightly, and aren’t avail­able with all pack­ages, but they let you set up lters for adult con­tent, with pre­set

lters and cus­tomis­able lists of URLs to block. Many of them also al­low you to limit the time your chil­dren spend on­line.


Many of the pack­ages we’ve re­viewed this year have mo­bile ver­sions avail­able. Some pack­ages in­clude mo­bile as stan­dard, but with many, pro­tect­ing your smart­phone or tablet re­quires an ad­di­tional sub­scrip­tion. Gen­er­ally avail­able for An­droid, some also work on iOS.

It’s ar­guable how use­ful these prod­ucts are, as mo­bile op­er­at­ing sys­tems are well locked down, and apps are mostly avail­able only through of cial stores, so it’s far harder to get a mo­bile virus. We don’t rec­om­mend pri­ori­tis­ing mo­bile pro­tec­tion over desk­top pro­tec­tion, and the mo­bile per­for­mance of each prod­uct has not been tested.


We’ve listed the of cial man­u­fac­turer’s pric­ing for each prod­uct, but you may be able to nd a bet­ter deal on­line by buying a boxed copy of a prod­uct. Don’t worry if the prod­uct you buy seems out of date, such as the 2017 ver­sion; all soft­ware will up­date to the lat­est cur­rent ver­sion on in­stal­la­tion. You can used boxed soft­ware for fu­ture up­dates, too, rather than hav­ing to re­new your li­cence through the man­u­fac­turer af­ter a year.

“The last time we tested se­cu­rity suites, the low­est-rated prod­uct man­aged a pro­tec­tion rat­ing of 88.1%; this year, the worst prod­uct man­aged just 48%”

Newspapers in English

Newspapers from Australia

© PressReader. All rights reserved.