Security programs/features and performance hit?

I've been looking at a lot of different security software recently and I have wondered if there is a way to measure any performance hits introduced with different software? I know about benchmarking programs like PCMark 7 and PassMark Performance test, but would these actually be able to tell me anything? I don't see anybody using programs like these to test their setups except for AV Comparatives (which uses PCMark, which I guess answers my question). I only see claims like "My system felt slow with it installed so I got rid of it." To me, "noticing" that a certain product slows down your pc could be because of reasons other than it actually slowing it down (i.e. disliking a program because of something you read). Do you guys use any methods or programs to test the performance impact a certain program has on your pc? I tried running both PCMark7 and PassMark Performance Test in a VirtualBox VM and both had major issues.

 

Yes, if you don't use benchmarks, it's subjective and each person has different perception of "slowdown" and different tolerance. That's why you see in the same thread people complaining about slowdown and others saying "it's light as a feather".

I use the task manager to see CPU and disk access and the explorer perception. For example, no matter how much i like Comodo, when i install it, clicking around, opening folders etc, is a liiitle bit slower than without Comodo or with NoVirusThanks. Could be placebo effect? Of course. But i am convinced it's so. I associate that with the sound the folder do when opening. Don't ask me details. I do it quickly, that's the secret.

 

Various test organizations do performance testing, AVTest does, click each one and it lists how heavy it is on their test system;

http://www.av-test.org/en/tests/home-user/windows-7/sepoct-2012/


Also Raymond.CC does a yearly performance test on all aspects, click the dropdown to change aspects. I have stated for years MBAM is heavy on a system, and his test results show it is actually one of the heaviest things you can install. Note the versions though, he tested Bullguard2009, which I felt was hideous, and heavy. I find 2013 very light, and fast, but it was a complete re-write for the most part. So factor the versions, and watch for his updated tests to be released;

http://antivirus.raymond.cc/launch.html

(much of this test is irrelevant because he used older product versions, search around, he might have a newer test)

 

I'm not just talking about antivirus software, but any software in general. Things like Shadow Defender, EMET, Appguard, etc.

 

Keep in mind that there are many different potential bottlenecks on a system, and different programs will tax different areas of a system. That's not to mention the software environment. So different programs are going to perform differently on different systems, just like a game that relies more on CPU may perform better or worse on your system than a game that relies more on GPU. There may even be things that you can't measure with your system resources, like response from cloud servers.

You can use this, though: http://www.passmark.com/products/apptimer.htm -- just make sure to run each app multiple times, since it'll be put in memory cache after the first time it's run. You can also find page load timers around.

So while subjective impression may be prone to bias, it's probably your best bet. The biggest thing to watch out for when trying different security apps, though, is leftovers from previously uninstalled software. Many can leave behind things like drivers, which can still cause conflicts with software you install later. It's not uncommon to see someone that has trouble running any other AV/FW after uninstalling a trial without cleaning up a bit. Just remember to look for uninstall tools to run after the regular uninstall (and I wouldn't trust uninstallers like Revo, as I've seen it cause problems that the program's own uninstaller didn't).

You sure do like to slam products that you don't use.

With multiple products installed, including MBAM Pro, apptimer registered a difference of 5 hundredths of a second (0.05); not a huge drag, and I think MBAM Pro alone only added about .02 or .03. Then again it seems to run mostly from RAM, and I have pretty fast RAM.

 

Actually I ran MBAM for about a year before figuring out that was what was slowing down my fast gaming PC. Then I gave my license to my father inlaw, who then complained for weeks his PC seemed slower. I removed it, and it was faster for him. So I put it on my daughters computer, she later complained it was slower. I finally removed it and sold my license on Ebay. MBAM really does seem to slow things down, and tests done by independant labs seem to confirm this. Raymond.CC found MBAM to be the almost the heaviest possible thing for applaunch and other aspects. MBAM free is fine, but the realtime protection seems to slaughter systems.

 

I agree: irrelevant. For example, the Webroot AV he tested is a version prior to the merge with PrevX.

I don't know that Raymond site but he may need to read and reflect more before recommending things IMO. I searched some of his 2012/13 articles and I saw indications to "TopTenReviews" and "Virus.gr" as if they were among the best test labs... and many other similar issues.

 

I don't doubt your experience. However, it seems that there's a greater number of others around here that don't experience the same, at least currently. Any software can and will negatively affect a system; it's just a matter of how many systems it affects and how the developer/company responds. Any number of other things (software and hardware) can also affect the software.

There's a difference between saying that it's your experience and may have been fixed in newer versions, and saying that it's definitely heavy and will "slaughter" your system in absolute terms.