It's true: Antivirus programs can slow your computer, sometimes by quite a lot.
Yet we've found several antivirus programs with system slowdowns so small that you'll never notice them unless you're doing something processor-intensive, such as playing a high-end game or rendering video footage.
|Passive slowdown||Full-scan slowdown||Quick-scan slowdown|
|Avast Free Antivirus||4.4%||29.2%||53.2%|
|AVG AntiVirus Free||5.3%||12.8%||25.6%|
|Bitdefender Antivirus Free Edition||2.2%||29.2%||N/A|
|Bitdefender Total Security||19.2%||32.7%||30.1%|
|ESET Smart Security Premium||2.0%||11.8%||N/A|
|Kaspersky Security Cloud Free||-0.9%||5.8%||1.2%|
|Kaspersky Total Security||11.8%||26.5%||21.6%|
|McAfee Total Protection||8.9%||96%||38.6%|
|Microsoft Windows Defender||N/A||20.3%||12.8%|
|Trend Micro Maximum Security||3.9%||56.9%||35.3%|
How we test AV software performance impact
We tested six paid Windows antivirus programs in the summer and fall of 2020 on a Lenovo ThinkPad T470 notebook computer. We tested five free Windows antivirus programs on an Asus X555LA notebook in the spring of 2019, and a sixth free Windows program, Kaspersky Security Cloud Free, in the fall of 2019.
We used our home-grown Excel benchmark test, which measures how long it takes the CPU to match 20,000 names with 20,000 addresses on a spreadsheet, to gauge system impact. For Kaspersky Security Cloud Free, we used a previous version of the same test that ran on OpenOffice.
We measured system speed before any third-party antivirus program was installed; after an antivirus program was installed, but not during any active scans; during a full scan; and during a quick scan. Each test was run three times for each program, and the results averaged.
The numbers recorded were the completion times of the name-and-address matching tasks. Slowdowns were calculated using the pre-installation completion times as baselines and calculating the percentages by which each task increased the amount of time needed to complete the matching task. Baselines were taken separately for each program's set of tests.
Among paid antivirus programs, ESET was not the fastest in all three categories that we measured: passive system impact, full-scan system impact and quick-scan system impact.
That may be because ESET doesn't have a quick-scan option. The lightest quick-scan impact among paid programs was held by Kaspersky Total Security at 22%, a percentage that was nearly twice as much as ESET's full-scan impact. Kaspersky did have the second-lightest full-scan impact in the bunch, however.
Kaspersky Security Cloud Free was the undisputed winner in the free categories, beating all rivals in passive, full-scan and quick-scan impacts. It actually sped up the system a bit after installation, indicating that it may be more efficient that the built-in Microsoft Windows Defender, which runs by default but deactivates itself if a third-party antivirus program is installed.
Kaspersky Security Cloud Free's full-scan system impact was a barely noticeable 6%; its quick-scan impact was a barely measurable 1%.
Most other programs were less consistent in their rankings. For example, McAfee Total Protection had a relatively light background impact, slowing down the Lenovo laptop by only 9% after installation. But McAfee's full scan hit our test laptop like a ton of bricks, slowing it down by 96%.
You'd probably notice that full-scan slowdown even if you were just checking email. You wouldn't notice ESET's 12% hit unless you were intensively using the processor already.
Likewise, Avast Free Antivirus (4.4%) had less of a background impact than its sibling AVG AntiVirus Free, (5.2%). But Avast's full-scan slowdown was the heaviest at 53%, while AVG's was less than half that at 25%.
It's hard to determine a second place among the paid programs. Two of them, Bitdefender Total Security and Kaspersky Total Security, had relatively high post-installation background system impacts, of 19% and 12%, respectively, yet traded that for fairly light impacts — 33% and 26% — during system scans.
In other words, you'd be trading a small amount of permanent background noise for huge slowdowns during full scans, which you could schedule to happen only at night while the computer is idling.
Establishing the second-place finisher among the free programs is even more muddled. Microsoft Defender Antivirus, aka Windows Defender, can't really be measured for background system impact because it's built right into Windows 8.1 and Windows 10. But it had a very respectable full-scan slowdown of 20% and a quick-scan one of 13%.
Meanwhile, Bitdefender Antivirus Free Edition doesn't have a quick-scan option. Its background impact is only 2%, the best after Kaspersky. But its full-scan slowdown is 29%, tied with Avast Free Antivirus for last.
It is clear that Avast is in last place in the free bunch. Not only was it tied with Bitdefender Free for the largest full-scan slowdown, but it had the largest background impact (after AVG AntiVirus Free) and a whopping quick-scan slowdown of 53%, the most among all the free programs.
How antivirus programs are marketed
Most antivirus brands offer three or four different paid Windows programs, all sharing the same basic protections but adding a few more extra features with each increase in price.
For example, Bitdefender offers Bitdefender Antivirus Plus, Bitdefender Internet Security and Bitdefender Total Security, and each costs about $20 more per year than the previous one.
But only the last has all the bells and whistles that Bitdefender offers, such as a password manager, webcam protection, parental controls or a system optimizer. Meanwhile, Bitdefender Antivirus Free Edition uses the same malware-detection engine as its paid siblings, just with far fewer options and extra features.
When deciding which antivirus program to buy (or download for free), evaluate which one has the features you need. For more on this, check out our guide on how to choose antivirus software.