Author Topic: New Antivirus performance benchmark Sept 30 2010  (Read 10834 times)

0 Members and 1 Guest are viewing this topic.

Offline wonderwrench

  • Sr. Member
  • ****
  • Posts: 223
New Antivirus performance benchmark Sept 30 2010
« on: October 08, 2010, 08:07:08 PM »
Passmark has updated their Antivirus performance benchmark to include most of the latest products. Avast 5 Free is tested but Pro and IS are missing from the test. Also they tested 5.0.594 not 5.0.667 for some reason. Avast 5 Free takes second place, only beaten by Norton AV. Symantec paid to have the testing done BTW. Passmark is a very reputable company so IMO the results are valid. Your view may differ. http://www.passmark.com/benchmark-reports/index.htm
Main Box*i7 930*GB X58A-UD3R*3x4 gig Patriot DDR3 1600 EL*EVGA GTX 460 1 gig*Intel X25-M G2 80 gig*WD 2TB Green*ASUS DRW-24B3LT*Samsung SH-S223L*LG WH14NS40*Corsair AX750*Rosewill Challenger case*Windows 8 Pro 64 bit*Avast 8 Free 8.0.1482*MBAM Pro*Firefox 19.0.1*NoScript

Offline DavidR

  • Avast Überevangelist
  • Certainly Bot
  • *****
  • Posts: 88895
  • No support PMs thanks
Re: New Antivirus performance benchmark Sept 30 2010
« Reply #1 on: October 08, 2010, 08:40:08 PM »
Symantec paid for similar tests to be done at the Dennis Technology Labs (between 07/07/2010 and 22/07/2010) also and are also reported in the forums.

Considering Dennis Technology Labs one was meant to be for security suites, avast free came 4th beating many security suites and topped the scan speed tests.

In the Overall Protection scores avast free came 4th, where many of the paid security suites didn't exceed the average score.
Quote
The only free product that performed above average was Avast! Free AntiVirus 5.

So it seems to be following a pattern, avast is performing very well against the competition, even in the free version ;D
Windows 10 Home 64bit/ Acer Aspire F15/ Intel Core i5 7200U 2.5GHz, 8GB DDR4 memory, 256GB SSD, 1TB HDD/ avast! free 24.2.6105 (build 24.2.8918.824) UI 1.0.799/ Firefox, uBlock Origin, uMatrix/ MailWasher Pro/ Avast! Mobile Security

MasterTB

  • Guest
Re: New Antivirus performance benchmark Sept 30 2010
« Reply #2 on: October 08, 2010, 09:11:31 PM »
Don't get me wrong, I'm starting to believe the love for Avast!s speed but there's something that's being bothering me.
When I used Nod32 then moved to Eset Smart Security I remember that it scanned All files, not just a list of predefined extensions, and it was still faster than most.
I know things have changed (I dropped Eset for it's lousy firewall and Outlook limited most email and spam features) but Avast! doesn't scan all files and so, to compare that with an AV that does it's kind of unfair, isn't it?
Unless you can tell me for sure that I can enable the resident protection FOR ALL FILES not just the ones predefined and Avast will still be as fast, I don't believe this performance tests that much.
Besides, they are Dependant on hardware and not all of us run the same hardware.

Martin.-

Offline DavidR

  • Avast Überevangelist
  • Certainly Bot
  • *****
  • Posts: 88895
  • No support PMs thanks
Re: New Antivirus performance benchmark Sept 30 2010
« Reply #3 on: October 08, 2010, 09:50:10 PM »
Well speed shouldn't sacrifice protection as in order to be granted some AV tests awards, etc. you must be scanning files that are of a certain type and that is commonly what is used by AVs as their default minimum scan. These are the types of files that present an a) immediate risk (executable files) and b) those considered at risk of infection. Then you have the other levels of protection, right up to paranoid.

So for these tests my guess is that the testers aren't stupid and try to select a scan level which is comparable across the board.

Don't forget that for the most part on-demand scans are going to be scanning files that would otherwise be dormant or inert. Files that are active on being run are obviously going to be scanned by the on-access scanners.

They are dependant on hardware, but all the tests are done on the same system specs so there is a direct comparison, that should be roughly the same if it is a high spec or low spec system. The only difference being some AVs are resource hogs so on a low spec system they might be even slower by comparison to an AV that is relatively light on resources (like avast).
Windows 10 Home 64bit/ Acer Aspire F15/ Intel Core i5 7200U 2.5GHz, 8GB DDR4 memory, 256GB SSD, 1TB HDD/ avast! free 24.2.6105 (build 24.2.8918.824) UI 1.0.799/ Firefox, uBlock Origin, uMatrix/ MailWasher Pro/ Avast! Mobile Security

MasterTB

  • Guest
Re: New Antivirus performance benchmark Sept 30 2010
« Reply #4 on: October 08, 2010, 11:28:57 PM »
Yes but, take this test for example: http://www.raymond.cc/antivirus/
Here is a real life (if you can call it that) test that takes into account several daily tasks and how AV's perform and impact on a given System.

BUT: The test here doesn't specify what types of settings where set for each AV, assuming they are at Default settings -which vary from product to product- hence the given results for each test must be taken with a grain of salt because not all AV's perform exactly the same tasks, they don't all have web shields, network shields, mail shields, etc. Not to mention that some don't even have any settings (Like MSE).

Still, it is clear that Avast! is one -if not the- lightest AV's in the market. I can say for certain that I hardly feel the Suite running on my system, not that I could actually, since my system can easily perform a full system scan while playing HD videos running in economizer mode (that is a speed of 984 Mhz. on the CPU) and not feel the impact.

"So for these tests my guess is that the testers aren't stupid and try to select a scan level which is comparable across the board." I don't dare to affirm that... people are stupid and do stupid things, even testers. I've read tests that testing HIPS software allow Malware to run (disregarding the software's advice) just to see if it is able to stop it later -WTF??-. (this was a funny note, kind of sarcastic even  ;D)

But point taken, Avast! is fast, no doubt about it.

Martin.-

Dch48

  • Guest
Re: New Antivirus performance benchmark Sept 30 2010
« Reply #5 on: October 09, 2010, 06:07:46 AM »
The results look pretty much the same as what my experiences have been. Avast! and Norton were both fast and lightweight and MSE was comparatively a slug.

Offline wonderwrench

  • Sr. Member
  • ****
  • Posts: 223
Re: New Antivirus performance benchmark Sept 30 2010
« Reply #6 on: October 09, 2010, 06:45:56 PM »
So for these tests my guess is that the testers aren't stupid and try to select a scan level which is comparable across the board.

The manual scans are run by right click context menu so I would guess all products settings are at their defaults at least they should be. If not the test is useless. Testing must be done to reflect what the average user will see which is default settings. Same goes for protection testing done by other companies. Then and only then can we use the results to make an educated decision as to which products to use and or recommend.

BTW personally I almost went with NIS 2010/2011 on my systems as Symantec is really on the ball lately. Never before have their products been less of a hog. Avast has always been good to me and I ended up picking Avast 5 IS over NIS because of this forums. No other security community can even come close to what we have here, no way. NIS might have slightly better protection and both are light on resources but if I went NIS I'd have have no support at all IMHO. Besides Symantec left a bad taste in my mouth years ago and I swore NEVER AGAIN! I think I made the right choice.

Bill
Main Box*i7 930*GB X58A-UD3R*3x4 gig Patriot DDR3 1600 EL*EVGA GTX 460 1 gig*Intel X25-M G2 80 gig*WD 2TB Green*ASUS DRW-24B3LT*Samsung SH-S223L*LG WH14NS40*Corsair AX750*Rosewill Challenger case*Windows 8 Pro 64 bit*Avast 8 Free 8.0.1482*MBAM Pro*Firefox 19.0.1*NoScript

Offline DavidR

  • Avast Überevangelist
  • Certainly Bot
  • *****
  • Posts: 88895
  • No support PMs thanks
Re: New Antivirus performance benchmark Sept 30 2010
« Reply #7 on: October 09, 2010, 07:21:20 PM »
Unfortunately there is no comparative information of exactly what is scanned using the right click context scan of each of the products. Whilst some might take the approach that some files don't represent a risk and aren't scanned on that scan, all I know is what is scanned by avast and no other AV.

The avast ashQuick.exe scan is the most thorough of all of the avast scans, it scans all files, opens and scans all supported archives (see note) and uses the highest sensitivity. So if any of those aren't the same there will be some disparity.

Note: Avast has some of the best archive support the more packers you support, the more files are going to be scanned and potentially the longer the scan. So here is another area where if all AVs don't support the same archives you will get more disparity.
« Last Edit: October 09, 2010, 07:24:11 PM by DavidR »
Windows 10 Home 64bit/ Acer Aspire F15/ Intel Core i5 7200U 2.5GHz, 8GB DDR4 memory, 256GB SSD, 1TB HDD/ avast! free 24.2.6105 (build 24.2.8918.824) UI 1.0.799/ Firefox, uBlock Origin, uMatrix/ MailWasher Pro/ Avast! Mobile Security

Offline Lisandro

  • Avast team
  • Certainly Bot
  • *
  • Posts: 67195
Re: New Antivirus performance benchmark Sept 30 2010
« Reply #8 on: October 09, 2010, 07:30:39 PM »
Isn't it Symantec biased test?
Worth reading?
I'm lazy ;D
The best things in life are free.

Offline wonderwrench

  • Sr. Member
  • ****
  • Posts: 223
Re: New Antivirus performance benchmark Sept 30 2010
« Reply #9 on: October 09, 2010, 08:16:39 PM »
Isn't it Symantec biased test?
Worth reading?
I'm lazy ;D
Yes the test was payed for by Symantec but knowing Passmark's track record I doubt the results are biased. I would guess Symantec picked Passmark because of their ability to create the scripts/programs and test environment needed to test performance metrics completely and accurately. I get the idea Symantec realized their products were huge resource hogs and figured the best way to prove they were eliminating the problem was to hired Passmark to prove it. I'm glad Symantec hired Passmak as these tests will surely benefit everyone in the long run. The companies who's products that scored poorly will surely take note. The end result will be better security products for everyone.

Bill
Main Box*i7 930*GB X58A-UD3R*3x4 gig Patriot DDR3 1600 EL*EVGA GTX 460 1 gig*Intel X25-M G2 80 gig*WD 2TB Green*ASUS DRW-24B3LT*Samsung SH-S223L*LG WH14NS40*Corsair AX750*Rosewill Challenger case*Windows 8 Pro 64 bit*Avast 8 Free 8.0.1482*MBAM Pro*Firefox 19.0.1*NoScript

Offline DavidR

  • Avast Überevangelist
  • Certainly Bot
  • *****
  • Posts: 88895
  • No support PMs thanks
Re: New Antivirus performance benchmark Sept 30 2010
« Reply #10 on: October 09, 2010, 08:49:03 PM »
Well the test in itself may not be biased, the it is the piper that calls the tune, e.g. sets the parameters of the test, e.g. what they want tested.
Windows 10 Home 64bit/ Acer Aspire F15/ Intel Core i5 7200U 2.5GHz, 8GB DDR4 memory, 256GB SSD, 1TB HDD/ avast! free 24.2.6105 (build 24.2.8918.824) UI 1.0.799/ Firefox, uBlock Origin, uMatrix/ MailWasher Pro/ Avast! Mobile Security

Offline Lisandro

  • Avast team
  • Certainly Bot
  • *
  • Posts: 67195
Re: New Antivirus performance benchmark Sept 30 2010
« Reply #11 on: October 09, 2010, 09:17:58 PM »
The end result will be better security products for everyone.
Hope so.
But, indeed, other tests are more reliable and have more influence over the vendors, like, av-comparatives (they're conducting a performance test and the results will come by the end of the year).
The best things in life are free.

Offline wonderwrench

  • Sr. Member
  • ****
  • Posts: 223
Re: New Antivirus performance benchmark Sept 30 2010
« Reply #12 on: October 10, 2010, 04:47:55 AM »
The end result will be better security products for everyone.
Hope so.
But, indeed, other tests are more reliable and have more influence over the vendors, like, av-comparatives (they're conducting a performance test and the results will come by the end of the year).

What do you mean more reliable Passmark is not testing effectiveness just performance. Passmark business is testing performance and stability of both software and hardware and the development of software and hardware that can be used in the field to test the same. av-comparatives testing will surely be unbiased but I doubt they have the resources to do as through and accurate a job as Passmark. The more testing the better so I can't wait to see what av-comparatives comes up with. Sure protection is top priority but there is no excuse for poorly written code that slows systems to a crawl. This test eliminates several otherwise good products from even being considered. I was a big Trend Micro fan back in the early to mid 2000's. Their products were light had good protection and were highly configurable. Each year their products got more bloated and less configurable. I was a private beta tester for them during this time and could use any of their products for free. I raised hell every year trying to get them to wake up but it was useless. I finally gave up around 2006 as I could no longer use their products even for free. I tested many AV products looking for a replacement and ended up using Avast Home. It was light and did the job in its day. Today we have Avast 5 its light, highly configurable and has good protection. Not many if any other product can claim the same, free or paid. Today with the help of Passmark, av-comparatives etc it will be much easier to choose a AV product.

Bill
« Last Edit: October 10, 2010, 04:51:10 AM by wonderwrench »
Main Box*i7 930*GB X58A-UD3R*3x4 gig Patriot DDR3 1600 EL*EVGA GTX 460 1 gig*Intel X25-M G2 80 gig*WD 2TB Green*ASUS DRW-24B3LT*Samsung SH-S223L*LG WH14NS40*Corsair AX750*Rosewill Challenger case*Windows 8 Pro 64 bit*Avast 8 Free 8.0.1482*MBAM Pro*Firefox 19.0.1*NoScript

Offline Lisandro

  • Avast team
  • Certainly Bot
  • *
  • Posts: 67195
Re: New Antivirus performance benchmark Sept 30 2010
« Reply #13 on: October 10, 2010, 07:51:09 PM »
What do you mean more reliable Passmark is not testing effectiveness just performance.
I'm saying that people generally trust other tests like av-comparatives and not Passmark, which they never heard about...

Sure protection is top priority but there is no excuse for poorly written code that slows systems to a crawl.
I'm not saying anything against performance tests.
I just doubt *any* test that Symantec wins as they pay for the testing companies, just that.
Well, I trust in independent tests as they remain independent...

Each year their products got more bloated and less configurable.
The worst service that Norton SystemWorks all-in-one-everything-bloatware did to the software industry.

Today we have Avast 5 its light, highly configurable and has good protection.
Fully agree. Seven or eight years using it :)
The best things in life are free.

Dch48

  • Guest
Re: New Antivirus performance benchmark Sept 30 2010
« Reply #14 on: October 10, 2010, 11:17:56 PM »

BTW personally I almost went with NIS 2010/2011 on my systems as Symantec is really on the ball lately. Never before have their products been less of a hog. Avast has always been good to me and I ended up picking Avast 5 IS over NIS because of this forums. No other security community can even come close to what we have here, no way. NIS might have slightly better protection and both are light on resources but if I went NIS I'd have have no support at all IMHO. Bill
This is not correct. The support on the Norton/Symantec forums is as good as it is here and Symantec employees respond and post help and information very frequently.