Help - Search - Members
Full Version: New Antivirus comparative
Kaspersky Lab Forum > Beta Testing > KIS\KAV 2015
Pages: 1, 2
kllm
Av-comparatives

¿Do you Know Gdata Antiviruskit?. It has a the best results, better than Kav 5.
ekerazha
February 2006 ( http://www.av-comparatives.org )

#1 - AVK
#2 - F-Secure
#3 - Kaspersky

Kaspersky had a "low" detection rate (97,90%) for "other malware" (flooders, nukers, ddos, exploits, activex, rootkits, irc malware etc.).
Lucian Bara
Will Kaspersky score higher after the new heuristic engine is implemented? Take a look at the last polymorphic tests. kaspersky 32%, Nod32 62%.
ekerazha
QUOTE(lucianbara @ Mar 2 2006, 01:10 PM)
Will Kaspersky score higher after the new heuristic engine is implemented? Take a look at the last polymorphic tests. kaspersky 32%,  Nod32 62%.
*


That was the Retrospective/ProActive Test of November 2005.
gil
QUOTE(kllm @ Mar 2 2006, 11:59 AM)
¿Do you Know Gdata Antiviruskit?. It has a the best results, better than Kav 5.

that's explain : gdata has dual scans using the KAV and BitDefender engines.

and

QUOTE
Please link ONLY to our main site www.av-comparatives.org and not to the other subpages.
It's forbidden to use/provide our test results/documents on other sites without our permission.
JohnH
Thank’s for the update, but remember the copyright statement:
It is forbidden to provide the tables or results in full or in parts on other sites! Please just refer a link to www.av-comparatives.org”

Think of the responsible editor (Kaspersky Lab).
ekerazha
QUOTE(gil @ Mar 2 2006, 01:17 PM)
that's explain : gdata has dual scans using the KAV and BitDefender engines.


Yeah... but this could mean 2 things:

1) KAV engine lacks some aspects of the BitDefender engine.
2) AVK defs are globally better than the KAV ones.
gubo
GDATA's detection rates have always been excellent using two great engines.
I've tested it for a while but it slows down serious even fast machines.
Reload
QUOTE(ekerazha @ Mar 2 2006, 01:25 PM)
Yeah... but this could mean 2 things:

1) KAV engine lacks some aspects of the BitDefender engine.
2) AVK defs are globally better than the KAV ones.
*


1)
Because AVK uses two engine's it has best off both products
But using two engine's makes the product a bit slow... Everything
is processed two time...

2)
Because AVK use the scan engine's from two company it also uses
the virus def's from those company's
Don Pelotas
If using two engines are better than using one, then why does Kaspersky and F-Secure have the same kind of detection in this test? F-Secure uses 4 engines, a modified Kaspersky among them!

It is not as simple as that, there are many issue's to solve and as those who have tried using AVK or F-Secure knows, it usually means a slower pc with virtually no more detection. smile.gif

P.s. I haven't even mentioned the proactive defense in 6.0 yet.
Don Pelotas
QUOTE(ekerazha @ Mar 2 2006, 01:16 PM)
That was the Retrospective/ProActive Test of November 2005.
*

Heuristics are also used in the on-demand test.
saso
i must say that i always liked the test from av-comparatives (and i still do), but i must also confess that in the last time i cannot lose the feeling that, specially this on-demand tests are not as good as they ware. i have the feeling that the malware sample database is somehow limited unsure.gif
ekerazha
QUOTE(Reload @ Mar 2 2006, 01:52 PM)
1)
Because AVK uses two engine's it has best off both products
But using two engine's makes the product a bit slow... Everything
is processed two time...

2)
Because AVK use the scan engine's from two company it also uses
the virus def's from those company's
*


1) KAV engine could lack some aspects of the BitDefender engine.

2) AVK defs could be globally better than the KAV ones... I think they don't use "virus def's from those company's", I believe they have their own defs.
ekerazha
QUOTE(Don Pelotas @ Mar 2 2006, 02:30 PM)
Heuristics are also used in the on-demand test.
*


Maybe, but "kaspersky 32%, Nod32 62%" are from the test of November 2005 biggrin.gif
Lucian Bara
Yes, but that test was made using old bases. That means it uses the heuristic engine more than the bases.
Don Pelotas
QUOTE(ekerazha @ Mar 2 2006, 03:59 PM)
Maybe, but "kaspersky 32%, Nod32 62%" are from the test of November 2005  biggrin.gif
*

I know that, just wanted to make it clear, if someone perhaps thought otherwise. smile.gif
ra8ul
its always better to check speeds and detection rate with 2 engines avk has an edge
kav is faster while scaning,boot times and ondemand compared to f-secure and avk.
Don Pelotas
QUOTE(ekerazha @ Mar 2 2006, 03:57 PM)
1) KAV engine could lack some aspects of the BitDefender engine.

2) AVK defs could be globally better than the KAV ones... I think they don't use "virus def's from those company's", I believe they have their own defs.
*

1. It could, but very unlikely, much more likely to be a combination of some signatures in BD, but not in Kav and perhaps some heuristic detections. On the other hand, i'm sure that there other viruses (not included in this test) detected by Kaspersky, but not by BD.

I'm equally sure that in the real world environment Kaspersky users should not fear anything from AVK, the results are practicly the same and simply means that in this particular test, Kaspersky again is at the top, which it has been in every one so far.

Keep in mind this is an on-demand scan test and results could be very different in a real world scenario with malware actually executing and being detected by the real-time protection.

2. No, the signatures come from Kaspersky & BitDefender.

So in my opinion this test shows you that Kaspersky is still at the top and that you can feel good about your choice of AV. smile.gif
Lucian Bara
BD has indead more signatures in their bases ~220.000 and kaspersky has only ~180.000
Julian
that's because kav has a better unpacking engine then bd wink.gif

btw: avk produces very much false positives.
Don Pelotas
QUOTE(lucianbara @ Mar 2 2006, 05:06 PM)
BD has indead more signatures in their bases ~220.000 and kaspersky has only ~180.000
*

Which is means nothing, Kaspersky still detects more, the number of signatures is completely irrelevant, it 's the detection that matters. As an example, AntiVir has 300000 signatures (or thereabout), but detects less. wink.gif
Blackcat
F-Prot for Windows has 242745 malware signature files at the present time but give me KAV any time.

As already stated, total number of signatures are not a good feature for comparing different AV's.
Blackcat
F-Prot for Windows has 242745 malware signature files at the present time but give me KAV any time.

As already stated, total number of signatures are not a good feature for comparing different AV's.
ekerazha
QUOTE(Don Pelotas @ Mar 2 2006, 05:03 PM)
1. It could, but very unlikely, much more likely to be a combination of some signatures in BD, but not in Kav and perhaps some heuristic detections. On the other hand, i'm sure that there other viruses (not included in this test) detected by Kaspersky, but not by BD.

I'm equally sure that in the real world environment Kaspersky users should not fear anything from AVK, the results are practicly the same and simply means that in this particular test, Kaspersky again is at the top, which it has been in every one so far.

Keep in mind this is an on-demand scan test and results could be very different in a real world scenario with malware actually executing and being detected by the real-time protection.

2. No, the signatures come from Kaspersky & BitDefender.

So in my opinion this test shows you that Kaspersky is still at the top and that you can feel good about your choice of AV. smile.gif
*


These were hypotheses.

If KAV would be always better than BitDefender, having Kaspersky with or without BitDefender shouldn't produce differences... but it produces differences, so Kaspersky lacks some aspects of BitDefender (at engine-level or defs-level, I don't know).

We have that Kaspersky is often (not always) better than BitDefender.
saso
here are some on-demand statistics from me wink.gif they ware made at the end of january 2006 on something over 400000 possible malware files.

WARNING this statistics are only informational, av products are not directly tested and compared for detection. with other words this statistics may very well be buggy, for example old program versions might be used or bad misconfiguration of settings or maybe a bit old signatures... the number is the % of detected files from all files, so the higher the number the better the detection smile.gif

PS. i just reinterpret the results, don't do the scanning. so please don't ask if i can add or retest some scanner smile.gif

Kaspersky 97,92634
McAfee 97,36117
Symantec 92,19306
Panda 80,97749
DrWeb 80,72556
AVG 80,23495
Frisk 79,50037
AntiVir 77,95399
Norman 75,96117
MkS_vir 72,10537
Trend 69,64101
AVAST! 68,42693
Sophos 67,11633
InoculateIT 65,81416
NOD32 58,57626
IKARUS 49,50867
VET 49,02073
Don Pelotas
QUOTE(ekerazha @ Mar 2 2006, 05:35 PM)
These were hypotheses.
*

And yours are not. rolleyes.gif

I didn't say Kaspersky is always better than BitDefender, i do believe it is better, but that is another story.
QUOTE(ekerazha @ Mar 2 2006, 05:35 PM)
If KAV would be always better than BitDefender, having Kaspersky with or without BitDefender shouldn't produce differences... but it produces differences, so Kaspersky lacks some aspects of BitDefender (at engine-level or defs-level, I don't know).
*

There are differences in this test, when comparing the combo of BD/Kav in AVK versus Kav alone, so it would be very reasonable to assume the difference are in sigs/heuristics and not engine, since BD alone scores significantly less than Kaspersky in the test (93.08% versus 99.57%), if we exclude dos virus, which has no meaning to most users.

You do know that the differences are minute don't you?(between Kav & AVK) smile.gif
ekerazha
QUOTE(Don Pelotas @ Mar 2 2006, 05:57 PM)
And yours are not. rolleyes.gif

Yeah, you have misunderstood. I've said mine were hypotheses smile.gif

QUOTE
I didn't say Kaspersky is always better than BitDefender, i do believe it is better, but that is another story.

There are differences in this test, when comparing the combo of BD/Kav in AVK versus Kav alone, so it would be very reasonable to assume the difference are in sigs/heuristics and not engine, since BD alone scores significantly less than Kaspersky in the test (93.08% versus 99.57%), if we exclude dos virus, which has no meaning to most users.

You do know that the differences are minute don't you?(between Kav & AVK) smile.gif
*


There are no differences detecting dos or not-dos viruses.
However... I don't know where differences are, but I know that KAV+BitDefender has a better detection rate (minute or not minute) than KAV alone... so somewhere (yeah... in a little part) BitDefender is better than KAV. That's all.
Lucian Bara
Maybe in the HIVE heurisitc scanner?
simlet
Extendia Anti-virus Pro is almost an exact replica of GData AVK, but is cheaper. Has anyone ever tried this program?
saso
QUOTE(ekerazha @ Mar 2 2006, 06:06 PM)
Yeah, you have misunderstood. I've said mine were hypotheses smile.gif
There are no differences detecting dos or not-dos viruses.
However... I don't know where differences are, but I know that KAV+BitDefender has a better detection rate (minute or not minute) than KAV alone... so somewhere (yeah... in a little part) BitDefender is better than KAV. That's all.
*


yes but also kav + mcafee has a very good detection smile.gif and any other combination of multi av engine. this is why microsoft did a very smart and "nasty" move when taking over the sybari, since it use/support more or less all av engines... so why microsoft av is the best? wink.gif

PS. sybari technologies is as far as i know used my microsoft only for coorporate enviroments.
ekerazha
QUOTE(saso @ Mar 2 2006, 06:14 PM)
yes but also kav + mcafee has a very good detection smile.gif and any other combination of multi av engine. this is why microsoft did a very smart and "nasty" move when taking over the sybari, since it use/support more or less all av engines... so why micorosft/sybari av is the best? wink.gif
*


In an ideal world, KAV + "every engine" doesn't have a better detection than KAV alone. Until this, there will be solutions with a better detection than KAV smile.gif
saso
QUOTE(ekerazha @ Mar 2 2006, 06:17 PM)
In an ideal world, KAV + "every engine" doesn't have a better detection than KAV alone. Until this, there will be solutions with a better detection than KAV smile.gif
*


it is not that much about the engine, in this case it is mostly about signatures... and this: "KAV + "every engine" doesn't have a better detection than KAV alone", is simple impossible.
ekerazha
QUOTE(saso @ Mar 2 2006, 06:19 PM)
it is not that much about the engine, in this case it is mostly about signatures... and this: "KAV + "every engine" doesn't have a better detection than KAV alone", is simple impossible.
*


Yeah... so KAV should improve its signatures. "Impossible is nothing" wink.gif
saso
QUOTE(ekerazha @ Mar 2 2006, 06:21 PM)
Yeah... so KAV should improve its signatures. "Impossible is nothing" wink.gif
*


it is again not that much about improving the signatures, it is only mor or less about getting every single malware sample in the world and adding signatures for it.

"Impossible is nothing", maybe, but this would only be possible if ALL av vendors would agree to share ALL malware samples, it is maybe sad but in real world this is impossible mostly because of the money smile.gif
ekerazha
QUOTE(saso @ Mar 2 2006, 06:28 PM)
it is not about improving the signatures is is noly about getting evry single malware sample in the world and adding signatures for it. "Impossible is nothing", yes, but this would only be possible if ALL av vendors would agree to share ALL malware samples, it is maybe sad but in real world this is impossible mostly because of the money smile.gif
*


If an av vendor has found some malware, then you can find it too wink.gif
saso
QUOTE(ekerazha @ Mar 2 2006, 06:30 PM)
If an av vendor has found some malware, then you can find it too wink.gif
*


if that is true then so can you... so tomorow you will send all undetedted malware samples to KL? or do you need a day or two for this? biggrin.gif
Don Pelotas
QUOTE(ekerazha @ Mar 2 2006, 06:06 PM)
Yeah, you have misunderstood. I've said mine were hypotheses smile.gif
There are no differences detecting dos or not-dos viruses.
However... I don't know where differences are, but I know that KAV+BitDefender has a better detection rate (minute or not minute) than KAV alone... so somewhere (yeah... in a little part) BitDefender is better than KAV. That's all.
*

Hey no worries, we're just debating and we do not have to agree, it's actually more interesting if we don't........correct? wink.gif

Actually excluding dos does (wow, is there an ekko in here unsure.gif ) means something, for BD a 2.50% drop in detectiorate, for Kav only 0.20%, but if these two %'s doesn't mean anything, then the 0.27% difference in detection between Kaspersky & AVK should mean even less shouldn't it? tongue.gif

All i'm trying to explain, is that the differences are virtually zero, it just means there are both great choices with regards to detection, having them installed is quite another matter, try to installed AVK 2006 full and see what i mean, there is no contest IMHO. smile.gif
ekerazha
QUOTE(saso @ Mar 2 2006, 06:32 PM)
if that is true then so can you... so tomorow you will send all undetedted malware samples to KL? or do you need a day or two for this? biggrin.gif
*

This is not my job smile.gif This is the job of the Kaspersky team wink.gif
ekerazha
QUOTE(Don Pelotas @ Mar 2 2006, 06:33 PM)
All i'm trying to explain, is that the differences are virtually zero

Imho, this isn't "virtually zero". It is a little "difference" but it is still a difference smile.gif And a a +0.1 is better than a -0.1 smile.gif
Don Pelotas
QUOTE(ekerazha @ Mar 2 2006, 06:42 PM)
Imho, this isn't "virtually zero". It is a little "difference" but it is still a difference smile.gif And a a +0.1 is better than a -0.1 smile.gif
*

It is in practical terms, but you can keep on splitting hair if you wish and then you will finally be "right". cool.gif
saso
QUOTE(ekerazha @ Mar 2 2006, 06:37 PM)
This is not my job smile.gif This is the job of the Kaspersky team wink.gif
*


oh come one, be a good boy, share your supremacy with us, less fortunate people rolleyes.gif
Lightning
So what because you have found a "test" that have kaspersky anvivirus a bit lower comparing with others?

I have in my hands 7 tests since 2002, where Kaspersky is the test winner since august 2004 and until today. 2006 beta also tested

I give you a little sample :

113334 virus samples used for the test..


1. Kaspersky Personal Pro Version 5.0.390- 99.46%

1. Kaspersky 2006 beta Version 6.0.15.222- 99.46%

2. F-Secure 2006 Version 6.10.330 - 96.92%

3. CyberScrub Version 1.0 - 96.62%

4. eScan Virus Control Version 2.6.522.9 - 95.21%

5. McAfee Version 10.0.27 - 94.80%

6. BitDefender Version 9 - 90.75%

7. Nod32 Version 2.50.41 - 88.79%

8. AntiVir Personal Version 6.32.00.51 - 86.55%

9. MKS_VIR 2005 - 86.16%

10. Norton Professional Version 2006 - 85.17%

( Source www.virus.gr )

I can give more information to anyone who may want it
At the other tests (i saied that i have tests since 2002, and Kaspersky ranks 1st since 2004 and until today) At the while between 2002 - 2004 Kaspersky was 2nd with F-Secure first.

Regards
Lightning

P.S. My personal opinion. Kaspersky is the best!
And the best tester is the personal taste.
I had a customer with norton AV and "4" trojans on his system that cound not be disinfected/deleted/ etc.
I used Kaspersky and detected AND REMOVED 24 .. not 4 tongue.gif
saso
one think i would like to add about my comments in this discussion is that, they are based on this on-demand scan comparisons and so they are true only as far as we are talking about on-demand scans using mostly the signatures for detection. so talking about heuristics, proactive defense, fast adding of new signatures, system resource usage, support of archives and packagers,... is a whole different story smile.gif
Lightning
Saso forgive me i forgot to tell..
In ALL 7 tests i can give you, all 50 antiviruses tested were set to the maximum available settings of protection. smile.gif
ekerazha
QUOTE(Don Pelotas @ Mar 2 2006, 06:51 PM)
It is in practical terms, but you can keep on splitting hair if you wish and then you will finally be "right". cool.gif
*

rolleyes.gif
In practical terms AVK has a better detection rate.
ekerazha
QUOTE(saso @ Mar 2 2006, 06:58 PM)
oh come one, be a good boy, share your supremacy with us, less fortunate people rolleyes.gif
*

Should I really reply to a 3yo child?
ekerazha
QUOTE(saso @ Mar 2 2006, 07:16 PM)
one think i would like to add about my comments in this discussion is that, they are based on this on-demand scan comparisons and so they are true only as far as we are talking about on-demand scans using mostly the signatures for detection. so talking about heuristics, proactive defense, fast adding of new signatures, system resource usage, support of archives and packagers,... is a whole different story smile.gif
*

We are talking about detection... that is a detection test, not a "fast adding of new signatures, system resource usage" test.

P.S.
Both real-time and on-demand use "mostly the signatures for detection", and both use "heuristics" and "support of archives and packagers".

Be a good 3yo child biggrin.gif
TonyW
QUOTE(ekerazha @ Mar 2 2006, 05:37 PM)
This is not my job smile.gif This is the job of the Kaspersky team wink.gif
They get samples in a lot of cases from users so if you do come across malware that is NOT yet detected, help them, and us, by sending it in.
ekerazha
QUOTE(TonyW @ Mar 2 2006, 08:10 PM)
They get samples in a lot of cases from users so if you do come across malware that is NOT yet detected, help them, and us, by sending it in.
*


Maybe you should ask this to http://www.av-comparatives.org wink.gif
TonyW
QUOTE(ekerazha @ Mar 2 2006, 07:12 PM)
Maybe you should ask this to http://www.av-comparatives.org wink.gif
I believe they already do this. wink.gif
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Invision Power Board © 2001-2014 Invision Power Services, Inc.