New virus.gr test on the way...

[Don Pelotas]

Originally Posted by Marcos

In other words, the more an AV produces the higher it ranks

Hey that was actually funny.......i guess you spent some on this one.

[bigc73542]

This last test just reinforces my belief that this test is conducted by third graders.

[pykko]

mm strange test results, indeed...

[TeraInnovations]

Hello everyone.
The results really surprised us as well - and we are asking Virus.gr to remove the results from his website. These results are clearly flawed, and we do not want more advertising for ViruScape 2006 - anywhere,. To REALLY prevent this, we are removing the download link to ViruScape from our website once and for all.

I am assuming that a majority of the detections are false positives - there are fatal flaws in the scanner and it should NOT perform that well.

I hate to bash our own product, but, let us be the guinea pig and say that those tests are horrible

There is a good chance that we will not be releasing a second version of our product, because we are in the middle of negotiations and may be selling the second version of our program to a different company. Current ViruScape users will not be affected and will continue to get their updates.

[Durad]

TeraInnovations,

The 147184 virus samples were chosen using VS2000 according to Kaspersky, F-Prot, Nod32, Dr.Web, BitDefender and McAfee antivirus programs. Each virus sample was unique by virus name, meaning that AT LEAST 1 antivirus program detected it as a new virus.

ViruScape 2006 detected 93.87%

It mean that ViruScape 2006 detected 138161 unique samples.

Can you tell us does your product detect that much of unique samples?

[TeraInnovations]

ViruScape does have that many unique definitions (many more than that), but that does not matter - nor should this thread continue about ViruScape at all.

http://www.terainnovations.com - as you can see, we have removed our website.

[Firecat]

I would like to assure you that the tester from virus.gr himself is doubting the results of Viruscape 2006. He himself is very surprised as to how Viruscape detected so much.

It is strange to receive such a reaction from Tera Innovations. Perhaps the theories about how Viruscape managed this level of detection are indeed true. However, I am not going to reveal such information here. Besides, I am not sure about it either.

[IBK]

Originally Posted by Firecat

It is strange to receive such a reaction from Tera Innovations. Perhaps the theories about how Viruscape managed this level of detection are indeed true. However, I am not going to reveal such information here. Besides, I am not sure about it either.

i guess i know what you mean, it was my first thought too

[FRug]

hm, do i hear "checksumming on unreplicated samples" ?

[IBK]

no, another thing. but i will also not repeat it again.

[pykko]

Well, what I spotted in this test regarding NOD32 detection rates is they have constantly been improved and this is a good thing to see from the test results.

NOD32 detection rates:
2-12 October 2003 - 75.76%
10-25 August 2004 - 82.68%
2-16 April 2005 - 85.66%
14-22 December 2005 - 88.79%
15-25 August 2006 - 95.14%

[Inspector Clouseau]

Don't discuss detection rates and improvements based on such rubbish tests. I think even ESET will agree that you shouldn't make any conclusion out of such type of "tests". It's absolutely pointless and a big waste of time even to read the reports of virusp av tests. Since day one of his "av tests" he didn't learn a single bit how to do it better, it becomses even worse!

Originally Posted by Inspector Clouseau

Since day one of his "av tests" he didn't learn a single bit how to do it better, it becomses even worse!

[Inspector Clouseau]

Originally Posted by HiTech_boy

I don't see the reason why my reply sounds funny to you?

[pykko]

Originally Posted by Inspector Clouseau

Don't discuss detection rates and improvements based on such rubbish tests. I think even ESET will agree that you shouldn't make any conclusion out of such type of "tests". It's absolutely pointless and a big waste of time even to read the reports of virusp av tests. Since day one of his "av tests" he didn't learn a single bit how to do it better, it becomses even worse!

OK! Sorry. I'll remove it from the tests I check.
Anyway I know it is made not so professional and some of the samples used are rubbish, non-working, etc... so no reliable.

Originally Posted by Inspector Clouseau

I don't see the reason why my reply sounds funny to you?

Your reply is not funny , it is just well-said truth (but this is my personal opinion)

[Inspector Clouseau]

Originally Posted by pykko

Anyway I know it is made not so professional

Are u trying to give credits to the tester now?

[IBK]

Originally Posted by TeraInnovations

Although we regret to leave the antivirus market, our ViruScape 2.0 code has been acquired by another antivirus company
http://www.terainnovations.com - as you can see, we have removed our website.

which other company bought viruscape? please let us know, i hear that some peoples already start again to rumour that viruscape was probably really just a fake of a vx trader to get more samples (to e.g. trade with virusp, which could explain the results in virusp tests). like some here at wilders said already years ago i think.

[pykko]

Originally Posted by Inspector Clouseau

Are u trying to give credits to the tester now?

Oh, yes! He is my sponsor and he gave me money to pay my schoolar taxes.

[AnthonyG]

Originally Posted by Firecat

It is strange to receive such a reaction from Tera Innovations. Perhaps the theories about how Viruscape managed this level of detection are indeed true. However, I am not going to reveal such information here. Besides, I am not sure about it either.

I was thinking this very same thing myself last night.

But I thought I would leave it for one of the pro's to mention this fact first though.

But I can remember the first mention of the Viruscape Antivirus on the forum a few month back. And the storm it created. But the representitive from TI handled it very well. And personally gained my admiration for his product for the way he did handle this.

But to see the same firm now slating their own product in the mannor they have last night, which is only a few months later from this incident is very unusal.

Especially after also seeing in these months some members who have tried Viruscape post some good encouriging things about it too.

Very strange indeed!.

Thanks
Anthony

[Smokey]

In the meantime I love every post about virus.gr and the subsequent heated, indignant reactions and discussions.
Nice stuff to read and enjoy on a rainy, boring sunday noon.

[pykko]

Originally Posted by Smokey

In the meantime I love every post about virus.gr and the subsequent heated, indignant reactions and discussions.
Nice stuff to read and enjoy on a rainy, boring sunday noon.

It's sunny here. Do you want me to loan you 10 pounds of Sun?

[TeraInnovations]

Our detection rates were NOT caused by any relations with virus.gr. There are a great deal of false positives in v1 (i.e. - it detects every delphi dll).

We cannot disclose who bought us, but they did not buy any of the v1 code or our samples.

It is a mere coincidence that virus.gr releases the results as we are about to take down our website -- the transaction just finished last week.

Also - if we detect 93%, did kaspersky trade viruses with virus.gr every day to get to 99%? If Kaspersky finds 99%, why can't we find 93% (even though it is based a lot on false positives)?

[IBK]

no, kav gets the samples from vxer without having to exchange or ask .
'false positive' is not the correct definition, lets call them 'crud files'.

[TeraInnovations]

Looking at some other scan results - it appears that the high numbers are more related to mis-detections - detecting non-dos files as dos viruses for example.

I emailed him yesterday to take down the results, but have gotten no response.

Something I found while browsing thru the forum on virus.gr:

Author: virusp
"RE: ViruScape: How is their detection rate? Hungry for Knowlledge
Posted: 22.05.2006 20:03

Very low, it also crashes during the scan."

We have added samples, but not enough to change from a very low to a 93% - maybe he reported the wrong antivirus as ViruScape. This is what I'm guessing.