Avast WEBforum
Consumer Products => Avast Free Antivirus / Premium Security (legacy Pro Antivirus, Internet Security, Premier) => Topic started by: treker96mk2 on June 12, 2006, 12:05:43 AM
-
VPS: 0624-0 211kb
what was it?
-
About 25,000 new detections added, that's all...
-
not listed in history?
-
1. The web servers are just being synchronized
2. Please realize that (as always) that the history page only lists new virus NAMES. Avast often finds a LOT viruses using the same name. So you can't really judge the size of the VPS update according to the number of names added...
-
I thought after reading many reply to my posts where I asked for better protection of avast, that Avast didn´t need more virii signatures..
I don´t understand why many ppl flamed me for saying that it would be nice that avast detection rate get better.
Im happy that avast signatures are getting better, maybe I´ll use it again if it continue this way.
Greetings.
K.
-
OK
-
About 25,000 new detections added, that's all...
That's all, you say...
Holy S**t. That is excellent.
-
Hi Klavier,
I think this means that avast is getting better. Haven't you noticed lately that it is leaner (it does not weigh that heavy on the cycles anymore), it is meaner (detects more).
Always remember that you do multi-scanning, an extra non-resident scanner like ClamWin (frequently updated with slightly different virus base), use online scanners of another product that do not conflict with avast (full scan bi-weekly) that could be Bitdefender online, use DrWeb hyperlink pre-scanner plug-in inside your browser to pre-scan all your links (very frequently updated, only second after KAV's). This way I am not saying you scan covers all, but you are very well protected, and avast is the basis for all this security.
I think it is a good product.
polonus
-
This reminds me now...
How do you get an accurate number of signatures that avast! has in it's database?
I know that if you put an asterisk in the Virus Database... section it gives you the number of 55,899 as of today. But I have read here in this forum before that the number is not accurate for several reasons, like generic detections or one signature covering several viruses and such.
I personally think that having a total, shown in the way that many other antiviruses do would be something that would be good for avast! to display in the main program. It's kind of the "I have a bigger **** then you" of antiviruses, you know. Sure, it might not mean a whole lot, but many people do go by those numbers.
-
Hi polonus. I also think that Avast is good, and I am happy that it improves. The reason that I now am using BD on my desktop, is because I felt that Avast could improve some more, I think it still has a lot of potential to develop, and when I can see in some tests, like av.comparatives that Avast is in the same league (sorry for my spell.. don´t know how to write that) that Kaspersky, NOD or BD, Ill use it again and for good.
As I said before, Im happy that avast improves, and with my previuos posts, I don´t want to criticize avast, I wanted that avast improve.
K.
-
This reminds me now...
How do you get an accurate number of signatures that avast! has in it's database?
I know that if you put an asterisk in the Virus Database... section it gives you the number of 55,899 as of today. But I have read here in this forum before that the number is not accurate for several reasons, like generic detections or one signature covering several viruses and such.
I personally think that having a total, shown in the way that many other antiviruses do would be something that would be good for avast! to display in the main program. It's kind of the "I have a bigger **** then you" of antiviruses, you know. Sure, it might not mean a whole lot, but many people do go by those numbers.
I agree.
-
How do you get an accurate number of signatures that avast! has in it's database?
I know that if you put an asterisk in the Virus Database... section it gives you the number of 55,899 as of today. But I have read here in this forum before that the number is not accurate for several reasons, like generic detections or one signature covering several viruses and such.
How do you count potential viruses in the generic detections for new variants, answer you can't so it really can't be measured. Even if it were estimated people would complain about using this method of counting, there is no way to make a direct comparison between AVs as there is no standard in either virus/malware naming or counting. Not to mention some are trying to detect different things.
As men say it's not the size, it's what you do with it ;D yes it is initially an attraction having a big one, but if you can't use it efficiently the attraction will ware off. They say Norton has a big one (virus database) but that only rates a Standard rating at av-comparatives.org when avast rates Advanced with a smaller one ;D
So numbers/size isn't everything ;D
-
David,
I can't tell if you your bragging or making excuses ;D ;D
I don´t understand why many ppl flamed me for saying that it would be nice that avast detection rate get better.
@Klavier
But many agreed, or made the point that there are other things to consider too. We can't guarantee anything but open discussion here.
The real point is that one av will be best in some ways this month, another next month and so on. It makes a lot of sense choose the one that does the job you expect over the long run and stick with it through the fluctuations.
-
There's something a bit ridiculous about the number...
According to the latest on-demand scanning test conducted by AV-Comparatives.org
- avast! has approximate 55,000 malware records in its database, F-Prot has 232,823 malware records but avast! got "ADVANCED" level while F-Prot got "STANDARD"
- avast! has approximate 55,000 malware records in its database, BitDefender has 269,149 malware records but both avast! and BitDefender got the same level "ADVANCED"
I have no intention to discredit some AV but I think F-Prot has too exaggerated about its malware records. In my humble opinion, I think the current version of F-Prot is just a mediocre scanner, I think it's not better than eTrust or AVG if you care about the number and F-Prot's proactive detection is also not good. ::)
-
Isn't it all in the way they are counted? Avast! counts families of malware as a detection while others count each member of the family.
But that brings up a question. If 25,000 new detections were just added and we now have a total of 55,899 did we really just increase are detections by almost 70% !?
-
Isn't it all in the way they are counted? Avast! counts families of malware as a detection while others count each member of the family.
Yes, I know but it's so sceptical why they do that way, I think it's probably all about "marketing gimmick", IMHO.
But that brings up a question. If 25,000 new detections were just added and we now have a total of 55,899 did we really just increase are detections by almost 70% !?
:o
-
Yes, I know but it's so sceptical why they do that way, I think it's probably all about "marketing gimmick", IMHO.
I'm sure that's it.
-
But.. if avast with its 55000+ sig. records detects like between 90 or 93 % of av-comparatives and maybe others test.. with 25.000 (more than 50% new signatures)... Avast should detect like 130 or 140%!!! that sounds odd.
It would be nice more info. from avast about that update of 25.000
it almost duplicates avast detection? Considering that avast detects families as 1 virii name.., in practice, 25.000 would be like 200.000 new viruses?
Regards,
K.
-
I don't know, but I think these 25,000 new detection signatures are so-called generic signature or generic detection for Trojan-like malware.
See this thread: http://forum.avast.com/index.php?topic=14273.0
-
That gives more light on the detections added, nice for alwill!
-
If I'm not wrong, it should be generic signature for Trojan-like malware, I've noticed avast! now detects some malware samples I sent to Alwil a few weeks a go as Win32:Trojan-gen. xxx.
-
Some scanners probably don't detect this sample (but don't heavily rely on such online scanners).
-
But that brings up a question. If 25,000 new detections were just added and we now have a total of 55,899 did we really just increase are detections by almost 70% !?
Not at all... :)
I mean, look e.g. at the latest AV-comparatives.org on-demand test (since it's so popular quoting it here) - February 2006.
The test consisted of (unique) 474,759 samples, of which avast detected 444,293.
So, even in February, avast detected at least 444,293 "viruses" - even though in reality, it is much more, actually... (IBK's archives, while comprehensive, are by no means "complete" - they can't really be).
Thanks
Vlk
-
Ok, thanks. I think I've got it.
If nothing else I learned to leave the malware counting to the experts. :)
-
Ok, thanks. I think I've got it.
If nothing else I learned to leave the malware counting to the experts. :)
Me too
Cheers ;)
-
I don't know, but I think these 25,000 new detection signatures are so-called generic signature or generic detection for Trojan-like malware.
See this thread: http://forum.avast.com/index.php?topic=14273.0
Let's hope that there are many new generic signatures for Trojan-like malware because that is really the only area in which avast! is lacking at this time. If it can improve in detecting trojans than that will be incredible.
Keep pumping 'em out (signatures that is) avast! Team. It is becoming very obvious just how much avast! has been improving it's overall detections over the last year or so. Keep up the good work.
-
I'm hoping to see more signatures like Win32:Ardamax-gen, Win32:Swizzor-gen, Win32:SpyBot-gen and Win32SdBot-gen4 (and etc with numbers).
McAfee is making generic signatures for nearly every family and they're pretty good in general because of this.
-
[So, even in February, avast detected at least 444,293 "viruses"
Just a curiosity, why doesn't you count the virus like the 'other' antiviruses?
Won't make the things easier and improve your marketing? ::)
-
And so... is this? When an VPS update does not show in Avast web page (happens with certain frequency) is it because of soooooo many virus detected that makes it virtually "impossible" to be written in the home page?
-
And so... is this? When an VPS update does not show in Avast web page (happens with certain frequency) is it because of soooooo many virus detected that makes it virtually "impossible" to be written in the home page?
Well, not exactly - when the VPS update is not listed on the history page, this is because there were not new samples names(witch is the case with the last VPS update, the update contained 25000 new detections, but these detections are so-called generic detections, like the others in this topic mentioned) added or just the VPS update fixes some False positive.
[So, even in February, avast detected at least 444,293 "viruses"
Just a curiosity, why doesn't you count the virus like the 'other' antiviruses?
Won't make the things easier and improve your marketing? ::)
Yes, this is very interesting question ::)
Or won't it be better to fully remove the virus encyclopedia from the program (I know that this should be in the WishList topic, but...) - I mean when an ordinary user opens the encyclopedia and see only 56000 sample names, and after that compare avast! with other AV, in many cases the user will choose the AV with more virus signatures showed in the program. So this is a bit confusing or let it say missleading for some users, in my opinion. For example if I am not wrong for the same reason VBA32 removed their Virus list from their product.
I mean, look e.g. at the latest AV-comparatives.org on-demand test (since it's so popular quoting it here) - February 2006.
The test consisted of (unique) 474,759 samples, of which avast detected 444,293.
So, even in February, avast detected at least 444,293 "viruses" - even though in reality, it is much more, actually... (IBK's archives, while comprehensive, are by no means "complete" - they can't really be).
BTW: Vlk, have you already added the missed samples from the February AV-comparatives test ? ::)
-
BTW: Vlk, have you already added the missed samples from the February AV-comparatives test ?
Not really - and the main reason is that we don't have them... ;) :)
-
Not really - and the main reason is that we don't have them...
Read: "maybe we did, but it's hard to say since we don't know which ones they are" ;)
-
[So, even in February, avast detected at least 444,293 "viruses"
Just a curiosity, why doesn't you count the virus like the 'other' antiviruses?
Won't make the things easier and improve your marketing? ::)
I just wanted to echo these questions from Tech again and would really look forward to hearing an answer from an Alwil Team member on this one. :)
Thanks,
Dave
-
still Avast! keep missing these i sent long time ago :(
-
BTW: Vlk, have you already added the missed samples from the February AV-comparatives test ?
Not really - and the main reason is that we don't have them... ;) :)
Not really - and the main reason is that we don't have them...
Read: "maybe we did, but it's hard to say since we don't know which ones they are" ;)
OK, thanks for the answer :)
How about your opinion about the Virus List? ::)
-
You really want 100 000+ records added like Win32:Trojan-000001 [Trj]? ;D
-
Rather than hearing from all the wannabee business managers of avast and the numbers they think avast should be advertising ... not one of which will do one d*mned thing to keep my system safe tomorrow and all the tomorrows to come perhaps we can hear a bit more about Dwarden's concern about missing information provided to avast.
How does the avast team respond to a report that it is missing infections that were reported to it some time ago?
-
Submissions are not forgotten.
They just aren't in LIFO or FIFO order, contrary to most of the user's expectations. :-\
-
kubecj,
I think most of us can understand that response ...
However, if a user says that there are infections still undetected that were reported "a long time ago" I have to guess they are speaking of more than a few weeks.
How would you want Dwarden to deal with specific longstanding issues that have not been addressed?
-
Submissions should be added the next day. Anything else is just failing.
-
You really want 100 000+ records added like Win32:Trojan-000001 [Trj]? ;D
No, of course ;D
I mean your opinion about this:
Or won't it be better to fully remove the virus encyclopedia from the program (I know that this should be in the WishList topic, but...) - I mean when an ordinary user opens the encyclopedia and see only 56000 sample names, and after that compare avast! with other AV, in many cases the user will choose the AV with more virus signatures showed in the program. So this is a bit confusing or let it say missleading for some users, in my opinion. For example if I am not wrong for the same reason VBA32 removed their Virus list from their product.
-
You really want 100 000+ records added like Win32:Trojan-000001 [Trj]? ;D
No, not really, you can count it as just one... You're taking an extreme situation as example. Maybe the virtue will be in the middle of the extrems possibilities.
Submissions are not forgotten. They just aren't in LIFO or FIFO order, contrary to most of the user's expectations. :-\
This will never help AFTER the user is infected... :-[
Submissions should be added the next day. Anything else is just failing.
At least, the ITW ones should have this protocol...
-
kubecj,
I think most of us can understand that response ...
However, if a user says that there are infections still undetected that were reported "a long time ago" I have to guess they are speaking of more than a few weeks.
How would you want Dwarden to deal with specific longstanding issues that have not been addressed?
well they keep adding them over time ...
most of them aren't important to be added instantly ...
but there was small amount which really bother me (let say real danger) ... like one or two still missing
-
Hi polonus. I also think that Avast is good, and I am happy that it improves. The reason that I now am using BD on my desktop, is because I felt that Avast could improve some more, I think it still has a lot of potential to develop, and when I can see in some tests, like av.comparatives that Avast is in the same league (sorry for my spell.. don´t know how to write that) that Kaspersky, NOD or BD, Ill use it again and for good.
As I said before, Im happy that avast improves, and with my previuos posts, I don´t want to criticize avast, I wanted that avast improve.
K.
I do not agree friend >:-)
Having some 17 years working with computers -from the old-good 8-bit machines- I think avast! is an EXCELLENT antivirus that can without any problem share first place with Kaspersky's, BitDef., Norton, NOD, Norman, Panda, etc. and even get ahead all of them.
I think avast! needs to mutate to a comprehensive [Internet] Security Solution with proactive defense (like KIS6, ProcessGuard or TinyPersonal Firewall PRO) and a very-good internal firewall (like Tiny).
Keep up this excellent work and please consider getting a step forward Security Suite
-
Avast has in the last months added a lot of the missed samples (over 16500 samples so far) and will continue to do that.
Of course samples submitted from users should be added with higher priority.
btw, samples must be added by priorities, no av company adds within some few days all samples that are submitted to them, esp. not if they are not coming from users, but from e.g. testers or other big collections [imo]
@alwil: if you need an updated list of samples still missed, just let me know ;) - at the moment you should have enough samples on the ftp to add ;)
-
Avast has in the last months added a lot of the missed samples (over 16500 samples so far) and will continue to do that.
avast! certainly has been improving greatly over the last year or so with detections. I have noticed that myself after studying IBK's tests recently. From IBK's On-demand comparative from February 2006, I have posted an image which shows just how much avast! has improved with detections. And that was way back at the start of 2006. And with all those signatures added just recently, avast! will certainly be one of the top AV's in no time.
Credit of this image goes to IBK. IBK, if you would prefer me to remove this image, let me know and I will do that.
(http://img147.imagevenue.com/loc260/th_13951_compare.jpg) (http://img147.imagevenue.com/img.php?loc=loc260&image=13951_compare.jpg)
-
And a change from Standard to Advanced in the last test. This speaks to the "more than just detection rate" we argue so often.
-
no av company adds within some few days all samples that are submitted to them, esp. not if they are not coming from users
Antivir, AVG and Ewido are doing this. I know because I have sent them samples and they have been added with 24 hours, in the case of Antivir and AVG, with a personal email in reply, not an automated message.
These were samples of new Feebs and Zlob variants.
These are obviously dangerous viruses- the avast! website lists Feebs as one of the latest threats, and a real threat- obtained from my email inbox or from web browsing.
These were not some exotic zoo viruses, so leaving them undetected would have left users of those products vulnerable to attack.
I still say not adding a sample within 24 hours is failing: it denies protection from that virus to users of the AV. If Kaspersky and others can do it in hours, then obviously that's the gold standard. 24 hours is in fact perhaps over generous- AV companies should do it quicker. Certainly making excuses for taking longer is not going to impress anybody who takes the time to submit a sample.
I've heard the excuse that avast! receives a huge number of emails, but I wonder: do Antivir and AVG receive fewer? Does avast! receive a disproportionate number of virus samples?
In fact the number of samples sent each day is a golden opportunity to improve detection. What more could an AV company ask than to be supplied with virus samples?
As to my submissions to avast!, sadly the response was a little less swift: my Feebs sample (a 'latest threat') is still undetected after 10 weeks.
-
no av company adds within some few days all samples that are submitted to them, esp. not if they are not coming from users
Antivir, AVG and Ewido are doing this. I know because I have sent them samples and they have been added with 24 hours, in the case of Antivir and AVG, with a personal email in reply, not an automated message.
These were samples of new Feebs and Zlob variants.
Yes, I can confirm this, AVG is really fast at adding new malware to its database, I now have some Zlob trojans (at least 3 samples) that AVG detects while avast! still not. I have sent both AVG and avast! a samples and AVG always added them in its next updates.
I think adding a lot of Trojan-gen.xxx signatures from time to time is not that good. :P
=====> http://forum.avast.com/index.php?topic=7160.msg58870#msg58870
-
I said ", esp. not if they are not coming from users", because otherwise the samples I sent e.g. to AVG and AVIRA would not be still undetected after several months or in some cases years.
-
@IBK
Is this because your samples pose a greater threat than those typically submitted by users, or is it more a desire on the part of the AV companies to do well in the tests?
@alwil team
How are the additions prioritized? Damage potential? Ease of infection? Something else?
-
well i'm still waiting for automated web based submission system where i can see easily status of submission
plus it can prevent duplicate mess (if 2 or more users upload file with same hash it will inform later upload this file is already uploaded and awaiting investigation OR being investigated OR clean OR damaged something OR infected yet not added to VPS OR INFECTED and already detected)
this could also solve issue of emailing question why file XY is not yet added or forum questions and complains about why this wasn' added when etc...
i know patience but ... this is year(s) old scheme and still nothing
-
well i'm still waiting for automated web based submission system where i can see easily status of submission
Very good suggestion indeed... 8)