New Update F-Secure beta version updates

[Jonny Quest]

Definitely. In an architecture of engines, changing one engine is a very big deal (much bigger than what people think). It introduces duplicates in detection and protection and significant gaps from vectors not covered.

F-Secure (if they wanna stay relevant) will need to work very hard for over a year to be able to close the gaps this switch has introduced.

And then how hard would it be to win the people back over, like the old conception of McAfee most of of us had compared to where it's at now? I did end up in getting a Total Protection license for 3 devices since Best Buy had an excellent deal on them. I got over my initial qualms after I used the Antivirus for a couple of days. McAfee also has a lot more resources and options regarding Identity Monitoring than F-Secure's sub par ID Monitoring.

 

[Trident]

And then how hard would it be to win the people back over, like the old conception of McAfee most of of us had compared to where it's at now? I did end up in getting a Total Protection license for 3 devices since Best Buy had an excellent deal on them. I got over my initial qualms after I used the Antivirus for a couple of days. McAfee also has a lot more resources and options regarding Identity Monitoring than F-Secure's sub par ID Monitoring.

It all depends on how F-Secure will play their cards. I see they are already implementing anti-bot (most likely looking at ports, protocol, domain reputation and potentially the reputation of the files creating a process).
This gives F-Secure some control and ability to close the Avira gaps.

But for beat results, F-Secure will have to study the Avira engines in-depth and will need to come up with a lot more than anti-botnet heuristics and sensors. Possibly a Yara rules engine (so malware hunters can quickly push rules) and many others.

It will be hard to compete with McAfee because McAfee already has the framework in place. They’ve developed all their engines from scratch and each one of them is built to close the gaps of the rest.
In identity monitoring, even Norton can’t compete with McAfee.

 

[Szellem]

And then how hard would it be to win the people back over, like the old conception of McAfee most of of us had compared to where it's at now? I did end up in getting a Total Protection license for 3 devices since Best Buy had an excellent deal on them. I got over my initial qualms after I used the Antivirus for a couple of days. McAfee also has a lot more resources and options regarding Identity Monitoring than F-Secure's sub par ID Monitoring.

I don't want to ruin the topic, but did McAfee work out?

 

[Jonny Quest]

I don't want to ruin the topic, but did McAfee work out?

So far so good It's only been about a week, but I feel better about the company in general, McAfee, in that their more progressive using their own framework, as @Trident described above (and not the Avira SDK in F-Secure's case). F-Secure seems to be focusing more on the VPN, is getting that figured out, than any new advancement that may actually help improve our device security in a significant way.

I will still keep it installed on a Desktop I rarely use to occasionally help out on the forum, and see if they do start making some improvements, at least until next May when my Total license expires.

 

[Trident]

So far so good It's only been about a week, but I feel better about the company in general, McAfee, in that their more progressive using their own framework , as @Trident described above (and not the Avira SDK in F-Secure's case). F-Secure seems to be focusing more on the VPN, is getting that figured out, than any new advancement that may actually help improve our device security in a significant way.

I will still keep it installed on a Desktop I rarely use to occasionally help out on the forum, and see if they do start making some improvements, at least until next May when my Total license expires.

I believe that with botnet detection they are just getting started. F-Secure is not planning on being just an Avira skin-on-top. They’ve analysed the Avira weaknesses and they are taking steps to release proprietary technology. They have less staff and resources (no doubt there) but if the staff is experienced and knowe what they do, they can still achieve a lot. Specially now, that ChatGPT 4 can accelerate developments to unseen levels.

 

[Jonny Quest]

Excellent, good points as always @Trident I like the optimism

 

[Jonny Quest]

As long as I'm running F-Secure on at least one of my PC's, I'll update this thread
This was originally posted on Oct 9th. I won't post every beta update, but maybe one then the public release version with its changelog notes.

Release 25.10 beta 2 (6.7.152)​

Improvements:

• Minor optimizations for Browsing Protection, Banking Protection, Trusted Shopping and Family Protection

Bug fixes:

• Device Protection malfunction smart task shows proper details when the problem occurs because of too little free disk space (PBL-15312)
• Clicking on some areas of Scan UI infection card does not open infection details (PBL-15336)
• In some VPN location change scenarios VPN could connect to other location than shown in UI (PBL-15339)

 

[Jonny Quest]

F-Secure version 25.10 public release.

FSecure Knowledge

www.kb.f-secure.com

Version 25.10​

Released November 2025

VPN websites bypass​

Specific websites or domains can be set to bypass VPN tunnel. This may be helpful in case some website blocks access over VPN and user wants to avoid turning off VPN to access such site.

Bug fixes:​

• Device Protection malfunction smart task shows proper details when the problem occurs because of disk space
• In some VPN location change scenarios VPN could connect to other location than shown in UI
• Family Protection cannot be set up when LanmanWorkstation service is not running
• Changing Windows user name crashes UIs before the computer is restarted
• Excluded paths were not handled correctly in Scam Protection
• User interface consistency improved in various interfaces

 

[Moonhorse]

F-secure prices gone up while protection goes worse than before, i would use f-secure in my daily pc usage but it doesnt seem improving anytime soon. Freedome vpn is good and i would decide to use it again

 

[Lord Ami]

Release 25.11 beta 1 (6.8.184)​

Native ID monitoring:

• ID monitoring is now implemented natively in the Windows app
• This improves the performance of the feature while also adding support for dark mode and accessibility features

Known issues:

• Main UI may crash where there is no network connection

 

[Jonny Quest]

F-secure prices gone up while protection goes worse than before, i would use f-secure in my daily pc usage but it doesnt seem improving anytime soon. Freedome vpn is good and i would decide to use it again

I agree, but the Avira scanning engine isn't to bad, in general, but both Avira and F-Secure do suffer the bigger FP numbers (if that's ever been a issue for users?). It's all cosmetic, some glitch fixes and VPN related fixes, nothing of significance like, "we improved the Behavior Detection by including..." Or, "we've enhanced our browser/Scam protection technology by including..." type of improvements.

This post as well as the others by @Trident on this page number help to give some perspective to F-Secures plight.

New Update Post in thread 'F-Secure beta version updates'

Oct 10, 2025

Yep, I agree I didn't realize it was incorporated into the first v25.2 release Maybe to try help make up for or help with Deepguard's underwhelming replacement, Behavior Protection/Avira Sentry?

Definitely. In an architecture of engines, changing one engine is a very big deal (much bigger than what people think). It introduces duplicates in detection and protection and significant gaps from vectors not covered.

F-Secure (if they wanna stay relevant) will need to work very hard for over a year to be able to close the gaps this switch has introduced.

• Trident

• 
• 

If there's a significant beta release, I'll post it if @Lord Ami doesn't post it first, otherwise, as for me, it will only be for the public release version.

 

[Trident]

They can’t really improve the behavioural protection much because they have 0 control over it. All that avira gives them is one wrapper.

This wrapper will:
-Initialise the engine
-Pass some parameters to the engine, such as output logs destinations and so on
-It will take serialised JSON information from the engine, which will include things like detection name (something internal that mqy not necessarily be displayed to users), the offending process, its PID and potentially some metadata.
-Upon the engine detecting something, a bridge will trigger an alert from F-Secure.

In this stricture, the only thing that belongs to F-Secure is the bridge. There is not much to improve there.

In terms of scam protection, using APIs with a token-based charge is not smart because one text message can easily soawn over 1000 tokens (and more).
For F-Secure to offer that, they need their own models designed for low and predictably cost.
They will eventually, but it must be kept in mind that theirbR&D finances are very limited.

 

[Jonny Quest]

They can’t really improve the behavioural protection much because they have 0 control over it. All that avira gives them is one wrapper.

This wrapper will:
Initialise the engine
-Pass some parameters to the engine, such as output logs destinations and so on
-It will take serialised JSON information from the engine, which will include things like detection name (something internal that mqy not necessarily be displayed to users), the offending process, its PID and potentially some metadata.
-Upon the engine detecting something, a bridge will trigger an alert from F-Secure.

In this stricture, the only thing that belongs to F-Secure is the bridge. There is not much to improve there.

In terms of scam protection, using APIs with a token-based charge is not smart because one text message can easily soawn over 1000 tokens (and more).
For F-Secure to offer that, they need their own models designed for low and predictably cost.
They will eventually, but it must be kept in mind that theirbR&D finances are very limited.

What can be done then, are they a lost cause, stick a fork in them they're done? And yet, a normal common sense user who isn't downloading the internet everyday could be very secure in using this AV, I know I would be. Especially in tandem with GlassWire free, which I used for the notifications and information, as the Firewall module is off, and wouldn't interfere with F-Secure's Firewall sensor.

An issue for me was the performance issue, system impact in opening apps, even Brave. Now if they could do something about that?

 

[Trident]

What can be done then, are they a lost cause, stick a fork in them they're done? And yet, a normal common sense user who isn't downloading the internet everyday could be very secure in using this AV, I know I would be. Especially in tandem with GlassWire free, which I used for the notifications and information, as the Firewall module is off, and wouldn't interfere with F-Secure's Firewall sensor.

An issue for me was the performance issue, system impact in opening apps, even Brave. Now if they could do something about that?

They can do a lot about that, they can put the Avira scan process in their own host. They can develop their own caching, kernel driver (mini-filter), as well as logics when and where the engine is called. For example, they could scan Brave once and then never again until an update occurs.

However, if the slowdown happens somewhere in the hook injection logics or other behavioural monitoring functions, they wouldn’t be able to do much. Potentially if the engine allows them to define exclusions (not sure as Avira will provide the documentation only to customers), they can exclude some trusted processes from monitoring.

There are some actions that could be taken, specially in cooperation with Avira but they all require significant restructuring and payments.

They need to run debugging to identify the bottlenecks and then take it from there.

I wouldn’t say they are dead but they just don’t have the spending power some others possess.

 

[Jonny Quest]

I think @bazang mentioned previously this is going to be, is now a niche AV, used mainly by those who support them *especially in the Finland region (*my words).
They still have ISP providers who use their rebranded version (including in the US) and some education centers in Europe (read that in a F-Secure forum post).

 

[Jonny Quest]

They can’t really improve the behavioural protection much because they have 0 control over it. All that avira gives them is one wrapper.

This wrapper will:
-Initialise the engine
-Pass some parameters to the engine, such as output logs destinations and so on
-It will take serialised JSON information from the engine, which will include things like detection name (something internal that mqy not necessarily be displayed to users), the offending process, its PID and potentially some metadata.
-Upon the engine detecting something, a bridge will trigger an alert from F-Secure.

In this stricture, the only thing that belongs to F-Secure is the bridge. There is not much to improve there.

In terms of scam protection, using APIs with a token-based charge is not smart because one text message can easily soawn over 1000 tokens (and more).
For F-Secure to offer that, they need their own models designed for low and predictably cost.
They will eventually, but it must be kept in mind that theirbR&D finances are very limited.

They can do a lot about that, they can put the Avira scan process in their own host. They can develop their own caching, kernel driver (mini-filter), as well as logics when and where the engine is called. For example, they could scan Brave once and then never again until an update occurs.

However, if the slowdown happens somewhere in the hook injection logics or other behavioural monitoring functions, they wouldn’t be able to do much. Potentially if the engine allows them to define exclusions (not sure as Avira will provide the documentation only to customers), they can exclude some trusted processes from monitoring.

There are some actions that could be taken, specially in cooperation with Avira but they all require significant restructuring and payments.

They need to run debugging to identify the bottlenecks and then take it from there.

I wouldn’t say they are dead but they just don’t have the spending power some others possess.

It sounds like they really painted themselves into a corner, to help save on expenses when they made this Avira SDK change. But, they probably knew that going in, that they needed to do something for a company that was losing money, especially on the consumer side.

 

[stonjean633]

It sounds like they really painted themselves into a corner, to help save on expenses when they made this Avira SDK change. But, they probably knew that going in, that they needed to do something for a company that was losing money, especially on the consumer side.

They used to be good. I hope they survive and bounce back.

 

[Jonny Quest]

They used to be good. I hope they survive and bounce back.

I like that optimism, but I'm not counting on it. I'm running McAfee on one laptop and Avast free on another, deciding which one(s) going to be my all around F-Secure replacement. Eset is to expensive (even though there are other discount vendors), with to many settings I don't need, or wouldn't use.

 

[Trident]

It sounds like they really painted themselves into a corner, to help save on expenses when they made this Avira SDK change. But, they probably knew that going in, that they needed to do something for a company that was losing money, especially on the consumer side.

ZoneAlarm for example (not to be confused with Check Point) consists of 5 guys that take the Check Point engines and bundle them.

TotalAV is the same, they’ve got 2-3 guys in marketing and 2-3 devs.

I see the F-Secure future to be something similar.

However if Gen pumps the prices up through the inclusion of premium Avast-ripped features, F-Secure will switch back to Bitdefender (which will happily issue reduced-price contracts to attract business).

F-Secure if they don’t up their game with bells and whistles A-la Gen Digital, risks circulating from OEM to OEM like a florist, amidst declining user bases and revenues, as a change of engines is an earthquake.

Check Point has over 70 proprietary engines and in the beginning of the full transition to Sophos, they messed up big time and it took them more than a year to stabilise the entire architecture.

 

[Trident]

Speaking of their scam protection: