Teacher
Professional
- Messages
- 2,670
- Reaction score
- 806
- Points
- 113
It's been a long time since there were any articles, but since the other day I had a chance to go a little deeper into testing new anti-detection tools, I decided to describe for you in more detail what's what, and what all these webgl and other things are responsible for that can be turned on and off. For those who have been working with all this for a long time, the article may seem obvious, but do not forget that there are those subscribers who do not understand "what is reverse frequency response".
Let's start with useragent - here, in general, everything is trite simple, if you put a useragent that differs from your browser version (whether it's chromium in most anti-detections, or firefox in some adspower), then fuck you; if you put a different axis (not the one that is based on), then it also sends you fuck it, because you can see a banal font difference. You can even check it on pixelscan.net, which, by the way, is not a special indicator of the trustworthiness of the setting. Why? Because all modern antidetects (even dolphin) are cut with the expectation of deceiving it, although in fact they deceive only you, because you pay for their garbage.
Let's move on to the rest of the prints, which are a little more complicated.
WebGL / WebGL info (WebGPU) - this parameter, if you move away from working with antifraud, is used by sites to determine your video card for the correct processing of those site elements that are processed directly on your device. And if you take our native-favorite antifraud, then one thing is important to us-that the WebGL information (in simple terms , the displayed name of the video card) matched her fingerprint. Even Google checks this when working with it, so if you work with logs, it's better to turn on substitution, even if in the absolute majority of anti-detection it's noise, but if you compare it with other parameters , it's not so bad, but more on that later.
AudioContext - the same thing, only with an audio card/microphone. In general, now this fingerprint is bad everywhere, so when working with anti-detection, this parameter can in theory burn you and interfere with successful work, but in practice - everyone doesn't care, and if you take everyone's favorite chase - it doesn't know a fig except for webgl. In general, it is better not to touch this parameter at all - the noise curves will give you away more strongly than the real print.
ClientRects is a parameter that is used to determine how to position page elements in the browser. It directly depends on the display resolution and window size, so if you are working with the log and need to change it, it is better to enable it, otherwise you will be burned and you will not see half of the pages. But again, it is better to play with this parameter before working with the anti-detection - there is a chance that it replaces it crookedly.
Canvas-canvas is a system fingerprint that consists of fonts/browser versions(which is why the left user agents are burned), and other bullshit that describes your system as a whole. Talking about it is not figs, but if the canvas is noise, then when registering banks, for example, it will send figs. And Google for example on it in Fig.
So how do I check the adequacy of these parameters? In short, as I said, all checkers lie because the anti-detection tools focus on them. But there is one checker-CreepJS on github, which is not particularly fucked up (at least for now). So, the original device in a clean browser window has a trust rate of about 65. And how to work with substitutions already depends on the purpose of your work, but I think the essence is clear. You should also look at the Frequency Check on pixelscan - if you have a parameter that is too unique ("no data"), then this is freaking noise, because the prints of identical devices are absolutely identical, which means it can't be unique for sure. There is also the AmIUnique website, but it sometimes gives out such garbage that even a salt store that steals your dollars will seem super adequate.
Important - when working with these sites, never continue working in the same profile, make a clone of it, because these sites hang fraudulent cookies on you and thus you do not sregat or work out anything.
Let's start with useragent - here, in general, everything is trite simple, if you put a useragent that differs from your browser version (whether it's chromium in most anti-detections, or firefox in some adspower), then fuck you; if you put a different axis (not the one that is based on), then it also sends you fuck it, because you can see a banal font difference. You can even check it on pixelscan.net, which, by the way, is not a special indicator of the trustworthiness of the setting. Why? Because all modern antidetects (even dolphin) are cut with the expectation of deceiving it, although in fact they deceive only you, because you pay for their garbage.
Let's move on to the rest of the prints, which are a little more complicated.
WebGL / WebGL info (WebGPU) - this parameter, if you move away from working with antifraud, is used by sites to determine your video card for the correct processing of those site elements that are processed directly on your device. And if you take our native-favorite antifraud, then one thing is important to us-that the WebGL information (in simple terms , the displayed name of the video card) matched her fingerprint. Even Google checks this when working with it, so if you work with logs, it's better to turn on substitution, even if in the absolute majority of anti-detection it's noise, but if you compare it with other parameters , it's not so bad, but more on that later.
AudioContext - the same thing, only with an audio card/microphone. In general, now this fingerprint is bad everywhere, so when working with anti-detection, this parameter can in theory burn you and interfere with successful work, but in practice - everyone doesn't care, and if you take everyone's favorite chase - it doesn't know a fig except for webgl. In general, it is better not to touch this parameter at all - the noise curves will give you away more strongly than the real print.
ClientRects is a parameter that is used to determine how to position page elements in the browser. It directly depends on the display resolution and window size, so if you are working with the log and need to change it, it is better to enable it, otherwise you will be burned and you will not see half of the pages. But again, it is better to play with this parameter before working with the anti-detection - there is a chance that it replaces it crookedly.
Canvas-canvas is a system fingerprint that consists of fonts/browser versions(which is why the left user agents are burned), and other bullshit that describes your system as a whole. Talking about it is not figs, but if the canvas is noise, then when registering banks, for example, it will send figs. And Google for example on it in Fig.
So how do I check the adequacy of these parameters? In short, as I said, all checkers lie because the anti-detection tools focus on them. But there is one checker-CreepJS on github, which is not particularly fucked up (at least for now). So, the original device in a clean browser window has a trust rate of about 65. And how to work with substitutions already depends on the purpose of your work, but I think the essence is clear. You should also look at the Frequency Check on pixelscan - if you have a parameter that is too unique ("no data"), then this is freaking noise, because the prints of identical devices are absolutely identical, which means it can't be unique for sure. There is also the AmIUnique website, but it sometimes gives out such garbage that even a salt store that steals your dollars will seem super adequate.
Important - when working with these sites, never continue working in the same profile, make a clone of it, because these sites hang fraudulent cookies on you and thus you do not sregat or work out anything.
