Side channel attacks are a very active field of research right now (and have been since a few years). We'll probably see more of these, especially in hardware.
However, I don't consider RSA (with any keystrength) cracked because of these things. Because side-channel weaknesses can usually be mitigated without modifying the algorithm, and they don't actually have any relevance for the mathematical properties.
It's still important to know about them, because they have real world application and can be a risk.
When will people bother to actually read the documents? Last time the documents said very specifically that they never actually managed to do what they claimed to have already done.
This time they did not admit they are incapable of doing this. I guess they intend to get another grant so they can continue playing with microphones.
Has it occurred to you that they are speaking of detecting the differences between 4096 zeros and ones, in advance, on a system they may have never tested in a soundproof room, that may be running other software at the same time on the same core(s) as the encryption is being ran on?
You want to know a good way to prevent this from effecting you? Use a dozen different applications that each use a tiny bit of CPU time constantly. And if you think they are after you, throw in a benchmark so they get a wall of nonsense that encryption would look scrambled in.
One problem with this...the newer ransomwares don't let you boot into safe mode, thus making the virus tool useless...if you can't download it or install it from a usb in safe mode, sorry your f#+×÷d!!
Comments
However, I don't consider RSA (with any keystrength) cracked because of these things. Because side-channel weaknesses can usually be mitigated without modifying the algorithm, and they don't actually have any relevance for the mathematical properties.
It's still important to know about them, because they have real world application and can be a risk.
When will people bother to actually read the documents? Last time the documents said very specifically that they never actually managed to do what they claimed to have already done.
This time they did not admit they are incapable of doing this. I guess they intend to get another grant so they can continue playing with microphones.
Has it occurred to you that they are speaking of detecting the differences between 4096 zeros and ones, in advance, on a system they may have never tested in a soundproof room, that may be running other software at the same time on the same core(s) as the encryption is being ran on?
You want to know a good way to prevent this from effecting you? Use a dozen different applications that each use a tiny bit of CPU time constantly. And if you think they are after you, throw in a benchmark so they get a wall of nonsense that encryption would look scrambled in.
Stop buying this nonsense.
** Post edited due to vulgarity by Goldstein