3 ms·

You're comparing keys from two different cryptosystems, which is a mistake. A RSA key ("1024 bit keys") needs to be longer than a AES key ("256 bits") to provi

You're comparing keys from two different cryptosystems, which is a mistake. A RSA key ("1024 bit keys") needs to be longer than a AES key ("256 bits") to provide a given amount of security, since more conditions are placed on an RSA key in order to get its favorable properties. In 1999 you could break a 512-bit RSA key with a supercomputer, in 2015 you can do it with 4 hours and $75 on EC2: http://arstechnica.com/security/2015/10/breaking-512-bit-rsa-with-amazon-ec2-is-a-cinch-so-why-all-the-weak-keys/ http://arstechnica.com/security/2015/10/breaking-512-bit-rsa... A 768-bit RSA keys has been factored in a large academic effort: https://eprint.iacr.org/2015/1000.pdf https://eprint.iacr.org/2015/1000.pdf When you pick a key length, you want it to be long enough to provide security for your communications as far into the future as practicable, and the classical efforts to break RSA are getting uncomfortably close to 1024-bits. Maybe in a few decades it would be practical for some adversaries to break even perfectly made 1024-bit keys. However, I'm not a cryptographer, so someone please correct me if I'm wrong.

baldfat 11y ago
> 1024 bit keys are endangered? Only in some broken encryption schemes. He didn't compare anything.
- gherkin0 11y ago
  > He didn't compare anything. No, he did: >>> 1024 bit keys are endangered? .... You can't even break 256 bits 1024 bit key = most likely something like RSA 256 bit key = some symmetric algorithm or ECC He's implying a 1024 bit RSA key should be safe because a 256 bit key from some other algorithm is.
  - baldfat 11y ago
    > Only in some broken encryption schemes RSA 1024 is a broken encryption scheme (2010). In Jan 2010 there was already the concern it was broken or quickly to be broken. http://arstechnica.com/security/2010/01/768-bit-rsa-cracked-1024-bit-safe-for-now/ http://arstechnica.com/security/2010/01/768-bit-rsa-cracked-... in March 2010 http://www.techworld.com/news/security/rsa-1024-bit-private-key-encryption-cracked-3214360/ http://www.techworld.com/news/security/rsa-1024-bit-private-...
    - gherkin0 11y ago
      I mentioned your first link, the factoring of a 768-bit RSA key, in my first comment. RSA isn't a "broken encryption scheme": it's a perfectly fine one. It's just that you need to use key lengths long enough to defend against the computing power available to current and expected adversaries. That paper is just a demonstration of modern computing power, which just shows that you should use longer keys. I took his statement of "Only in some broken encryption schemes" to be a misunderstood reference to stuff like this: http://arstechnica.com/security/2015/10/how-the-nsa-can-break-trillions-of-encrypted-web-and-vpn-connections/ http://arstechnica.com/security/2015/10/how-the-nsa-can-brea... or a side channel attack. Your second link has garbage sensationalized headline. It actually describes a side channel attack that has nothing to do with cryptography algorithms. It's basically equivalent to a clever way of looking over someone's shoulder.

baldfat 11y ago

> 1024 bit keys are endangered? Only in some broken encryption schemes. He didn't compare anything.

gherkin0 11y ago
> He didn't compare anything. No, he did: >>> 1024 bit keys are endangered? .... You can't even break 256 bits 1024 bit key = most likely something like RSA 256 bit key = some symmetric algorithm or ECC He's implying a 1024 bit RSA key should be safe because a 256 bit key from some other algorithm is.
- baldfat 11y ago
  > Only in some broken encryption schemes RSA 1024 is a broken encryption scheme (2010). In Jan 2010 there was already the concern it was broken or quickly to be broken. http://arstechnica.com/security/2010/01/768-bit-rsa-cracked-1024-bit-safe-for-now/ http://arstechnica.com/security/2010/01/768-bit-rsa-cracked-... in March 2010 http://www.techworld.com/news/security/rsa-1024-bit-private-key-encryption-cracked-3214360/ http://www.techworld.com/news/security/rsa-1024-bit-private-...
  - gherkin0 11y ago
    I mentioned your first link, the factoring of a 768-bit RSA key, in my first comment. RSA isn't a "broken encryption scheme": it's a perfectly fine one. It's just that you need to use key lengths long enough to defend against the computing power available to current and expected adversaries. That paper is just a demonstration of modern computing power, which just shows that you should use longer keys. I took his statement of "Only in some broken encryption schemes" to be a misunderstood reference to stuff like this: http://arstechnica.com/security/2015/10/how-the-nsa-can-break-trillions-of-encrypted-web-and-vpn-connections/ http://arstechnica.com/security/2015/10/how-the-nsa-can-brea... or a side channel attack. Your second link has garbage sensationalized headline. It actually describes a side channel attack that has nothing to do with cryptography algorithms. It's basically equivalent to a clever way of looking over someone's shoulder.