Also no one likes to talk about NSA backdoors into all computer chips that specifically target the random number generators. What it does is limit the amount of random numbers so the secure keys that get generated using the RNG function (md5, dsa, rsa, etc) can be broken within seconds instead of years. And to what extent does this mean they can force all chips to spit out just ones or just zeros for any period of time? I wouldn't doubt that they have that control as well by now.
This GCP experiment is entirely based on AMD and Intel chip hardware designs and they have known flaws when it comes to RNG. Of course iOS and Windows will play along and allow the hardware exploits to be used in the software (they are closed source). But Linux is open source and TPTB got to that too. The creator of Linux forced a binary dependency (closed source, back like 15 years ago) from a govt agency and its for the RNG component. Nothing is random if you are using computer chips to measure the randomness.
Shut it all down and start over.
Quoting: AcidBass With due respect, you don't know WTH you're talking about
generated using the RNG function (md5, dsa, rsa, etc)
Quoting: AcidBass MD5 is a hash algorithm.
DSA and RSA are public-key encryption algorithms.
None of those are RNGs. You just discredited yourself at the start.
This GCP experiment is entirely based on AMD and Intel chip hardware designs and they have known flaws when it comes to RNG.
Quoting: AcidBass Also wrong. They are not relying on CPU intrinsics. From the Princeton preprint:
Briefly, the network employs research grade, commercial random bit generators [Orion, Mindsong, PEAR]. The devices process quantum-level electronic noise (post-barrier voltage from electron tunneling in diodes or field effect transistors; or Johnson noise) to generate a bit stream with binomial probability of 1/2, at rates of several thousand bits per second.
Quoting: Bancel and Nelson [
link to global-mind.org (secure)]
The creator of Linux forced a binary dependency (closed source, back like 15 years ago) from a govt agency and its for the RNG component. Nothing is random if you are using computer chips to measure the randomness.
Quoting: AcidBass After your earlier B.S. I have to ask you for a source for that.
The Linux GNU C Library
random() uses a simple LFSR (linear feedback shift register), no "binary dependency", but is
never trusted for security-critical or research work. It's also ridiculously well-known never to use an OS/stdc "rand()" for anything secure or random, as it's notorious for bad statistical behavior...
Security and research projects often rely on trusted cryptographic software PRNGs e.g. based on a stream or block cipher (RC4, AES-CTR), or the slower Blum algorithms.
For instance, the security of Blum-Micali is provable based on the difficulty of computing discrete logarithms:
[
link to en.wikipedia.org (secure)]
It's not hardware dependent.
