[QUOTE=OmniConsUme;50337534]Cue FBI now having another freakout.[/QUOTE]
you need a machine that has the two random sources. You consumer PC doesn't.
I thought this was gonna be the Mario 64 guy again for some reason
"Current methods of generating random numbers can produce predictable results." ITS NOT A FUCKIN RANDOM NUMBER GENERATOR THEN IS IT
[QUOTE=Britain;50340080]"Current methods of generating random numbers can produce predictable results." ITS NOT A FUCKIN RANDOM NUMBER GENERATOR THEN IS IT[/QUOTE]
It's pseudo-random. It's technically predictable but pretty difficult to do so. From what I know they often use unix time to seed the random number meaning that to predict the values you need to be able to accurately pinpoint when the number was generated to the millisecond.
Wait, how do random number generators work?
[QUOTE=t h e;50342893]Wait, how do random number generators work?[/QUOTE]
Pseudo random number generators in computers generally take the current time since the UNIX epoch and run it through an algorithm. It's not really random, but it's good enough if you can't determine the exact time in milliseconds that the number was generated
[QUOTE=AJ10017;50342928]Pseudo random number generators in computers generally take the current time since the UNIX epoch and run it through an algorithm. It's not really random, but it's good enough if you can't determine the exact time in milliseconds that the number was generated[/QUOTE]
Not really, because if you can get multiple numbers generated from the same seed you can easily predict more numbers.
That's if it's not cryptographically secure, anyways.
[QUOTE=t h e;50342893]Wait, how do random number generators work?[/QUOTE]
Like this: [url]http://www.mscs.dal.ca/~selinger/random/[/url]
This is how GNU C's random() function is implemented (not the exact same but the code in the example produces the same output),
All the code does is do mathematical operations to the seed number to create new numbers.
I do lots of work with random numbers and simulation, we usually use the [URL=https://en.wikipedia.org/wiki/Mersenne_Twister]Mersenne Twister[/URL] algorithm. Random number generation isn't a bottleneck for our systems - I think that the lads down in crypto will benefit more from this new algorithm.
[QUOTE=Britain;50340080]"Current methods of generating random numbers can produce predictable results." ITS NOT A FUCKIN RANDOM NUMBER GENERATOR THEN IS IT[/QUOTE]
Random is only random out of pure ignorance of what makes it not random
[video=youtube;9rIy0xY99a0]http://www.youtube.com/watch?v=9rIy0xY99a0[/video]
[editline]20th May 2016[/editline]
I have a GUID generator bound in autohotkey
[QUOTE]9a7513fb-acd8-859a-43e9-6253aca0033d
452ddf23-8822-7f96-4402-51c658ff70a3
8d7a578c-0615-ba8e-4e7d-c44f7ba4ff32
893ea1e8-5f06-a5ac-487a-a40b8abf5406
eac3ab94-98fd-5e87-4dc8-cca23e2fd40c
4f59080f-f8a1-db96-4c5a-86a66b0faa63
a68b25f3-f551-a2a7-49cf-f4f4603c83a1
46481aa8-bfbc-0e95-4fe7-a7b47fc8e8db
e364d4a2-c481-a899-4eed-6339361e0a42
b1edd135-7aed-4599-4f1f-5ba424146c9b
f04265cd-6d78-0f9c-4c46-d0611267b115
[/QUOTE]
even this isn't 100% truly random
[QUOTE=t h e;50342893]Wait, how do random number generators work?[/QUOTE]
Depends on how "random" you want the generated values to be. Simple random number generators would take a simple value like the current time in seconds since epoch and run it through an algorithm such as one mentioned by Im Crimson here.
[QUOTE=Im Crimson;50343226]Like this: [url]http://www.mscs.dal.ca/~selinger/random/[/url]
This is how GNU C's random() function is implemented (not the exact same but the code in the example produces the same output),
All the code does is do mathematical operations to the seed number to create new numbers.[/QUOTE]
If you want more secure random number generation, you would need to gather a lot of [URL="https://en.wikipedia.org/wiki/Entropy_(computing)"]entropy[/URL] (eg. data from the "real world" such as user's mouse movements, amount of running processes, background noise from a webcam feed, etc.). This data is then mixed together into a seed value, which is given to the random number generator.
Modern Linux systems do a lot of entropy collection in the background, so randomness on there is very acceptable.
Servers tend to have issues with entropy though, since they lack many of the more common entropy sources. No HID inputs, no microphones, no webcams, no accelerometers, etc.
This can be noticeable rather easily when generating high-grade RSA encryption keys - for instance.
Sorry, you need to Log In to post a reply to this thread.