'badpassword' 10 days
'bad1password' 37 years
'bad password' 546 years
'bad 1password' 333 thousand years
'bad password 1' 18 million years
'bad 123password' a billion years
As with all password checkers, there are problems here.
But password security is much more about using bcrypt or scrypt (in case someone gets your database) and having timeouts and lockouts on password entry (to prevent brute-forcing through public channels).
Obviously you shouldn't enter your real passwords there, but the site could have a valid use for discovering the strength of passwords you would generate for some hypothetical service that is actually not used.
It doesn't even have merit there because it's figures are completely wrong:
First of all it doesn't take into account GPU cracking techniques - in fact it uses an average desktop PC as the processing power gauge (completely ignoring that password crackers would be running on highly powerful systems bespokely configured systems),
Next it doesn't take into account modern dictionary attacks - sometimes people will have a seemingly secure password but it's actually a common term and thus included in password dictionaries.
And to top it all off, there's no details about how long it would take against different encryption cyphers (the best cyphers are designed to be computationally expensive - but sadly a lot of sites are still using SHA128/256, or worse yet, MD5).
It's rare that I'm this rude about other peoples projects, but that site is total garbage. In fact it's worse than garbage because it deludes people into thinking their passwords are secure when often they're not. It's alarming how quick hackers can rattle off password attempts these days - to quote a recent Ars article[1] "Using a commodity computer with a single AMD Radeon 7970 graphics card, it took him 20 hours to crack 14,734 of the hashes, a 90-percent success rate." and that's an extremely low spec machine compared to most.
This is questionable. If I enter 15 random alphabet characters, it gives me 49 years. 15 numbers is 2 days. 15 characters of punctuation is 743,000 years.
If the cracker was trying to brute force a password, then they would have to try every available character in every valid position. Unless I'm misunderstanding something, and n-length password would take a similar amount of time to guess as any other n-length password with the same restrictions. The cracker doesn't know going into it that I only have letters in my password, so he has to use any valid character.
Perhaps a more useful system would be one that would let websites enter their password restrictions and tell them how secure their users' password can be with those restrictions.
> The cracker doesn't know going into it that I only have letters in my password, so he has to use any valid character.
Typically hackers will run through quicker combinations to pick off the low hanging fruit. It doesn't matter if they can't crack a few hundred 15 char characters with extended ANSI characters as by that point they'll already have tens of thousands of passwords from others who's passwords contained less entropy.
I don't get the fascination with MD5 cracking speeds. If you use different passwords for each site, a 7 character lowercase password would take 10 years to exhaust the password-space if a site limits you to 20 attempts per second. As long as you are using different, random passwords for each site, you can just change for password for InsecureApp once InsecureApp gets hacked.
I think we should be educating people on using different passwords for services rather than convincing them to use very long, complex passwords.
I use so many different services/sites that having a different password for each one is unrealistic. Here's a small sample: Personal email, work email, version control at work, Github, bank, university systems, personal computer, work computer, Facebook, Hacker News, Reddit, Steam, Skype, ... I could go on. In order to keep track of all of these unique passwords, I would need some kind of protected database like 1Password [0] (which I do have), or a notebook with them all written down. At this point, however, cracking/guessing a single password or obtaining a physical notebook has given someone access to all of my accounts.
What I typically do is give "important" services unique, random passwords (bank, personal email, work email; things that can do damage), and use one of a set of passwords for everything else. If one password gets found, then maybe only my Reddit, Skype and a few forum accounts are compromised. This has the advantage that I use the unique password frequently enough to memorize them, and non-unique passwords are used in a few places which results in them also being used enough that they are memorized.
I'm not a password cracker, but from what I read, my test password of "superman15" could probably be cracked a fair bit faster than 10 days, as indicated by the site.
Good site! I recently switched all my passwords over from about 3 varieties of a ~8 character string to ~30 characters that's unique (and easy to remember for each site.) I'm unsure why it's taken me so long to realise that I could have a super secure password like:
My name is bob smith from California USA and I am trying to login to: xxx
Comments
Easy test: If you type you password in on random sites, it's not secure at all.
Also:
But radix had that password (and thousands of others) in an hour[1][1] http://arstechnica.com/security/2013/05/how-crackers-make-mi...
Some tests:
As with all password checkers, there are problems here.If you want a password checker, this one is still the best I've seen: http://dl.dropboxusercontent.com/u/209/zxcvbn/test/index.htm... (it correctly identifies all my above examples as terrible passwords)
But password security is much more about using bcrypt or scrypt (in case someone gets your database) and having timeouts and lockouts on password entry (to prevent brute-forcing through public channels).
My god, people would use this? I was hoping it was one of the sites where as soon as you started typing, it flashed "wtf are you thinking you moron".
Obviously you shouldn't enter your real passwords there, but the site could have a valid use for discovering the strength of passwords you would generate for some hypothetical service that is actually not used.
It doesn't even have merit there because it's figures are completely wrong:
First of all it doesn't take into account GPU cracking techniques - in fact it uses an average desktop PC as the processing power gauge (completely ignoring that password crackers would be running on highly powerful systems bespokely configured systems),
Next it doesn't take into account modern dictionary attacks - sometimes people will have a seemingly secure password but it's actually a common term and thus included in password dictionaries.
And to top it all off, there's no details about how long it would take against different encryption cyphers (the best cyphers are designed to be computationally expensive - but sadly a lot of sites are still using SHA128/256, or worse yet, MD5).
It's rare that I'm this rude about other peoples projects, but that site is total garbage. In fact it's worse than garbage because it deludes people into thinking their passwords are secure when often they're not. It's alarming how quick hackers can rattle off password attempts these days - to quote a recent Ars article[1] "Using a commodity computer with a single AMD Radeon 7970 graphics card, it took him 20 hours to crack 14,734 of the hashes, a 90-percent success rate." and that's an extremely low spec machine compared to most.
[1] http://arstechnica.com/security/2013/05/how-crackers-make-mi...
"the site could have a valid use for discovering the strength of passwords you would generate"
That's what I did - generate a new password using my usual method.
Apparently "it would take a desktop PC about 13 duovigintillion years to crack your password" - is that long enough?
This is questionable. If I enter 15 random alphabet characters, it gives me 49 years. 15 numbers is 2 days. 15 characters of punctuation is 743,000 years.
If the cracker was trying to brute force a password, then they would have to try every available character in every valid position. Unless I'm misunderstanding something, and n-length password would take a similar amount of time to guess as any other n-length password with the same restrictions. The cracker doesn't know going into it that I only have letters in my password, so he has to use any valid character.
Perhaps a more useful system would be one that would let websites enter their password restrictions and tell them how secure their users' password can be with those restrictions.
> The cracker doesn't know going into it that I only have letters in my password, so he has to use any valid character.
Typically hackers will run through quicker combinations to pick off the low hanging fruit. It doesn't matter if they can't crack a few hundred 15 char characters with extended ANSI characters as by that point they'll already have tens of thousands of passwords from others who's passwords contained less entropy.
I don't get the fascination with MD5 cracking speeds. If you use different passwords for each site, a 7 character lowercase password would take 10 years to exhaust the password-space if a site limits you to 20 attempts per second. As long as you are using different, random passwords for each site, you can just change for password for InsecureApp once InsecureApp gets hacked.
I think we should be educating people on using different passwords for services rather than convincing them to use very long, complex passwords.
I use so many different services/sites that having a different password for each one is unrealistic. Here's a small sample: Personal email, work email, version control at work, Github, bank, university systems, personal computer, work computer, Facebook, Hacker News, Reddit, Steam, Skype, ... I could go on. In order to keep track of all of these unique passwords, I would need some kind of protected database like 1Password [0] (which I do have), or a notebook with them all written down. At this point, however, cracking/guessing a single password or obtaining a physical notebook has given someone access to all of my accounts.
What I typically do is give "important" services unique, random passwords (bank, personal email, work email; things that can do damage), and use one of a set of passwords for everything else. If one password gets found, then maybe only my Reddit, Skype and a few forum accounts are compromised. This has the advantage that I use the unique password frequently enough to memorize them, and non-unique passwords are used in a few places which results in them also being used enough that they are memorized.
[0] https://agilebits.com/onepassword
a good way to create password dictionaries :D
I'm not a password cracker, but from what I read, my test password of "superman15" could probably be cracked a fair bit faster than 10 days, as indicated by the site.
Most interesting feature of the site is the counting of years :). I got to 29 million nonagintillion years before I reached Infinity :).
Good site! I recently switched all my passwords over from about 3 varieties of a ~8 character string to ~30 characters that's unique (and easy to remember for each site.) I'm unsure why it's taken me so long to realise that I could have a super secure password like:
My name is bob smith from California USA and I am trying to login to: xxx
'abcdefghijklmnopqrstuvwxyz' will take "48 quintillion years"! Sweet!