3 ms·

I can't believe you're defending "an incorrect assumption" regarding your previous generator, especially after you wrote a very good article showing how badly w

I can't believe you're defending "an incorrect assumption" regarding your previous generator, especially after you wrote a very good article showing how badly wrong you were. Wait, let me restate that. You're Engineering the Disruption of Real Money Gaming and yet you say of some PRNG "no reason the ... implementation ... should not be good enough to not have to worry about"? You're not just wrong, in fact you're "not even wrong", that's how wrong you are. Plus that sentence has three negatives! Jules Winnfield said it better than I could. Riffing from him, what you were doing was, compared to what you should have been doing: ain't the same fuckin' ballpark, it ain't the same league, it ain't even the same fuckin' sport. Here's something I came up with in 60 seconds. It's far better than what you didn't want to worry about: ID Quantique hardware RNG [1] mixed with Intel RdRand [2] Yes, you're right. That's not seed-able, so you can't create a reproducible sequence of values (e.g., for testing). That's an advantage. You want to test, you replace hardware with software. But if you want to disrupt real money gaming you don't use some crappy software in some random library that wasn't designed for the purpose. Once again, I can't believe your defense of how badly you failed was the "reputation" that Google has. Real money is involved and you thought it was OK to rely on some crappy PRNG created by Google? For something that is at the heart of, the quintessence of what you are attempting to accomplish? EPIC FAIL!!! [1] http://www.idquantique.com/random-number-generation/ http://www.idquantique.com/random-number-generation/ [2] https://en.wikipedia.org/wiki/RdRand https://en.wikipedia.org/wiki/RdRand

PhantomGremlin 11y ago
Oh, yeah, downvote the messenger instead of respecting the message, which is that a third-rate amateur approach to random number generation just doesn't cut it. This was known as far back as 1999, when similar flaws bedeviled online poker sites.[4][5] Anyone going into the business should have known that from the start, instead of stumbling into it. And definitely shouldn't be trying to retroactively justify such gross incompetence. It's not Google's fault that these guys designed a crappy PRNG into a real money application. There is always the potential for cheating when real money is involved: 1980 Pennsylvania Lottery scandal[1] people added weight to supposedly random ping pong balls scratch lottery tickets weren't random[2] a "trick worked about 90 percent of the time" American Coin Machine Company[3] video poker machines were programmed to cheat players out of jackpots people were cheating online poker as far back as 1999, due to a flawed PRNG[4][5] Those are just a few cases that I could recall. I'm sure there have been many many more. True randomness is just the start in what it takes to prevent this. People who want to "disrupt real money gaming" shouldn't be making amateur mistakes like this. [1] https://en.wikipedia.org/wiki/1980_Pennsylvania_Lottery_scandal https://en.wikipedia.org/wiki/1980_Pennsylvania_Lottery_scan... [2] http://www.wired.com/2011/01/ff_lottery/ http://www.wired.com/2011/01/ff_lottery/ [3] https://www.google.com/search?q=American+Coin+machine+company&ie=utf-8&oe=utf-8 https://www.google.com/search?q=American+Coin+machine+compan... [4] http://www.nytimes.com/1999/09/13/business/compressed-data-e-gambling-software-gave-players-an-edge.html http://www.nytimes.com/1999/09/13/business/compressed-data-e... [5] https://www.cigital.com/papers/download/developer_gambling.php https://www.cigital.com/papers/download/developer_gambling.p...
- mmalone 11y ago
  For the record, we are not and at no point have we ever used the Node.js PRNG to generate random numbers for our gambling games. That is an entirely separate, audited, and carefully controlled system. This code was unrelated to gambling. In fact, it was not even consumer facing.
dang 11y ago
Your comments here are breaking the HN guidelines. HN comments need to be civil and substantive. We detached this subthread from https://news.ycombinator.com/item?id=10601984 https://news.ycombinator.com/item?id=10601984 and marked it off-topic.
- PhantomGremlin 11y ago
  Sorry. Not a badge of honor to have a subthread detached.

PhantomGremlin 11y ago

Oh, yeah, downvote the messenger instead of respecting the message, which is that a third-rate amateur approach to random number generation just doesn't cut it. This was known as far back as 1999, when similar flaws bedeviled online poker sites.[4][5] Anyone going into the business should have known that from the start, instead of stumbling into it. And definitely shouldn't be trying to retroactively justify such gross incompetence. It's not Google's fault that these guys designed a crappy PRNG into a real money application. There is always the potential for cheating when real money is involved: 1980 Pennsylvania Lottery scandal[1] people added weight to supposedly random ping pong balls scratch lottery tickets weren't random[2] a "trick worked about 90 percent of the time" American Coin Machine Company[3] video poker machines were programmed to cheat players out of jackpots people were cheating online poker as far back as 1999, due to a flawed PRNG[4][5] Those are just a few cases that I could recall. I'm sure there have been many many more. True randomness is just the start in what it takes to prevent this. People who want to "disrupt real money gaming" shouldn't be making amateur mistakes like this. [1] https://en.wikipedia.org/wiki/1980_Pennsylvania_Lottery_scandal https://en.wikipedia.org/wiki/1980_Pennsylvania_Lottery_scan... [2] http://www.wired.com/2011/01/ff_lottery/ http://www.wired.com/2011/01/ff_lottery/ [3] https://www.google.com/search?q=American+Coin+machine+company&ie=utf-8&oe=utf-8 https://www.google.com/search?q=American+Coin+machine+compan... [4] http://www.nytimes.com/1999/09/13/business/compressed-data-e-gambling-software-gave-players-an-edge.html http://www.nytimes.com/1999/09/13/business/compressed-data-e... [5] https://www.cigital.com/papers/download/developer_gambling.php https://www.cigital.com/papers/download/developer_gambling.p...

mmalone 11y ago
For the record, we are not and at no point have we ever used the Node.js PRNG to generate random numbers for our gambling games. That is an entirely separate, audited, and carefully controlled system. This code was unrelated to gambling. In fact, it was not even consumer facing.

dang 11y ago

Your comments here are breaking the HN guidelines. HN comments need to be civil and substantive. We detached this subthread from https://news.ycombinator.com/item?id=10601984 https://news.ycombinator.com/item?id=10601984 and marked it off-topic.

PhantomGremlin 11y ago
Sorry. Not a badge of honor to have a subthread detached.