3 ms·
High-entropy randomness generator using calibrated soundcard
- acqq 16y agoI believe it results in high quality bits but I understand it doesn't get enough attention because it's not sexy like "I'm using the radioactivity" and doesn't appeal to lazy -- the equipment must be calibrated. Still it's worth reading, at least this part should matter even to people who wouldn't like or know ow to do the calibration: '“Whitening” Considered Unhelpful'
- dazzawazza 16y agoPeople who require good quality random numbers (like gambling outfits) like off the shelf natural random number generators because a) you can do due diligence to see if the company meets its claims * b) you don't have to become and expert in measuring randomness c) sexiness is good when people will constantly claim your system and RNG are fixed biasing the house * * . Being able to point to a well known and verified company that generates real random numbers helps. Having said that, having worked for a poker startup I did find it fascinating. * doesn't mean your RNG is random but it's more likely to be random them a homespun solution, unless you are an RNG expert. * * of course the RNG isn't fixed biasing the house.... the games themselves are fixed but people fail to see that. In poker the house has no real reason to fix the RNG as they rake money faster than you can print it.
- losethos 16y agoGod says... twisted Legal second refuse leasing thirsteth steer indulgent congregations drunkards hoar decrepit Accept ended trees demonstrated neglect
- bhickey 16y agoThis is proving to be an interesting read, but I have much different design goals in mind than the author. I'm interested in simulation, and this sort of RNG is useless for that purpose. I would go as far to say this approach is useless for any scientific purpose. With great state comes great reproducibility. (I couldn't resist!) In the abstract the author suggests that PRNGs rely on "on unprovable assumptions about “one-way functions”" Any PRNG worth using should have solid number theoretic footing. For example, Multiply with Carry generators are based on the decimal sequence produced by dividing integer by a very large prime.
- vessenes 16y agoReading this, I'm curious how much could be automated with software. There's quite a lot of 'plug in this, measure that voltage, manually reset this' in the docs. I compare this to just subscribing to quantum entropy over the internet and conclude you'd need a lot of private random bits before this approach makes sense in terms of time. I found really fascinating the comments and complaints about dumping bits through SHA-1 (which they do) and then testing randomness of the SHA-1 bitstream. In short, they think it's stupid to test randomness using just heuristics on SHA-1. Presumably because SHA-1 is actually pretty good already, and so it obscures some of the upstream quality.
- acqq 16y agoThe whole point of determining the lower bound is measuring the sensitivity of the input of the sound card. If you don't know the exact sensitivity of your device you have to measure it. The author evaluates the quality of the different sound cards inputs as a byproduct: note that he for example measures: "(Creative Labs) Extigy has astonishingly poor input sensitivity. It’s about 5 times worse than a typical cheap sound card, and about 100 times worse than a decent 24-bit soundcard, as you can see from table 4." The sensitivity is the first column of his table, demonstrating some products being unusable for this purpose and obviously unusable for any quality recording too. It would certainly be good to have the database of the measurements for different products. It's amazing how little we know about the consumer products being sold unless we measure them.