Slashdot Mirror
Physicist Uses Laser Light As Fast, True-Random Number Generator
MrKevvy writes
"An Ottawa physicist is using laser light to create truly random numbers much faster than other methods do, with obvious potential benefits to cryptography: 'Sussman's Ottawa lab uses a pulse of laser light that lasts a few trillionths of a second. His team shines it at a diamond. The light goes in and comes out again, but along the way, it changes. ... It is changed because it has interacted with quantum vacuum fluctuations, the microscopic flickering of the amount of energy in a point in space. ... What happens to the light is unknown — and unknowable. Sussman's lab can measure the pulses of laser light that emerge from this mysterious transformation, and the measurements are random in a way that nothing in our ordinary surroundings is. Those measurements are his random numbers.'"
Finally a reason for socially inept people to buy diamonds!
The world's smartest bug zapper www.zapstats.com/kickstarter
I don't believe such a thing can possibly exist.
9 9 9 9 9 9 9 9 9 ....
You don't KNOW it's not random...
Recycle PCs and build a wireless community network www.hillsborough.org.nz
There. I said it.
They lost me at "microscopic energy".
"Tell me doctor, with all of your defenses, are there any provisions for an attack by killer bees?"
A diamond bends light (a laser in this case) via reflection rather than quantum vacuum fluctuations? Maybe I'm missing something...
Anyways, I find hardware based cryptography much more scalable and attainable than tfa.
http://en.wikipedia.org/wiki/Hardware_random_number_generator
Also, I've never heard of a hacker trying to reverse engineer an encryption algorithm to break into a system, 0 day IIS exploits on the other hand...
I mean, what about a diamond in the middle attack? If you manage to replace it with well known and tweaked diamond, with known quantum effect (you see, i could use funny words too), then all the systems would be jeopardized.
Random enough for me. Truly.
For every benefit you receive a tax is levied. - Ralph Waldo Emerson
Light goes in, light comes out. You can't explain that.
Obligatory xkcd: http://xkcd.com/221/
I just use the rand() function in Excel. Way less hassle than firing a laser through a diamond...
The Old One Doesn't Play Dice
Since this is a random number generator deriving random number off of the uncertainty principal all I want to know is where can I get one.
"the measurements are random in a way that nothing in our ordinary surroundings is"
Nonsense. They are random in precisely the same way that a good bouncy roll of the dice are. They are random in precisely the same way that a temperature measurement of a cup full of boiling water 10 seconds after it is poured is. They are random in precisely the same way that the sound coming out of a piezoelectric microphone taped to a car window travelling at 60 MPH is. They are random in precisely the same way that the noise of a reverse-biased silicon junction is.
Perhaps the author meant to say "the measurements are random in a way that no pseudorandom number generator algorithm is."
This guy better be real careful with his analysis..... A lot of things you'd think are not correlated actually are. For instance, the intensity of starlight arriving at two separated telescopes has been found to be quite correlated!
I also wonder if he's pushing the principle of induction a bit far.... He may find the light is random, but what if a cosmic ray goes through the diamond and all the atoms go Boooonnnnngggg! in resonance together, won't that mess up his perfect randomness? I don't see much use for a random generator that might randomly spazz out into putting out patterns.
Finally a reason for socially inept people to buy diamonds!
I dunno about that. Diamond video cards were okay.
"I like to lick butts!" by MobileTatsu-NJG (#32700246) (Score:5, Informative)
Not me, but the post. Why is it in IT/Management when it should be in Science?
The newspaper article is not giving any information that is not already included in the summary.
The paper is published in Optics Express, the abstract can be read here. The full article is behind a paywall unfortunately. The author claim that this concept could deliver random numbers at a rate of 100 GHz which is quite fast compared to other true random number generators out there that are based on thermal noise, radiation or other processes.
"An Ottawa physicist is using laser light to create truly random numbers much faster than other methods do, with obvious potential benefits to cryptography"
Even faster, use neutrinos!
http://blogs.discovermagazine.com/crux/2011/11/22/faster-than-light-neutrinos-confirmed-in-one-way-yes-in-another-no/
Or? Maybe the answer is random? Truly random!
Do not look at random numbers with remaining eye.
Who would win this election: Andrew Weiner vs Andrew Weiner's weiner.
A while back, the Simtec Entropy Key was making the rounds among Debian Devs, and claims to be exploiting quantum effects in the P-N junctions to be a true RNG.
They seem serious and I tend to trust paranoid Debian developers' opinions, but ultimately I don't have enough knowledge myself to make a confident judgment call. I'd be curious about more opinions.
Misleading titles? Inflammatory blurbs? Keep in mind that Slashdot is a tabloid.
A lot of tools have diamond blades.
I agree that the numbers are random, in the sense that they're subject to chance, but how confident are they that they know the sampling distribution? That is, can you use this method to generate a random sample a with uniform distribution, or a gamma distribution, or anything else you'd like to use random numbers for?
With quantum observation errors, I wonder if they're assuming the sampling distribution is normal, in which case they'd have to do some work to convert it to give the kind of output that rand() gives. Problems would likely show up in the tails of the distribution (near 0 and 1). TFA doesn't mention any of the statistical issues, only the physics ones.
This reminds me of the researcher at SGI (Remember them?!) who used his Indigo's camera to take pictures of his lava lamp, from which he would hash random bits.
'Sussman's Ottawa lab uses a pulse of laser light that lasts a few trillionths of a second. His team shines it at a diamond. The light goes in and comes out again, but along the way, it changes. ... It is changed because it has interacted with quantum vacuum fluctuations, the microscopic flickering of the amount of energy in a point in space. ... What happens to the light is unknown — and unknowable.
Sounds very much like xray crystallography which discovers all kinds of interesting things about the crystalline matrix.
Would be hilarious if they discover via non-random results there is, after all, some inherent crystaline like order to the quantum vacuum. Or even funnier if they knew it all along, and some TLA agency paid them to try and pass it off as random, cloaked in a lot of new age zero point energy stuff.
"Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
Can it generate pi? Wow.
Shine on you random diamond.
Unlike easily predicted phenomena like radioactive decay and thermal noise?
Lacking <sarcasm> tags,
When it comes to true random devices, I've coded some micro-controllers to add random numbers based on key-presses from humans, picture someone pressing the button when a 24 mhz timer runs mad, no human that I know of - can repeat press the button so accurately that it hits the same number at a 0.00001th of a second more or less.
When no human interaction is required, I use an insanely accurate temperature sensor, no temperature, not even placed in a professional fridge with 0.01c accuracy can get the same results each time. Mix this with a spinning/running timer, and you've got yourself a winner! ;)
What this world is coming to - is for you and me to decide.
use Monte Carlo method
Hey don't blame me, IANAB
Oh yeah baby... back with my VLB Diamond Viper 4MB of VRAM... and a 486DX-2 66 with 16MB I was styling'. The chicks just couldn't stay away.
Cwm, fjord-bank glyphs vext quiz
The generation of random numbers is too important to be left to chance.
Ian Ameline
The Commodore 64 could produce random numbers by sampling the white noise generator in the SID audio chip. They probably weren't as random as shining a laser through the diamond but I wonder if the difference is enough to matter...
Is this really something that could be reduced in size to something that fits in a computer? The article didn't say what scale his equipment is - square kilometer or square inch?
Looking for a job in Portland, Oregon?
Big advantages of this is that it requires no outside information source, inexpensive and could be miniaturized to fit on an extension card. Then we all could put a random card next to our graphics card in our machines.
Finally a reason for socially inept people to buy diamonds!
Industrial grade diamonds are cheap. They are already found in various consumer gadgets that geeks may already have. :-)
"Matter is built on flaky foundations. Physicists have now confirmed that the apparently substantial stuff is actually no more than fluctuations in the quantum vacuum."
http://www.newscientist.com/article/dn16095-its-confirmed-matter-is-merely-vacuum-fluctuations.html
Everything is random.
"The Higgs field is also thought to make a small contribution, giving mass to individual quarks as well as to electrons and some other particles. The Higgs field creates mass out of the quantum vacuum too, in the form of virtual Higgs bosons. So if the LHC confirms that the Higgs exists, it will mean all reality is virtual."
And if it's all virtual who or what is running the simulation? Or maybe it's self generating "I am because I think I am"
"If any question why we died, Tell them because our fathers lied."
Now we need a cool name for it. How about:
Zero Point Entropy
"Nine times out of ten, starting a fire is not the best way to solve the problem." - my wife
Temperatures
Randon EM interference.
Throw it through your typical pools.
A TRNG in hardware wouldn't be that hard to add, or expensive. Just feed it all data from constantly varying sources like those above, even HDD speeds, fan speeds, bus inputs (USB especially!), timers, battery power, so many sources.
Better yet, if you wish to prevent cases of random continuous strings (such as that 999999 joke up there), you could add a very simple register that holds the last value, or couple values, and the current value cannot be the same or it rerolls.
Add as many registers as you want for less continuous sequences (up to whatever the size of the data is)
Of course, they are also useful, so don't eliminate too many of them.
"Truly random numbers". I don't believe such a thing can possibly exist.
It is far easier to believe when you are dealing with quantum physics rather than classical physics.
Oh yeah baby... back with my VLB Diamond Viper 4MB of VRAM... and a 486DX-2 66 with 16MB I was styling'. The chicks just couldn't stay away.
You must have been rolling in the dough back then to have 16MB ram. 8MB about broke my bank.
I just use the rand() function in Excel. Way less hassle than firing a laser through a diamond...
But not nearly as much fun. :-)
Thanks for posting about the Simtec Entropy Key. At only $56 (Qty 1) for a FIPS-140-2 Level 3 compliance type device based on quantum tunnels is pretty amazing. Just the buzz words, are worth that for any system advertised as secure.
"the measurements are random in a way that nothing in our ordinary surroundings is"
Nonsense. They are random in precisely the same way that a good bouncy roll of the dice are.
No. The bouncy dice are describable by classical physics. Our inability to predict is based upon our imprecise understanding of the path of the dice, their rotation, air density and movement, the geometry of the area landing in and bouncing about in, the understanding of the materials of the dice and objects it is bouncing against, etc.
In contrast this new method utilizes effects of quantum physics. That is inherently far less measurable and predictable.
Yeah I was a spoiled brat.
Cwm, fjord-bank glyphs vext quiz
What happens to the light is unknown — and unknowable.
Or was, at least until someone raised the idea of using it for purposes that will lead to incentivizing people to figure it out. Unlocking this mystery will presumably soon become a lucrative endeavor pursued by many. The question now is how long it will be until someone (or some supercomputer) starts to see patterns in the data produced.
Cheers to novel ideas that inadvertently advance scientific understanding.
What happens to the light is unknown — and unknowable.
It's knowable or else we couldn't measure it to generate random numbers.
Whether it's predictable is another matter entirely, and I'm almost positive that it isn't.
Finding God in a Dog
Dilbert did it first, and better.
http://dilbert.com/strips/comic/2001-10-25/
Measure the time between service calls from my mom about how to fix her Windows box. Random and short interval to boot!
Socially inept people are the only ones who ever had to buy diamonds.
Think about it.
You must have been rolling in the dough back then to have 16MB ram. 8MB about broke my bank.
Well, I had a T5200 portable which had 14MB RAM (its maximum) along with a 100MB disk. It had a 20MHz 386 which had the protected-mode bug which was only supposed to affect 16MHz chips (maybe Toshiba just overclocked it) and a 387 chip, too. It also only had the lousy orange plasma VGA display, because they didn't release the color VGA LCD until a year later. Damn thing was built like a tank, and survived repeatedly being accidentally dropped onto concrete - impossible to kill the thing. It cost a few thousand, but hardly broke the bank.
Those who can make you believe absurdities can make you commit atrocities. - Voltaire
diamonds are a geek's best friend
Those measurements are HIS random numbers. ;)
ps: IANAC
If there are infinite universes, then there is a universe where engineers build random number generators, and they always produce this sequence. The scientists struggle to come up with some theory as to why random number generators can't be built. It's not because they aren't random, it's just that they happen to produce this sequence whenever they're built in that universe.
For extra grins and giggles, there is a universe next door where the 9,9,9... random number generator on display at the science museum in the capitol of Earth mysteriously produces a 6 one day, and the world is thrown into a panic that it portends the apocalypse.
For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?
If bouncy things were not competitive as a source of random information, then someone would be able to predict lottery powerball drawings, which are indeed governed by bouncy balls.
This is my sig.
Attributed to Boltzmann, "There are many kinds of order, but only one kind of disorder"
Randomness would have the joint probabilities of every possible pair of outcomes over every possible pair of spacings to approach 1/2 exactly. Let's see that data set please.
I picture a physicist struggling for 5 years to get a consistent response shining a laser through a diamond when someone makes a flippant remark about a random number generator...
Randomness is often considered to be an unpredictable pattern, but what is and isnt predictable is dependent on knowledge, so its not an absolute truth.
Randomness is really just an unrecognized pattern, its subjective , what seems random to one person may not be to another.
e.g. Im sure there are people (maybe children) that consider the tides to be random, they dont have the knowledge and experience to see the pattern.
So, any sufficiently advanced pattern is indistinguishable from randomness.
"Any sufficiently advanced technology is indistinguishable from magic." - Arthur C. Clarke
You're both spoiled! I had 4Mb and only 2Mb on the card and thought I was kicking! Funny now when I have 8Gb on my netbook and the same on my desktop to think about how I spent more for a tiny couple of Mb of RAM back then than I did for my whole netbook now. I'll never forget though the first time I loaded up my Voodoo in my spanking new 133Mhz and loaded Unreal, I swear me and my friends just watched that opening demo for ages going "ooooh!". Man we were easily entertained back then.
as for TFA I never understood why getting a random number always seemed to be so hard, just take all the scores off the sports page, multiple by that last winning powerball numbers, then divide by how many pepperonis you got on your last pizza. Easy peasy!
ACs don't waste your time replying, your posts are never seen by me.
Anyone else think that this will eventually result in decrypting enthusiasts figuring out the laws governing such fluctuations and unlocking the secrets of the (observable) universe ?
Well they have a good reason to as this NSFW poster illustrates. Personally I don't see why DeBeer's don't just cut out the bullshit and make that the slogan. I'm sure many of us would appreciate the honesty.
ACs don't waste your time replying, your posts are never seen by me.
In classic mechanics, the universe is held to be deterministic, so nothing is "truly random." But in quantum mechanics there is uncertainty, so true randomness does exist.
P.S. The Heisenberg uncertainty principle is actually a mathematical theorem. So assuming you believe in mathematics, you should only reject quantum uncertainty if the underlying premise that energy exists in discrete quantities does such violence to your intuitions, that you cannot accept the mountains of empirical evidence in its favor. But if that is the case you are saying that energy exists at infinitely discrete increments. But this just reintroduces the idea that energy levels can be unknowable.
Democracy Now! - your daily, uncensored, corporate-free
I can put away that cup of really hot tea.
Steve's Computer Service, Hobbs, NM
Truly Random? that doesn't sound correct from what they are saying, more like impossible with current technology to predict as we can't actually measure the fluctuations. So this to me doesn't appear truly random at all.
OK, if we want to talk about fun hardware:
A DX2 66 486 with 64MB RAM, two 1GB SCSI disks and a CDROM with a 4MB VRAM card. But, that was just the desktop machine. It only clocked in at about $8K (work really sprung for that one) Then there was the decked out Indigo 2. Don't recall the RAM, but the MIPS 4400 upgrade was around 8K alone, and that was small potatoes compared to the $25K 256 layer Z-buffer video card that was added in. That's right - $25K for a video card that today is probably outpaced by pretty much anything you pull out of the recycling pile. And it was the low price of $25K because we bought 2 in a bundle with the upgrades. Originally they went for $38K.
Of course, all of those prices are totally blown away by the $8K 430MB WORM drive we purchased. To truly get how expensively stupid this purchase was, you have to understand how WORM drives operate. They basically had their own controller internally that worked with the internal hardware to position the write/read head as you progressed along the spiral. The problem was, there was no segmentation of the disk, no error correction, no guide tracks, or anything else. So, the entire process was based on the head placement mechanism being in the right place at the right point of the spin to write/read the data. The problem was, these parts would wear, so a disk was good across about 250-400 read-write cycles of the drive. Read that again - the drive could only be used less than 250 times reliably between the writing of a disk and the current reading. After 250, it got dicey, after 400, you could no longer read it. Oh, and just to compare it to today's BD disks, a WORM disk at the time sold for roughly $100 a piece in lots of 100.
The cesspool just got a check and balance.
Off the top of my head, I think this would work as a definition of what's expected of "random": the limit of the proportion of the number of a number's occurrence to the total number of samples taken as the number of samples taken approaches infinity for all numbers in the sample space are equal, and there does not exist a dominant strategy for guessing the next sampled number. Basically, each number has an equal chance of appearing, and there is no optimal way to predict any one number, which is all anyone would ask of a fair RNG.
FYI, that photo was inspired by (but definitely not taken from!) an old Family Guy episode.
I feel fantastic, and I'm still alive.
you just HAD to be "that" guy didn't you....
Sharks with lasers... and diamonds!
Damn, 4 Mbit is tiny. I feel sorry for you,
Also, I hate to break it to you, but 1 GB on a notebook is nothing these days.
Its good for a one time pad.
our ordinary surroundings is." - even more random than HP's business strategies?
A lot of tools have diamond blades.
Huh? +4 informative? I've been absent from /. for a good 5 years, but in my day that comment would have looked like
A lot of tools have diamond blades.
1/ are they truly random numbers, that is do they pass all the statistic tests of distribution? If I remember correctly quantum phenomenon are unpredictable, but are they truly random in the distributive sense, or do they have some numbers that are more likely to appear than others? It would seem that even if quantum phenomena are truly random, nonetheless the physical structure of the diamond would give a bias to one set of phenomena than another.
2/ This kind of randomness is actually pretty useless for cryptography, especially as these randoms will be impossible to verify or debug. One of the most important properties of the random function is that it gives a predictable series so that you can start your program knowing in some degree the values. To generate more 'randomness' you change the seed to the function, not the function itself.
I remember once that a certain company thought about using the microphone as a random number generator, assuming that the noise about the computer was 'random'. What they forgot is that most parts of the computer make a very cyclic noise (for example the fan), which led to some very bad security holes. What this experiment is doing could be nothing more than a certain optical noise that might not be in fact be statistically random.
unknowable
Phhhsh, physicians and your imprecise science.
I believe you guys are throwing the term "unknowable" a bit lightly. I'm pretty sure its knowable, and even predictable, just very very VERY complicated. But not infinitely complicated. Therefore I demand you provide a rigorous demonstration that the process of it is unknowable(or unpredictable, exclude the semantic piss)
I remember very clearly a theorem during college that was proven beyond doubt is demonstrable, but was never demonstrated yet. Mathematics FTW!
that's brilliant !
-- no sig today
A, there you are sergeant pedantic...
ics
that's
This is exactly the effect driving /. .
Why are these things always advertised as having huge benefits? There is absolutely no need for new true random generators, the ones we have are completely sufficient. And they are cheap, unlike this discovery that would also take at least a decade to become usable in practice and very likely be far, far too expensive. What the practical problem is at this time (and has been for some time now) is that people do not understand how to use entropy generators right.
I am sick and tired of these stupid stories.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
So how many microseconds until the complete works of Shakespeare show up?
If Pandora's box is destined to be opened, *I* want to be the one to open it.
Fo many simulations , you required large number of random numbers, hence this research
"An Ottawa physicist is using laser light to create truly random numbers much faster than other methods do"
Of course it didn't break the bank. If you wanted to do that you should have thrown it through the window instead of dropping it onto concrete.
I think you meant Brillant, Paula
Damn that site used to be funny (read: it's not anymore)
My Dad bought a NeXT cube when I was a teenager.
At some point, he decided to get a RAM upgrade. He had the choice between 32MB and 64MB, and he went with the 64MB, but it wasn't much of a choice, since they *cost the same amount*. For some reason, the upgrade cost was per SIMM, irrespective of the actual amount of RAM on it.
As I recall, it was several thousand dollars. But that was hardly relevant for a computer that cost $15000 out of the box (CDN).
To this day, it was one of the finest, most responsive machines I've ever used (and is responsible for my switch to OS X). It could do so much on 25MHz; it makes me sad about how little I can do with 2.8GHz. :/
It cost a few thousand, but hardly broke the bank.
Even today "a few thousand" for a computer is a hell of a lot of money for most people, as you're well into the secondhand car league there. Twenty years ago you must have been able to buy a new car for that.
To have a right to do a thing is not at all the same as to be right in doing it
Just connect it with a gun trigger, point to your head and see what the world is made of!
...until a pixelated circle of ones in a field of zeroes shows up. In which case, you should not open snail mail letters -- they're likely to say your Dad is not your Dad.
What happens to the light is currently unknown — and currently unknowable under present theories.
You've managed to combine link-speak with xkcd-speak, both of which I hate. Now get off my lawn!
Oh great, so now my random numbers will be propping up conflict in Africa. Don't buy blood crypto! :-)
Ah, NeXT ... the machine that gave my my first access to a UNIX command-line.
Didn't get to play on the console much, but a friendly Jesuit in the physics department allowed me access to his research machine so I could see UNIX since the comp-sci department didn't have anything at all.
Fond memories of that machine ... by the time I had a Linux box I was pretty good to go with bash and a couple of other things.
Lost at C:>. Found at C.
Even today "a few thousand" for a computer is a hell of a lot of money for most people, as you're well into the secondhand car league there. Twenty years ago you must have been able to buy a new car for that.
A new car??? Not even close in Finland. At that time, cars had in excess of 100% taxes, leading people to keep them running a long time. So used cars were not much cheaper than new ones, and the price curve was very long and declined rather slowly. In the early 1990s, even a clunker which was already 20 years old would have cost rather more than a brand new T5200. Don't assume that car prices in all countries are similar to those in the US, especially used car prices.
FYI, my current car has more than 320000km on the clock, and is in very good condition; I hope to keep it well past 500000km. BTW, car taxes are slightly less than 100% now, part of it being a large "registration fee" component to avoid EU issues. There is, of course, an additional 23% VAT on the car and on the registration fee.
Those who can make you believe absurdities can make you commit atrocities. - Voltaire
...I thought RNG quest spawns were bad enough as it is. I can hardly wait 'til Blizz adopts this for WoW.
Actually, radioactive decay is totally random (that's why Schrödinger used it in the famous cat-in-a-box gedankenexperiment. The half-life just tells you the overall rate, but that no more help than knowing how often you flip a coin -- no help with heads or tails.
Lacking <sarcasm> tags,