How long would we have to wait to see a bitcoin address collision? a LONG time.

[americanpegasus]

Our planet has approximately 7 billion people on it.  Let's ignore the problems of population growth for the moment and assume we will always have 7 billion people.   
 
Now let's also assume that starting this very second, all those people (our global population of seven billion) each begins to click the button to generate a new bitcoin address every second. 
 
"WHY ARE YOU GENERATING SO MANY FUCKING ADDRESSES!  You're going to cause a bitcoin address collision!" yells the supreme bitcoin chancellor of Earth.  "STOP IT!"  ಠ_ಠ
 
Let's pretend everyone's in on the joke, and they all ignore him. 
    ( ͡° ͜ʖ ͡°)
  
 
With every human on Earth, all generating a new address every second it would still take about 6.5 billion years in order for there to be a 50/50 chance of a bitcoin address collision.  For there to be approximately a 99% chance of a collision, the assholes would need to keep up this buffonery for about 11.7 billion years total. 
 
Looking ahead to what scientists think will be the timeline of our planet: 
  
 

• 500–600 million:  Estimated time until a gamma ray burst, or massive, hyperenergetic supernova, occurs within 6,500 light-years of Earth; close enough for its rays to affect Earth's ozone layer and potentially trigger a mass extinction.
• 800 million:  Carbon dioxide levels fall to the point at which C4 photosynthesis is no longer possible.  Free oxygen and ozone disappear from the atmosphere. Multicellular life dies out.
• 1 billion: The Sun's luminosity has increased by 10 percent, causing Earth's surface temperatures to reach an average of ~320 K (47 °C, 116 °F). The atmosphere will become a "moist greenhouse", resulting in a runaway evaporation of the oceans.
• 1.3 billion: Eukaryotic life dies out due to carbon dioxide starvation. Only prokaryotes remain.
• 2.3 billion: The Earth's outer core freezes, if the inner core continues to grow at its current rate of 1 mm per year.  Without its liquid outer core, the Earth's magnetic field shuts down, and charged particles emanating from the Sun gradually deplete the atmosphere.
• 2.8 billion:  Earth's surface temperature, even at the poles, reaches an average of ~420 K (147 °C, 296 °F). At this point life, now reduced to unicellular colonies in isolated, scattered microenvironments such as high-altitude lakes or subsurface caves, will completely die out.
• 3.5 billion:  Surface conditions on Earth are comparable to those on Venus today.
• 5 billion: With the hydrogen supply exhausted at its core, the Sun leaves the main sequence and begins to evolve into a red giant.
• 7.59 billion: The Earth and Moon are very likely destroyed by falling into the Sun, just before the Sun reaches the tip of its red giant phase and its maximum radius of 256 times the present day value.  Before the final collision, the Moon possibly spirals below Earth's Roche limit, breaking into a ring of debris, most of which falls to the Earth's surface

 
  
(Source: Wikipedia, https://en.wikipedia.org/wiki/Timeline_of_the_far_future)  

 
Summary:  I'm pretty sure there will never be a bitcoin address collision.  My math might not be perfect, but it shouldn't be off by more than an order of magnitude.  You are welcome to fact check me and I'll correct it if necessary.

[anderson00673]

Never happen.  The world governments are spending their resources blowing each other up, not on space exploration.  Maybe if another alien species, like the body snatchers or something were to take over and keep the block chain going, then we might have a chance for a collision.

[jonald_fyookball]

I get slightly different numbers.  

24*365*3600*7*150*1,000,000*((1,000,000,000)^4)=

3.311 * 10^52,

whereas 2^160 = 1.46 * 10^48.

so instead of it taking 150 billion years, it would take more like 150 million years.  but who's counting?

[americanpegasus]

I get slightly different numbers.  

24*365*3600*7*150*1,000,000*((1,000,000,000)^4)=

3.311 * 10^52,

whereas 2^160 = 1.46 * 10^48.

so instead of it taking 150 billion years, it would take more like 150 million years.  but who's counting?

 
  
I took a second look and the math was pretty FUBAR the first time.  Thanks.  I limited them to only generating one new addresses a second to keep everything else working right.  

[futureofbitcoin]

No one ignores the supreme universal bitcoin chancellor! 

[americanpegasus]

No one ignores the supreme universal bitcoin chancellor!  

 
  
 
NOW YOU WILL FEEL THE WRATH OF DICK BLOCKCHENEY

[achow101]

Well if you are blockchain.info's android wallet with their screwed up RNG from a few months back, then you are going to generate a collision pretty much guaranteed for every single address generated. 

[Amph]

there is a way to know if a collision happened, i mean blockchain will tell you that? because otherwise the one who will generate it, because of heavy luck, will not say a word about it, and no one will know if a collision has occured or not...

[ajareselde]

In a normal circumstances it's safe to say that bitcoin address collision will never happen. The only way for it to be a reality is if in some
point in time some kind of vulnerability gets discovered, allowing address generation not to be as random as it is supposed to be.

cheers

[Kprawn]

When I start to think about this, I also consider this video ---> https://www.youtube.com/watch?v=ZloHVKk7DHk

He explains it so clearly, you could not even consider it as a risk factor.... The wonderful world of math and Bitcoin, makes this a mission impossible.

So on the OP's list, you have a better chance of dying from a catastrophic event, than loosing your Bitcoins. Now that is more scary...  

[chaoman]

you don't gamble very much do you? It's possible to have a collision in the next minute.

[ajareselde]

When I start to think about this, I also consider this video ---> https://www.youtube.com/watch?v=ZloHVKk7DHk

He explains it so clearly, you could not even consider it as a risk factor.... The wonderful world of math and Bitcoin, makes this a mission impossible.

So on the OP's list, you have a better chance of dying from a catastrophic event, than loosing your Bitcoins. Now that is more scary...  

You mean; "than to gain someone elses bitcoins.." but anyways, loosing bitcoins is way more probable due to uninformed people
getting hacked or scammed, than it is to loosing them from address collision, and this is where the attention should be pointed.

cheers

[Lauda]

you don't gamble very much do you? It's possible to have a collision in the next minute.

You don't like mathematics do you? There is a huge difference between possible and probable. While a collision is possible at the moment it is highly unlikely.
Even if we are going to see the first collision in 500 years, it doesn't matter. Nobody can honestly expect any money system to last that long. Even if Bitcoin becomes widely adopted, eventually it will be replaced by something better (just as Bitcoin is replacing current systems) or it will have to be extensively upgraded.

there is a way to know if a collision happened, i mean blockchain will tell you that? because otherwise the one who will generate it, because of heavy luck, will not say a word about it, and no one will know if a collision has occured or not...

Well if you get lucky you would notice an added balance to your wallet. Otherwise you can manually check the first time the addresses has been seen in the blockchain. I do not think that others can know if you've generated a existing address though. Only the person(s) involved would know at first.

[erre]

There is still a very small chance this could happen tomorrow. The same chances as getting annilhated from a black hole I think, but still not impossible.

[Amph]

there is a way to know if a collision happened, i mean blockchain will tell you that? because otherwise the one who will generate it, because of heavy luck, will not say a word about it, and no one will know if a collision has occured or not...

Well if you get lucky you would notice an added balance to your wallet. Otherwise you can manually check the first time the addresses has been seen in the blockchain. I do not think that others can know if you've generated a existing address though. Only the person(s) involved would know at first.

so there are chances(yeah very slim and all, i know), that it has happened already and no one know about it, if there aren't ways to know about it, maybe for a small amount, for a big one someone might have created a thread already...

[DannyHamilton]

- snip -
With every human on Earth, all generating a new address every second it would still take about 6.5 billion years in order for there to be a 50/50 chance of a bitcoin address collision.- snip -

I get slightly different numbers.  
- snip -

While an address collision is so unlikely that it can be considered, for all practical purposes, impossible...

An address collision is also meaningless unless there are actually bitcoins associated with the address.

Since there can't be more than *2,099,725,475,406,004 (approximately 2.1 X 10¹⁵) addresses at any moment in time that simultaneously have any unspent outputs associated with them, please recalculate your odds taking into consideration the chance of a meaningful collision occurring.


* See https://en.bitcoin.it/wiki/Controlled_supply  and   https://bitcointalk.org/index.php?topic=675321.msg7647669#msg7647669

[afriezalie]

It'll take a very long time even all of us generates a new address every one transaction because bitcoin address have very many combination. Maybe when we achieve it, bitcoin isn't popular anymore or maybe we have died.

[Lauda]

so there are chances(yeah very slim and all, i know), that it has happened already and no one know about it, if there aren't ways to know about it, maybe for a small amount, for a big one someone might have created a thread already...

Exactly. It is not something that is impossible, it is just very improbable (unlikely). I do not think that the system has a means of figuring out if someone generated a collision. So, unless someone comes out publicly stating that he/she generated a collision, we might never know.
Also, OP forgot to consider what Danny suggested. If a collision happens with a unused/dead/empty address then it is really pointless. The chance to generated a meaningful collision should be much smaller.

[AtheistAKASaneBrain]

What would happen if this actually happened? Like what would the consequences be of an address being generated as new, actually being one with a previous history and record? what would happen if you did a transaction? would the funds of said address suddendly pop on your wallet? weird stuff to think about, its almost like a glitch in the matrix.

[GODLIKE]

Our planet

Very interesting.
And I suppose this makes 99.9999999999999999% of transactions safe.

But... there's a butt.

I was thinking to post a thread regarding this exact matter MINUTES AGO.
If I didn't consult the forum right now, I would have posted it.
How many chances are those, that somebody opens a thread answering my question before I post it?