Bitcoin Forum

Bitcoin => Pools => Topic started by: zatoichi on December 30, 2013, 07:54:00 PM



Title: mining pool nonce range assignment?
Post by: zatoichi on December 30, 2013, 07:54:00 PM
I've read a little about pools but nothing brought up this area. Do any (or all) pools assign nonce ranges to avoid wasteful duplication?

My goodness. Zero replies. I'm assuming this also means zero pools using this technique. This seems incredibly wasteful. How wasteful is it? I'm glad you asked.

To demonstrate the wastefulness I will have to use my Bitcoin system simulator. It's cute. I downloaded it as an app. The software is credited to aliens beings beyond the edge of the universe. My favorite setting is a mining pool of 2^32-1 computers with random nonce generation. You'd think that many computers would hit the solution on the first cycle. Unfortunately the probability is not one, it is just close to one. How close? I'm glad you asked.

The major factor is the random number generator. The alien simulator comes with 2 option. The first being the random number generator designed by the aliens and tested over 14 millennia. The second is the human designed one compromised by the NSA. Anyway were going to set the simulator to use the alien version. Okay we're ready to go. I just press the "Mine" button. There.

It takes a few seconds to run a simulation of 2^32-1 computers. It runs faster on my Samsung smartphone but my iphone is so cool I always use it instead. The app network activity indicator always comes on. I think it connects to this computer on Forbidden Planet for processing help. Sometimes the app makes these really weird noises. Somewhere is the documentation it says the aliens can add a Bitcoin to their wallet just by thinking about it. Also noted is the reason they went beyond the edge of the universe. Apparently there were some bad vibes starting to come out of the Sol system. They theorized some kind of whilrpool after the last Bitcoin was mined. With that bad news they fueled up their ships (with the bad news) and headed in the other direction. Their ships actually are fueled by bad news since nothing travels faster1. Anyway it's finished.

Before I reveal whether the solution was found on the first cycle, I'd like to ask what you think the chances are finding the solution on the second cycle if the first cycle is a dud. You guessed it. The probability is not one but very close. In fact, there is a very, very slight chance that the solution might not be found in your lifetime, using random nonce assignment. That probability is close to zero but is not equal to zero. So you never know.

Wow, we were lucky. We got it on the first try. Let's check out the resource utilization feature. Cool, we would have blacked out Europe and half the mid-east. Hey, 2^32-1 computers suck up a lot of energy. Neat! I just noticed because of that huge power surge, we also ended up with a Higgs bosun in our wallet. Unfortunately it disappeared almost instantly. I wonder if Bitcoins have a half-life. There are definitely some entropy concerns. Just for fun let's see what the second cycle would have done on top of that. Whoa, back to the industrial revolution. Gotta check the third cycle. Damn, stone ages. Don't want to go there.

So with these obvious big risks, why use random nonce assignment? The alternative guarantees first cycle solutions (if you got 2^32-1 computers lying around). You just assign nonces incrementally until exhausted at 2^32-1. Crank out one cycle and you're done.

Anyway happy mining and avoid those random nonce pools. Don't forget to download that app and you can follow the aliens on Twitter @monstersfromtheid

1 Bad news theory of space travel attributed to D. Adams (2nd Guide)

Wow! How many posts here have footnotes? Or humor for that matter. (Bitcoin: the Universe of Grinding Illogic)[pronounced "buggy"]


Title: Re: mining pool nonce range assignment?
Post by: -ck on December 31, 2013, 02:39:00 AM
No pool repeats work.


Title: Re: mining pool nonce range assignment?
Post by: eleuthria on December 31, 2013, 03:19:12 AM
This is not obvious from cursory research. Do you have a link to a description of how this is accomplished? It would be greatly appreciated.

Pools change a value in the coinbase message for each worker/piece of work they assign.  Unless the pool wrote their own software and did it HORRIBLY wrong, there is absolutely no way that any two workers will be hashing the same values.


Title: Re: mining pool nonce range assignment?
Post by: zatoichi on December 31, 2013, 03:32:55 AM
Thanks. I actually tried a new search "bitcoin mining pool repeat work" and got some good information. Sometimes you just have to know the terminology. Whooda thunk it.