Bitcoin Forum
May 25, 2024, 10:14:24 PM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: [1]
  Print  
Author Topic: We count time in units of sync from everyone to everyone - relativity compatible  (Read 359 times)
BenRayfield (OP)
Sr. Member
****
Offline Offline

Activity: 316
Merit: 250



View Profile
April 12, 2015, 07:21:06 PM
 #1

We can accept that everone's clock may run different speeds, especially if your clock is broken or your computer is overclocked which means its motherboard clock wave vibrates more times per second so you compute more than others around you.

The time used to predict movements of planets is very close in sync to the kind of time we use because they are subject to the same force of trying to sync with eachother else pay a higher energy cost of vibrating out of sync. Through the Internet, we also pay a higher cost of complexity, repeating the same calculations in many contexts, if we are out of sync with computers near us, and we save even more computing resources the greater the sync reaches the the rest of the world.

Bitcoin syncs on 10 minute intervals, when a new root node of each blockchain is agreed on. To change it after that is a political process among those who hold value in that network, and it has been done before when Bitcoin's algorithms were agreed by many to not be the best algorithms when multiple blockchains branched and they had to decide which to follow, or they could have split their held value in half more or less each only spendable on different sides of that divide.

There have also been examples in the push toward closer connections to stock trading systems, but that is a goal of putting your systems out of sync ahead of others while others would probably still choose to sync with those stock systems when they can.

Lack of sync in our calculations has huge effect on the world. Sync is more than just counting time.

Which kind of time do you mean? Relativity counts time by each Sender and Receiver in the network, how fast that path is received. Relativity does not count how fast all parts of the network can each send the smallest piece of unique information to everyone in the network and receive all those from everyone. When that is done, we count 1 clock tick of network sync.

Of course our most extreme members' speeds and distances affect how fast we can global sync, and they will slow down all our calculations if we wait on them. Do we want to wait until Australia and USA can each send a bit to eachother and the other receive it, before our sound card calculates the next wave amplitude? It wouldnt be fast enough to hear, even at lightspeed, because light crosses Earth 23 times per second, so at most we could (by the nyquist frequency saying you can do it on the hills and valleys of each wave) play 11.5 hz, and you cant hear much below 20 hz, not to mention that the computers internals do billions of cycles (hills/valleys in the motherboard wave) per second. So we dont wait for light to reach us to continue computing. It would be impractical even for the speed of a pocket calculator. You might even calculate too fast using an ancient mechanical computer, if its parts can move more than 23 times per second.

So instead of global sync, I think we should formalize a definition of time, in the context of syncs of information flow, which defines global sync as lagging behind each smaller and faster piece of the network which agrees to local sync, all the way down to computers as they exist today.

Why does Kinect and many phone cameras have 1/5 second lag between you moving and the game reacting? Because people dont care about sync. They should, because playing games that way sucks. Among gamers, lag is a cussword. Go lag yourself. You are lagging our game. No laggers allowed, you see before people form into groups to play such games. Its also important for scientific networks. If we are to get the full benefit of the Internet, we must define what relation between speed and sync we want in each part, what depends on what groups of other things all finishing, or how many of them is enough to proceed? Then what depends on us and is waiting? Sound familiar? The world of business is plagued by sync problems too.

Instead of asking how many milliseconds have passed since 1970, wouldnt it be better for each syncgroup to keep its own integer for how many cycles have passed? For example, each Bitcoin network as a syncgroup counts time as the length of its blockchain, 1 more every 10 minutes more or less. Milliseconds since 1970 is a syncgroup of certain physical clocks across sealevel which are adjusted toward eachother when they slightly differ.

What do you all think of this definition of time by global sync of smallest amount of information being able to (if that path is desired) move from everywhere to everywhere?

CIYAM
Legendary
*
Offline Offline

Activity: 1890
Merit: 1078


Ian Knowles - CIYAM Lead Developer


View Profile WWW
April 12, 2015, 07:28:42 PM
 #2

Has been a long time since I've seen one of your posts.

Always entertaining but not quite sure exactly what your idea is (as with most of your posts).

Some sort of improvement for block timestamps?

With CIYAM anyone can create 100% generated C++ web applications in literally minutes.

GPG Public Key | 1ciyam3htJit1feGa26p2wQ4aw6KFTejU
person
Sr. Member
****
Offline Offline

Activity: 315
Merit: 250



View Profile WWW
April 13, 2015, 01:37:44 AM
 #3

What do you all think of this definition of time by global sync of smallest amount of information being able to (if that path is desired) move from everywhere to everywhere?

Had to read 3 times before thinking that OP is proposing - I believe - to use global latency as an unit of time (Huh)

So instead of global sync, I think we should formalize a definition of time, in the context of syncs of information flow, which defines global sync as lagging behind each smaller and faster piece of the network which agrees to local sync, all the way down to computers as they exist today.

A key aspect of an agreeable definition of time by multiple parts is the ability to independently verify it.
If the definition of time derives from the same event you are trying "to time", it turns into a mere "event counter" and not anything usable in a scenarion you need to sync.

Either that or I still have no idea what you are proposing... Smiley
Pages: [1]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!