Topic: The real testnet3 difficulty is equal to one (Read 56 times)

I know that. Testnet coins are worthless by definition. Monopoly money, in other words. However, I think NameCoin should also start with no additional coins in the first place. It was supposed to be about names. Names, that could be purchased for example with Bitcoins. And then, if you want some decentralized Domain Name System, where you can purchase domains with Bitcoins, then why you need more than 21 million coins for that purpose? The simplest answer is: you don't.

So, I think testnet coins are worthless, but the full Script is still supported, often as standard (there are transactions, which are non-standard even in testnet3, and can be included only by testnet3 miners, but it is hard to reach those cases, and often you can solve that problem with another type of transaction, that could still be standard, even on mainnet).

However, even if you start with some hypothetical test network, where the total supply would be zero, and where no coins could ever be created, then I guess you could land in the same situation, as testnet3 is going to reach here and now. Also, "real application usage" is only about monetary value. Your publicly shared tests are "real application usage", even if you use zero satoshis everywhere, or if you put some worthless test coins. It is still about publicly sharing data, and that is still allowed, even if you don't touch any coins, that are connected with some real world money.

Nice topic, will check that out.

Edit: I think this discussion should be continued where it started, so it is probably a good idea to close this topic, and redirect people here: https://bitcointalksearch.org/topic/interesting-behaivor-of-testnet-difficulty-5466740

Testnet should never be used as network base for real application usage:

Acording to https://en.bitcoin.it/wiki/Testnet

15 days ago that happened again the diffculty when to minimum:

Interesting behaivor of TESTNET difficulty

Maybe you wonder what is going on, when it comes to testnet3, and a lot of blocks every few seconds. Does it mean that suddenly, a lot of miners joined this testnet, and started mining? The short answer is: no.

In general, people noticed one important fact, when it comes to testnet3: after 20 minutes, the difficulty can drop to one. If you combine it with the fact, that during difficulty adjustments, it is possible to reset the global difficulty into the minimum, you may notice why attacking is possible. If you add "two hours rule", and a "time warp attack", you will get the full picture.

Also if you notice, that the only important thing is the chainwork, and a lot of weak blocks can always win in the long-term, over some smaller amount of strong blocks (for the same reason, people use shares in the first place), you will conclude it with the sentence from this topic: "the real testnet3 difficulty is equal to one".

Any higher difficulty is just a result of competition between miners, where they cannot generate blocks in a single milliseconds or faster. Because it is profitable to rise that bar a bit higher, and not be beaten by some small CPU miners, that could be lucky enough to put a block between some bigger players. In theory, big miners could mine easier blocks than you can see, but then they will enter the area, where network latency, and block propagation starts to matter.

So, what does it mean in the context of testing? So far so good. There will probably be some blockstorms, but people are used to that on testnet3. So, what will happen next?

1. We will run out of coins in the whole testnet3 network soon. That means, there will be more and more halvings, and some people will join the competition, just for the sake of mining the last satoshis.
2. The basic block reward will be zero, and will stay there forever. Of course, some people may try to get the block with the last satoshi, but sooner or later, the chain will be pushed forward. And with every chain reorganization, it will be harder and harder to get it, so that precious block will have so huge chainwork, that people will eventually give up, and honestly continue the chain (also because blocks after that will be full of transaction fees).
3. Testnet3 will be the first network, that will be based only on fees, and will be quite close to the mainnet. It will suddenly be the only network, with a more than decade history, that will enter the era of "no coins generated from the thin air". That means, some people will switch to signet, or to some other test networks, but others will stay, because of some unique properties of testnet3.
4. Test coins will be used only as a spam filter, exactly as intended. Because no coins does not mean no UTXOs, new coins will still be created, but they will be used only to transfer data (maybe also with other chains, and some other things like peg-ins and peg-outs may be tested as well, because the full Script support is still there, no matter how many coins will land on those UTXOs).
5. A lot of zero satoshi outputs will fly inside testnet3. Because it is the only amount, that can be easily obtained, and miners can agree to produce them, if they will have some room in their blocks. As long as testnet3 will not enter "UTXO flood era", it will be possible to get some zero coins, and test some scripts, there could exist some pages offering to process such transactions for free, for some Proof of Work, or in other ways (like: the contest for the most interesting Script).
6. Testnet3 will enter "UTXO flood era", and then, some proposals will be tested more seriously, before activating features on mainnet, that will allow faster Initial Blockchain Download, or faster synchronization of the chain between non-archival nodes. Compressing history will be more and more important, because of the flood.

Of course, more things will happen. I tested some of them on localhost, and some points above are just conclusions from our local games with garlonicon and others, but now, we will see them in production, so I wonder, how good are some of my predictions, and what will really happen in practice. What do you think?

I know this topic could land in some other board as well (and maybe it will be even moved, if the discussion will lead us there), but I am curious about technical aspects of testnet3 entering that phase, more than economical or social. What protocols will be tested there? Would a network with zero total supply lead us into the same place? Will testnet3 be a new NameCoin, but closer to Bitcoin, and without Merged Mining?