You are viewing a single comment's thread from:

RE: SPS Governance Proposal - Adjust Rewards Based on Card Level

in #splinterlands2 years ago

I read your post and I agree with you about bots. If they can beat chess and go they can beat Splinterlands (which is currently much simpler than chess or go). I have two comments on that:

  1. The plan is to make the game more and more involved with more player input over time. I know it's still a while out, but the item and spell cards that are part of the land expansion are a good example of that. That type of thing will increase the complexity for bots by many orders of magnitude and hopefully give human players a big advantage, at least for a while, by which time hopefully we can add even more things to the game. The advantage we have over chess or go is that chess and go don't change and add new features, but Splinterlands does (it's just been very slow, but will be getting faster).

  2. I'm not sure it's necessarily a bad thing if bots are ultimately better than human players. Many people will still just enjoy playing the game and they will be matched up against opponents at a similar skill level - especially as the game grows and there are more players at all levels. So at the highest levels people may use bots and compete on the best bot software (which is also kind of cool), and then at other levels human players can compete against other players (whether bot or human) at their skill level so they can enjoy the game and work on improving. So I'm not sure that it's necessarily bad or good - it's just different and has its own pros and cons.

Sort:  

Nice reply!

The first part sounds promising.

Concerning the second part:
Yes, actually, I am also fascinated by some chess programs, 'masterpieces' of software and (partly) AI.
And I also respect people who will write better and better Splinterlands bots in future (yes, that's indeed also "cool"). :)

A pure bot championship would be actually fascinating, too: who is able to write the best, most sophisticated Splinterlands bot? In chess there are competitions only for software ...

Especially in chess, bots are great tools to practice one's skills, analysing games and learning to understand chess better.
(Interesting also for Splinterlands could be the idea that some chess programs are having certain modes in which they intentionally don't always play the best possible move to make the games against human opponents more exciting and variable.)

However, one difference is that in chess (or GO) in (human) tournaments and public servers bots are strictly forbidden. That means human players are never forced to play against a bot if they don't want. They can intentionally challenge a chess program (if they really think they would be able to beat 'God') :-) or just practice with it, but in human tournaments they can be sure to face human opponents only.

Concerning Splinterlands that could mean, there could be different kinds of competitions in future, for example "bot only", "human only" and "mixed". Then at least everybody would have a choice. And I know very well that even if 'we' would try to implement these different kinds of tournaments it would be very difficult to verify that a 'human' is really a 'human'.

 2 years ago  Reveal Comment

Why should anybody play themselves, if bots do the same without wasting time and energy? In the end with this vision Splinterlands might become an inhabited ghost town where bots play against bots. A pure staking mechanism but no game anymore.

If the daily games are too many for a human, just make them less.
You also don't tell a soccer player like Lionel Messi to play 5 soccer games per day. :)
Create conditions within the game which humans can fulfil.

I wrote a longer post about it some time ago.

Yes, we are all investors (I am also one since 2018), but in my opinion long term every investor earns most if the game is developing well and attracts as many as possible satisfied new players.
And the big majority of human players prefers to play against other humans ...

On point 1 and 2
Would it make more sense to wait for land to implement then propose this change?
Depending on how far land is from implementation it could be factored in then.

On the first point, would it be a good idea to commit to something like 1 new ruleset per month? And maybe deprecate some older ones in favor of newer ones after a while to not make it super complicated for the human players after a while.

That should give bots a hard time keeping up. The question is, would we tolerate the rapid changes better, as humans?