Loading...
  OR  Zero-K Name:    Password:   

Post edit history

ZK Developer Support of AIs

To display differences between versions, select one or more edits in the list using checkboxes and click "diff selected"
Post edit history
Date Editor Before After
1/25/2017 10:05:52 PMUSrankaeonios before revert after revert
Before After
1 [quote]Realistic, but I wish to add something: 1 [quote]Realistic, but I wish to add something:
2 \n 2 \n
3 Have a Human tourney running at the same time so those who couldn't build bots, or want to fight humans get a chance. 3 Have a Human tourney running at the same time so those who couldn't build bots, or want to fight humans get a chance.
4 \n 4 \n
5 After the Human and AI victor is chosen, the two will duel each other in maps chosen by PRNG, stable only ofc. 5 After the Human and AI victor is chosen, the two will duel each other in maps chosen by PRNG, stable only ofc.
6 \n 6 \n
7 I would like to see this happen. It would be a great goal for many players, go in human tourney and get to top, then fight your AI oppponent, or build an AI strong enough to takedown a human[/quote] 7 I would like to see this happen. It would be a great goal for many players, go in human tourney and get to top, then fight your AI oppponent, or build an AI strong enough to takedown a human[/quote]
8 \n 8 \n
9 Apparently you've missed all the games where various people beat 2x circuit AIs on CCR (as well as some close attempts at 1v3). Building an AI that's strong enough to beat an advanced human player is extremely difficult, let alone an AI strong enough to beat a tournament winner. Even ZKGBAI can't do that in 1v1, and it's by far the strongest 1v1 AI that's yet been developed. 9 Apparently you've missed all the games where various people beat 2x circuit AIs on CCR (as well as some close attempts at 1v3). Building an AI that's strong enough to beat an advanced human player is extremely difficult, let alone an AI strong enough to beat a tournament winner. Even ZKGBAI can't do that in 1v1, and it's by far the strongest 1v1 AI that's yet been developed.
10 \n 10 \n
11 [quote]Creating a new AI that builds units and sends them to the enemy is something most beginner programmers should be able to achieve in a few hours. What makes it easy is that you don't need to familiarize yourself with any existing code(!!), and you only have to learn a few callouts/callins which are either generated or can be looked up easily in any modern Java IDE.[/quote] 11 [quote]Creating a new AI that builds units and sends them to the enemy is something most beginner programmers should be able to achieve in a few hours. What makes it easy is that you don't need to familiarize yourself with any existing code(!!), and you only have to learn a few callouts/callins which are either generated or can be looked up easily in any modern Java IDE.[/quote]
12 \n 12 \n
13 I think you're seriously underestimating the task. Even coming up with a proper scouting algorithm is quite difficult, not to mention other necessary things like parsing metal spots from GRP, choosing a good start pos and other such things. It took me a couple of days just to understand KGB enough to make any real progress on it, and that was back when it was a wee AI that hardly did more than what you've described. Building an AI that's actually competitive with circuit or KGB as they stand now is a pretty epic undertaking. 13 I think you're seriously underestimating the task. Even coming up with a proper scouting algorithm is quite difficult, not to mention other necessary things like parsing metal spots from GRP, choosing a good start pos and so on. It took me a couple of days just to understand KGB enough to make any real progress on it, and that was back when it was a wee AI that hardly did more than what you've described. Building an AI that's actually competitive with circuit or KGB as they stand now is a pretty epic undertaking.
14 \n 14 \n
15 [quote]The hardest (and most time consuming/unfun) part is probably to compile Spring with the .jars you need. I'd like to see that requirement taken out of the picture. It's certainly not something you want to present as a barrier to newbie devs.[/quote] 15 [quote]The hardest (and most time consuming/unfun) part is probably to compile Spring with the .jars you need. I'd like to see that requirement taken out of the picture. It's certainly not something you want to present as a barrier to newbie devs.[/quote]
16 \n 16 \n
17 You can copy the necessary libraries from NullOOJavaAI which ships with all precompiled engine versions. :P 17 You can copy the necessary libraries from NullOOJavaAI which ships with all precompiled engine versions. :P