We know and understand that this release is very close to the first tournament but unfortunately, it had necessary fixes that were required for the Tournament.
Luckily, not much has changed in terms of the game play, it was mostly bug fixes.
The good news is that new submission on the Player Portal should now work!! We’ve also added a reference bot !
Now, this bot will by no means win anytime soon, however, it can give you some idea of how your own bot is doing or give you an intro if you don’t have a bot yet.
Good coding and good luck!
Enhancements
Reference bot added!
Bug fixes
Spamming commands no longer makes you cling to walls
Player placement fixed
Running matches on the Player Portal has been fixed
is using numOnLevel instead of bot.CurrentLevel meaning that instead of the current level, the number of bots on the current level is being used as the index. Is this intentional?
Hi @kobus-v-schoor, thanks for pointing that out, I’ve realised I made a mistake.
The intention is that it must get easier to advance to the next level when more bots have done so. So for example, the first bot to advance will need 100 collectables. The second bot will only need 80. Third 60 and fourth 20. Just so we don’t have bot still stuck on level 0 when the game is over.
But I see that will not be the case, so I’ll pop in a fix. We’ve also had a look at the PR and did decide to bring it in. It’ll be out soon.
Thanks @Izak for clarifying. If I’m understanding correctly, this will this also mean that a strong bot will have a much easier time winning once it gets ahead (if it’s alone (first) on a level, it will be able to advance very quickly to the next)? Or will this be limited to bots that are on lower levels only?
Also please note that I’ve seen that the issue I’ve mentioned in this post seems to still be present in the engine:
This means that a misbehaving bot (intentional or not) can potentially fill up the queue and prevent other bots from submitting a command
No, a strong bot will have the hardest time because each level will reset the required collectables. So if your bot is the first to advance on each level, it will need to collect the maximum number of collectables each time. But if your bot is the second to advance on each level, it will need only i.e., 80 collectables to do so each time.
The aim is to help bots who fall behind not to get stuck on a level unecessarily.
Thank you very much @Izak. Just want to ask about something else, would it be possible to add a flag to the state update to indicate that a bot’s command has been skipped because it was last in the queue? It’s technically possible to infer it but there are some edge cases where it’s not possible to reliably detect it.
We won’t be adding a flag like that for now, but you should be able to determine from your bot state whether a command was executed or not based on the bot’s movement state machine and location.