Topic: News about AI
WorldSavior |
Posted at: 2017-05-23, 18:45
Are you sure that you build enough buildings? Since especially the military buildings of the barbarians really need tons of blackwood, I cannot imagine how one wood hardener could be enough. I rather made the experience that two wood hardeners are not enough
I cannot find out what that is. fernery=reedyard? I think that an early reedyard is only important if you want to build ships early, or if you want to build a lot of ships in a short time, or if you want to build an insanely huge economy in a rather short time Wanted to save the world, then I got widetracked Top Quote |
Nordfriese |
Posted at: 2017-05-23, 19:09
I nearly always beat the AI, so yes, I do think I build lots of military buildings. But I usually play on maps with few trees and infertile ground, so 1 wood hardener matches my log income. But I never run out of blackwood (except for about the first 20 minutes, in which I need especially much for a quick start). I usually build a warehouse when my territory becomes quite large, and I build 1 wood hardener next to every HQ, warehouse and port.
Yes, I still call the reed yard "fernery". That is what it was called when I started playing Widelands. And I build it for the last reason – I like to start by building a very strong economy, so I like to always have some thatch reed in reserve. The fernery isn´t that expensive, so I can afford it. The first buildings I build are ALWAYS wood hardener – lime kiln – fernery – 2 lumberjacks – 5 rangers – quarry – well. By the way, why did you suggest 2 hunters and gamekeepers? 1 each is enough in the beginning for me. Top Quote |
Tibor Topic Opener |
Posted at: 2017-05-23, 20:00
I think this should be tested. Pick about three distinct maps, test my lists, some alternative lists, for all three teams and see how they perform. Otherwise it is just guesswork Top Quote |
WorldSavior |
Posted at: 2017-05-23, 21:10
Well, the AI is not a good benchmark, because it's so pathetic
And running out of it slows you down. That should be avoided
Wow, so sometimes you have a giant number of wood hardeners? Could also be possible to build less
That's debatable. It costs 5 wood, and barbarians need tons of wood for buildings, so they need tons of wood in the beginning. And I usually build so many more important buildings in the beginning that the HQ cannot put all those wares outside at the same time. So building the reedyard can block you a little bit...
2 lumberjacks? I really recommend you to build much more. 5 are not even enough, maybe not enough by far And 2 quarries should be the minimum... Well, a tribe can only be as strong as the weakest part of it. And I think that 2 lumberjacks form a very weak wood industry, so your whole tribe might be very weak, if my theory is right
Because barbarians usually start with at least two mines (or much more). And each mine needs one gamekeeper, so here you go
Yes, improving of the AI needs testing, for sure. But I would be bad at that, I can only watch matches of the current AI and analyse them. Okay, I think I'm not that bad at analysing those matches, but I cannot really test new AI branches
Crater, Full Moon, The Nile
Well, I'm pretty sure about the way how the AI should play, for example: Crystal mines should be build only when they are needed (that's usually quite late) But I don't know so well how this can be implemented And of course: Even if the basic economy will be perfect, the other parts of the economy can spoil the whole tribe. So you could only judge which basic economy is the best if the rest of the AI is perfect, right? Complicated. Wanted to save the world, then I got widetracked Top Quote |
Tibor Topic Opener |
Posted at: 2017-05-24, 06:09
Well, no in-depth analysis is needed here - just overall performance. E.G pick map, tribe and run two AI-only games for 3 hours with my list and your list and evaluate, findings might be:
Then you might decide to tweak your list for better performance. When I compare trunk vs. my branch I look after overall performance, I dont care about individual buildings, their count and when they were built... All this testing would take a lot of time so this is why I am not eager to test other variation of lists... Top Quote |
GunChleoc |
Posted at: 2017-05-24, 13:44
How about having multiple levels of economy, rather than just the basic one? This way we could experiment with multiple stages. If all required buildings have been built for one stage, move on to the next stage. So, the AI could require 1 sawmill in stage 0, but 2 sawmills in stage 3,... of course, this could cause a problem when it's impossible to build all buildings for a stage on the current map, so the rules when to move on should be somewhat relaxed, or be a list of building priorities or some such. Busy indexing nil values Top Quote |
Tibor Topic Opener |
Posted at: 2017-05-24, 15:28
This would be quite different to how AI is built now, basic economy is only a small help when IA should know what buildings to build first and which to avoid on the beginning... As I can see, even for such simple concept it is difficult to find consensus and volunteer testers Top Quote |
GunChleoc |
Posted at: 2017-05-24, 17:03
Well, it's just an idea - I don't know that much about the AI code. If it doesn't work for you, forget that I said anything Busy indexing nil values Top Quote |
Tibor Topic Opener |
Posted at: 2017-05-25, 07:23
I think it is too much work and too unsure result. Moreover I would be really glad if I could finalize what I have now and offer it for review Top Quote |
GunChleoc |
Posted at: 2017-05-25, 07:41
Yep, you've been working on it for a long time. I'll see what I can do about testing. Busy indexing nil values Top Quote |