Stacking Limits (Game Design)

Game and computer related topics only. Tech questions and answers.

Trolls will be waterboarded

Moderators: chijohnaok, Kameolontti

Forum rules
No trolling, no stalking, no bullshit.
Message
Author
Kameolontti
His Most Gracious Majesty, Commie of the Year
Posts: 7880
Joined: Sat Feb 28, 2009 12:44 pm

Re: Stacking Limits (Game Design)

#11 Post by Kameolontti » Tue Apr 10, 2018 5:21 am

AngryOldMan wrote:One thing about stacking in general is that it adds a tremendous amount of overhead to the AI player. The number of combinations that have to be examined grows exponentially once you allow even 2 units to occupy the same hex.

I'm pretty sure that's why so many WWII grand strategy games don't allow stacking at all.

OTOH, there is one positive tradeoff with allowing stacking for the AI player. Pathfinding can be greatly simplified as you can often move 'through' your own units. Things humans are extremely good at like moving a bunch of units out of the way to let another one through are much more difficult to program than you might imagine.
Fully agreed.

Density vs. breadth is a real challenge for AI because the human opponent is so quick at analyzing any weakness in the AI's formation. Spread too thin and player will punch through, too dense and player will encircle.

These days everything else about making a hex based strategy game is trivial and straight forward but AI is the heel of Achilles.

That said there's so much processing power on today's average gaming rig that it's a lot lesser issue than it was, say, 10 years ago. Not only is there a lot more processing power per processor but you've also regularly got 4 or more processors that can be threaded and then there are additional tricks to count a lot of things fast.
Screw you nero

jwilkerson
Sergeant Major
Posts: 12508
Joined: Mon Sep 15, 2008 9:50 pm

Re: Stacking Limits (Game Design)

#12 Post by jwilkerson » Sun Apr 15, 2018 9:38 am

How to determine stacking limits?

Look at real life frontages for the period and area you are trying to model.

Stacking limits between Alexander's time, Napoleon's time and WWII have changed.

And within the period, stacking limits for given terrain, season are different ... like mountains, desert, prairie, woods. Gather up the empirical data before making final decision based on bottoms up theory model. You are trying to model history,
Ugum Bugum Uber Alles - Iddi Ut Amine Dada !!

Anthropoid
Sergeant Major
Posts: 16789
Joined: Tue Feb 01, 2011 3:06 pm
Location: marching home to [url=https://www.youtube.com/watch?v=qAzVYCs4BMY]Erica[/url]

Re: Stacking Limits (Game Design)

#13 Post by Anthropoid » Sat Apr 21, 2018 10:26 pm

jwilkerson wrote:How to determine stacking limits?

Look at real life frontages for the period and area you are trying to model.

Stacking limits between Alexander's time, Napoleon's time and WWII have changed.

And within the period, stacking limits for given terrain, season are different ... like mountains, desert, prairie, woods. Gather up the empirical data before making final decision based on bottoms up theory model. You are trying to model history,
Yup.

Kameolontti
His Most Gracious Majesty, Commie of the Year
Posts: 7880
Joined: Sat Feb 28, 2009 12:44 pm

Re: Stacking Limits (Game Design)

#14 Post by Kameolontti » Thu Apr 26, 2018 7:35 am

jwilkerson wrote:How to determine stacking limits?

Look at real life frontages for the period and area you are trying to model.

Stacking limits between Alexander's time, Napoleon's time and WWII have changed.

And within the period, stacking limits for given terrain, season are different ... like mountains, desert, prairie, woods. Gather up the empirical data before making final decision based on bottoms up theory model. You are trying to model history,
That's the more simulatory approach. The more 'gamey' approach is that you simply try, try and try and keep changing balancing and other factors until it feels right as a system.

Essentially, economics, tech, cash flow - map sizes, turn length - all things considered it has to feel right and then you have a good game.

Simulators don't need to be 'good games', they need to be accurate and have a good interface.

It's possibly to mix both. Make a good game that also simulates certain aspects with certain accuracy.
Screw you nero

Anthropoid
Sergeant Major
Posts: 16789
Joined: Tue Feb 01, 2011 3:06 pm
Location: marching home to [url=https://www.youtube.com/watch?v=qAzVYCs4BMY]Erica[/url]

Re: Stacking Limits (Game Design)

#15 Post by Anthropoid » Fri Sep 20, 2019 10:51 am

Been back at Hearts of Iron IV with Black Ice mod here lately and I was reminded of this shiz.

Sadly, the perfect grand strategy game remains elusive . . .

AngryOldMan
Staff Sergeant
Posts: 4071
Joined: Fri Sep 19, 2008 2:29 pm

Re: Stacking Limits (Game Design)

#16 Post by AngryOldMan » Fri Sep 20, 2019 4:58 pm

Anthropoid wrote:Been back at Hearts of Iron IV with Black Ice mod here lately and I was reminded of this shiz.

Sadly, the perfect grand strategy game remains elusive . . .
I find I always go through the same 3 phases with a strategy game:
1) Learning phase - just trying to figure out all the rules and game mechanics. Lots of frustration, lots of googling and forum posts. Maybe some YouTube videos.
2) Proficient phase - finally understand how everything works and start to optimize my strategy. This is the fun part as everything that used to be hazy is now clear and I know exactly what to do and how to deal with challenges I'm likely to face.
3) Fed up phase - either the game gets too repetitive or I get too good at it and win too easily for it to provide any more satisfaction.

And at that point, I start the search for the next 'perfect grand strategy game'.

If I can get a month out of phase 2 I feel like I got my money's worth.
---------------------------------
If those who can do, and those who can't teach, what does that say about a sex education teacher?

C_S
Sergeant Major
Posts: 12008
Joined: Wed Mar 16, 2011 1:49 pm

Re: Stacking Limits (Game Design)

#17 Post by C_S » Sat Oct 05, 2019 3:09 pm

I'd think stacking would improve AI b/c the computer can crunch numbers better than I can regarding which hex to 'hit im where he aint'. "With the mostest" In Toaw I can't beat the computer unless I have more stuff where it needs to be. It's good gaming. Shrug. I've always reckoned stacking gives advantage to the automated opponent.
This post is not a legal document under any circumstance. It has not been prepared at all and is written offhand in jest while mentally impaired.
Do not act or rely on any of this information without first seeking the advice of a psychoanalyst.

Anthropoid
Sergeant Major
Posts: 16789
Joined: Tue Feb 01, 2011 3:06 pm
Location: marching home to [url=https://www.youtube.com/watch?v=qAzVYCs4BMY]Erica[/url]

Re: Stacking Limits (Game Design)

#18 Post by Anthropoid » Thu Oct 24, 2019 4:41 pm

C_S wrote:I'd think stacking would improve AI b/c the computer can crunch numbers better than I can regarding which hex to 'hit im where he aint'. "With the mostest" In Toaw I can't beat the computer unless I have more stuff where it needs to be. It's good gaming. Shrug. I've always reckoned stacking gives advantage to the automated opponent.
Computer crunch numbers GREAT! They are perfectly accurate; fast as fuck (boi!); and rigidly rule-bound, meaning they will not overlook ANYTHING they have been told to keep an eye on or include in their crunching.

Problem is: they have to be told what, how, and when to do EVERYTHING, and they have ZERO capacity to comprehend "why" anything.

The difference between a great computer opponent and an "okay" one is likely to be many thousands of lines of source code including lots of conditional algorithms. All these 'factors' that all those lines of code are covering have to be well conceived by the designer and also tested in a sufficiently broad range of play conditions to see if they 'work' effectively.

Programmers and testers are the roadblock, not the technology. That is my view at this stage.
AngryOldMan wrote:
Anthropoid wrote:Been back at Hearts of Iron IV with Black Ice mod here lately and I was reminded of this shiz.

Sadly, the perfect grand strategy game remains elusive . . .
I find I always go through the same 3 phases with a strategy game:
1) Learning phase - just trying to figure out all the rules and game mechanics. Lots of frustration, lots of googling and forum posts. Maybe some YouTube videos.
2) Proficient phase - finally understand how everything works and start to optimize my strategy. This is the fun part as everything that used to be hazy is now clear and I know exactly what to do and how to deal with challenges I'm likely to face.
3) Fed up phase - either the game gets too repetitive or I get too good at it and win too easily for it to provide any more satisfaction.

And at that point, I start the search for the next 'perfect grand strategy game'.

If I can get a month out of phase 2 I feel like I got my money's worth.
Same here!

AngryOldMan
Staff Sergeant
Posts: 4071
Joined: Fri Sep 19, 2008 2:29 pm

Re: Stacking Limits (Game Design)

#19 Post by AngryOldMan » Fri Nov 08, 2019 1:53 pm

C_S wrote:I'd think stacking would improve AI b/c the computer can crunch numbers better than I can regarding which hex to 'hit im where he aint'. "With the mostest" In Toaw I can't beat the computer unless I have more stuff where it needs to be. It's good gaming. Shrug. I've always reckoned stacking gives advantage to the automated opponent.
Let's say you have a front that's 20 hexes long and 10 units that can each reach all of the front hexes and let's say stacking is 3 deep.

So the number of potential combinations is 120 in each cell. With 10 cells that's on the order of 120^10. It's not that simple but that gives you an idea of the order of magnitude for comparison purposes.
With no stacking it drops to 10^10. That's a staggering difference.

It is true that computers can crunch a lot of numbers but not that many. Not in a reasonable amount of time. (Biggest complaint about AI: too slow).

I know this from experience because I've implemented an AI in a strategy game with stacking (well many really because I had to start over from scratch about 10 times).
Brute force is NOT an option.
---------------------------------
If those who can do, and those who can't teach, what does that say about a sex education teacher?

AngryOldMan
Staff Sergeant
Posts: 4071
Joined: Fri Sep 19, 2008 2:29 pm

Re: Stacking Limits (Game Design)

#20 Post by AngryOldMan » Fri Nov 08, 2019 2:46 pm

One interesting grey area in the whole stacking vs. non stacking thing is the idea of whether units can move through friendly units. In WarPlan, you can move through your own units. I feel like this greatly offsets some of the disadvantages on non stacking. Not only does this greatly lessen the traffic jam issue but it also makes it much easier to attack with multiple units from the same hex by swapping them out.
---------------------------------
If those who can do, and those who can't teach, what does that say about a sex education teacher?

Post Reply