a995b4 No.31560
How does one balance the different genres?
Right now, I am working on a single player game involving skills and attributes of multiple player controlled guys, plus items and equipment that increase and decrease this. On top, I then have abilities of characters that affect these stats of other characters.
How do I balance? My first thought was to load it all into spreadsheets and tweak numbers and formulas in there before adding to the game code. Does that make sense? I am thinking I could run game move simulations in the spreadsheet?
____________________________
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
893c05 No.31563
Beta testing is probably the best option. There's a lot more to it than tweaking numbers though. Unit X might be superior when played a certain way against Unit Y, unless Unit Y uses Z strategy. Shit's complicated. Probably best to tweak as best as you can, and just hope nobody finds an exploit. If they do, you're gonna end up in the whole nerf/buff cycle. Also there's no need for 100% balance, it just doesn't need to be terribly unbalanced. People will develop strategies to counter minor unbalances.
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
a995b4 No.31565
>>31563
How would I get the initial balance right? Or is it simply a matter of setting some "finger in the air" numbers and just tweak like crazy for 100's of games? Balance is extremely important to my game, so I want to try and be scientific about it. I know that in the end gameplay and experience wins over statistically accurate models, but if my game has 1000's of gameplay options, I think that doing this statistically will cover more scenarios quicker than just tweaking as I beta test.
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
893c05 No.31573
I doubt there's a formula you can use to ensure the initial figures are balanced. Too many variables to consider. I think the only real automated solution would be AI vs AI testing. But AI would not play the same way as humans.
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
f47436 No.31590
>>31560
>I am thinking I could run game move simulations in the spreadsheet?
Absolutely. Excel/VBA is a very good fit for what you are talking about.
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.
bdf380 No.31705
I've seen people taking a specific part of the throwing random action a million time by a CPU. They usually save the result in a hash table in which they key is a list of action/item/card... used in order in the test and look toward the ratio of success/failure by each key. It's extremly effective for card game apparently.
Sometime, when there's no absolute solution, you can only find your answer by doing empirical test.
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.