Pre-Order the Game!

Desura Digital Distribution

Friday, May 23, 2014

Plan for Beyond Beyaan

It've been a while, but it has been a productive month in terms of design tweaks.  Arpad and I has been discussing Master of Orion 1 and its good and bad aspects.  Arpad helped illustrate the flaws in AI, and after he explained them, I tried a new game on Impossible difficulty.  If you recall, my last attempt failed disastrously.  This time, I'm effortlessly crushing the AI.  I now feel dirty knowing AI's exploits, Impossible is no longer "Impossible".  Our plan for AI is to have AI be actually smart, and reduce AI cheating as much as possible.

So our goal is to not exactly replicate MoO 1, but take the best and improve on the bad.  With mostly input from Arpad, the following are list of changes that Beyond Beyaan will have:

Space Combat - The map will be a hexagonal grid.  This is to improve logistics for missiles/torpedoes, minimize distance discrepancies, and generally balance things out.  This is mostly to help the AI.  It will still be vertical columns of tiles, so it shouldn't differ too greatly from original.

Diplomacy - All races start with neutral relation toward each other.  This will help boost the weaker races (Mrrshans, Alkari, and Darloks especially) while not weakening other races (Humans will have different bonuses as explained below)

Personality - AI won't have "Leader Personality" like in MoO 1, they hurt more than help the AI.  So each AI will factor in their race's strengths and act on them, instead of trying to do something counter-intuitive.

New Races - Since Kickstarter has resulted in success for the three extra races, Arpad and I discussed back and forth on how those will fit in a MoO 1 universe.  We have tentatively set forth their attributes as explained below.

Leaders - Likewise, I promised leaders in Kickstarter, but how to retain the simplicity from MoO 1?  Arpad suggested that they be a random event, with a randomized duration.  I liked this idea, and we fleshed it out.  There will be 22 different leaders, revolving around 11 different attributes that can help or hinder your empire.  For example, there are two financial leaders, one good, one bad.  Good will add a certain amount of income, while bad will reduce amount of income.  You cannot hire or dismiss them.  Each leader only have one attribute that they improve/degrade.  It's possible to have multiple leaders at the same time, based on random event roll.

Both New Races and Leaders will be checkbox options that you can disable when starting a new game, if you do not like those features.

We discussed extensively on how to balance the races in MoO 1.  Psilons, Meklars, and Klackons were the strongest, while Mrrshans, Darloks, and Alkari were the weakest.  We wanted to balance them, but still have each be unique from other races.  We've settled on the below (Left is the original bonuses/attributes, if it don't list a technology field, it is "Average", on right are the Beyond Beyaan's version of the race).  This is subject to balancing and testing, but we feel that this will even things out and still retain an interesting gameplay.  Also the new races were problematic, but hopefully they are unique enough with the listed attributes.



If you have concerns or suggestions for any of the discussed features, please let us know in the comments!

12 comments:

  1. Can you tell us how the AI tricks and how this can be exploited? Just to understand AIs better.

    ReplyDelete
    Replies
    1. The first one from Arpad's list below was the game-changer for me (I used most of the other advices already). In the game I mentioned, as Psilons I expanded fast, and had twice the planets as Meklars, but yet Meklars out-researched me somehow. The only way for me to not lose planets was the missile exploit. I was able to stall them long enough to stalemate battles that I couldn't win otherwise (planets without missile bases for example), and destroy their fleets with minimal losses. With this exploit, the game became easy :(

      Delete
  2. Here are a few to get you started. SPOLIER ALERT!
    1. AI tends to pull back ships from missiles, which it can avoid. So build up bases on the likely and known planets that are going to be attacked by AI. Use good computers and good missiles. Design 6 (roughly) identical medium or large ships. Each with only 1 missile from each range. Say range 5, 6, 7, 8 missiles on each. Also high warp and high combat speed. Nothing else. Maybe one with battle scanner. Once you see an enemy ship with your battle scanner, screen shot it for future decisions. Send only one of each of these ships to defend each planet in danger. When AI ship is in range, fire a single missile at that ship. For example, if AI ship is in range 7 and you fire missile, it will only fire range 7+ missiles, so you can fire the rest later. Fire all planetary missiles against most dangerous AI ship stack. Those can't be avoided by the AI and they will keep getting hit as long as they stay in battle. Your 6 ships can fire up to 6*4*5=120 times, which is more than the maximum of 50 turns per battle, so your planet may not get hit depending on AI speed and number of stacks.

    2. Attack any AI planet that is not defended by large fleet by 6 different huge ships only. Each ship should have high warp, high combat speed and max biological weapons, which you know or suspect that will affect the AI. Nothing else on these ships. At least one of them should reach the planet to release the biological weapon. If you can bring population down to 0, the colony is destroyed. If you don't have such biological weapon, then use the best bomb you have, but that should be a good bomb. Also, bombs are more cost effective on small and medium ships. Biological weapons may work on small/medium ships as well, but you will need many ships in that case.

    3. Use ion stream projector or neutron stream projector form multiple different ships. Works best with 6 huge ships, which are strong on defense, high combat speed and warp, preferably auto repair or damage control, maybe warp dissipator, cloaking device, lightning shield, or repulsor beam. As long as your ships are not destroyed, they take out any fleet and any number of bases.

    4. Black hole generator on 6 different large ships. They take out huge Meklar fleets with minimal losses.

    5. Purposefully give up planet with lots of factories to incoming enemy transport. Send out your transport to make it easy for AI to capture it. Then send back your transports to recapture it next turn. The reward is up to 6 captured technology, which helps to even out the tech disadvantage. Don't do it against Bulrathi unless you are high in pop.

    6. Make contact with all other races. Trade tech and do espionage as often as possible. But when you assign spies to do espionage, do not spend too much on spying on that race. Spend high on spying when you don't have any spies at that race. When you capture tech, always get Computers if possible. Second preference: propulsion. Staying strong in computers is important for successful spying. Never spy on Darloks. Make screen shots on each race's tech in each turn, so you will know what to steal next turn. In espionage you go deep. In research you go wide.

    7. Spread as fast as you can.

    8. Obtain range 5 or 6, then develop warp speed and combat speed. Quickly scrap all ships once a better warp is obtained. Having high warp is the most important thing to own and use. AI has no response and AI does not use warp well.

    9. Get improved space scanner or advanced space scanner ASAP, so you will see where the AI is going. Until then, use screen shots and compare 2 consecutive turns to estimate AI fleet directions and ETA.

    10. Get Planetology techs and get them early. They all increase production. Especially Improved Eco Restoration.

    ReplyDelete
    Replies
    1. 11. Don't worry about wars. AI tends to be slow to attack. Defend planets where AI attacks or attack AI at the closest point ASAP, especially if you are already at war. Since you only can have 6 ship designs, you must choose if you want to defend or attack. You can't do both in the same time against a well developed AI. Attack is better than defense, unless you are weak and small. Try to attack only when you are ready: typically 6 fast huge ships with biological weapons attack a planet and wreak havoc.

      12. Unless you play human, never agree to trade.

      13. Play huge galaxy, where AI is lost.

      14. Every time a new technology is obtained, design a new ship that is as expensive as possible. Change ship production to that ship at all planets. Even if you have 0% assigned to ship building. Get close to finishing these ships, but only finish when you really need them. But then design new ships as needed and build only the newly designed ships. If you time well, all come out in the same turn and they can be all top tech and even diverse. Very useful for small ships too. No need to pay upkeep on a small group of small ships.

      Delete
  3. Few corrections to the blog post:
    Sakkra planetology was excellent.
    New Silicoid propulsion is average and weapon is poor.

    ReplyDelete
  4. Just curious why 190% instead 200% bonus on population production and growth?

    ReplyDelete
  5. Klackon and Sakkra bonus was too high in MOO. This is our first attempt to tune it. It may change further after testing.

    ReplyDelete
  6. :) Oh, I see. One more thing, about the leaders, would they randomly generated or predefined? I mean, will Chug the Planetologist (+ pop growth) always be the planetologist or sometimes the scientist or something else? Also if some player has +science leader can other players get +science leader but with different name and appearance?

    ReplyDelete
  7. Randomizing leader names is better not to do. Instead, we shall try to come up with suggestive leader names, which give away their qualities, so players will learn fast who is doing what, instead of looking it up every game. It will also add personality to the game and vocabulary to the players to discuss with each other. So "Chug" is always Planetologist, but Chug is not the only Planetologist. Chug working should not change the probability for any player including the same player to get another Planetologist, but with different name. I think that same graphics for all plenetologists is better than different ones: for simplicity, orientation of players, lowering confusion, saving on artwork, and overall perceived value of the game. The main take home lesson from MOO is that less can be perceived as more if done right. On the long run modding will be greatly supported, so fans can make more pictures, etc.
    Do you want to join our team to help speed things up?

    ReplyDelete
  8. Thanks for thorough answer! If I can help I'm willing to join.

    Oh, one more thing regarding the blog post, hexagons in tactical combat, great idea!

    ReplyDelete
    Replies
    1. Great! Welcome to our team. Please send me email to arpad.kelemen at gmail.com and I get you into our discussions. Our current issue is if we are to switch to C++ or stay with C#. There are many things to consider and we don't blog all details here.
      Also, thanks for the compliments and your long-term interest in the BB project!

      Delete
  9. This comment has been removed by the author.

    ReplyDelete