Jump to content

  • Log In with Google      Sign In   
  • Create Account

Banner advertising on our site currently available from just $5!

1. Learn about the promo. 2. Sign up for GDNet+. 3. Set up your advert!


Member Since 25 Nov 2005
Offline Last Active May 22 2015 03:42 AM

#5167834 Need help with a fuzzy Scoring algorithm.

Posted by LorenzoGatti on 19 July 2014 - 01:04 PM

You should probably use rank statistics rather than averages. Assuming raw scores are positive and better if higher:

  • normalized score= weight*(number of lower scores than the player)/(total number of scores-1).
    0 if the player is the worst of them all, maximum if he is the best, half maximum for a median (not average) performance. Suitable for distributions that are known to be skewed and clustered, e.g. batting average in a season of baseball.
  • normalized score= weight*(player score-minimum score)/(maximum score-minimum score)
    0 if the player is the worst of them all, maximum if he is the best, half maximum at the midpoint between minimum and maximum raw scores. Suitable if raw scores are quite linear but their range is unpredictable, e.g. number of goals in a season of football or number of yards gained in a season of American football.

#5167569 Speed of fleets

Posted by LorenzoGatti on 18 July 2014 - 04:57 AM

If you had a game where you just built up a fleet (active choices of content of fleet), then send it to attack somewhere (active choice of target), but you had no further input into the system, how long would it be fun?


Combat,                 automatic.

Fleet movement     automatic.

Battlefield strategy automatic.


It would get very boring, very quickly.

Not really. For example, GalCiv has automatic and abstract combat (number vs number),with no battlefield "strategy" and automatic fleet movement, and on top of that automatic ship building and no ship choice. It is still interesting because you need to direct fleet movements and attacks to the right places depending on enemy actions.

#5166945 How to avoid "stacks of doom" in 4X?

Posted by LorenzoGatti on 15 July 2014 - 03:41 AM

Most good 4x games I've played used economy as a deterrent to large forces.

If you place 'all of your units' to defeat the enemy's strongest base, you leave yourself opened for them to overtake all of your economic settlements. You do capture one planet, but end up losing 10-20. After a few turns, your opponent has managed to use these planets in such a way that their fleet is now larger than yours.

That's the best way to do it imo.

I'm perfectly fine with concentration of forces to force a victory, because, most times, this is a bad strategy in a good 4X game...

This penetration scenario could, unfortunately, be symmetrical: a very powerful "strongest base" that can challenge a great fleet means that the opponent is already open to large-scale invasion because other places are inadequately defended, exactly like the player. Enemy large fleets could be simultaneously ravaging the undefended interior of the other empire without necessarily meeting, possibly ending the game by randomly conquering the enemy capital.


Boring penetration and raiding with a single powerful force needs to be a worse strategy than more fun and challenging small battles of attrition and large-scale offensives of distributed small forces:the game should reward allocating the vast majority of ships to defense, so that "hunting parties" capable of conquering a planet are going to

  • start small and grow slowly, as surplus ships are produced and places in the safe interior of the empire are demilitarized. Only an overwhelming economic lead (i.e. endgame cleanup) should allow building fleets from scratch instead of slowly reinforcing a large number of planetary defense forces and drafting ships from them when the risk is acceptable.
  • struggle to keep pace with the growth of enemy forces. Too much defense is conservative and safe; too much offense should be a danger.
  • meet their enemy counterparts, which should have a comparable strength, and be thinned out (even if they win) before they become stupidly strong.

This can be obtained through a combination of different means.

  • Making ships slow enough, as the OP dislikes. What's important is avoiding fast long distance travel for strong fleets, so that they cannot defend the whole empire effectively; maybe short travels could be fast enough to not be annoying.
  • Encouraging minor deep space battles in random contact locations along front lines, with planetary assaults as the result of a small (and not too quick) military campaign rather than as the standard type of battle. There is a fundamental difference between ships that are constantly located in some interesting place with a chance of being detected and intercepted and ships that effectively disappear when they start a trip and reappear at their destination thanks to the vast emptiness of space.
  • Rewarding possession of planets etc. only after some time has passed, encouraging players to hold and defend locations rather than advancing and attempting to conquer objectives faster than they lose them elsewhere.
  • Combat rules that reward slight superiority proportionally more than overwhelming superiority, e.g. guaranteed significant losses for the winner. For instance, splitting a fleet to fight two battles with a 2:1 ratio could be better than fighting one battle with a 4:1 ratio and the other with the survivors of the first (maybe because there is a necessary delay allowing the enemy to send reinforcements).

#5165631 Speed of fleets

Posted by LorenzoGatti on 08 July 2014 - 02:13 PM

One of the points of slow ships could be giving them important and exclusive abilities, turning them from part of a fleet to strategic weapons or strategic resources. For example, colony ships that can terraform planets and should often have a fleet as an escort, or Death Star-like planet scale weapons. But it is not necessary, and other types of strategic problems could be better

#5164541 Speed of fleets

Posted by LorenzoGatti on 03 July 2014 - 05:36 AM

Many observations made so far point in the same direction: speed differences make little sense and can be eliminated as a source of pointless annoyance.

  • FTL propulsion can have the same pseudo-speed for every ship because it's based on the same technology: for example, a succession of "jumps" involves a maximum or fixed distance and a constant preparation time for each jump, with the only difference between ships being energy cost and device size.
  • Realistic propulsion can be designed to give the same "standard" acceleration, and therefore the same travel time on any route, to any ship.
    If both engines and robust materials are very good and relatively cheap, the limiting constraint for acceleration is not squashing passengers to death (rather than affording powerful engines or not breaking the ship apart) and it doesn't vary by ship type. Improved engines are going to be smaller, cheaper, more efficient, but not "faster".

If FTL speed differences (like in Star Trek) are needed, they can be simplified to a few tiers and used to forbid ships with different speeds to belong to the same fleet. Serious speed differences are going to imply different strategic movements in any case.


If limiting thrust of realistic engines according to available fuel is a normal occurrence, it can be assumed that the fleet automatically allocates available fuel to maintain the same travel time for every ship, rather than slowing down specific fuel-challenged ships. Computations are easy: fuel is proportional to ship mass.

#5163680 Alpha Zone on Targa Texture just won't show alpha

Posted by LorenzoGatti on 29 June 2014 - 01:40 PM

What about the creation of the OpenGL texture? The alpha channel could be lost after loading the file correctly.

#5162732 Idea for samurai/ninja game

Posted by LorenzoGatti on 25 June 2014 - 02:30 AM

So question: Would this again be a good idea for a single player game with some multi-player aspects? Something like a Halo or Call of Duty or something like that?

Multiplayer modes add a large amount of extra work because they add significant technical challenges, such as running servers and finding them, designing communication protocols, synchronizing clients while minimizing lag, handling player identities and maybe accounts, etc.
All these problems can be solved, usually in effective and efficient ways, but not for free.
I regret mentioning that in a strategy wargame player vs player battles are particularly appropriate; it's a standard feature but, in your case, probably a luxury that should be left out even under ideal circumstances.

Or would it just be best to make something like a single player that is like Tenchu and Way of the Samurai combined?  Both of those games have a mission based part role playing aspect to them.

Both these game series appear to have relatively complex fighting, complex character models and large 3D environments, putting them in the horribly expensive.tier of development budgets.
If you survive the initial setup effort of being ready to make 3D assets (which might be a worthwhile learning/recruiting experience) and getting a 3D engine running (which might be relatively easy and inexpensive with ready-made ones) you are left with the choice between a small and short game, a simplified and ugly game, or both.
Instead, as already touched by others, you should focus on cheap technology and development tools.
For example, if you choose a RPG angle rather than a wargame angle, tile-based 2D map editors should be suitable for churning out large amounts of Japanese-style scenery (after drawing some highly reusable tiles), and something like RPG Maker could serve as a tile map editor and allow cheap implementation of large amounts of characters, dialogues and fights, sacrificing the sort of action found in games like Tenchu.

Or it could also take on a Fable type life.  The only thing would be, based on what character the player decides to play, the missions and quests would be slightly different.  Am I hitting on something a bit more viable now?


No. Something like Fable is even more horribly expensive to make than a "standard" action-adventure game, and more difficult because it's more sophisticated. More mechanisms can go wrong, requiring more testing and more iterations, and it's very likely that more assets would be wasted on optional and mutually exclusive levels instead of used once in mandatory levels.


Please, give me as much advice as you can.  For me some of it is hard to hear, but it is helpful in understanding the scope of what I can do.


Your scope should be the smallest that leaves you with an acceptable game, and your expectations for a minimum viable product should be proportional to your skill and experience. What's the simplest and cheapest entertaining thing that a samurai or ninja could do?


So please, anything, any thoughts at all.  Now that I think about it, it would make an awesome single player game in the console arena.


Development for consoles is another difficult and expensive luxury. Choose one easy platform (most likely, Windows PC leveraging game-making tools).


Please let me know if I am now headed in the right direction.  You have my thanks!!


You are still headed in a tragically ambitious and confused direction, but at least you seem open to suggestions.

#5162271 Idea for samurai/ninja game

Posted by LorenzoGatti on 23 June 2014 - 01:53 AM

Combining large-scale battles, personal combat and adventuring is overly ambitious and incoherent: some of these aspects can be reduced to a collateral minigame (e.g. send spies between battles), reduced to a well-integrated part of the main game (e.g. exploration and stealth with specific units and limited purposes within a RTS), or implemented as variants of normal gameplay (e.g. a RTS turning into micromanaged melee combat with a handful of complex hero units and/or into RPG-like exploration with large maps and no reinforcements).


Your character types appear a bit irrelevant, because the only real one is the samurai, able to face all challenges and take any initiative. A ninja is, by definition, insignificant cannon fodder, while a ronin can be either hired by a noble house and effectively "promoted" to be the same as a regular samurai or independent and uninvolved in the plot (except perhaps as a minor enemy, not unlike bandits or savage animals).

Since your story is about noble houses struggling for power, ninja clans, monasteries, town leaders, various kinds of powerful free agents etc. cannot be important factions; they should appear as someone's allies.


Making a "learning tool" of your game is going to lead to ugly conflicts and compromises between fun and historical accuracy. For example, you might discover that rifles are a dominant strategy in battle and need to be limited or nerfed in unrealistic ways.


Regarding making a MMO, see the first item in the forum guidelines. Moreover, you are describing an eminently single-player game (your character vs. the rest of Japan: death or glory) which could be extended to small multiplayer battles without a plot. In a MMO design, not all players could be able to be the samurai general, and being a lowly soldier or ninja would suck.

#5161260 Attribute and Formula Help

Posted by LorenzoGatti on 18 June 2014 - 03:24 AM

I think one key information missing is what range of damage you'd like to see (minimum to maximum).

The range of damage values is only a scaling factor, so simple to manage that I forgot it in the example formulas in my post.

The starting points are

  • A set of benchmarks: the fight of unit A vs. weaker unit B should end with A's victory with probability p, often p=1, and its duration of n turns should follow a distribution with a given mean N and a given minimum and maximum, possibly approximate.
  • A range of hit point totals for different units: highly compressed like in Warhammer to make defense and "saving" hit points a big deal, extremely wide like in D&D to make minor damage insignificant, or something in the middle.

Given these constraints, it's obvious that, if unit B has b hit points, A's typical attack should do b/N average damage per turn, independently of how the stats and equipment of A and B determine damage. 

Different ways to do b/N damage per turn are also independent from stats:

  • Fixed and certain damage X every turn: X=b/N. Only an option if hit point ranges are large or N is very small, or the necessary rounding to an integer will distort damage values.
  • Random damage: unit A's damage can be a random variable with an average of b/N. Winning probability p and the distribution of fight duration n can be computed from the distribution of damage.
  • A chance to miss: if unit A hits with probability q, doing X damage, q*X=b/N. Critical hits are a slight generalization to X or Y damage rather than X or nothing: q*X+(1-q)*Y=b/N. With both misses (probability 1-q) and critical hits (probability c conditional on hitting) you simply add three branches for X,Y and 0 damage: q*(c*X+(1-c)*Y) =b/N. And so on; X and Y can, of course, be random variables.
  • Random delay: if unit A attacks (on average) every m turns doing X damage each time,  X/m=b/N. In most games X<b (no 1-hit kills) implying m<N.

#5161003 Attribute and Formula Help

Posted by LorenzoGatti on 17 June 2014 - 02:01 AM

You need to decide how the combat rules should behave, then formulas that fit the requirements will be easy to find.

For example, consider critical hits:

  • Why do you want higher damage variance? How is low variance bad? Barring strange effects, higher variance gives a chance to the weaker party, and every game needs to find the proper balance between confidence and uneasiness (for strong characters) and between despair and gambling (for weak characters).
  • Wouldn't rolling virtual dice for damage, with the possibility of choosing the probability distribution of damage output quite freely for a desired expected value (in particular, larger variance can be obtained by summing fewer larger dice), be more flexible than a mixture of "high" and "low" damage?
  • Are you OK with a bimodal distribution? Would the players notice?

Or consider special (magical) attack and defense:

  • Against a given reference opponent (e.g. Str 50, Def 50, Mg 5), increasing Mg is equivalent to increasing Str and Def. What is the equivalence? Does it depend on your stat values? Or on the opponent's stat values?
  • For example, you might want a very linear system in which a +k Str or Def increase is always as useful as +1 Mg, regardless of your stat values, and against any enemy: expected magical attack damage could be (Str+k*Mg)/(Def+k*Mg).
  • Or you might want a strongly level-based system in which lower Mg values don't stand a chance against higher ones, with Str and Def mattering only in the same Mg tier: expected magical attack damage could be min( (Str-Def)*(attacker Mg - defender Mg), 0)
  • How many points of Str, Def, or Mg should equipment be worth? Should it vary according to stat values?

#5158061 Horror Story Help?

Posted by LorenzoGatti on 04 June 2014 - 04:50 AM

Rituals are only a detail of the plot, and in a game they should be designed according to practical criteria because they are completely arbitrary:

  • What is suitable for depiction? For example, chanting in made-up languages, chalk diagrams and candles might be rather family-friendly, violent human sacrifices in the middle of a disgusting sex ritual probably not. What pretty graphics and special effects are doable? How can they be adapted to be part of a magical ritual?
  • What is suitable for player interaction? Assuming the cult has to be disrupted, means could include sneaking an "impure" person inside a sacred perimeter, damaging an artifact, distracting or attacking cultists at a critical moment, delaying cultist travel, suppressing information, destroying or defiling a place of power, etc. The genre of your game determines the genre of the rituals; if the action is fun, any excuse about how magic works is acceptable..

On the other hand you should reflect on the higher metaphysical level on which your plot is supposed to make sense:

  • What are gods?
  • Are there even higher powers, e.g. generic "good" and "evil"?
  • What does it mean for a god to die, and revive?
  • What sort of magic is possible? (Reviving dead gods is probably one of the smallest niches of magic, probably something that never happened before; there has to be some larger body of tradition and knowledge.)
  • Where does magical power come from?
  • How can mere humans interact with gods?

For example, requiring the death of another (enemy?) god to resurrect the old dead one might range from appropriate because of cosmic balance and limitation mechanisms, to a reasonable plot gimmick to make resurrection difficult, to preposterous because the death of the god was a very special occurrence that has nothing to do with other gods.

#5155181 Roads (connection between villages/cities)

Posted by LorenzoGatti on 22 May 2014 - 02:28 AM

The first idea is to make some "build road" option after clicking on any village/city which would show DESTINATION ("neighbour" slots/villages/cites/nodes only) and by clicking you can build/upgrade a road/railroad. Also, my though is that maybe it should not be allowed to join any village with any village because it would look very messy (crossing of roads), so maybe some "distance limit" for the build road feature?


My suggestion to limit nonsensical road construction is relying on the Delaunay triangulation of village locations: allow roads that are an edge of the Delaunay triangulation, and forbid all others.

It seems that you are able to move, add and delete villages during initial map creation to obtain a nice Delaunay graph (with appropriate numbers of neighbours, sufficiently uniform edge length, etc.)

#5150283 Cars differentials, open, locked and preload

Posted by LorenzoGatti on 29 April 2014 - 03:48 AM

You are completely failing to explain the actual problems you have, apart from being perplexed by a new and unfamiliar subject or mixing up different models of different designs of differential. Are the cars in what I suppose to be a a racing game doing anything inappropriate?  


There is a wide range of useful locking and torque distribution policies you can adopt in your software-simulated differential, without even worrying about actual mechanical implementations.

As long as the behaviour of your differential is physically sound (e.g. you don't drive wheels with friction in the clutch, with preload springs, or with torques out of thin air) and it conforms to some basic principles (unlocked for small wheel velocity differences, locked when one wheel appears to be slipping) you can do anything you want with parameters such as a "preload torque": unlike actual physical quantities such as wheel angular velocities they are only an abstract description of what the differential does.

#5149881 Magic advancement system

Posted by LorenzoGatti on 27 April 2014 - 09:50 AM

Don't confuse different aspects of the magic system that are actually rather independent, particularly strategic improvement (acquiring new spells or improved variants of old spells), tactical spellcasting (e.g. gathering energy, performing rituals, hitting targets, etc.) and intermediate time scales (e.g. "preparing" spells in advance, D&D style, or keeping magical objects ready).

What is it that should feel more "magical" than learning from a trainer? What do you really need in your game?

Making study a boring activity that is left off screen and skipped over, but actual spellcasting player-skill-intensive, cool and complex, is as legitimate as making learning spells a varied collection of mystical revelations and discoveries, themselves an important part of the plot, but casting spells simple and dependable like mundane melee or ranged combat.

#5149868 DSP. Reverb

Posted by LorenzoGatti on 27 April 2014 - 08:33 AM

 The reverberation chapterof a a great textbook by Stanford professor Julius Orion Smith, "Physical Audio Signal Processing", is particularly comprehensive (feedback delay networks appear to be one of his research interests).