doncha hate it when you forget to put your name on a post? That was me above

As far as randomness in units hitting, games already has this. Starcraft has it. The problem is, it doesn''t make any difference. When you have a batttle of 50 vs 50 units, if a unit randomly misses once in a while it really makes no difference at all. Law of averages means that most of the time things are going to even out.

The same is mostly true of critical hits. The more units you have in a battle, the less this stuff matters.

Also, I would say that groups getting orders wrong is WAY too much. Games are frustrating enough with bad pathfinding, now not only do they pathfind poorly, they might be looking to go to the wrong place! The first time that happened to me I would probably say "screw this game" and move on.

In RTS games you have direct control over your units. You can tell them exactly where to fire, where to move, etc. The more you break that the less like an RTS game it is. You *can* break it to some degree, but you really want to be careful with it.

Randomness as a whole is better suited to things like starting positions, the level you are playing on, etc.

I see randomness as a simplified way to represent the many variables involved in complicated systems. It''s not unrestricted randomness, but randomness combined with statistical models. Since players cannot possibly control all the variables in certain systems without making the game unplayable, any game that wishes to model these variables must resort to statistical models. The trick is in coming up with models which represent the expected behavior of a given system.

A quick note: it is possible to have very significant random elements in a game without damaging the challenge for the different sides involved.

The card-game bridge for instance starts with the random dealing of cards. This dealing alone generally decides who will earn most hands in the end, regardless of the quality of the players. The trick is that at the start of a game the parties negotiate the assumed outcome of the game. The team who wins the negotiation must fulfill the negotiated goals. The other team must stop those goals. So even if the random pushes the advantage left or right, the victory conditions are also pushed in the same direction by further negotiation.

Another thing:

Panzer General had lots of random, both type 1 and type 2 and it was a great game. One flaw with it was that it quickly degenerated in savegame hell, both in single player and in multiplayer play by email. As long as one resisted the urge to reload the game was excellent, even if type 2 random hit hard (say snow comes early in the Barbarossa campaign - if the snow delayed the victory by a week or two or even caused a defeat, the war continued but on a different course).

First of all, some pratical issues with randomness: randomness makes bugs VERY difficult to fix because you are unlikely to get a similar enough situation during debugging. Also, it increases program complexity without necessarily adding any advantage to gameplay.

Second, it would really bother me to see a battlecruiser die to two or three marines. This all equals out in the end; however, since every once in a while a battlecruiser would die to two or three of my marines.

Weather is a perfect example of the kind of randomness I would support. A kind of randomness I would not support is if a marine was firing into a crowd of zerglings and missed.(It''s an implementation issue, I grant you, that missing one unit doesn''t hit another close by, but once again, it adds difficult to debug program complexity as well as immersion destruction) The kind of "luck" in whether or not a unit hits is the kind that is in AoE--fire a projectile from a ballista, and if its target or some other unit is still there when the projectile lands, it hits.

Not adding randomness to Civilization is absurd, but adding it to Starcraft(except as part of the bonus for height advantage) is superfluous.

I''d just like you to analyze for a moment the historical situations you proposed as an example for randomness:

In Midway, the planes could have been ordered to fly a different pattern. The luck is derived from the commander''s choices, and even in a game like Starcraft, such a thing could and does happen. Simply moving your observer one way instead of the other might make you miss. Is this random? Somewhat, but it doesn''t need to be added. It''s an implicit part of the somewhat realistic fog of war.

The Hood is an example of your randomness, and so I concede that it happens in reality. But is it more fun is really the question? I imagine her superiors would be rather frustrated while her opponents would be rejoicing.

Since you mention that Custer was stupid to not bring along his gatling guns, you obviously recognize that it was not luck that caused him to die, it was his bad planning. Not bringing along an observer because your enemy hasn''t used any wraiths or ghosts in the past is probably not a good idea, and if you lost because of that you would be just as stupid as Custer.

Had the President at the time not made the almost unforgiveable mistake of moving the US''s entire pacific fleet to Pearl Harbor a few months prior(as a political display of power), only a few ships could have been destroyed. If a commander would not like a sneak attack to destroy his entire fleet, he would make it impossible for a single sneak attack to destroy his entire fleet, perhaps by not having his fleet in the same place. The same "luck" could be brought to bear by good planning. (btw, they were out on "maneuvers" or practice at the time)

So even without making the universe scheme for or against someone(preferably both (-: ), apparent luck can be achieved whenever control over both the tactical and strategic level is afforded the player.

So even with a deterministic universe(we mostly live in one), "luck" in a system''s output can be simulated simply by not knowing the inputs, and that is the kind of luck I prefer.




Oh, and by the way, Earth 2150 is a fun game that implements logistics to a small extent--ammo runs out, units overheat(as a result of enemy action, not normal operations), electronics are overloaded(again by enemy action), and comms are jammed.

Determinism could be the ability to determine randomness.

A game is fun whenever there is a factor of chance/luck, otherwise you wouldnt play the game.

Take Counter Strike as an example:
A bad player could actually score some really good hits
in the head on a better player and do a pretty good score, say
30 vs 40. He had a good day, and the better player got some challenging competition. That is what makes CS a very good game, its unpredictable, but not in the long run.

Say you build one infantry unit with a 50% chance of killing an enemy with a single blow.
Say you and your enemy builds one such unit each and engage in battle. The chance is 50% of you winning if the situation is equal.
Now say you build 500 infantry units and you enemy build 400 units. In an equal situation your army would win by about close to 100% certainty because of THE HUGE AMOUNT of rolls. Making the randomness go away.

Add leadership and tactics and you add some more randomness to the situation, or should I say: you are adding more skill required components (SRC).

The more SRC you got in your game, the more intrigueing and fun it will be for the average RTS player to play, and it will offer alot more OPTIONS without losing either way -> randomness vs determinism.

Wars work this way, you might not win the battle, but you can still win the war.

This is what real life is all about and it is fun.

Hmm, a lot of deep thought going on here. I have a simpler answer if you would like: randomness is bad because it is neither teleological nor causal, and events of these two types are better recieved by the human brain because they can be used to construct a narrative model of the game world. Err, maybe it''s not that simple... Teleological means an event that happens because it is a necessary lead-up to the destined ending. Narrative means that sequential events are logically related. (EG. NOT The queen died and the king died. BUT INSTEAD The queen died, so then the king died of grief.) But the basic idea is simple - random events frustrate the human instinct to find patterns in experienced events.

Flareloc

Some interesting points. I hadn''t really thought about the debugging aspect of randomness, and you''re right, it would make it more diffucult to find flaws. However, I do disagree about randomness not adding to gameplay. When a game is formulaic to the point that you can tell with absolute certainty that an event will happen, then the game really boils down to manuever. Military scientists once thought like this, and entire battles could be won merely by positioning your troops so that the other side thought his situation was hopeless. Gustavus Adolphus and Frederick the Great put an end to this way of thinking though.

Lets look at the counter examples you gave:

Midway- many hidden variables were at work that could not have been reduced. Even if the Americans had decided on a different flight pattern, who''s to say the Japanese Fleet would not have been in a different area? So changing search patterns might have made discovery a little more or less likely, it was still basically just luck that the Americans found the Japanese first...and in another piece of crucial luck, caught the Japanese with their pants down (the Japanese carrier was refueling and reloading, with tons of highly explosive and volatile substances sitting right on top of the deck...it was not certain the Americans could have blown up[ the carrier so easily if this were not the case)

Custer- Well, Custer decided to leave his gatlings with the reserve unit, though why is anyone''s guess. I said it was stupid because he could have brought them with him. Why? He thought it was better to leave them behind (and the prevailing thought of the time was that Gatling''s were wasteful of ammunition). You''re right though, in some ways this was more of a decision based outcome, but there''s really no reason why Custer couldn''t have brought them along.

So, let me give two other examples instead. At Operation market Garden, the Germans found allied battle plans in a crashed Allied glider, and yet the Germans thought it was disinformation. Luck was on the sides of the Germans for getting the battleplans, but poor decision making cost them a decided advantage. In an opposite circumstance at Antietam, Northern forces discovered Lee''s battleplans, but due to sheer ineptitude on McClellan''s part they were unable to take advantage of them. Again, luck was with the North, but they just couldn''t really capitalize on it as well as they could have. Now imagine what could happen in a game if you could somehow "read" the other players movements (when you think about it, most RTS wouldn''t really let you benefit from such an event as having battleplans, since there is really no planning involved other than in the player''s head)

Pearl Harbor- from what I''ve read, a German spy had actually given the Japanese detailed information on the composition and daily activities of the American fleet. Yes, the carriers had gone out on exercises, but it was unscheduled, and therefore the carriers were supposed to have been there but weren''t.

I was thinking of your "hidden inputs" notion, and I think it''s a very good one. I think far too much information is known by the players. What often seems random is actually just a lack of knowledge of variables as Kylotan also mentioned. I think this hidden information should go deeper than just not knowing certain variables, but should extend to a great deal of objects.

So is random really random? Or just the inability to determine hidden variables? If it is the latter, the end result to the player will still be that it is random, though it may be easier to debug such coding so in that sense it is good. Also, I think that in the type1 random factors, troop quality or leader quality can help reduce "bad luck". But I still wonder about the truly random elements.

I''m still undecided if I want to include elements of the 2nd type of random variable. I think that while interesting and statistically it is not unfair because it can happen to either side, it can have too great an effect on players plans. But then again, being able to deal with the unforseen is an attractive option as well. Hmmmmmm

quote: Original post by Anonymous Poster
I was thinking of your "hidden inputs" notion, and I think it''s a very good one. I think far too much information is known by the players. What often seems random is actually just a lack of knowledge of variables as Kylotan also mentioned. I think this hidden information should go deeper than just not knowing certain variables, but should extend to a great deal of objects.

So is random really random? Or just the inability to determine hidden variables? If it is the latter, the end result to the player will still be that it is random, though it may be easier to debug such coding so in that sense it is good. Also, I think that in the type1 random factors, troop quality or leader quality can help reduce "bad luck". But I still wonder about the truly random elements.

I think that the "hidden" inputs should not be necesarily hidden. Any information that can affect the outcome should be discoverable. It doesn''t have to be easy to discover it, just possible. Otherwise, as you say (Dauntless?) it still appears random and arbitrary to the player.