|
On March 17 2010 14:27 fmagrave wrote:
I think the probability model of a miss rate system would be.
Probability(Target will die in N shots)=(N-1 choose k-1)*p^k*(1-p)^(N-k)
where
p is hit rate
k is the number of shots unit A needs to kill unit B (assuming 100% hit rate).
I agree! That's why I posted it two pages ago.  (Edit: Three pages. And man this really didn't have to start a new page, sorry about that.)
|
I think the range idea is absolutely awesome and would love to see that implemented for something different :D
|
And about:
"A bit of bad luck will cost you the game."
If u have an army aventage but u are force to attack up the tarrain - u just DO NOT. u fall back fo a second , regroup and then attack again. That just another reason for high ground misssing mechanics. It works both ways - for defender and agressor. You need to stay focus all the time. If u cry about the fact that the smaller , but standing at the high ground army destroyd u - well its just and only your foult not a game mechanics.
|
On March 17 2010 13:08 stink123 wrote:
Another issue that blizzard cares deeply about is the intuitiveness. They want to make the game easier to pick up and play, and a high ground miss chance is much less intuitive than requiring sight above a hill before you can shoot them. Not everyone notices the "miss" animation above their units in a battle, and there's no easy way to teach (in game) that the miss chance is 30%.
You don't have to teach the player that the miss chance is 30%. You just need to tell them that units positioned on high ground have an advantage. Players will learn or rather feel the actual value by playing. Majority of values in all sorts of games are hidden from the player and they just intuitively learn them over time by playing. That's normal.
|
your main reason is, that high ground advantage would give more expanding/teching options. but as a zerg player i cannot agree.
1.) expanding: if you expand (to your nat) you loose your high ground advantage, so how are you supposed to hold an expansion with the help of high ground?
2.) you tell us teching is impossible? in zvp and zvt i can see A LOT of teching. just like in bw! there are voidray, phoenix rushes, fast tech to immortals, and in zvt fast banshee, and so on.
the only rush that could counter it is a roach rush, and i think the defender bonus is still large enough to hold a roach rush with cannons/bunkers and the choke and tech up.
so are you only talking about tvt pvp tvp? if not, please give an example how teching is denied in a certain situation!
|
Has anyone also considered the possibility of other positional advantages that might add some flavor to the game? Things like cover and flanking are the first things that come to mind.
Cover: Basically the fact that a target is not fully exposed makes it harder to hit. Things like buildings, terrain features, even other units provide cover. The idea here would be to make positioning of troops and choice of target more of a conscious decision on the part of the player. "Do I try to focus fire down the ghosts/marrauders in the back of the pack and risk missing with my ground forces, or do I try to take out the meatshields that are on the front line?" " Do I try to take a position from the front, or do I need to go around and try to flank it and reduce a cover bonus?"
Flanking: Something that I think would add a lot of flavor to the game and perhaps deemphasize single massive armies. The idea here is that there is some reward to engaging an opponent on multiple sides perhaps a flat damage increase. There could be interesting ways to make this apply a bit differently to multiple races, perhaps you wouldn't get a bonus against a protoss unit if its shields were still up. Maybe ultralisks and or thors would have more damage reduction on attacks from the front, but be vulnerable to attacks from the sides and behind.
In general the idea is to generate more decisions for a player to make, other than right now it is "he is building unit x, which means I need to add unit y to my forces in equivalent supply." Some questions that might come up are: "Do I really want to assault that position with ground forces, or do I really need to hit it with some air and ground in order to reduce a cover/highground advantage." Do I really want to clump all of my forces into a single large army and reduce the overall effectiveness of my units because they have to shoot through the space that other friendly units occupy?" "Do I have the possibility of being flanked and how does that influence where I need to move my troops and have a backup force?"
Anyway just some thoughts, maybe I'll elaborate later on.
|
France2061 Posts
On March 17 2010 20:04 spad12 wrote:
Has anyone also considered the possibility of other positional advantages that might add some flavor to the game? Things like cover and flanking are the first things that come to mind.
Cover: Basically the fact that a target is not fully exposed makes it harder to hit. Things like buildings, terrain features, even other units provide cover. The idea here would be to make positioning of troops and choice of target more of a conscious decision on the part of the player. "Do I try to focus fire down the ghosts/marrauders in the back of the pack and risk missing with my ground forces, or do I try to take out the meatshields that are on the front line?" " Do I try to take a position from the front, or do I need to go around and try to flank it and reduce a cover bonus?"
Flanking: Something that I think would add a lot of flavor to the game and perhaps deemphasize single massive armies. The idea here is that there is some reward to engaging an opponent on multiple sides perhaps a flat damage increase. There could be interesting ways to make this apply a bit differently to multiple races, perhaps you wouldn't get a bonus against a protoss unit if its shields were still up. Maybe ultralisks and or thors would have more damage reduction on attacks from the front, but be vulnerable to attacks from the sides and behind.
In general the idea is to generate more decisions for a player to make, other than right now it is "he is building unit x, which means I need to add unit y to my forces in equivalent supply." Some questions that might come up are: "Do I really want to assault that position with ground forces, or do I really need to hit it with some air and ground in order to reduce a cover/highground advantage." Do I really want to clump all of my forces into a single large army and reduce the overall effectiveness of my units because they have to shoot through the space that other friendly units occupy?" "Do I have the possibility of being flanked and how does that influence where I need to move my troops and have a backup force?"
Anyway just some thoughts, maybe I'll elaborate later on.
That sounds like Company of Heroes and other Relic games. It's a nice mechanic, but those games revolve entirely around micro and tactics. I don't think such an elaborate system would be appropriate for Starcraft, especially since the game hasn't been developed with that in mind from the get-go. Artificial cover and flanking bonuses (Starcraft had a "natural" flanking bonus) may work in low unit count, squad-based games, but it's the sort of thing that'll be unmanageable for players in a macro game like SC2, with so many units involved. And how are you going to find cover for your 150 food army anyway? Walls or potholes aren't going to protect them.
The high ground advantage and chokes/open ground provide enough tactical options.
|
Starcraft does (according to the battle.net site) have a cover bonus. Mapmakers just don't use enough doodads these days for us to ever see it.
|
Even though Idra (A professional gamer) has voiced his opinion against the random factor he uses the blanket statement of many progamers are for it, I'd like to see some documentation on that. Personally miss chance is bad and he did prove that it's POSSIBLE to miss 10 shots in a row. I still prefer the idea of -1 range while firing up hill, it's more realistic(firing up a cliff would reduce your range after all) and it would probably solve the problem of high ground advantage since the units on top get a few free shots in. Combine that with the need to scout up hill first and high ground seems to have a strong defensive advantage.
|
not to get off topic, but theres alot of chance stuff thats been taken away. mines are gone, no lurkers means no possibility of stop lurkers, no reavers mean no scarabs (though i guess the article mentions this one). and other mechanical stuff like perfect vs imperfect splits, and even glitchy units to some extent; MBS, smart casting, etc etc.
i think this is the direction that blizzard has chosen to go in. not only do they want to streamline the battles, they're striving for the player to have perfect execution each and every time. this is how they make it "accessible" to new players. so i don't think blizzard is going to change this, it certainly doesn't make sense for them to do so in the direction they're going in.
|
Just to complete the list of alternatives in the OP. Instead of penalizing units that shoot from low ground, it is of course also possible to give bonuses to the high ground units. The net effect of this is of course similar, but there are some subtle differences. (Also bonusses are generally more fun than penalties, right?) This gives the following three additional alternatives.
Increased damage for high ground units
Units firing from higher ground do +x% damage
This is very similar from the reduced damage for units firing from lower ground option discussed in the OP. It has the same drawback of having a very nonlinear effect across the board.
Increased range
Units firing from higher ground have increased range
From all options mentioned this might actually make the most sense in terms of real world logic. (Not that that should be an argument) Similar arguments on pro and con side as for the decreased range mechanic.
One subtle difference is that the player on the high ground is choosing the affected units. This means that defender can choose units that would maximize the effect this mechanic (increasing the effectiveness of the mechanic). Whereas, in the penalty case the player on the lower ground chooses with what unit he attacks, obviously in a way that he is minimally affected by the penalty, possibly nullifying the effect. (Sieged tanks firing up a cliff are not going to be too worried about the decrease in range.)
This makes the a range bonus (IMO) slightly more attractive than the range penalty option. In my opinion this option could also be interesting as an alternative.
increase firing rate
Units on higher ground fire fast
Logically, this makes even less sense to me than the reduced firing rate. For the rest similar arguments apply. The effect is linear, etc.
To me the reduced firing rate and increased range options seem the most attractive. I'm not convinced by the pro randomness arguments in the OP. IMHO unpredictability in starcraft should result from small differences leading to big effects (chaoticness for the mathematically inclined), rather than from an RNG. The reaver scarab thing is an example of this. As far as I know scarab path finding is 100% deterministic (i.e. no RNG is involved). This means that whether a scarab lands or not is a 100% in control of the players. Even if it is effectively impossible to really predict the outcome, some players (thru instincts gathered by experience) will be better at landing their scarabs then others.
|
In reality nothing is absolutely deterministic: ΔxΔp≥ℏ/2
Honestly, with RNG the thing that separates the truly great players from the good ones is how they will respond when RNG does not act in their favor. The goal for an RNG system for a high ground advantage would be a system that has some RNG, but the RNG effect is not so great that a player cannot recover from a bad RND streak.
Everything has a finite probability associated with it. Sometimes that probability is so small that we approximate it to be deterministic.
|
Just to clarify, are the reasons Blizzard removed high ground miss-chance and doodad miss-chance because:
1. They don't like probability based mechanics.
2. They think it makes the game unintuitive.
Are those the only reasons?
|
The old mechanic also wouldn´t work as intuitivly with Brush and Watchtowers. It also doesn´t emphatise the advantage of air-superiority as much.
|
On March 18 2010 02:22 spad12 wrote:
In reality nothing is absolutely deterministic: ΔxΔp≥ℏ/2
[offtopic]
That is a popular misconception about quantum mechanics. QM is deterministic. Only the semi-classical approximation to QM is non-deterministic. In any case that is completely irrelevant for starcraft, since on the level of computer electronics ℏ=0, and everything is determinstic. That includes any RNG.
|
In trying to balance three entirely different races, Blizzard has to make some compromises. One of those compromises is making maps very uniform. Everything is symmetrical, has a choke, a natural expansion, etc.
If high ground was made significantly more powerful, you'd have disparity between maps where your choke is a ramp and where is is flat. This would be an additional factor that would need to be balanced or all maps would have to be standardized. The maps are boring enough as it is.
|
On March 16 2010 21:39 Manifesto7 wrote:Show nested quote +On March 16 2010 21:24 no_re wrote:
To me this article seemed full of holes, bias and exaggerations =[ Which... you don't care to point out? Other than the silly lottery analogy (which people are missing the forest for the trees on) I don't see where the bias is. None the people who have complained about stating so strongly that there is a problem have presented evidence that there is no problem.
Evidence for the fact that there is no problem with SC2's current cliff dynamic is easy to present.
First, if you believe there's a "problem" with the cliff dynamic because you want players to have a strong high ground, the no vision approach is an incredibly strong mechanic. One player can decimate an entire base from high ground all because his opponent had no air units to provide vision. Then, alternatively based on that, if the low-ground player has vision he can completely negate this high ground advantage (which arguably adds another layer of strategy). (If fact, I believe this makes high ground and its simple vision counter TOO extreme if anything. I'd prefer a cliff advantage for SC2 that's less powerful without vision and more powerful with vision than how the game currently has it configured.)
Second, if you believe there's a "problem" with the cliff dynamic because players need more automatic defensive advantages for tech-tree investment, you can easily argue that such a problem comes from some other dynamic present in the game (before we blame the current cliff advantage). Most obvious would be the incredibly weak nature of defensive structures in SC2. Defensive structures have traditionally been the very things that were supposed to counter early-game units while being easily countered by late-game units. If this obvious, early-game role for defensive structures does not exist in SC2, perhaps we should petition to have that fixed before we ask Blizzard to "fix" SC2's cliff dynamic to provide a similar effect. (Plus, one could argue that not every map should begin a player's defensive position on high ground.)
Even more troubling was the way the article dismissed criticisms that said his idea would result with random wins. Daigomi acted like these criticisms were easily and obviously wrong when he provided no truly strong arguments to justify that claim (and, as others have pointed out, many of his arguments were very weak). While I'll admit he could be right that StarCraft 2 has enough back and forth combat to make any random wins next to impossible (I'm no expert on how chance manifests itself in games), many of my doubts about that remain (I could list them if you're interested).
Most troubling, however, was the way he didn't include a very wide list of alternatives. Making chance behave "more like it should" by putting a chance limiter on his miss percentage (as a couple people have pointed out in this thread) is a good way to have what he wants while keeping the game much more predictable. Also, removing damage-to-armor bonuses when a unit attacks high ground (as one person suggested) or introducing a new balance system into the unit damage system for high ground in the first place (to give strengths and bonuses to specific units when attacking to/from high ground) would be easier to balance than a flat, %-based reduction but it would otherwise function the same. However, he didn't mention either of these possibilities (or others that people mentioned).
All in all, for claiming to be impartial with this issue and claiming to consider everything important, this article failed to actually meet that claim. I thought it was a great dissertation otherwise. I don't see why it needed to pollute itself with obvious exaggerations at those few points.
|
Your assertion about "Miss Chance and Damage Reduction" is incorrect.
/.../
My claim
-50% miss: Three shots kill a marine
-50% less damage: Three shots kill a marine
Actually, as has been pointed out before, the required number of hits to kill a unit is affected proportionally to the miss chance, so that if it takes n shots to kill a unit with 0% miss chance, it will take n/(1-p) shots (on average) to kill the same unit with p% miss chance.
Your calculations are wrong in that you assume that the damage dealt by a tank is always 35, which is not the case. The first hit on a marine deals 35 damage, but the second hit deals only 5 (because the marine only has 40-35=5 hit points left), so that the average damage dealt is 20 per hit, and, with 50% miss chance, 10 per shot, not 17.5 as in your calculations. Correcting the numbers in your calculation, we get 20*0.5n>=40 -> n=4, as expected.
|
South Africa4316 Posts
On March 19 2010 12:12 Tiptup wrote:Show nested quote +On March 16 2010 21:39 Manifesto7 wrote:On March 16 2010 21:24 no_re wrote:
To me this article seemed full of holes, bias and exaggerations =[ Which... you don't care to point out? Other than the silly lottery analogy (which people are missing the forest for the trees on) I don't see where the bias is. None the people who have complained about stating so strongly that there is a problem have presented evidence that there is no problem. First, if you believe there's a "problem" with the cliff dynamic because you want players to have a strong high ground, the no vision approach is an incredibly strong mechanic. One player can decimate an entire base from high ground all because his opponent had no air units to provide vision. Then, alternatively based on that, if the low-ground player has vision he can completely negate this high ground advantage (which arguably adds another layer of strategy). (If fact, I believe this makes high ground and its simple vision counter TOO extreme if anything. I'd prefer a cliff advantage for SC2 that's less powerful without vision and more powerful with vision than how the game currently has it configured.)
That's exactly my argument. The current cliff mechanic is either game changing or useless, where it should rather be reasonable advantage all the time. However, I would argue that the current mechanic will more often be useless than overpowered. Players know what advantage the higher ground offers, so having a few air units present for big battles will become the norm, completely nullifying the advantage. So yeah, I agree, you need a more stable mechanic that gives an stable advantage all the time.
Second, if you believe there's a "problem" with the cliff dynamic because players need more automatic defensive advantages for tech-tree investment, you can easily argue that such a problem comes from some other dynamic present in the game (before we blame the current cliff advantage). Most obvious would be the incredibly weak nature of defensive structures in SC2. Defensive structures have traditionally been the very things that were supposed to counter early-game units while being easily countered by late-game units. If this obvious, early-game role for defensive structures does not exist in SC2, perhaps we should petition to have that fixed before we ask Blizzard to "fix" SC2's cliff dynamic to provide a similar effect. (Plus, one could argue that not every map should begin a player's defensive position on high ground.)
This doesn't contradict my point at all, it simply shows that there are multiple reasons for the problem, something which I will once again agree with wholeheartedly. Yes, defensive structures are weak, but so is the higher ground.
Even more troubling was the way the article dismissed criticisms that said his idea would result with random wins. Daigomi acted like these criticisms were easily and obviously wrong when he provided no truly strong arguments to justify that claim (and, as others have pointed out, many of his arguments were very weak). While I'll admit he could be right that StarCraft 2 has enough back and forth combat to make any random wins next to impossible (I'm no expert on how chance manifests itself in games), many of my doubts about that remain (I could list them if you're interested).
You say that no strong arguments were made, but you do not show why the arguments were weak. Some people have raised criticisms, most of which I have addressed. How specifically were the arguments weak?
Most troubling, however, was the way he didn't include a very wide list of alternatives. Making chance behave "more like it should" by putting a chance limiter on his miss percentage (as a couple people have pointed out in this thread) is a good way to have what he wants while keeping the game much more predictable. Also, removing damage-to-armor bonuses when a unit attacks high ground (as one person suggested) or introducing a new balance system into the unit damage system for high ground in the first place (to give strengths and bonuses to specific units when attacking to/from high ground) would be easier to balance than a flat, %-based reduction but it would otherwise function the same. However, he didn't mention either of these possibilities (or others that people mentioned).
If this is the most troubling aspect of the article, then I am unperturbed. What you are saying I did not cover simply fell outside of the scope of the article. The purpose of the article was to clear up some of the misconceptions regarding miss chances to allow a constructive debate to take place. As such, it was not crucial that I cover all the alternatives, as they are not part of the misconceptions. The ones I did look at were mentioned very briefly to give readers an idea of what the alternatives were, but it was never intended to be a comprehensive examination of the alternatives. If you believe that these require more attention, feel free to write an article on them, and I will feature it on the front page.
All in all, for claiming to be impartial with this issue and claiming to consider everything important, this article failed to actually meet that claim. I thought it was a great dissertation otherwise. I don't see why it needed to pollute itself with obvious exaggerations at those few points.
I'm glad you thought it was good. Most of the criticisms so far have been misunderstandings, or people not thinking the issue through thoroughly (or in the case of maths, applying the wrong formulas). However, in between people have raised many interesting points, and I have enjoyed reading the debate in this thread.
Just a quick note, I made one reply a few pages back which briefly addressed some of the criticisms. If you haven't read it, just quickly check it out before replying 
|
i think adding high ground advantage will also fix the mass roach problem with ZvZ
|
|
|
|
|
|
EPT
