One of the key differences of miss chance vs. other systems is touched on in various examples, but I would encapsulate it like this: high ground miss chance advantage has a psychological effect that the other systems don't. (In the spirit of listing features, let's not outright call that a shortcoming.) It is scary to attack high ground in SC1 because you don't know exactly what's going to happen. I love watching the pros dance back and forth on Heartbreak Ridge. From the SC2 I've watched and played, this particular element is almost entirely missing. In my vision of Starcraft at its finest (the RTS, right?)--a battle of wits requiring focus and interface control--the tactical palate is greatly enriched with the psychological element of miss chance. Whether that, or SC1 defender advantage as we know it, is appropriate to the style of SC2 is a different question. I have faith that creative top players can invent new tech-based strategies regardless, and a solidifying build order metagame will help inform them. I really like watching players exploit minimal defense for econ and tech growth though. I think that heightens the fundamental RTS tension of military threat vs. long term power. Which is the most fun for connoisseur spectators.
[SC2B] Missing the Point - Page 10
Forum Index > News |
EatThePath
United States3943 Posts
One of the key differences of miss chance vs. other systems is touched on in various examples, but I would encapsulate it like this: high ground miss chance advantage has a psychological effect that the other systems don't. (In the spirit of listing features, let's not outright call that a shortcoming.) It is scary to attack high ground in SC1 because you don't know exactly what's going to happen. I love watching the pros dance back and forth on Heartbreak Ridge. From the SC2 I've watched and played, this particular element is almost entirely missing. In my vision of Starcraft at its finest (the RTS, right?)--a battle of wits requiring focus and interface control--the tactical palate is greatly enriched with the psychological element of miss chance. Whether that, or SC1 defender advantage as we know it, is appropriate to the style of SC2 is a different question. I have faith that creative top players can invent new tech-based strategies regardless, and a solidifying build order metagame will help inform them. I really like watching players exploit minimal defense for econ and tech growth though. I think that heightens the fundamental RTS tension of military threat vs. long term power. Which is the most fun for connoisseur spectators. | ||
Legionnaire
Australia4514 Posts
Decrease any armor factor for units on low-ground. These options are probably inferior to the others, but just pointing out that there are many ways to skin a cat. Something has to be done at least. | ||
Kiwikawa
Germany3 Posts
2/(1-p) Meaning: With a miss chance of 0.5 (=50%) the expected number of shots is 4. Math + Show Spoiler + Consider a model where a tank shoots at one marine at a time. The tank shoots as long at the marine which he is firing at is alive. When the tank kills the current marine he proceeds with the next. We are looking at the shoots of the tank. We can classify each shoot as 1. a shoot which didn't kill the marine and after which the marine was hit 0 times. 2. a shoot which didn't kill the marine and after which the marine was hit 1 times. 3. a shoot which killed the marine (meaning that he was hit a second time) We can consider these three possibilities as states of a markov chain. The matrix of this chain is [(p, 1-p, 0), (0, p, 1-p), (p, 1-p, 0)]. This chain is ergodic with the stationary distribution [p/2, 1/2, (1-p)/2]. The value (1-p)/2 is the proportion of the shoots which killed a marine with respect to all shoots. This means a proportion of (1-p)/2 of all shoots killed a marine. The mean recurrent time of state 3 (= the mean number of shoots you need to kill a marine after the shoot which killed the last marine) is 1/[(1-p)/2] = 2/(1-p). And this is the expected number of shoots you need to kill a marine. Notice that for p=0 (no miss chance) you need 2/(1-p) = 2/1 = 2 shoots to kill a marine. | ||
Deviation
United States134 Posts
On March 22 2010 00:47 Kiwikawa wrote: Hi, if p is the miss chance and a marine need to be hit two times to be killed, then the expected number of shots you need to kill a marine is 2/(1-p) Meaning: With a miss chance of 0.5 (=50%) the expected number of shots is 4. Math + Show Spoiler + Consider a model where a tank shoots at one marine at a time. The tank shoots as long at the marine which he is firing at is alive. When the tank kills the current marine he proceeds with the next. We are looking at the shoots of the tank. We can classify each shoot as 1. a shoot which didn't kill the marine and after which the marine was hit 0 times. 2. a shoot which didn't kill the marine and after which the marine was hit 1 times. 3. a shoot which killed the marine (meaning that he was hit a second time) We can consider these three possibilities as states of a markov chain. The matrix of this chain is [(p, 1-p, 0), (0, p, 1-p), (p, 1-p, 0)]. This chain is ergodic with the stationary distribution [p/2, 1/2, (1-p)/2]. The value (1-p)/2 is the proportion of the shoots which killed a marine with respect to all shoots. This means a proportion of (1-p)/2 of all shoots killed a marine. The mean recurrent time of state 3 (= the mean number of shoots you need to kill a marine after the shoot which killed the last marine) is 1/[(1-p)/2] = 2/(1-p). And this is the expected number of shoots you need to kill a marine. Notice that for p=0 (no miss chance) you need 2/(1-p) = 2/1 = 2 shoots to kill a marine. Can you create a probability curve of my chances of getting into the SC2 beta next? ^_^ | ||
Kiwikawa
Germany3 Posts
On March 22 2010 03:53 Deviation wrote: Can you create a probability curve of my chances of getting into the SC2 beta next? ^_^ Unfortunately this is out of reach. + Show Spoiler + btw, instead of using the admirable concept of markov chains you also can simply use the (multiple of the) mean of a geometric distribution. Just for the sake of completeness... (...not that we can ever reach completeness...) | ||
obesechicken13
United States10467 Posts
I too hope Blizzard reintroduces the miss chance concept. It makes the most sense to me and as long as the randomness is controlled it will be fine. For example, the blademaster in WC3 has a 15% chance of a critical hit. This value is true even If it get's 10 critical hits in a row, because after the ten hits, it will have a lower chance of getting successive critical hits meaning that the final percentage of critical hits it gets in a game will be about 15%. Think of it as applying the "law of averages" like when you're rolling a dice and you notice that five hasn't come up after 20 throws. You think five should come up soon, and is more likely to come up next than say... six. Blizzard turned this into math. Though to be fair, a blademaster getting 10 crits in a row would kill two of any hero, leading to a win so this is a bad example T_T. | ||
Tiptup
United States133 Posts
On March 20 2010 06:17 Daigomi wrote: 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? I don't really believe your primary arguments were weak. in fact, on the whole I'd say my concerns about random issues are now more on your side now than anything should blizzard go in that direction (they don't really bother me as much now). I believe your arguments were lacking only in the sense that they failed to fully achieve the way they were presented. Stronger statements in the article like making "the misconceptions" into a very unquestionable-looking list and saying things like "they are all almost entirely wrong" need stronger proof than what I read. In particular, some of the arguments to reject certain ideas, like referring to poker or a lottery, were highly questionable. The same went for my criticism of the lack of alternatives mentioned. It's not that your treatment of alternatives was not vigorous enough for your purposes, it was the tone of the article, as if everything it had to say were the final word of truth, that made the lack of alternatives into a criticism from me. As for details, the following is why I'd say your arguments can at least be questioned (starting from the "It is the same as damage reduction, just worse" section and going down): 1. "Easier to balance" doesn't usually mean "better." A game that is more difficult to balance is often more difficult to understand when playing and that can often be good. A complex game system allows different players to have more or less skill with it. The best games in the world are complex in this way. As such, while straight damage modifier would give us a more complex game, that could potentially be beneficial for the sake of fun. 2. You don't need a unit to miss ten times in a row in StarCraft 2 to greatly throw off its predictable unit balance. Many of the unit matchups involve a low hit/kill ratio. Two Siege Tanks missing their initial volley at a group of Zerglings could have a large affect on whether they die or not. 3. In light of number 2, your poker analogy (as someone else pointed out) breaks down when you consider how many players will often have a string of bad luck in poker. If that were to happen at a critical moment in StarCraft, that would be game changing. 4. You're wrong to say that StarCraft 2 is not "all in." First, with a game like StarCraft 2, determining whether a player is all in or not is highly relative and depends a great deal upon the circumstances. On the whole, though, we can certainly say that StarCraft 2 is far more "all in" than in the original StarCraft. In the new game it seems like players are much more inclined to mass a large army at the center of the map and attack the biggest enemy targets (even as their own base is being destroyed by specialty units). This means that each battle is more critical. As such, I'd imagine that, for example, attacking up a ramp with your primary army might become problematic with a large miss chance. 5. A 1% loss or gain due to unpredictable unit balance would certainly not be a concern, but it seems plausible to me that certain unit matchups could sometimes result in much larger changes. Losing an extra 10-20% of your marines at the start of a single battle could have devastating effects. Whereas before your whole group of Marines would have won the battle, your whole group now gets hunted down by Zerglings. That kind of an upset would be dramatic for determining which side wins the game. 6. In StarCraft 2 it is very difficult to retreat due to the game's focus. As I said in number 4, the game is heavily centered on big army clashes. (Unique strategies, like harassment, now come less from the disorganized scope of the game, as in SC1, and more from specialty units designed for mobility and/or concealment.) It currently seems to me that if you lose even a slight amount on a battlefield, your enemy will usually push you into your base and finish you off right away. Sure, as you said, chokes and ledges would prevent an army on higher ground from following your retreating army right away, but that shouldn't slow down a pursuit by too much. StarCraft 2 units are very mobile and move much more smoothly than units in SC:BW. 7. In StarCraft 2 (like in StarCraft 1) it is very difficult to retreat a unit because of how fast combat moves. A little missed micro and you've lost a lot of guys. That would be pretty unforgiving in the case of some really bad luck. In fact, StarCraft 2 has combat that moves even faster than StarCraft 1 (if you ask me). At my skill level, when even closely watching a battle I'm losing half my guys before pulling them back. (Hah. It really sucks.) 8. I agree it would be fun to see unit matchups less than precisely predictable when "cheese" maneuvers are being carried out. However, as other people have pointed out (and you answered) cheese is already very unpredictable and random due to player choices and their ability to keep track of complex variables. There's really no need to add a further potential benefit based on chance (though I agree the affect might be fun). 9. Unpredictability for the sake of spectators is something already present in StarCraft 2. While seeing a scarab not explode as it should have is certainly a compelling event, I'd imagine there are already enough compelling events that we don't need random misfires built into the game. Plus, I find that games are more enjoyable when you're rewarded for being good enough to know what's going to happen when that result is a difficult thing to predict. 10. I prefer games where every result provided by the game is potentially predictable in some way. I may be weird but I find it more special when players (people's actions) are what provide random results in a game. It just seems to make a more ideologically pure game to me. I wouldn't want a first-person shooter to have random miss chances, so why must I be okay with the same thing being in StarCraft? Of course, that said, StarCraft isn't a game that has the luxury of being ideologically pure game and if a chance miss against high ground solves a lot of problems (and doesn't really add any) I'm fine with it, but that still wouldn't seem quite as fun to me for some reason (and I'd hope that others could understand my sentiment here). | ||
NicolBolas
United States1388 Posts
Why should there be a higher ground advantage? Without higher ground advantage, the player with the bigger army will almost always win the battle, as there are very few tactical opportunities for the player with the smaller army. This is an interesting statement, but one that I feel does not bear up under scrutiny. Your statement here suggests that the high ground advantage matters primarily for defending one's main base and natural expo from direct ground attacks. So I'll examine the statement from that perspective: the player wanting the HGA is trying to play defensively and hold their main/natural. Admittedly, since the Beta hit, I haven't been following SC1 very much. However, I seem to recall quite a few SC1 maps that had no high ground advantage in the spawn locations at all. Heartbreak Ridge, Medusa, etc. Some SC1 maps like Colosseum prefer to put expansions on the high ground rather than the main, while still allowing them to cover the choke point. And there are even some SC1 maps that went so far as to have inverse high ground advantage, where the natural was up a hill and thus the aggressor had the advantage. So I have to ask. If SC1 was capable of having reasonably balanced maps that reverse the high ground advantage, stripping it from the attacker and giving it to the defender, how important could having the HGA possibly be to SC1's style of play? This suggests that the "bigger army will almost always win the battle" phenomenon may be caused by something other than HGA that can be taken away. However, let's ignore that. Let's assume that you are correct, that in order to effectively be able to defend your main/natural, in order for a smaller army to beat a bigger one, you need to have some form of defensive advantage. And SC2 gives you one in the form of its high ground mechanics. The HGA in SC2 comes in two parts, like the HGA in SC1. Part 1 is what has been discussed: units on high ground cannot be attacked by units on low ground unless you can gain vision on the high ground. Part 2 is something that people seem to freely ignore, but is very important: melee units can't attack uphill. Nothing can defeat this advantage. From the perspective of a Terran player, this is naturally a meaningless statement. The units that a Terran player has none of can't do something. So what? From the perspective of a Zerg player like myself, this means quite a lot. Roaches have a range of 3. And even with the 3D perspective that reduces the effective size of cliffs, that doesn't leave the Zerg many options for attacking uphill. Far more units on the cliff are able to attack the Roaches than Roaches can attack up the cliff. This effectively means that the second HGA against a SC2 Zerg player is very damaging to Tier 1 Zerg. Zerglings and Banelings are of no value if they can't get into range to attack, and Roaches for all their high Hp and damage, simply aren't good enough. Therefore, I would submit that, no matter how many Overlords you have with vision over those cliffs, it won't matter much to whether or not you can get up a ramp. Which brings up another important point. Due to the lack of Tier 1 ranged attackers for Zerg, they also have few ways of actually using HGA. Yes, in Tier 2, Zergs can get Hydralisks which have the range needed to actually use cliffs. But until then? Zerg have little to go on. This is also around the time that you claim that the HGA in SC2 becomes ineffective. But my point is this: the utility of HGA, even when effective, is race specific. And the Zerg do not have the units to make effective use of it early on. So even if we were to make a more permanent HGA, you still would have to shuffle Zerg units around and rebalance everything to make it useful for them. SC2 has used this HGA mechanic for years now, while it has been in development. It is as central to the race balance as mining rates; any change to those requires adjusting the cost of every unit, usually in non-obvious ways. Because of that, changing the HGA mechanic at this late date is... unwarranted. Not unless you had some very, very compelling evidence that it would fundamentally and dramatically improve the quality of the game. A second reason why there should be a higher ground advantage has to do with tactics. And this is a very good point: combat tactics. The Part 1 of the current SC2 HGA is very binary. You either have it, or you don't. If the aggressor lacks vision, you get free hits. They can't even so much as shoot back, even if they have units that might be able to. Similarly, if the aggressor has vision, no more free hits. No more HGA, save for the effectiveness of melee/short-ranged units. However, the thing I think you forget is that, with one exception I'll get to in a moment, there's something you can do about it. Something must be providing the aggressor with vision, yes? That something, in almost all cases, can be killed. If you kill it, they no longer have vision, and your HGA is protected. This creates different tactics than SC1. Rather than simply relying on an advantage that can only be defeated through superior firepower, you must maintain your advantage. If you want to keep those pesky Hydralisks from killing your Siege line on a hill, you must invest in Vikings to kill off Overlords/Overseers. Not a lot, mind you; a couple would be sufficient. What this does is force you to expend effort to keep your advantage. It doesn't require a ton of money, but it does require spending APM and attention. You have to watch for Overlords or Observers. And you have to kill them or make them back off. You aren't allowed to simply get a bonus and sit on it; you have to do something. And that something is combat tactics. The one problem of course is that there is nothing you can do about a ComSat scan. That really throws a wrench into a mechanic that otherwise would create some interesting play styles. Terran players basically get a HGA-breaking ability; the only real balancing agent is the fact that using it takes ~270 minerals. And if it means the difference between their timing push failing or succeeding, I'm pretty sure they'll consider it an investment. ComSat aside, if you can effectively defend your high ground from units that give them vision, then your maintain a decisive HGA. This shifts the play from how many units the aggressor has to a different tactical level: control of the air. Going out to snipe Medivacs. Using detection to snipe Obvservers. And so on. In SC1, the HGA was a fundamental stopper: it halts the aggressor's attack. In SC2, it simply continues the fight by other means, forcing both sides to now fight over air control. Both sides have something important to gain, and both sides have a lot on the line. There would be less on the line if the defender knew that he would still have an advantage against ranged units regardless of the outcome. This is interesting. This is a new dynamic. And it is one that is worthwhile. | ||
ReaverDrop!
Canada81 Posts
No to miss chances, its annoying, it slows the game down, and as the intelligent poster above me said no miss chances introduces air battles, something that was hardly seen out of zvanybody in sc1. | ||
Daigomi
South Africa4316 Posts
On March 24 2010 10:59 Tiptup wrote: + Show Spoiler + Sorry I didn't reply to this right away. I've been playing a lot of the StraCraft 2 beta (and Demon's Souls for the PS3) in my spare time the last few days. On March 20 2010 06:17 Daigomi wrote: 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? I don't really believe your primary arguments were weak. in fact, on the whole I'd say my concerns about random issues are now more on your side now than anything should blizzard go in that direction (they don't really bother me as much now). I believe your arguments were lacking only in the sense that they failed to fully achieve the way they were presented. Stronger statements in the article like making "the misconceptions" into a very unquestionable-looking list and saying things like "they are all almost entirely wrong" need stronger proof than what I read. In particular, some of the arguments to reject certain ideas, like referring to poker or a lottery, were highly questionable. The same went for my criticism of the lack of alternatives mentioned. It's not that your treatment of alternatives was not vigorous enough for your purposes, it was the tone of the article, as if everything it had to say were the final word of truth, that made the lack of alternatives into a criticism from me. As for details, the following is why I'd say your arguments can at least be questioned (starting from the "It is the same as damage reduction, just worse" section and going down): 1. "Easier to balance" doesn't usually mean "better." A game that is more difficult to balance is often more difficult to understand when playing and that can often be good. A complex game system allows different players to have more or less skill with it. The best games in the world are complex in this way. As such, while straight damage modifier would give us a more complex game, that could potentially be beneficial for the sake of fun. 2. You don't need a unit to miss ten times in a row in StarCraft 2 to greatly throw off its predictable unit balance. Many of the unit matchups involve a low hit/kill ratio. Two Siege Tanks missing their initial volley at a group of Zerglings could have a large affect on whether they die or not. 3. In light of number 2, your poker analogy (as someone else pointed out) breaks down when you consider how many players will often have a string of bad luck in poker. If that were to happen at a critical moment in StarCraft, that would be game changing. 4. You're wrong to say that StarCraft 2 is not "all in." First, with a game like StarCraft 2, determining whether a player is all in or not is highly relative and depends a great deal upon the circumstances. On the whole, though, we can certainly say that StarCraft 2 is far more "all in" than in the original StarCraft. In the new game it seems like players are much more inclined to mass a large army at the center of the map and attack the biggest enemy targets (even as their own base is being destroyed by specialty units). This means that each battle is more critical. As such, I'd imagine that, for example, attacking up a ramp with your primary army might become problematic with a large miss chance. 5. A 1% loss or gain due to unpredictable unit balance would certainly not be a concern, but it seems plausible to me that certain unit matchups could sometimes result in much larger changes. Losing an extra 10-20% of your marines at the start of a single battle could have devastating effects. Whereas before your whole group of Marines would have won the battle, your whole group now gets hunted down by Zerglings. That kind of an upset would be dramatic for determining which side wins the game. 6. In StarCraft 2 it is very difficult to retreat due to the game's focus. As I said in number 4, the game is heavily centered on big army clashes. (Unique strategies, like harassment, now come less from the disorganized scope of the game, as in SC1, and more from specialty units designed for mobility and/or concealment.) It currently seems to me that if you lose even a slight amount on a battlefield, your enemy will usually push you into your base and finish you off right away. Sure, as you said, chokes and ledges would prevent an army on higher ground from following your retreating army right away, but that shouldn't slow down a pursuit by too much. StarCraft 2 units are very mobile and move much more smoothly than units in SC:BW. 7. In StarCraft 2 (like in StarCraft 1) it is very difficult to retreat a unit because of how fast combat moves. A little missed micro and you've lost a lot of guys. That would be pretty unforgiving in the case of some really bad luck. In fact, StarCraft 2 has combat that moves even faster than StarCraft 1 (if you ask me). At my skill level, when even closely watching a battle I'm losing half my guys before pulling them back. (Hah. It really sucks.) 8. I agree it would be fun to see unit matchups less than precisely predictable when "cheese" maneuvers are being carried out. However, as other people have pointed out (and you answered) cheese is already very unpredictable and random due to player choices and their ability to keep track of complex variables. There's really no need to add a further potential benefit based on chance (though I agree the affect might be fun). 9. Unpredictability for the sake of spectators is something already present in StarCraft 2. While seeing a scarab not explode as it should have is certainly a compelling event, I'd imagine there are already enough compelling events that we don't need random misfires built into the game. Plus, I find that games are more enjoyable when you're rewarded for being good enough to know what's going to happen when that result is a difficult thing to predict. 10. I prefer games where every result provided by the game is potentially predictable in some way. I may be weird but I find it more special when players (people's actions) are what provide random results in a game. It just seems to make a more ideologically pure game to me. I wouldn't want a first-person shooter to have random miss chances, so why must I be okay with the same thing being in StarCraft? Of course, that said, StarCraft isn't a game that has the luxury of being ideologically pure game and if a chance miss against high ground solves a lot of problems (and doesn't really add any) I'm fine with it, but that still wouldn't seem quite as fun to me for some reason (and I'd hope that others could understand my sentiment here). Fuck. I just wrote a huge reply that got swallowed up by an accidental close. Anyway, I'll summarise it for you. This means that I won't cover each point in great depth, but it doesn't mean they don't have the depth. I agree that the article was overstated, but, given the situation, it was necessary. While all the misconceptions have a grain of truth, like it is physically possible to miss 10 shots in a row, they are all either exaggerated or inconsequential. The position was overstated in order to avoid certain arguments(like "since it is possible to miss 10 shots in a row, miss shots should be excluded"). Instead, I wanted people to argue about the real issues here, the things that can have a feasible impact on the game rather than a possible impact in one game out of a million. 1. Easier to balance is an advantage, but decisions should not be based on it. When options are equal, then it should be considered. Artificial complexity raises the learning curve without raising the skill ceiling, so it is generally a bad thing. 2. Missing five times is still very unlikely. Three times are plausible, but if three times can lose you a battle, then you should know that you are taking a risk by engaging in the first place. This can be a calculated risk, but not all calculated risks pay off. 3. The misconception is that one piece of bad luck will cost you the game, which I argued is very unlikely. A string of bad luck can cost the game, just like in any other sport it can, but not only is it even more unlikely to have a string of bad luck that one case of bad luck, it is just as unlikely that the bad luck will be so important that it will determine a game. Also, when you come up with a situation where a miss plays a chance, try to imagine how a player would get in that situation in a real game and what effect it would have. How many games will two siege tanks be shooting at zerglings on higher ground, and have the outcome of the game depend on them hitting their first shots? I can't think of any games where this seems likely, and in the few games where it could occur, only 6.25% of them will be spoiled by the "bad luck". 4 & 6. Players are more likely to mass attack in the middle, but this is not relevant to HGA (although ironically, it could be fixed by HGA), since battles in the middle are not influenced by HGA generally. Battles occur up a hill in one situation, when the attacker has a larger force and is trying to overrun a smaller force. In such a situation, the attacker can always retreat early enough and kill anything following, while the defender can retreat and try to hold the ramp (an even bigger tactical advantage). 5. Related to #4, but lets quickly create a scenario: You have 30 marines holding the higher ground vs. hydras (roaches would prove my point even better). You lose 20% of your marines unexpectedly. It happened so quickly that you did not have time to retreat. If you were expecting to lose only 6 marines in the first 10 seconds, then this situation is impossible, since the expected shots to kill 6 marines is at 25% miss is 32, which, if fired perfectly, is only enough to kill 8 marines, not 12. In order to kill 20% of your army unexpectedly, at least 60% of your army needs to engage, in which case the hydras need to shoot 96 times in a row without missing to kill an unexpected 20%. Even if you expected to lose 80% of your marines (in which the impact of losing an additional 20% becomes questionable), the hydras would need to hit 120 times out of 128 shots to make this happen, or they would need to miss 6.25% of the time in a sample of 128 shots, something that is seriously unlikely. My point is not that it cannot conceivably happen, but that these situations which are easy to think of are very unlikely in game. Even more unlikely is that it happens so quickly that you cannot retreat. The only situation where a significant loss quickly is feasible is when marines attack tanks. However, the player with the marines would be aware of the risk in such a situation. 7. Covered already. The only units that really have a chance of quickly and unexpectedly killing units on higher ground in SC1 were tanks, lurkers, and reavers. Only tanks remain, and they attack very slowly. The only time when they can really unexpectedly wipe out your army is when they only need to hit once or twice, and in such situations the attacker will know that there is a risk involved. 8. As you yourself said, very few cheeses are affected by HGA. So basically, there are things adding a big element of randomness to cheeses, and then HGA adds a little bit more to a very small percentage of cheeses. This is definitely not a big concern. 9. Watching players make difficult predictions is exactly why it's important to have miss-chances. Damage reduction is easy. 10 marines beat 5 on HG but lose to 8. With misschances, 10 marines should beat 5 and lose to 8, but some lucky target firing against the 8 could lead to 10 vs 7, which is a new decision. Furthermore, your implication is that random misfires are built in to make it compelling is clearly false. Random misfires are built in for many reasons, it just happens to make it more compelling too. 10. The most competitive FPS games all have predictable miss-chances included, the only difference being that it is much more predictable in SC and has a much smaller impact. CS is the most obvious example, but even Q3 has the machine gun which is not accurate (otherwise it would be overpowered) and the plasmagun which could be said to have a misschance (much more so than any other gun, the PG is expected to miss a certain number of times). Good players cope with misschances, and take advantage of them, and as I said, this is a much, much smaller part of the game in SC. Compare the bad luck of a random miss-chance with the bad luck of scouting the wrong starting position. Which one is the most likely to happen, which one has the largest impact on the game, and which one adds the most to the game strategically? The fact that people don't complain about starting locations shows why the effects of miss-chance are grossly exaggerated, and why my article was a bit overstated to make people realise that. On March 24 2010 12:52 NicolBolas wrote: [/spoiler][/spoiler]I have to take issue with the entire impetus of this article. This is an interesting statement, but one that I feel does not bear up under scrutiny. Your statement here suggests that the high ground advantage matters primarily for defending one's main base and natural expo from direct ground attacks. So I'll examine the statement from that perspective: the player wanting the HGA is trying to play defensively and hold their main/natural. Admittedly, since the Beta hit, I haven't been following SC1 very much. However, I seem to recall quite a few SC1 maps that had no high ground advantage in the spawn locations at all. Heartbreak Ridge, Medusa, etc. Some SC1 maps like Colosseum prefer to put expansions on the high ground rather than the main, while still allowing them to cover the choke point. And there are even some SC1 maps that went so far as to have inverse high ground advantage, where the natural was up a hill and thus the aggressor had the advantage. So I have to ask. If SC1 was capable of having reasonably balanced maps that reverse the high ground advantage, stripping it from the attacker and giving it to the defender, how important could having the HGA possibly be to SC1's style of play? This suggests that the "bigger army will almost always win the battle" phenomenon may be caused by something other than HGA that can be taken away. However, let's ignore that. Let's assume that you are correct, that in order to effectively be able to defend your main/natural, in order for a smaller army to beat a bigger one, you need to have some form of defensive advantage. And SC2 gives you one in the form of its high ground mechanics. The HGA in SC2 comes in two parts, like the HGA in SC1. Part 1 is what has been discussed: units on high ground cannot be attacked by units on low ground unless you can gain vision on the high ground. Part 2 is something that people seem to freely ignore, but is very important: melee units can't attack uphill. Nothing can defeat this advantage. From the perspective of a Terran player, this is naturally a meaningless statement. The units that a Terran player has none of can't do something. So what? From the perspective of a Zerg player like myself, this means quite a lot. Roaches have a range of 3. And even with the 3D perspective that reduces the effective size of cliffs, that doesn't leave the Zerg many options for attacking uphill. Far more units on the cliff are able to attack the Roaches than Roaches can attack up the cliff. This effectively means that the second HGA against a SC2 Zerg player is very damaging to Tier 1 Zerg. Zerglings and Banelings are of no value if they can't get into range to attack, and Roaches for all their high Hp and damage, simply aren't good enough. Therefore, I would submit that, no matter how many Overlords you have with vision over those cliffs, it won't matter much to whether or not you can get up a ramp. Which brings up another important point. Due to the lack of Tier 1 ranged attackers for Zerg, they also have few ways of actually using HGA. Yes, in Tier 2, Zergs can get Hydralisks which have the range needed to actually use cliffs. But until then? Zerg have little to go on. This is also around the time that you claim that the HGA in SC2 becomes ineffective. But my point is this: the utility of HGA, even when effective, is race specific. And the Zerg do not have the units to make effective use of it early on. So even if we were to make a more permanent HGA, you still would have to shuffle Zerg units around and rebalance everything to make it useful for them. SC2 has used this HGA mechanic for years now, while it has been in development. It is as central to the race balance as mining rates; any change to those requires adjusting the cost of every unit, usually in non-obvious ways. Because of that, changing the HGA mechanic at this late date is... unwarranted. Not unless you had some very, very compelling evidence that it would fundamentally and dramatically improve the quality of the game. And this is a very good point: combat tactics. The Part 1 of the current SC2 HGA is very binary. You either have it, or you don't. If the aggressor lacks vision, you get free hits. They can't even so much as shoot back, even if they have units that might be able to. Similarly, if the aggressor has vision, no more free hits. No more HGA, save for the effectiveness of melee/short-ranged units. However, the thing I think you forget is that, with one exception I'll get to in a moment, there's something you can do about it. Something must be providing the aggressor with vision, yes? That something, in almost all cases, can be killed. If you kill it, they no longer have vision, and your HGA is protected. This creates different tactics than SC1. Rather than simply relying on an advantage that can only be defeated through superior firepower, you must maintain your advantage. If you want to keep those pesky Hydralisks from killing your Siege line on a hill, you must invest in Vikings to kill off Overlords/Overseers. Not a lot, mind you; a couple would be sufficient. What this does is force you to expend effort to keep your advantage. It doesn't require a ton of money, but it does require spending APM and attention. You have to watch for Overlords or Observers. And you have to kill them or make them back off. You aren't allowed to simply get a bonus and sit on it; you have to do something. And that something is combat tactics. The one problem of course is that there is nothing you can do about a ComSat scan. That really throws a wrench into a mechanic that otherwise would create some interesting play styles. Terran players basically get a HGA-breaking ability; the only real balancing agent is the fact that using it takes ~270 minerals. And if it means the difference between their timing push failing or succeeding, I'm pretty sure they'll consider it an investment. ComSat aside, if you can effectively defend your high ground from units that give them vision, then your maintain a decisive HGA. This shifts the play from how many units the aggressor has to a different tactical level: control of the air. Going out to snipe Medivacs. Using detection to snipe Obvservers. And so on. In SC1, the HGA was a fundamental stopper: it halts the aggressor's attack. In SC2, it simply continues the fight by other means, forcing both sides to now fight over air control. Both sides have something important to gain, and both sides have a lot on the line. There would be less on the line if the defender knew that he would still have an advantage against ranged units regardless of the outcome. This is interesting. This is a new dynamic. And it is one that is worthwhile. Your reply got swallowed to, so I'm keeping this short. There are multiple ways to include tactics to the game, higher ground is just one. A map like Rivalry would be just as strategic in SC2 as it was in SC1. The reason for the middle battles in SC2 is because the maps are built on SC1 HGA templates without having a HGA. Draw the new maps without higher ground (which is the way they play) and they are flat, lifeless maps that naturally lead to middle-clashes. The maps you use from SC1 do not contradict my point (which in no way is my central point), in fact they often support it. Colosseum had a safe expansion, but the outside expansion was both easier to defend thanks to the higher ground, and it offered defense for your main as well. Maps like Neo Requiem show why HGA is so important. ZvP turned into mass zealots vs mass zerglings on the map, because players could not get any teching done. Your statement that there are two aspects of higher ground is true in one way, and wrong in another. Melee units cannot attack across a wall, and higher ground forms a wall, just like space does. This makes higher ground no different than any other divide, which clearly gives map makers fewer strategic options in maps. Your points are true for divides in the map in general, but there is no specific advantage to higher ground (other than making the map prettier). What could be true is that introducing a further advantage to HGA could imbalance the game, but if we assume that it adds strategy to the game, then it would be worth balancing around this. What is left is proving that it would add strategy, something for which there is more than enough evidence available. Firstly, many top players believe that it is crucial to the strategy (we have to assume that they have some understanding of what is strategic, even if it is only an innate understanding). Secondly, battles are mostly large clashes that immediately determine the result of the game, something that never happened on maps like Loki in SC1, but would most likely still happen in SC2, even if Loki was remade (imagine the higher ground as walls on Loki). Thirdly, having HGA inherently gives more strategic possibilities to map makers, which should translate into more strategic play. Fourthly, there are all the reasons already mentioned in the article. There is just no reasonable argument that adding a reasonable HGA will decrease the strategic possibilities in the game. There might be arguments that it has other disadvantages, but it cannot be said that it will not add more possibilities. Regarding the fact that spotters can be killed, I never suggested removing the current mechanic from the game. Instead, I suggested improving the HGA, since the current mechanic is insufficient on its own. The problem with the current mechanic is that it already existed in SC1. In SC1 higher ground had two advantages: a 50% miss rate, and units were revealed very briefly when they attacked. In SC2, the miss rate was completely removed, while units are not revealed anymore. However, in SC1, attacking up to higher ground without a spotter made the 50% miss chance into an 80% disadvantage, since you attacked much less. Almost no higher ground battles took place in SC1 without some kind of spotter or a massive unit advantage, and sniping the spotter was still a strategy used to give an advantage to the defender. The problem with the sight mechanic alone is that it is just too easily overcome. Not only are there tons of available spotters (and trying to snipe a few obs, the weakest spotters in the game, in between carriers in SC1 will tell you how ineffective this strategy is), but most of the time, units can just run up to the higher ground while taking a few hits. That is what was done in SC1, players used a spotter to push the defenders back, and then charged up to the higher ground. In SC2, where there is no miss-chance and there are multiple units that don't even need chokes, this is even simpler. In SC2, the HGA is not a disadvantage to the attacker. Just like detection, it is something you simply have to have and once you have it, the effects are nullified. For every banshee you build to stop Zerg from getting sight, Zerg can build a mutalisk to get sight with. Requiring spotters adds interesting possibilities to the game, but on its own it is just not strong enough in enough situations to make the game tactically diverse. | ||
NicolBolas
United States1388 Posts
What could be true is that introducing a further advantage to HGA could imbalance the game, but if we assume that it adds strategy to the game, then it would be worth balancing around this. That is a very big assumption to go ransacking the game over. We're not talking about minor unit tweaks here; you're talking about drastically changing how all of Tier 1 Zerg works. Which drastically changes how air battles work vs. Zerg, since Tier 1 Zerg has no anti-air. If you move Hydralisks down to Tier 1, they must now necessarily be weaker. And something else has to go to Tier 2. If Roaches go up to Tier 2, then you have a Zerg Tier 1 that is elementally weak to upgraded Hellions, and mass Zealots become a significant problem. If you move the Banelings, then Zerg have no defense against mass Marines. If the Zerg have air defense in Tier 1, then entire strategies revolving around fast-teching to Banshees or Void Rays simply are unworkable. And so on. I'm not convinced that an assumption is worth it. I would rather that they focus on doing things that add strategy that don't require fundamentally rebalancing every unit in the game. Firstly, many top players believe that it is crucial to the strategy (we have to assume that they have some understanding of what is strategic, even if it is only an innate understanding). Many top players also believed that Hellions were useless just a few weeks ago. Live and learn. Secondly, battles are mostly large clashes that immediately determine the result of the game, something that never happened on maps like Loki in SC1, but would most likely still happen in SC2, even if Loki was remade (imagine the higher ground as walls on Loki). So were battles in SC1 in the early days. This is simply the result of an unevolved metagame. Thirdly, having HGA inherently gives more strategic possibilities to map makers, which should translate into more strategic play. There already is HGA; the question is whether it should be something you have to fight for or not. However, in SC1, attacking up to higher ground without a spotter made the 50% miss chance into an 80% disadvantage, since you attacked much less. I think this is somewhat indicative of the problem, because this is simply not globally true. This is true of some units, but not of others. For Siege Tanks in Siege Mode, it certainly is true. For Marines, or any ranged unit that depends on rate-of-fire, it is not. You're looking at this from a very Terran perspective, and a Terran Mech perspective besides. When looked at from a Protoss or Zerg perspective, this advantage is far less important if not entirely meaningless. This is a big part of the reason why Protoss and Zerg players used HGA far less in SC1 than Terrans, and it also shows part of the problem with this kind of thinking: Siege Tank-itus. StarCraft 2 should not be a game built around a single unit. It should not be a game where mechanics are bent around making a single unit the most valuable unit in the game, and "strategy" in the game should not be measured by the utility of this unit. This is how it was for 1/3rd of SC1, and that wasn't good. Just like detection, it is something you simply have to have and once you have it, the effects are nullified. For every banshee you build to stop Zerg from getting sight, Zerg can build a mutalisk to get sight with. Requiring spotters adds interesting possibilities to the game, but on its own it is just not strong enough in enough situations to make the game tactically diverse. What you said about detection is simply not true. Corsair/DT was a fairly common Protoss tactic vs. Zerg, even though Zerg are guaranteed from day 1 to have detection. It can work, and even if it doesn't, it forces the Zerg to invest in air defenses. It creates tactical and strategic diversity. Same goes with DTs vs. Terran. Scans are a problem, but you can force them to run out of scans. And without Turrets, the Terrans are incredibly open to DT harassment. So I'm not seeing how the current HGA fails to make the game tactically diverse. The current HGA is a disadvantage to the attacker; it simply isn't a permanent one. Nor should it be. | ||
Deviation
United States134 Posts
| ||
Daigomi
South Africa4316 Posts
On March 25 2010 03:19 NicolBolas wrote: That is a very big assumption to go ransacking the game over. We're not talking about minor unit tweaks here; you're talking about drastically changing how all of Tier 1 Zerg works. Which drastically changes how air battles work vs. Zerg, since Tier 1 Zerg has no anti-air. If you move Hydralisks down to Tier 1, they must now necessarily be weaker. And something else has to go to Tier 2. If Roaches go up to Tier 2, then you have a Zerg Tier 1 that is elementally weak to upgraded Hellions, and mass Zealots become a significant problem. If you move the Banelings, then Zerg have no defense against mass Marines. If the Zerg have air defense in Tier 1, then entire strategies revolving around fast-teching to Banshees or Void Rays simply are unworkable. And so on. I'm not convinced that an assumption is worth it. I would rather that they focus on doing things that add strategy that don't require fundamentally rebalancing every unit in the game. Firstly, you are making some very big assumptions here. While I conceded that it might change the balance of the game, I never said that it definitely would. SC1, for example, can be just as balanced on flat maps as it is on maps with higher ground. The short range roach worry you mention will be, by and large, true for all tier 1 units. Furthermore, I don't see how adding a HGA will be more damaging to roaches than any other units. Roaches have difficulty attacking up to higher ground, now it will be 25% more difficult, just like it will be 25% more difficult for marines or tanks or any lower ground units. The point is, the roaches range is its disadvantage, a 25% miss doesn't make its short range a bigger disadvantage. Relatively speaking its the exact same disadvantage. Many top players also believed that Hellions were useless just a few weeks ago. Live and learn. Yes, the top players can make mistakes, but when their opinions are supported by reasonable arguments then it is worth considering. So were battles in SC1 in the early days. This is simply the result of an unevolved metagame. I also considered this, but having played the game and read the opinions of others, I just don't buy it. It is not like the players playing now have suddenly forgotten the use of tactics, of retreating, or of taking advantage of the map. The game, in its current condition, is just not conducive to positional play. Also, the difference between our arguments is that I provided reasons for why it's not the unevolved metagame, while you claimed that this is the cause without providing any evidence. There already is HGA; the question is whether it should be something you have to fight for or not. The current HGA is, in most cases, irrelevant. Yes, there is an advantage, but it doesn't allow positional play because the advantage is nullified too easily. I think this is somewhat indicative of the problem, because this is simply not globally true. This is true of some units, but not of others. For Siege Tanks in Siege Mode, it certainly is true. For Marines, or any ranged unit that depends on rate-of-fire, it is not. You're looking at this from a very Terran perspective, and a Terran Mech perspective besides. When looked at from a Protoss or Zerg perspective, this advantage is far less important if not entirely meaningless. This is a big part of the reason why Protoss and Zerg players used HGA far less in SC1 than Terrans, and it also shows part of the problem with this kind of thinking: Siege Tank-itus. StarCraft 2 should not be a game built around a single unit. It should not be a game where mechanics are bent around making a single unit the most valuable unit in the game, and "strategy" in the game should not be measured by the utility of this unit. This is how it was for 1/3rd of SC1, and that wasn't good. The one race I've never been able to play is Terran. Saying that my view of HGA is terran is absurd. In SC, Terran got the biggest advantage from HGA, no doubt. However, Zerg used HGA to defend expansions or hold important positions on the map (think hydra play on HBR or lurker defenses on maps like Katrina), and Protoss used it more than enough with dragoons. Also, ironically, the revealing issue was the smallest problem for tanks. The only time when spotters weren't used in fights from lower ground up was when tanks were part of the attack, as they still get most of their shots off. Units with fast rates of fire were hurt by it since they only fire half the time. Tanks fired whenever they were reloaded. Correct me if I'm wrong, but I get the idea that when you think of HGA you think of it as helping you defend your main. HGA is much more than that, especially on creative maps. Take HBR for example, the whole map resolves, for all races, about taking the higher ground and using the higher ground. Hydras hold it against dragoons, dragoons hold it against tanks, any player that can take the higher ground can have an advantage. The same goes for so many maps, Loki and Katrina are two examples I already mentioned, but there are tons. HGA is not the same as defending a choke. A choke is usually a bigger advantage than HGA, which is why Zerg and Protoss used HGA less, because they couldn't out range units from higher ground, so it was more effective defending the choke than engaging in a higher ground battle. Also, saying that SC1 strategy revolved around siege tanks is a stupid statement. What you said about detection is simply not true. Corsair/DT was a fairly common Protoss tactic vs. Zerg, even though Zerg are guaranteed from day 1 to have detection. It can work, and even if it doesn't, it forces the Zerg to invest in air defenses. It creates tactical and strategic diversity. Same goes with DTs vs. Terran. Scans are a problem, but you can force them to run out of scans. And without Turrets, the Terrans are incredibly open to DT harassment. So I'm not seeing how the current HGA fails to make the game tactically diverse. The current HGA is a disadvantage to the attacker; it simply isn't a permanent one. Nor should it be. Corsair/DT was anything but a common tactic. Corsairs were effective because they denied zerg scouting in general, and in the beginning this lack of scouting was taken advantage of with DTs. Corsair/DT is a surprise tactic, and it works because the other player doesn't invest similar resources into stopping your corsairs. Sniping spotters in SC2 will not be the same, because players will invest similar resources on sniping vs maintaining. For each resource you waste on getting a sniper, I can waste the same resources on defending a spotter. In the end, the air units cancel each other out, and no HGA exists. The difference between your example and SC2 is that in your example they were surprise tactics, while in SC2 it will be expected. A better (although still flawed) example is marines vs lurkers in SC1. Zerg tries his best to snipe science vessels, but in general Terran can maintain his spotters, because Terran is prepared and knows the importance of maintaining a spotter. However, even this example doesn't quite show how easy spotting is, because science vessels are considerably more expensive than scourge, and could be sniped in an instant, something which won't be true in SC2. In SC2, spotters will be roughly as strong and as expensive as those units trying to snipe them, making successful sniping much harder. Note that I don't have a problem with the mechanic, I think it would be fun to see players try to snipe, but most of the time it will be as if there is no HGA. | ||
Xastros
Australia1 Post
I am one of the people that don't agree with having too much HGA. IMO it is wrong for the OP to assume everyone agrees that there should be more HGA. SC1 pros have been so used to having ramp and HGA that I feel the views here are too biased towards having HGA. In my opinion it should NOT be a right for you to build 2-3 zealots (vs zerg) to block a ramp and tech straight to corsairs. If you choose to tech with such few units you should be taking a MAJOR risk, not a minor one. Quoting the OP "Without higher ground advantage, the player with the bigger army will almost always win the battle" <---- Not really, scouting and consequently unit composition play a part also. "This leads to a very linear game development, where both players need to mass armies in order to stay in the game"<----- I agree in the fact that both players must mass to a certain degree but this makes decision making (on when to tech/expand) play an even more critical role. In my opinion a good strategy game should not allow for such extreme fast tech/exp options as BW did (see the 2 zlot block into sairs). You should NOT be able to completely skip a whole tier of units and go straight into air or cloaked units etc. If you want those units faster than your opponent or you want an expansion earlier than your opponent you must take a BIG RISK by sacrificing unit production. This in itself requires a lot of strategic decision making. In addition it would not necessarily result in a boring game because potentially it would result in more skirmishes and timing attacks etc rather than the game ending because someone rushed straight to DTs and the other didn't scout and build detectors. In fact there would be less room for 'cheese' tactics which IMO is better for all. "it prevents the game from turning into a macro competition where large armies clash in the middle of the map to decide each match" <---- No, as previously mentioned it could result in more timing attacks. attacking with your first 10 zerglings because your opponent has chosen to tech. In late game, strategic decision making in where an when to attack also decide the match (drops, backdoor etc) rather than both players just clashing in the middle (unlikely IMO). Having said all of that I do agree that there should be a DEFENDER'S advantage (not high ground advantage). I feel overall there needs to be (i) more effective (ii) cheaper and (iii) easier to get defensive structures. The reason I feel this is preferable than HGA is that defensive structures must be PAID for by the defender, not given by right. You want to fast tech or fast expand? Yes you can by building defensive structures that are more cost effective than mobile troops BUT you have paid for them in resources and map control. There should always be a sacrifice in choosing to tech or expand, and having more powerful defensive structures allows a player to choose to sacrifice mobile troops for static defenses. You can now tech or expand but you cannot attack your enemy. In the game's current state this is not viable because of how weak and hard to get defensive structures are. My proposal to fix this: 1. Make defensive structures more powerful or cheaper allowing a defender to have an advantage. This allows for options of fast tech and FE. 2. Make defensive structures available at the very start of the game without the need for forge, gate, spool, rax etc. This is very important because it must not be too much of a sacrifice to get these defensive structures as it ALREADY makes you sacrifice map control. Additionally since it doesn't require additional tech buildings to make you can always switch out of making defensive structures to making units at no cost other than the defensive structures already built. This fixes the problem of "I see my opponent has built a forge so I can freely expand". He hasn't invested 150 in a forge and then cannon so it is easier for him to switch to building units if he scouts and feels you are trying to take advantage by expanding without building units. Yes if you see an early cannon you could probably expand once but not safely a second time. In return your opponent has used the cannons to earn an early expand himself or an early tech so it is kind of even. I don't really see the purpose in making players build a forge before cannons. It doesn't help the game overall in any way IMO. Something to consider as an alternative to HGA, which I am deadly against because as previously mentioned, I feel strongly that you should have to earn or pay for your right to tech or expand. IMO this will result in a more entertaining game which is more battle and army based rather than tech and scouting based. It would also reduce the viability of the all in 'I WIN' tactics which I totally hate (eg fast DT vs anyone, or fast phoenix vs zerg before they get hydra/mutes). These are not fun games to play or watch. | ||
teekesselchen
Germany886 Posts
No Bonus at all like it is right now enforces every player to mass units all the time. No cool all-over-the-map spread game or certain fasttechs for special unit compositions exept for a few very save ones that are more like cheese than actually considered "normal play". -> Give a nice miss chance such as 33% against high ground -> Stronger or cheaper Tanks or weaker counters against it -> Giving something to Protoss and Zerg, for example buffing Cannons hp (in BW, way less of an armys units could attack simultaniously, so everything dies faster. Buffing HP would reduce this effect ofc) Makes more tactics available, no need for investing everything for instant units, cool stuff such as lots of expansions and harassment going on or cool fast techs and so on all this stuff everyone misses. | ||
RoosterSamurai
Japan2108 Posts
| ||
scott desu
United States34 Posts
In sc2, it feels like army positioning just doesn't have much of an influence on games anymore, which makes it feel like there is less strategy involved in decision making. Most game now are decided by attack timing and army composition. Maybe its what blizzard was intending, but I don't like it. | ||
Teejing
Germany1360 Posts
By reviving this op i would like to get to know if the general consensis on this subject. The goal behind this would be Blizzard importing the suggested changes of the op. I am aware of the fact that i could get temp. banned for beating on the dead horse, but to me this is one of the two bigger problems of sc2 mechanics (the other one being the lack of t3 casters) . | ||
generic88
United States118 Posts
| ||
| ||