On January 18 2014 19:30 SolidSMD wrote: @pure.wasted Can you give a decent example of when %miss chance is flawed?
Flawed in what sense? In the sense that it is luck and luck is bad? Or in the sense that it favors some units over others? I have a feeling that you mean the latter.
Imagine two scenarios. In the first, a pack of 10 Marines bumrush a Siege Tank on high ground. In the second, a pack of 5 Marauders does the same. They get hit once on the way there. Marines clump up more so they take more splash damage. It takes JUST over one tank hit to kill a Marine, it takes three tank hits to kill a Marauder. There's a 30% chance that 6 Marines will miss/2 Marauders will miss. Let's say that happens. He gets hit by the Tank again. Marines are already dying left and right, a bunch are in red, still not a single Marauder dead even with target fire. Both attackers want to retreat now. Marines still clumped as shit, take another big hit with lots of splash. Marauders retreat at almost identical speed, take less splash damage. With target fire from the tank, one Marauder dies. How many Marines you think died? Safe bet that it was a lot more than two, right? Especially if the tank is target firing? If more than two Marines died, then the Marauder player came out way ahead because his units were tough enough to survive a bunch of hits and fast enough to still get out.
Pretty straight forward, I think.
Imagine if there was no high ground mechanic, the outcome in the case would probably be pretty much the same, possibly even a better outcome for the marauders. I really think your logic is flawed on this, clumped up marines will not perform well vs siege tanks, period, high ground mechanic or not. Smart players will not risk such a move and will rather get a dropship if they plan to be very aggressive with bio.
easiest way to explain how % miss chance can be a problem is to imagine a metagame that could arise in which a toss makes a certain number of gates and attacks up a ramp in an all-in where the result is very very close and whether 1/4 shots miss or 3/4 shots miss determines whether the all-in works or not.
Unlikely to be a big problem though. Just saying: the most volatile games in RNG terms will be the all-ins.
On January 18 2014 19:30 SolidSMD wrote: @pure.wasted Can you give a decent example of when %miss chance is flawed?
Flawed in what sense? In the sense that it is luck and luck is bad? Or in the sense that it favors some units over others? I have a feeling that you mean the latter.
Imagine two scenarios. In the first, a pack of 10 Marines bumrush a Siege Tank on high ground. In the second, a pack of 5 Marauders does the same. They get hit once on the way there. Marines clump up more so they take more splash damage. It takes JUST over one tank hit to kill a Marine, it takes three tank hits to kill a Marauder. There's a 30% chance that 6 Marines will miss/2 Marauders will miss. Let's say that happens. He gets hit by the Tank again. Marines are already dying left and right, a bunch are in red, still not a single Marauder dead even with target fire. Both attackers want to retreat now. Marines still clumped as shit, take another big hit with lots of splash. Marauders retreat at almost identical speed, take less splash damage. With target fire from the tank, one Marauder dies. How many Marines you think died? Safe bet that it was a lot more than two, right? Especially if the tank is target firing? If more than two Marines died, then the Marauder player came out way ahead because his units were tough enough to survive a bunch of hits and fast enough to still get out.
Pretty straight forward, I think.
Imagine if there was no high ground mechanic, the outcome in the case would probably be pretty much the same, possibly even a better outcome for the marauders. I really think your logic is flawed on this, clumped up marines will not perform well vs siege tanks, period, high ground mechanic or not. Smart players will not risk such a move and will rather get a dropship if they plan to be very aggressive with bio.
The Siege Tank isn't the point. It can be a bunch of Hydras or a High Templar with Storm. The result is the same, when Marines get into range and don't do damage because they get unlucky, they die. When Marauders get into range and don't do damage, they just walk away with red health that can be healed back to full by Medics.
On January 18 2014 20:49 CutTheEnemy wrote: easiest way to explain how % miss chance can be a problem is to imagine a metagame that could arise in which a toss makes a certain number of gates and attacks up a ramp in an all-in where the result is very very close and whether 1/4 shots miss or 3/4 shots miss determines whether the all-in works or not.
Unlikely to be a big problem though. Just saying: the most volatile games in RNG terms will be the all-ins.
Yup. I mentioned this a few pages back. All ins are the worst by a mile. It'll also have a pretty noticeable effect on end game scenarios, when players have nothing to focus on but the control of their dwindling armies. Basically, any time that continuing macro isn't an option.
On January 18 2014 19:30 SolidSMD wrote: @pure.wasted Can you give a decent example of when %miss chance is flawed?
Flawed in what sense? In the sense that it is luck and luck is bad? Or in the sense that it favors some units over others? I have a feeling that you mean the latter.
Imagine two scenarios. In the first, a pack of 10 Marines bumrush a Siege Tank on high ground. In the second, a pack of 5 Marauders does the same. They get hit once on the way there. Marines clump up more so they take more splash damage. It takes JUST over one tank hit to kill a Marine, it takes three tank hits to kill a Marauder. There's a 30% chance that 6 Marines will miss/2 Marauders will miss. Let's say that happens. He gets hit by the Tank again. Marines are already dying left and right, a bunch are in red, still not a single Marauder dead even with target fire. Both attackers want to retreat now. Marines still clumped as shit, take another big hit with lots of splash. Marauders retreat at almost identical speed, take less splash damage. With target fire from the tank, one Marauder dies. How many Marines you think died? Safe bet that it was a lot more than two, right? Especially if the tank is target firing? If more than two Marines died, then the Marauder player came out way ahead because his units were tough enough to survive a bunch of hits and fast enough to still get out.
Pretty straight forward, I think.
Imagine if there was no high ground mechanic, the outcome in the case would probably be pretty much the same, possibly even a better outcome for the marauders. I really think your logic is flawed on this, clumped up marines will not perform well vs siege tanks, period, high ground mechanic or not. Smart players will not risk such a move and will rather get a dropship if they plan to be very aggressive with bio.
The Siege Tank isn't the point. It can be a bunch of Hydras or a High Templar with Storm. The result is the same, when Marines get into range and don't do damage because they get unlucky, they die. When Marauders get into range and don't do damage, they just walk away with red health that can be healed back to full by Medics.
One could argue that there are more marines and they attack faster, so the combined hit/miss will be much closer to 50/50 than for the marauders, so the high ground mechanic 'favors' marines over marauders in predictability. Your argument purely relies on marines being fragile units, the highground mechanic does not change that dynamic at all.
I really think this war on the miss chance is entirely overblown. The circular, "Why is randomness bad?" "Because it's random!" condemnation is a little flimsy and is hardly an impetus to change what already works.
Despite controlling just one hero in Dota where the impact would be much greater, the randomness is still acceptable and endorsed by the community at large because that randomness is averaged over multiple hits, over multiple kills, over multiple team fights (and even multiple games). Moreover, some heroes whose raison d'être is to hope for lucky breaks with their skills, are expected to do so and accounted for. Other factors are simply more important to the outcome of a game, like the draft, individual execution, team coordination, scouting, decision making, and even patch adaptation.
In Starcraft, the picture is even clearer. The miss chance is averaged over many units, over many battles. It's predictable and easily accounted for. It can be easily avoided. In truth, its unfavorable randomness having any bearing on the end result of even one game is almost unmentionably insignificant when you consider every other factor that goes into a match. In BW, Blizzard even mistakenly reported the miss chance as 30%, a number that became a nigh ubiquitous misconception in the community. The real miss chance is much greater, 46.875%, and yet that huge boost to defender's advantage was still predictable and still easily accounted for. You know what random element of luck has decided thousands more games than high ground miss chance ever could? Which direction you send your scout at the start of a game. Why is randomness good, then? What does the randomness bring to an esport? Spectator anticipation and excitement. In any big engagement where the miss chance will just become an average damage reduction as intended, it's an underlying mechanic that's viewed entirely perfunctorily or ignored. But when it's one unit that's the focus, even something as routine and mundane as a worker trying to scout at the start of a game, the miss chance becomes appreciable. Something to anticipate.
On March 05 2010 01:42 Bwenjarin Raffrack wrote: The only recent game I could think of where the miss chance made the difference:
Flash's vulture barely slips by on the ramp and gives him the intel he needs to decide to do a timing push to punish Stork's greedy build.
Some of you will look at this and say, "See? Pure randomness and luck let Flash win that crucial ace match. It doesn't belong in Starcraft." Others will say, "See? The crowd and commentators held their breath in utter suspense when that vulture was about to be taken out. This is what watching Starcraft is all about." And then others will say, "See? Stork ignored a fundamental strategical element of the game, didn't react fast enough and deny scouting better, and paid the price. Things like this are why Starcraft is such a strategically deep game."
Of course, the people in the live report thread just said, "2400!! FLASH BONJWA!!!"
Reviewing this old post of mine, coincidentally if the high ground advantage had been a 50% damage reduction instead, Flash's vulture would still have survived.
I've been watching BeastyQT play for several hours now, and the mod looks fantastic. His first use of manual mines was about an hour ago, three mines against 10+ Hydras. All Hydras died. He was impressed.
The last 5 games have been five for five in terms of having multiple engagements going on across the map at some point. The TvZ he's playing right now is amazing, 50 minutes of back and forth trading, with a huge Zerg army fighting around his third for about 15 minutes while his mech army pushes into the Zerg's main for 15 minutes... at the same time. While a flock of SciVessels flies around Irradiating shit.
On January 18 2014 19:30 SolidSMD wrote: @pure.wasted Can you give a decent example of when %miss chance is flawed?
Flawed in what sense? In the sense that it is luck and luck is bad? Or in the sense that it favors some units over others? I have a feeling that you mean the latter.
Imagine two scenarios. In the first, a pack of 10 Marines bumrush a Siege Tank on high ground. In the second, a pack of 5 Marauders does the same. They get hit once on the way there. Marines clump up more so they take more splash damage. It takes JUST over one tank hit to kill a Marine, it takes three tank hits to kill a Marauder. There's a 30% chance that 6 Marines will miss/2 Marauders will miss. Let's say that happens. He gets hit by the Tank again. Marines are already dying left and right, a bunch are in red, still not a single Marauder dead even with target fire. Both attackers want to retreat now. Marines still clumped as shit, take another big hit with lots of splash. Marauders retreat at almost identical speed, take less splash damage. With target fire from the tank, one Marauder dies. How many Marines you think died? Safe bet that it was a lot more than two, right? Especially if the tank is target firing? If more than two Marines died, then the Marauder player came out way ahead because his units were tough enough to survive a bunch of hits and fast enough to still get out.
Pretty straight forward, I think.
Imagine if there was no high ground mechanic, the outcome in the case would probably be pretty much the same, possibly even a better outcome for the marauders. I really think your logic is flawed on this, clumped up marines will not perform well vs siege tanks, period, high ground mechanic or not. Smart players will not risk such a move and will rather get a dropship if they plan to be very aggressive with bio.
The Siege Tank isn't the point. It can be a bunch of Hydras or a High Templar with Storm. The result is the same, when Marines get into range and don't do damage because they get unlucky, they die. When Marauders get into range and don't do damage, they just walk away with red health that can be healed back to full by Medics.
One could argue that there are more marines and they attack faster, so the combined hit/miss will be much closer to 50/50 than for the marauders, so the high ground mechanic 'favors' marines over marauders in predictability. Your argument purely relies on marines being fragile units, the highground mechanic does not change that dynamic at all.
No, my argument relies on miss chance being based on luck. I need my Marines to kill the Siege Tank as soon as they get in range of it. Which they can do with their damage output. Unfortunately, when they don't do this because of luck, they're fucked. Other units that fail to do this are less fucked because they have the HP to get away from an engagement that no player could have foreseen would go this way.
Not only all of that, but the main argument against giving lowground units a penalty to attack is that it hurts fast-hitting units like Marines too much. That's also not changing any game dynamics! It purely relies on Marines hitting often and thus armor being effective against them. Yet this is both the main and the ONLY argument against penalty to attack.
I don't even understand why you would argue about it, it will NEVER make any real difference. You just won't get the situation in which you have 10 marines and your overall dps is reduced 70%, it won't happen
On January 18 2014 21:37 The_Red_Viper wrote: I don't even understand why you would argue about it, it will NEVER make any real difference. You just won't get the situation in which you have 10 marines and your overall dps is reduced 70%, it won't happen
Once again, seems like devs didn't read my post. Can you guys change the Roach projectile animation back into acid saliva? This new flame attack looks pretty bad for Roaches, and when they kill stuff it bursts into flames like when it dies from Hellions/Colossi. :/
You can blame me for that one. ^^
Since we are using the orange/brown coloured Roach model I decided to try a different approach to its attack. The basic idea was to make the Roach more of a "fire beetle/ant/thingy", kinda like it holds a lot of chemicals inside of it and then spits them out while at the same time setting them on fire. The swarm version of a Firebat/Hellbat I suppose. (I guess Abathur saw potential in a special Roach strain on some volcanic world or something ^^)
Regardless, if the flame spit attack proves to be too unpopular I'll either revert back to the acid attack or think of something else.
Hey, no problem, I just think it doesn't fit.
I agree with Existor, Purple and Green are great and fits the Zerg style, Flame/orange not as much. And btw Existor, purple is already used by Devourers so I would just revert it back to the Green or find out some completely different type of projectile.
On January 18 2014 21:37 The_Red_Viper wrote: I don't even understand why you would argue about it, it will NEVER make any real difference. You just won't get the situation in which you have 10 marines and your overall dps is reduced 70%, it won't happen
How you know it?
Cause it is extremely unlikely, you would probably have to play 1 billion games to have this situation (probably even more) edit: this number is out of my ass, but i think it is the right direction
On January 18 2014 20:49 CutTheEnemy wrote: easiest way to explain how % miss chance can be a problem is to imagine a metagame that could arise in which a toss makes a certain number of gates and attacks up a ramp in an all-in where the result is very very close and whether 1/4 shots miss or 3/4 shots miss determines whether the all-in works or not.
Unlikely to be a big problem though. Just saying: the most volatile games in RNG terms will be the all-ins.
Except that during an all-in a lot of shoots are fired so it all averages out.
The larger the number of goons the more predictable the outcome. If you have 8 dragoons and count on them missing on the upper end of the average, you should reliably be able to predict the results with 99% certainty.
Also if you just move up the ramp and hope for the best you are pretty much a retard. For instance in BW the miss chance was 48% (and actually 0.45% or 1/256 on even ground) so a smart player would count on the goons killing the tank in slightly more shoots than usual, if it took a lot of RNG to be cost-efficient they just didn't do it.
Soo what you are talking about literally only makes a difference in 1% of games.
i dont know if this has been mentioned before, but i noticed new protoss units/buildings, like reaver, observatory, arbiter tribunal and sentinel arent affected by shield upgrades, they are always 0 shield and also because of this they cant be affected by arbiter +2 shield aura
and one more thing, id suggest removing energy from overseer to prevent feedbacking it, i know scout has this ability now but seeing how some units like dropships and banshes arent vulnerable to feedbacks anymore so it seems to be more broodwar-like with only casters having energy and overseer isnt a caster. also same concern about the swarm queen, shes not exactly a core t1 antiair unit anymore and seems to be much more vulnerable to feedbacks due to scouts having it and they fly, bascily u go corsairs in pvz like normal and add 1 scout to feedback queens... and they do get feedback without upgrade. unless ofcourse this would be intentional design, you cant feedback orbital or nexus can you?
On January 18 2014 20:49 CutTheEnemy wrote: easiest way to explain how % miss chance can be a problem is to imagine a metagame that could arise in which a toss makes a certain number of gates and attacks up a ramp in an all-in where the result is very very close and whether 1/4 shots miss or 3/4 shots miss determines whether the all-in works or not.
Unlikely to be a big problem though. Just saying: the most volatile games in RNG terms will be the all-ins.
Except that during an all-in a lot of shoots are fired so it all averages out.
The larger the number of goons the more predictable the outcome. If you have 8 dragoons and count on them missing on the upper end of the average, you should reliably be able to predict the results with 99% certainty.
Also if you just move up the ramp and hope for the best you are pretty much a retard. For instance in BW the miss chance was 48% (and actually 0.45% or 1/256 on even ground) so a smart player would count on the goons killing the tank in slightly more shoots than usual, if it took a lot of RNG to be cost-efficient they just didn't do it.
Soo what you are talking about literally only makes a difference in 1% of games.
How can you know how many games this will impact when Sbow has no metagame? You assume that there will be as few all in/pressure builds as in BW, but this is not BW.
On January 18 2014 21:37 The_Red_Viper wrote: I don't even understand why you would argue about it, it will NEVER make any real difference. You just won't get the situation in which you have 10 marines and your overall dps is reduced 70%, it won't happen
How you know it?
Cause it is extremely unlikely, you would probably have to play 1 billion games to have this situation (probably even more)
Same goes here. We just don't know. By pure stats, it will happen once in every 7 engagements. That might mean multiple times per game. By applicability to games, depends on the meta. If Bio becomes viable in Sbow, and Marine pressure builds are viable, then analogous situations will crop up often. If Bio pressure is not viable, then analogous situations may not crop up.
The only reason I brought up that example was because somebody asked me for a clear-cut example where miss chance impacted some units differently than others. I gave the first one I could think up without playing the game. You want a better example? Casters. High Templar, Ghosts, and Defilers on low ground aren't affected by high ground whatsoever. Miss chance disadvantages all "standard" units and advantages casters by not giving them any drawbacks whatsoever. Under miss chance, a Terran army that does not rely on casters will have a tougher time assaulting a Protoss base on high ground relative to a Protoss with Storms assaulting a Terran base on high ground, than they would if there was no miss chance. There, it only took until 5 AM, but I finally disproved the idea that miss chance is somehow fair to all ground units, and is therefore superior to other methods of defender's advantage, which are not equally fair.
Percentage miss chance was in WC3 as well, and though I like SC2 both of those games seemed to have a greater consistency of results and more genuinely dominant players, more so than SC2 which doesn't have that mechanic
Anything that leads to less random wins from players that are subsequently never heard from again is fine by me. Watching Bomber lose to Blink all ins to ladder heroes after scouting it just hurts my brain. More stability and reduced likelihood of fluke wins is a good thing imo.
EPT
