|
On July 03 2015 14:45 BlackLilium wrote:Hey guys, really happy you jumped in here to comment!  Show nested quote +On July 03 2015 06:38 Pontius Pirate wrote:
Chiming in to say I agree with adding the Starbowesque 50% miss chance in this mod, against the opinion of the other commenter. Will it be combined with vanilla SC2 high ground vision mechanics? And how do you feel about the increased cooldown between shots when fighting uphill? I didn't look at vision at all; what we have is the default "vanilla HotS vision"? Or did you mean "vanilla SC1 vision"? We can certainly have a look at that!
Most of my issue with it is one of feeling. It doesn't feel right to see units slow down their firing rate against a unit of equal distance uphill as on the same plane. It also misses out on the unique element of miss chances that allow a player to decide to take a risk with small numbers of units in minor engagements to chance fighting up a hill against similar-sized forces in the interests of picking off a few units. With a cooldown increase, it makes poking back and forth too reliable for picking off individual units. Basically, the optimal technique in almost all small engagement scenarios is that they're going to dash towards a cliff, fire on a unit with no more firepower than absolutely necessary, and retreat against any returning fire from other units not picked off, thus circumventing the consequences of the main effects of the cooldown increase. Yes, in major engagements, it will end up having a statistically similar (in fact, nearly identical) effect as a 50% miss chance, but I think you'll find that it does not accomplish the same goals in regards to minor scraps.
From playing Starbow a fair bit (mostly a long time ago, but I got a couple games in since 1.0), I've come to decide that SC1 vision is SC1 vision and SC2 vision is SC2 vision. I see no need to mess with a system that this particular game is built around that already works well for its particular, specific goals.
Show nested quote +On July 03 2015 06:38 Pontius Pirate wrote:
I think you went a bit far on some of the damage point changes, particularly for the Hellion and the Roach. Some units' attacks should feel a bit sluggish, in order to communicate a certain aspect of the unit itself. Admitedely, I was in a mindset "high damage point = bad" and just cycled through all units to fix that. We can certainly revert that in a Roach. Which other units you would prefer to see them with a more sluggish behavior? Ghost? In the hellion case however, it is meant to be a fast, agile unit that a player may want to micro extensively. Shouldn't it have a low damage point?
From that description alone, obviously the answer is yes. However, it has a very unique method of dealing damage. It's not like the Vulture, where a slow damage point would pretty much entirely ruin the whole point of the unit. Hellions can deal a ton of damage to a huge crowd of units, all at once, if they're positioned correctly (for you, not for your opponent of course!), so putting too much control in the hands of the Hellion user can lead to scenarios such as Hellions infinitely kiting Zerglings, units that are technically faster than them. Another issue is that they can zoom into a mineral line and hold their fire until the workers are pulled away and then unleash the hot death and get the workers at the exact moment that they line up, which is not a practical technique with their current damage point. I agree with the notion that harass units should generally be hair-trigger, instant response units, but I maintain that the unique splash pattern of the Hellion means it should be designed for finesse, rather than raw reaction time.
Unless of course, you plan to greatly redesign the unit. A redesign of moderate, but not severe, scope would be to institute gliding shot on Hellions, and set their acceleration to somewhere around 3.5 to 4, so that they can initiate their attack animation while decelerating, so timing your stop commands and attack commands will do much to determine whether you achieve a total BBQ or a slightly rare drone. This was almost exactly what the Jackal did during the WoL alpha, before the transition to the wheeled vehicle we all know and feel ambivalent towards.
|
On July 03 2015 18:13 Pontius Pirate wrote:
Most of my issue with it is one of feeling. It doesn't feel right to see units slow down their firing rate against a unit of equal distance uphill as on the same plane. It also misses out on the unique element of miss chances that allow a player to decide to take a risk with small numbers of units in minor engagements to chance fighting up a hill against similar-sized forces in the interests of picking off a few units. With a cooldown increase, it makes poking back and forth too reliable for picking off individual units. Basically, the optimal technique in almost all small engagement scenarios is that they're going to dash towards a cliff, fire on a unit with no more firepower than absolutely necessary, and retreat against any returning fire from other units not picked off, thus circumventing the consequences of the main effects of the cooldown increase. Yes, in major engagements, it will end up having a statistically similar (in fact, nearly identical) effect as a 50% miss chance, but I think you'll find that it does not accomplish the same goals in regards to minor scraps.
The whole point of removing randomness is so that the player does not have to take risks caused by it. The only risk factor is the opposing human's reaction.
That being said, you are bringing a valid point. Both mine, and RoomOfMush's propositions to high ground allow micro to reduce the negative effects. You can target single unit, snipe them off and retreat. The question is, if it is something to be desired or not? On one hand it does nullify the high ground, on the other - it rewards micro: move army in, fire a single shot, move army out. That's what I call "positional micro", as opposed to ability micro based on special unit abilities. So, the question is - what do we prefer: high ground to have a guaranteed impact (miss chance/damage reduction approaches) or encourage positional micro as a way to negate the hight benefit (slow fire/guaraneed 2nd miss)? I honestly don't know.
On July 03 2015 18:13 Pontius Pirate wrote:
From playing Starbow a fair bit (mostly a long time ago, but I got a couple games in since 1.0), I've come to decide that SC1 vision is SC1 vision and SC2 vision is SC2 vision. I see no need to mess with a system that this particular game is built around that already works well for its particular, specific goals.
So, unless someone objects - let's stick to what SC2 provides in term of unit vision
Regarding Hellions: we can definetely have a closer look at that particular unit in the future. Reevaluate what the unit should and should not be able to do. Being able to kite zerglings to death with a very intensive micro is not necessairy a bad thing (or is it?). However, I would leave it to a later discussion. Shall I add it to TODO list?
To my knowledge, however, it is hard to make ground units glide and act as if they were hovering. Unless someone knows a solution --- any mod that accomplishes it?
|
@Black
As for high ground, think about how unintuative gaming the highground is. If it encourages very strange micro to occur, people are going to find it incredibly silly and dislike it just for that reason alone. It breaks the logic of the game. Guaranteed impact gives a more abstract wargame feel to highground. It feels like a terrain modifier and is easier to both understand and rationalize what is happening.
I favor a cleaner visual language when having something strongly effect how units micro other than simply being up on a highground. If a unit were to shoot twice as fast on highground I'd prefer some kind of protossy time slow field on top of that cliff. If units were to miss every 3'rd shot, I'd prefer to see some kind of smoke and shroud effect that occurs.
Wargaming abstract logic vs matter of fact physics declaring logic.
Purely my preference for these sorts of things.
Depth of Micro doesn't say EVERYTHING should be zero damage point. Damage point should be on the table as well as backswing and turn rate for establishing a units personality. The damage point on a hellion is key to giving him his positional feel and allowing zergling surround micro to take place. On dragoons and hellions damage point works particularly well. Understand that the depth of micro encourages damage point to be played with and change how a unit micro's rather than leaving them at the default damage point. You can have a game where everything is zero damage point, no backswing, no turn rate to increase the agility of everything but then you would have a great lack of diversity in the personality of units.
|
When I was searching what other people have done in terms of high ground advantage I stumbled upon this marvelous mod:
http://www.teamliquid.net/forum/sc2-maps/335595-mod-high-ground-advantage
It shows that damage reduction can be implemented to handle armor nicely. It also avoids using triggers, which is a bonus. The downside is that it required inspecting every single ground unit and their ranged weapons - but that's something players don't see.
The mod requires updating to HotS (already on it). Integrating it with other changes to weapons (e.g. no-smart-fire) may require a bit of attention but it is definetely do-able. I contacted the author for permission, but he hasn't been around on TL for a year. The project is also officially abandoned by him.
Given that % damage reduction is again viable - do we agree that this is the best approach to high ground? No randomness, no micro that can bypass the effect.
|
I personally prefer the micro approach since it opens up another avenue for a player to shine without forcing you to do it.
But of course, my initial plan was too to reduce the damage of units attacking up a cliff so I am not against that either.
Reducing attack range sounds like a really nice idea because it just makes sense and is easy to understand. I never tried to do it because it sounds like something difficult to implement.
----------------------------------------------
I would also propose a balance change:
Change the Siege Tank. Make it more important in positional play, that means the following:
1) Buff the damage
2) Buff the splash
3) Make the Weapon Cooldown longer
4) Make the Siege / Unsiege time longer
This makes it more costly to use siege tanks because you have to use the timing just right, but attacking into a fortified siege position will cost you.
|
If damage reduction - how much? 30%? 40%? 50%?
50% miss chance is essentially 50% DPS reduction....
Regarding Siege Tank - it is already in the TODO list. Let's discuss it then, but I am thinking in the same direction as you are.
Experimental branch changes
- High ground advantage: Replaced 50% miss chance when shooting upwards, with 30% damage reduction. The solution correctly scales the armor of targeted unit as well.
- Reverted Damage Point on the units: Siege Tank (Siege Mode), Roach, Dark Templar, Immortal, Ghost.
Note, this is "experimental" - we can easily change it if you think different values would be better. I feel 30% damage reduction may be too little.
How do you feel about the Damage Points on other units?
For the air I think it is wise to keep it low. Same goes for Tank (tank mode), Stalker, Hellion (may require further balance adjustments later) as heavy micro on these may introduce some interesting games.
I am also leaving low damage point on Hydralisks, since they are so squishy. Also it is one of few ground-to-air Zerg units, and when shooting air you cannot wait too long!
|
In the hellion case however, it is meant to be a fast, agile unit that a player may want to micro extensively. Shouldn't it have a low damage point?
Your first priority should be to create fun, simple and unique micro interactions. If the Hellion can kite Speedlings infinitely while taking no damage in the proces that isn't fun. On the other hand if a "big unit" has a damage point that prevents it from doing "moving-based" micro that it otherwise would be capable of, then it shouldn't have a damage point in the first place.
Trying to fit lore in with damage point is extremely inefficient in my opinion. Use movement speed instead as the primary variable to accomplish that, but even then don't overdo it. E.g. I find the 1.8 movement speed of the Thor too low as well and I think it would benefit from both a higher speed and lower damage point.
Also, be careful about trying to make every single unit different when it comes to damage point and turn rate. Only differentiate units on those variables if you have identified a specific improvement you can make. Otherwise you should follow the general rule of 0 damage point on ranged unit as it simply feels a ton better (imo - i assume you agree with that).
My point here is that you shouldn't make a unit feel worse for the sake of diversity. Only add diversity (both in terms of strategies and micro) if it adds choices between fun interactions.
Tip 1 If you balance the game around 0 damage point on ranged units, it makes them better at kiting. This means that you need to increase the movement speed of melee units to compensate.
Tip 2: Don't make number changes based on what people tell you on the forum. Instead spent hours in the editor testing how a change to varius variables impact micro interactions. Don't make this only a "theoretical discussion". Learn by doing instead.
|
How much damage reduction is enough and how much is too low is not an easy question to answer.
In the end it boils down to what exactly you want to achieve with it. Should a player be punished for attacking up a cliff, or should it become impossible to make cost efficient trades when attacking up a cliff? 30% is already a punishment, anything above 0% is.
If it gets too high it might make the game more stale because players will not dare to attack, they will barricade themselfs on top of a cliff and wait until maxed out. Think swarm host + static D but for every matchup from both opponents.
In general I would tell you to just choose one value and go with it. Then, wait for people to play it and tell you how they feel about it. Make changes based on observations and tests.
|
On July 05 2015 04:21 Hider wrote:
Trying to fit lore in with damage point is extremely inefficient in my opinion. Use movement speed instead as the primary variable to accomplish that, but even then don't overdo it. E.g. I find the 1.8 movement speed of the Thor too low as well and I think it would benefit from both a higher speed and lower damage point.
Also, be careful about trying to make every single unit different when it comes to damage point and turn rate. Only differentiate units on those variables if you have identified a specific improvement you can make. Otherwise you should follow the general rule of 0 damage point on ranged unit as it simply feels a ton better (imo - i assume you agree with that).
My point here is that you shouldn't make a unit feel worse for the sake of diversity. Only add diversity (both in terms of strategies and micro) if it adds choices between fun interactions.
I hope to avoid adding senseless changes. If you spot one - let me know!
However, since we are affecting grand-scale components, such as economy and terrain influence, there may be wide-range changes affecting every single unit. For example: increase/decrease prices of all units by X%. Or change their speed. Or DPS... I don't know yet.
On July 05 2015 04:21 Hider wrote:
Tip 2: Don't make number changes based on what people tell you on the forum. Instead spent hours in the editor testing how a change to varius variables impact micro interactions. Don't make this only a "theoretical discussion". Learn by doing instead.
Don't worry, I spend more time in the editor than on the forums
But I do value the opinion of the community. I may disagree with some of the opinions, but it allows me to look at things from different perspective. For example - the high ground advantage and ways of achieveing it - it helped me search for new solutions and highlight problems that I didn't consider too problematic. The final solution is also not mine, the main credit is to urashimakt and his High Ground mod from 3 years ago. It is a pity he is no longer around here.
Speaking of which:
Experimental branch change
- Damage reduction when shooting uphill is set to 50%
- Bug fix: Siege Tank in tank mode didn't have uphill penalty applied correctly
It may seem high, but keep in mind that there are two primary stats of a unit: attack and health. Only one is affected. As a result (tested with single unit type in small numbers):
- 3:2 army size is (barely) enough to win an uphill battle. Usually micro decides who has the final blow.
- less than 2:1 army size allows you to win an uphill battle with equal losses.
Yes, it can create a bit stale fight if the players just focus on that one uphill battle. However, unlike swarm hosts, you cannot move a hill. High ground is very localized, and can be often bypassed by the enemy. It is also something that map-makers can control. Adding the fact, that we intend to work with HMH economy model, there usually should be more than one place where you may want to attack.
Secondly, being able to defend a spot with a bit smaller army promotes army splitting, positional play and more aggressive expanding.
In the end - that's the theory. As RoomOfMush rightfully says - in the end you need people to play it to evaluate.
Experimental branch change
- Colosus has "Rain from Above" trait which negates the uphill fight penality
- I am considering adding "Rain from Above" to Siege Tanks in Sieged mode
It gives these two units a specific, situational role - breaking an uphill sieged line. It fits the domain of the Colosi as a unit that bypasses cliffs.
We will have a look at the balance of that specific unit in the future. I believe that by granting it this specific trait, we will be able to apply a mild nerf for a straight battle, without nullifying the unit completely.
Finally, although a bit less important - it fits the lore and animation of the unit.
|
Secondly, being able to defend a spot with a bit smaller army promotes army splitting, positional play and more aggressive expanding.
Here is my issue with this approach in order to create a positional game. If you rely on map features for players to be able to split their army (so a lower army size can beat a higher army size), you are actually restricting map makers, as there needs to be hill in certain locations.
I think its far more advantageous to create a defenders advantage through macromechanics and abilities (Dark Swarm best example of this). This way you can have any typs of map you want and the positional playstyle will always be there when you need it.
Further, remember that high ground is a double edged sword too. Not only does it make a smaller army capable of beating a larger sized army. It also makes a larger sized army capable of snowballing a smaller sized army even harder if - for instance there is a drop in your base and you need to defend it by moving up a ramp. I think this is hardly the desired type of gameplay you are looking at.
Imo it's better to look at whether there are more efficient solutions than can provide the same advantages without the disadvatnages.
Colosus has "Rain from Above" trait which negates the uphill fight penality
I am considering adding "Rain from Above" to Siege Tanks in Sieged mode
I think you just started walking on a dangerous path by making new rules that increase the learning barrier. I am not saying you can never do that (because obviously its part of a mod), but you need to seriously evaluate disadvantage and advantages. Players won't be able to figure out that these rules exist by learning by playing unless you make a clear visual indication that its the case. They will just "have to know it", which imo is a nono when it comes to (simple) game design.
So you have to ask your self, do these new rules really add such awesome interactions while significantly improving the gameplay dynamic?
Isn't there a simpler way of giving the Colossus a unique role in the game?
As RoomOfMush rightfully says - in the end you need people to play it to evaluate.
By playing. Noone has any clue whether 20% or 40% are the right numbers when they speak from a purely theoretial discussion
|
On July 05 2015 22:12 Hider wrote:
Here is my issue with this approach in order to create a positional game. If you rely on map features for players to be able to split their army (so a lower army size can beat a higher army size), you are actually restricting map makers, as there needs to be hill in certain locations.
We added another map feature. We didn't remove anything.
Map makers may choose to use the feature or completely ignore it, e.g. by making a map completely flat. There are other, albeit less pronounced ways to positional game already present, such as chokes or differences in route length for defender vs attacker. All this is still in the arsenal of map maker.
I can guess (correct me if I am wrong) that you are worried that we will balance the mod with an assumption that map features elevated spots at certain positions, and if suddenly they are gone, map becomes broken. I hope it is not going to come to that. Nevertheless, certain map constraints are already present, such as an elevated main, or at least natural with a tight ramp connecting them. I don't think we are going to make matters worse with the high ground advantage.
On July 05 2015 22:12 Hider wrote:
I think its far more advantageous to create a defenders advantage through macromechanics and abilities (Dark Swarm best example of this). This way you can have any typs of map you want and the positional playstyle will always be there when you need it.
The existence of high ground mechanics does not limit us in terms of what abilities are present. We can have a look at abilities at later time (can't talk about everything at the same time). I wonder what macromechanics could be used to influence this. We have building placement... what else?
On July 05 2015 22:12 Hider wrote:
Further, remember that high ground is a double edged sword too. Not only does it make a smaller army capable of beating a larger sized army. It also makes a larger sized army capable of snowballing a smaller sized army even harder if - for instance there is a drop in your base and you need to defend it by moving up a ramp. I think this is hardly the desired type of gameplay you are looking at.
If you put a smaller army at a disadvantageous position (regardless of what game mechanic define such position) - you are bound to loose. After the change players will need to recognize new threats and avoid them. If it is proven impossible or too coin-flippy, we can then try to influence it.
Show nested quote + Colosus has "Rain from Above" trait which negates the uphill fight penality
I am considering adding "Rain from Above" to Siege Tanks in Sieged mode I think you just started walking on a dangerous path by making new rules that increase the learning barrier. I am not saying you can never do that (because obviously its part of a mod), but you need to seriously evaluate disadvantage and advantages. Players won't be able to figure out that these rules exist by learning by playing unless you make a clear visual indication that its the case. They will just "have to know it", which imo is a nono when it comes to (simple) game design.
The "Rain from Above" trait has an icon in the command card indicating its presence. Anyone who commands a Colossus at least once should be able to notice it quickly. Those few who never try Protoss should be able to learn from those who did.
The general rule that there is some high ground advantage is harder to communicate since there is no visual feedback.
Frankly, when I was at school, playing Broodwar, I had no idea that there is a miss chance for high ground! I learned it only few years later.
Isn't there a simpler way of giving the Colossus a unique role in the game?
Colossus is designed to be a simple unit (some say: "boring unit"). Build, move, shoot, die. No special abilities. The introduction of high ground mechanic gives a chance for Colossus to do something special - which by the animation and lore fits perfectly.
Anyway, other changes to Colossus - let's leave it at some later time, OK?
|
On July 05 2015 22:12 Hider wrote:
I think its far more advantageous to create a defenders advantage through macromechanics and abilities (Dark Swarm best example of this). This way you can have any typs of map you want and the positional playstyle will always be there when you need it.
Considering the stated goals of this mod I don't think adding additional abilities is really the way to go about handling this, especially when we have other options. Like BlackLilium I wasn't aware of the high ground mechanic in BW for some time after I started playing it so I don't think it's that big a deal here. At most, maybe you could do something that would slightly muffle the weapon sounds? If you hear your units weapons fire being quieter when firing up hill it's a subtle way of indicating that something is going on.
On July 05 2015 23:52 BlackLilium wrote:Show nested quote +On July 05 2015 22:12 Hider wrote:On July 05 2015 23:52 BlackLilium wrote: Colosus has "Rain from Above" trait which negates the uphill fight penality
I am considering adding "Rain from Above" to Siege Tanks in Sieged mode I think you just started walking on a dangerous path by making new rules that increase the learning barrier. I am not saying you can never do that (because obviously its part of a mod), but you need to seriously evaluate disadvantage and advantages. Players won't be able to figure out that these rules exist by learning by playing unless you make a clear visual indication that its the case. They will just "have to know it", which imo is a nono when it comes to (simple) game design. The "Rain from Above" trait has an icon in the command card indicating its presence. Anyone who commands a Colossus at least once should be able to notice it quickly. Those few who never try Protoss should be able to learn from those who did.
Well I also think it makes sense seeing as how the Colossus is a rather unique unit in the game as it can be attacked by anti-air and is one of the few where even when it is uphill it can still be seen by units on low ground. Given all of those unique behaviors for a ground unit, it not suffering from the low ground to high ground damage reduction makes sense and I don't think players will have a problem with it.
I also wanted to throw out a random suggestion when it comes to damage modifiers, all the ones in the game are X +Y, but what about X -Y? Having a unit that deals bonus damage to everything but a certain type of unit allows for a bit more flexibility in constructing matchups. I know we are tabling the discussion on Siege Tanks for right now, but I think that is a unit that would benefit from this as you can make it deal more damage to everything but light (for instance) to have its damage be high against everything, without having it absolutely murder light units.
|
In BW the high ground advantage had a pretty clear visual representation. When your units missed an attack the sprite of the projectile actually missed visibly and hit the ground or the cliff. I think it should be obvious to see that some of your goons shoot at the ground and your enemy not taking damage that something is going on.
The next unit I would like to discuss is the Ghost. In Blizzard SC its not a viable unit to produce. My suggestion:
* Make EMP a missle so that enemy player has a little bit more counter micro ability.
* Change the nuke to either make it much stronger (to break siege positions) or much weaker but cheaper at the same time (to allow you to use it for harassment). As it is nobody ever uses it.
* Either make them cheaper or make their normal attack stronger to make them more versatile and cost effective
* Change Snipe to deal 50 damage (like before the nerf) but it can not be used on massive units. Blizzard didnt want Snipe to be too strong versus ultralisks and brood lords. With 50 damage it can be used to harass worker lines, kill marines, banelings, etc.
|
I don't like the idea of an ability like snipe that can't be used on all targets. I can shoot a siege tank but can't shoot a Thor? I think this would be a good place for the negative damage bonus, snipe can do -X damage to massive. It even makes lore sense since the odds of a sniper round dealing critical damage to some giant thing is much lower than to a person or smaller machine. Like IRL, a 50 cal (anti-material) sniper rifle will turn a person to paste while only punching a hole in a tank.
|
The Core Setup - Closing Report
Uphill battle penality: The original proposed approach of 50% miss chance when firing uphill raised concern of high randomness in small battles. After some discussion and experimentation in the editor we discarded the uphill miss chance in favor of damage reduction. We found an mod High Ground Defender by urashimakt that implements it well - in particular, it scales the armor of the receiving unit accordingly.
There is a concern that such high damage reduction may lead to stale play and become a required element of every map. We don't share this opinion, although no evidence is available to decide what will actually happen. This will require actual games to be played.
In addition we introduce a "Rain from Above" trait which ignores the uphill penality. Currently we applied it only to Colossus. It gives the unit a unique in-combat trait. Further adjustment to the unit will be required. We considered RfA for Siege Tanks as well, however a concern was raised that RfA itself may be unintuitive, so we decided to keep in on Colossus only - at least at the moment.
Damage point: The original proposed approach set the damage point to almost all units to near-zero value. It was pointed however that some units do not absolutely need it, and the change seemed to be just for the sake of a change. We keep 0 damage point on small air units, as well as Hydralisk, Hellion and Siege Tank (tank mode). 0-point Hellion raises concerns that it may be too good against zerglings. This will require further investigation at a later time.
New topics in the TODO list:
- Investigate Hellion 0 damage point. In particular its function during harassment and interaction with zerglings.
- Investigate Ghost usability (request of RoomOfMush). This is however a detailed look on a single unit, not related to our current changes - thus I am putting it on a low priority.
The changes in the Experimental branch have been pushed to Stable branch.
|
Economy: HMH 5-4 or HMH 6-4? Other?
We need to decide which economy model are we going to use.
The default Hot Mineral Harvesting was designed with default HotS units in mind which gives us additional constraints. In particular:
- We tried to maintain mineral-to-gas ratio as much as possible. With Starcraft Improved however we can afford having a different mineral-to-gas ratio and rebalance the prices on some or all units.
- In early game we did not want to exceed 100% mining rate compared to Standard. That means, 8 workers in HMH should not mine faster than 8 workers in HMH. Why? As we have seen in Double Harvesting tournaments, higher incomes, around 110%, lead to 8-worker cheeses and all-ins that are hard to scout and stop. In Starcraft Improved there we can introduce mechanisms to slow down all-ins even with higher economy (higher unit production time, more costly basic buildings, etc.).
Things to consider:
- A) How much income do we want at 8 workers, compared to Standard?
- B) How much income do we want at 16 workers, compared to Standard?
- C) How far should efficiency drop at 16 workers compared to 8?
- D) What should be the maximum possible harvesting speed (at 24 workers)? This cannot be too high, since we want 3-rd round of workers to give even less income than 2nd round.
- E) How much each round of workers contribute in terms of their efficiency?
Options on the table
- Current HMH 5-4 75%. (W=0.6, H=2.686, HotH=3.17), where
HMH X-Y Z%
X=Harvesting amount of cold minerals (HotS default: 5)
Y=Harvesting amount of hot minerals
Z=efficiency at 16 workers (HotS default 100%)
W=wait time after harvesting
H=harvesting time in seconds (HotS default 2.786)
HotH=harvesting time of hot minerals
These numbers give us:
- A) At 8 workers it matches Standard
- B) At 16 workers it is 25% behind Standard
- C) Efficiency drop is 75% (as indicated in the name)
- D) Maxing up 30% behind Standard.
- E) 100% - 50% - 33%
- HMH 6-4 75% (W=0, H=3.5, HotH=2.67)
- A) At 8 workers it is 15% ahead of Standard. The workers sits quite long at the mineral patch (3.5s)
- B) At 16 workers it is 12% behind Standard. When paired, the workers sits significantly shorter at the mineral patch (2.67s)
- C) We maintain 75%. With HMH is actually easier to get even worse efficiency but we believe it would be too much!
- D) Maxing up at 17% behind Standard.
- E) 100% - 50% - 36%
- HMH 7-5 75% (W=0, H=3.1, HotH=2.76) - a bit more extreme solution, but has lower difference in harvesting times. This gives us saturated base income much closer to Standard, but early game is much higher.
- A) At 8 workers it is 43% above Standard - thus, lots of balancing will be required!
- B) At 16 workers it is 7% above Standard
- D) Maxing up at 1% above Standard
- E) 100% - 50% - 33%
- HMH 5-4 75% (W=0, H=2.34, HotH=2.76) - an idea we came up only recently. This is actually based on the current HMH and the difference is only in the harvesting times!
- A) At 8 workers it is 15% above the Standard
- B) At 16 workers it is 14% behind Standard
- C) The 75% efficiency is maintained
- D) Maxing up 20% behind Standard
- E) 100% - 50% - 33%
Overall, some early-time balancing will be required, but it should not have that strong feeling "omg it is so slow!" as the original HMH has. Hopefully.
All the above ideas assume that 75% efficiency is what we should aim for. Double Harvesting 3x3 has an efficiency of 81.5% and some people felt it was not impactful enough. 75% is also around the efficiency of Broodwar. At 75% efficiency it means that the new workers you produce help by a factor of 0.5, i.e. you are producing half-workers. We don't think that going below that is wise.
If you think that other number, between 75% and 81.5% would be better we can easily redo the calcuations with a different efficiency point.
Also, if you think any of the above suggestions should be tuned to achieve some additional effect or constraint - let us know.
|
I'll give you the same answer as before with the uphill damage scaling: This is really just theorycrafting. We can not discuss which values are right and which are wrong, because none of us knows really. You have to feel the numbers out over a long period of time with several games being played.
Dont waste your time giving it too much thought, just pick a reasonable value and then have people try it out.
--------------------------------
On another note, I would like to know what you intend to do with "the protoss situation"? Many people believe that protoss needs a drastic redesign, something that you dont want to do, so how do you adress the protoss balance?
|
Ultimately, everything is about numbers. What we can discuss however are the constraints and goals that we would like to see.
Should the income match Standard at low worker counts or high ones? How much influx of resources at the beginning can we afford (have seen that in games of DH10 and DH9) ? Do we want to permit workers staying much longer at the resource nodes, or not?
Also, one question that I didn't ask earlier is about the number of starting workers. I would go for 8 or 9, because that's the moment when builds start to diverge (9 pylon + scout or no scout, or going up to 10 pylon). Again - numbers, but there is some thought behind them.
The "protoss situation" I am putting on top of TODO list since it is a general, wide-range question that should have an answer early.
|
I would suggest you keep the income the same as HotS for now and only change it if it turns out to be imbalanced or produces boring games. With 8 workers being optimal saturation I would suggest starting with 8 workers.
But really, I think the balance changes are the most important to the majority of people.
|
On July 06 2015 20:46 RoomOfMush wrote:
I would suggest you keep the income the same as HotS for now and only change it if it turns out to be imbalanced or produces boring games.
But there is a problem. HMH income curve is flatter than of HotS. If we match HotS income at 8 workers, we fall behind at 16 by 25%. If we match HotS at 16, we are 33% higher at 8. My last suggestion is somewhere in between: approx. 15% higher at 8 workers, and approx. 15% lower at 16 - that's what I am currently thinking about.
On July 06 2015 20:46 RoomOfMush wrote:
With 8 workers being optimal saturation I would suggest starting with 8 workers.
I remember someone showing that there is a value in data that controls that. Where was it?
On July 06 2015 20:46 RoomOfMush wrote:
But really, I think the balance changes are the most important to the majority of people.
It's easiest to talk about balance changes as it affects the combat directly. Economy change infulences it indirectly and Hight advantage is very broad but situational at the same time.
We will get to balance, don't worry
|
|
|
|
|
|
EPT
