Jump to content
  • Advertisement
Sign in to follow this  
Hawkins8

An Innovative Combat System

This topic is 3653 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

Actually I borrowed this from a turn-based game I played. In this combat system, there are 3 tiers of player fighting techniques: 1st tier (auto-hit - minimum damages) This is the traditional sandwich type auto fight in all MMORPGs, your toon hits automatically when aggroed. You can't do much damage in this auto-fighting mode, it provides certain defense against the mobs. It's better than just stand still defenselessly when you are AFK or disconnected due to a network problem while being attacked by a mob. Anyway, this fight mode provides certain protection for toon but does little damage. The philiosophy is, you need to actively do something to cause damage, damage doesn't come automatically, no matter how sharp your weapon is. 2nd tier (specials - medium damages) Player specials are launched, this shows that the player behind does have a will to do some damage, so damage will be done, nothing is automatic, you actively launch a special, the special lands successfully and causes damages. This tier can also be commonly found in other MMORPGs such as Lineage II and WoW where you launch specials in the expenses of your mana pool. In this tier, the damage throughput is relatively stable and predictable, which is, your opponent's hit point bar drops quite regularly, say, 100 - 200 damages per hit with a 3000 hp pool. It consumes mana to cause this damage. Nothing special, almost exactly like what everyone does in another MMORPG. 3rd tier (death blow - maximum damages) Extremely powerful damage including insta-kill can be caused. However, you have to wait for a "time slot" to come in order to launch your death blow. In a party, the "time slot" comes just like a token pass, the token passes around each party members, when it's your turn to receive the token, it's your time to launch your death blow. To each player, such a token will appear as light bulb in your UI screen. When it lights up, you throw out your death blow. The trick is, there is a time delay between your throwing out of a death blow till it truly lands. Usually, the more powerful your spell is, the relatively longer the time delay will be. More like the pre-cast time delay in UO. This time delay will be displayed in both your friends and foes' UI screen, such that everyone in the battle may guess by both the color and length representing the category and time delay your death blow is. Someone in your opponent's party may thus choose to interrupt your death blow. The same applies to crucial healing, crucial buffing, curcial de-buffing and etc. Roughly, the more members you have in your team, the slower each one will get his token. While the tougher the mobs are, the faster a token is turned. Such that you need to optimize the number of team members in order to be token-efficient. This is basically the idea. And a battle may look like this: Your party runs into group of mobs (linked and grouped mobs), a battle thus is started just like in other games. The sandwich mode starts and your party members automatically melee fight their nearest mob. Now your party leader (tank) receives his token and thus casts a 3rd tier hate that most damage dealing will thus be tanked. The next token receiver will cast whatever he's specialized and powerful, and so and so. On the other hand, the mobs may do the same, the mob boss will launch a melee death blow, which can only be interrupted by, say, a Ninja but he just passed his token round. The only choice of the team is to put its hope in that the tank can either survive the blow, or he successfully evades the blow. A human tank may more likely survive the blow, while an elf tank may have to rely on his DEX to evade the blow. Ok, now the blow however lands successfully and knocks down your tank, before the mob boss gets his next token, your healer must rez the tank on time to avoid the boss' next death blow. If your team fails to bring up the tank, the next blow (if it's the same melee blow) will certainly insta-kill someone in your team. Something like that. I can't put all the details here, but basically it works just like this. In a nutshell, it is an attempt to introduce tactical fights into a battle such that tactical coorperation becomes possible, results will rely more on human skills instead of purely ingame skills. For instance, in order to pre-read a druid's death blows, you (as a player) need to be familiar with the druid's fighting tactics in order to make good guess on his next move and to stop him. This is even more crucial in PvPs and more challenging as human behaviors are difficult to predict, yet you need to predict correctly (the death blows) in order to win. [Edited by - Hawkins8 on June 13, 2008 2:46:43 AM]

Share this post


Link to post
Share on other sites
Advertisement
I think the predictive requirement is quite a nice addition, and I believe in some ways Age of Conan does this, in that one has to choose where to block, and predict where/when the enemy will block. (That's based on what I've heard, so correct me if I'm wrong there)

The big issue is that network latency makes timeslot based predictions hard to enforce effectively.

Share this post


Link to post
Share on other sites
With the advances in modern hardware, network latency is becoming less and less of an issue. I could see this working incredibly well if you could refine it into something a bit more detailed. You could easily control the speed of the combat (and its ebbs and flows).

Share this post


Link to post
Share on other sites
Quote:
Original post by Argus2
I think the predictive requirement is quite a nice addition, and I believe in some ways Age of Conan does this, in that one has to choose where to block, and predict where/when the enemy will block. (That's based on what I've heard, so correct me if I'm wrong there)

The big issue is that network latency makes timeslot based predictions hard to enforce effectively.


I played Lineage II and noticed that for each attack, the screen will show effectively the amount of damaged caused per hit, whether it is a critical hit and etc. So I think if the sandwich type battle details can be reported that frequently on a per hit bases, I think that it gives enough room for a smooth timeslot processing to drop in. While bandwidth keeps improving since Lineage II was launched.

Of course, in order to implement correctly, correct timing must be worked out such that players with slower connections can still effective using the timeslots. In the end the slower connections will only affect the response, rather than the outcome effectiveness. For example, slow connection usually causes a less than one second delay (from my experience in the turn-based design where this idea is borrowed). On the other hand, more powerful spells require more delay, say, from 3 to 8 seconds. Plus that tactically, not all spells require an asap response, actually some buffing and debuffing may actually require you to slow down your response in order to time correctly to give its effect. That says, you can implement the design to minimize the effect of a slow connection.

In my example given in my first post, the ninja should actually wait for a while till he sees (pre-read) the boss's action in order not to waste his round, as his job of debuffing or interruption should be focused on the big boss and the big boss only. Of course, in the end your game world should be ready before you can test out if the timing works or not.

[Edited by - Hawkins8 on June 16, 2008 1:05:06 AM]

Share this post


Link to post
Share on other sites
Quote:
Original post by Zooch
With the advances in modern hardware, network latency is becoming less and less of an issue. I could see this working incredibly well if you could refine it into something a bit more detailed. You could easily control the speed of the combat (and its ebbs and flows).


I think you may need to have a ready game world to test the timing out correctly. Before that a scheme can be worked out roughly. From my experience with both the real time battle system such as the one in Lineage II, and the turn-based system from where my this idea is borrowed, I think it should work fine. I roughly mentioned how the timing possibly works in my previous response.

The detail design scheme, however should be varied and depends on how you will arrange the class roles in a group battle in your design. A typical one can be like this,

The tank will receive his timeslot/token and cast a powerful protective buff, such that he can tank the big boss effectively.

Then the Ninja receives his timeslot, he may choose to wait for the big boss's action, or to cast a powerful de-buff to potentially de-buff all the buffs carried by the mobs by default when they entered the battle. The ninja's wait time will not be counted towards the timeslot token pass, but there will be a time limit, beyond which his turn will be considered as being missed.

In the design, this time limit is just abit long enough to allow the prediction of the big boss's action. Even when timing is not right, the Ninja can still launch a blind interruption but such a blind action may bear a chance of either interrupt successfully or being totally wasted.

In the case that your party is too bulky (too many members), even the Ninja successfully predicts the boss's blow, he doesn't have the timeslot on time to cast his interruption spell.

The bard sing a buffing song to increase the whole party's speed (of obtaining the timeslot quicker).

The sub-tank Samurai, with his self-sacrific spirit he casts a strong hate spell instead of a protection buff, he does so in order to cover up the Druid's group damaging spell.

The druid casts a powerful group damaging spell, as being protected by the Samurai it effectively reduces the other side's chance of interrupting this group damaging spell. But still, if the mob's side has a high end ninja, the ninja can still choose to interrupt the druid, thou by chance he might wrongly hit the Samurai.

Now mob's boss casts a melee death blow, the Ninja guesses so and thus casts his interruption. The game mechanic will calculate if the interruption time is correct or not, if the timing is correct the boss's melee blow will be interrupted right after the ninja's spell pre-cast wait time.

The healer will have to cast a blind heal to the main tank or the Samurai, in case the tank or the Samurai is damaged, his healing action will thus take effect. However, if he chooses to heal the main tank, while the Ninja successfully interrupted the boss, most likely his healing action will be wasted but it's quite worth it.

Since the mob side has a Ninja (or anything) which critically hinders the player party's tactics, the treasure hunter in the player's party casts a critical hit death blow which bears a chance of critically hitting the Ninja, or when it is with the correct buffing or de-buffing from other teammates, the treasure hunter can actually (have a chance) insta-kill the target Ninja.

The player's ninja will continue to focus on the big boss, as at the moment noone can withstand a single blow from the big boss. Now the main tank receives his second timeslot/token, as he knows that he's not tough enough to tank the big boss yet, he casts a second strong protective buff which adds accumulative effect to the first one.

So and so till he successfully accumlates up to 3 layers of strong protection shields, he's confident that he can tank any single death blow from the big boss. Now he can cast an extremely strong hate for single foe, to firmly attract the big boss from attacking any teammates but himself. The ninja is thus freed to do his other de-buffing tasks, while the healer still needs to keep a close eye to the main tank, however, he now has enough time to take care of others besides the main tank, as the party's protection sheild is successfully set up.

The 3-layered protection shield will last just for a period of time, the tank needs to keep the shield up to 3-layers. Because the shield is built layer by layer, it will be worn off layer by layer, the tank thus needs to keep his 3-layerd shield on a layer by layer basis timely. If somehow the mobs successfully de-buff all the 3 layers of the shield, the whole party will be put to an extreme danger. The tank, the ninja and the healer will have to try again to re-build the 3 layerd shield. Usually if this happens in the middle of the fight, the shield re-building will be, by design, more difficult then how it's first built. You need more luck and more cooperation to get you shield up again the second time.

The battle will go on just like that, in a highly cooperative way.


If unfortunately that the main tank is killed, then the whole party will have to rely on the Ninja to interrupt, and the Samurai to sacrifice himself to firmly attract and to effectively evade the big boss's attacks in order to steal time for the healer to ressurect the tank up and to allow the 3-layer shield to be built. If either the Ninja or the Samurai is insta-killed by the big boss, most likely your party will be defeated.

[Edited by - Hawkins8 on June 16, 2008 1:12:29 AM]

Share this post


Link to post
Share on other sites
In a typical battle, you need to build up your protective shield such that damages can be tanked effectively, as if the healers in your team are killed, noone can heal and rez which usually means that your team is defeated. On the other hand, you may need to kill first not the boss but those cause effective debuffing to your shield.

That's why a tactical discussion is necessary before you fight a boss, and before such a discussion, you need to do your home work on the tactics usually available to this particular boss, such that you can predict his behavior in the battle. You thus need a good community and a good knowledge base such that you can effectively kill the different kinds of mobs using different kinds of tactics. Moreover, a good desgin may need you to form a correct and optimized group in order to face your enemies, while you may have to work out different combinations and number of teammates to deal with the different kinds of enemies in both PvM and PvP. Thou I am not sure if this design works in reality as it's said here or not, which requires a ready game engine to find out (you can hire me to do this for you, if you have like 20M to burn. hehe...).


For example, in PvP you may form a team of 10 druids with each casting a group damaging spell at the very first timeslot to kill everyone. To counter this, you may need to form a team with 10 ninjas to interrupt then kill them one by one. In design time, you may have to work out the effect of such a kind of non-standard grouping combinations. It may be difficult to balance, but I don't think that it's impossible. Moreover, even in siege you need to form teams tactically, to effectively kill the NPC bosses, to effective counter your enemies' teams, you form special tactical groups to do different things, such as the 10 druid group mentioned above, such that you can kill many before you are finally taken down by the ninjas. Anyways, you don't simply zerg to win like today's MMOs.

[Edited by - Hawkins8 on June 16, 2008 2:02:30 AM]

Share this post


Link to post
Share on other sites
Sounds like wacka mole as opposed to a strategic mechanic.

I am not sure it would fit well with the flow of a PVP game, but may be ok with the PVE part of an MMORPG.

An interesting a creative idea, but I think it needs some work.

Share this post


Link to post
Share on other sites
Quote:
Original post by Stangler
Sounds like wacka mole as opposed to a strategic mechanic.

I am not sure it would fit well with the flow of a PVP game, but may be ok with the PVE part of an MMORPG.

An interesting a creative idea, but I think it needs some work.


Since the idea is borrowed from a turn-based game, if it works for PvE, it should also work for PvP. More interestingly, the tactics used in PvP is totally different from that in PvM.

In PvP, the tank will use another set of skills. He will no longer be able to build a 3-layered shield. Yet he will build another kind of shield, the shield needs only one timeslot to build, the shield will be broken if hit for 3 times. So the tank will cast powerful group hate followed by this shield.

Moreover, he has to make good trade off between his endurance and DEX (adjusted by gear), because he has to rely on evasion/dodging to keep his shield up which will be broken by 3 hits.

While the sub-tank will use single foe hate to attract whoever (say, Ninja) capable of debuffing the 3-hit shield.

The tank(s) will thus cast shield and group hate to protect the relatively weak classes. That's how PvP is done, theoretically in a real time battle. Practically it's already done in the turn-based MMO.

As for 1-on-1 PvP, it's even simpler to design. The timeslot token will flow slower such that you have to make extremely good use of it together with other 2nd tier spells.

Moreover, in clan wars and siege, you no longer rely purely on zerging to win. You need to form tactical groups to tackle with the mob bosses. You need to form player killer groups to take down enimies's mob bosses killing groups. The other side will then form killer groups for player killer groups...etc. And a highly cooperative and tactical team can take down like some 10 other player groups.

Which says, one side with 100 players forming good teams can beat down the other side with 300 but with less tactical and cooperative players. Quite unlike in today's MMOs where your 300 players will surely win by zerging, no matter what.

[Edited by - Hawkins8 on June 17, 2008 8:18:36 PM]

Share this post


Link to post
Share on other sites
A typical group vs group PvP goes like this;

Before entering the group battle, your bard is pre-equiped with a buff spell which will freeze a battle with its head counts, that is, once a battle is engaged, others can no longer join either group, nor can they attack those in either group. However, if a teammate is killed, 1 of your friends may join your team to replace him (your enemy can form killer groups to kill your stand alone friends thou).

1st timeslot:
main tank casts group hate (such that the group is protected)
sub-tank attracts 1st Ninja
healer casts blind heal for main tank
...


2nd timeslot:
main tank casts his 3-hit shield (he himself is protected)
sub-tank attracts 2nd Ninja (if any), Or any tough damage dealer failed to be group-attracted by the main tank
healer may choose to heal sub-tank as main tank is self-protected, or continue to heal/blind heal the main tank
...


3rd timeslot:
main tank casts buff to boost up his own DEX, OR another group-hate to attract those failed to be attracted by his first group hate
sub-tank casts 3-hit shield for self protection
healer will heal whoever in need
...


Something like that. To simply put, by tactics you need to first kill the healer, to effectively do so you need to find a good chance when the tanks hates are off effect. However, in between the time such a chance appears, you need to keep your own healer's survival, to do so you need somehow to disable (by kills or hates) the other side's damage dealers (DDs). Sometimes, you may need to kill the Ninja first to keep your own shielding effective.

Anyways, there's virtually no fix rules in PvP, it all depends on situations, say how effective your tanks are, how effective and sneaky your DDs are, your Ninjas are, and so forth.

Your team combination can also vary to be more offensive with more DDs, more protective with more tanks and healers. You need to keep testing out the "good" combination suitable but only for your current enemies, your enemies will do the same to adapt the situations, so to speak.

[Edited by - Hawkins8 on June 18, 2008 1:57:03 AM]

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

Participate in the game development conversation and more when you create an account on GameDev.net!

Sign me up!