[ QUOTE ]
The question is one fundamentally of team dynamics, and the optimal performance of such dynamics. Assuming that a team dynamic is one that needs a meatshield/scranker is to already bias the test towards having a melee AT. It will illuminate Tanks vs scrapper slightly, but not much else.
We have to get past our I1-I4 preconceptions, and test the TEAM dynamics, with regards to what tankers bring, rather than just how we perform the old roles. Half the contention is the roles have changed.
[/ QUOTE ]
While I understand what you're saying, I think we're looking at too broad a test here. If we can't define the roles we want to test, we end up with an open-ended number of tests, and no one has time for dozens or even hundreds of testing missions.
We're going to need definition of what roles need to be tested, if we're going to do any tests at all. This, I believe, was Tom's original question. Having read through the thread, my take on this is as follows:
<ul type="square">[*]Meatshield. You're the guy that attracts the attention of all the mobs and tries to hold it. My previous post pretty well covers this one. [*]Scranker. We'll need some definition of what we mean on this. My take on it is that you are the first one into most mob groups, but rather than taunting anything you just attack the most dangerous mobs, expecting that your team can handle any aggro from the rest.[*]Selective Meatshield. This is the role Statesman described. Taunt what you know you can handle, to free the team to handle the rest of the mobs. You're still probably the first one into most fights, you'll probably end up with more aggro than the scranker, and chances are your build will be more defensive than the scranker so your damage is less.[/list]Are there other roles to test? Is there anything else a tank can do?
Skip
My Char. List and Market Transactions
HeroStats Developer
Legion of Valor
Iron Eagles
I have to ask then, what are you expecting out of these tests? What outcomes lead to what conclusions?
If we do nothing but rehash the old roles, how does that address the issue of "redefinition"? (Not that I am saying we shouldn't test the old roles, I'm just trying to get a clear picture of what we are trying to test)
It seems to me that whatever new role a tanker now has, it must be defined in terms of the team dynamic, rather than the team dynamic being defined in terms of it.
I agree that testing such a thing would be impractical. But I question the utility of a test that is limited in scope to just testing tanks and other ATs performing the roles you described. That only answers the question: What makes the best meatsheild, scranker or partial meatshield?
[ QUOTE ]
I have to ask then, what are you expecting out of these tests? What outcomes lead to what conclusions?
If we do nothing but rehash the old roles, how does that address the issue of "redefinition"? (Not that I am saying we shouldn't test the old roles, I'm just trying to get a clear picture of what we are trying to test)
It seems to me that whatever new role a tanker now has, it must be defined in terms of the team dynamic, rather than the team dynamic being defined in terms of it.
I agree that testing such a thing would be impractical. But I question the utility of a test that is limited in scope to just testing tanks and other ATs performing the roles you described. That only answers the question: What makes the best meatsheild, scranker or partial meatshield?
[/ QUOTE ]
I am of the opinion that we have a chance to get the DEVs attention if and only if we can prove one assertion:
In the test cases (and it needs to be cases that cannot have holes shot in them with charges of bias) the following was true:
The team will suffer fewer defeats and gain xp faster if another AT is substituted for a given type of tank.
If we can present case after case where this is true using the same team setup and the same missions, I think we have a shot.
For the purpose of this kind of testing, the *role* of the tank is important from an anecdotal standpoint, but not from a statistical standpoint. What is important to the DEVs, I think, is the hard numbers.
Now, it is possible that this could be true below 32 or above 32 or from 22-40. It's possible that it might not be true for stone tanks, but true for Fire and Ice. It's possible that it might not be true vs S/L enemies, but true vs others.
Conditional negative results are as important as positive ones, because it points out imbalances between the sets.
I'm pretty sure that the datamining that the DEVs do looks for two things--how many people are playing an AT and how fast they gain xp. If either are true, they look for ways to fix the problem without unbalancing the rest of the game.
The general philosophy has been stated that if people level too fast, they get bored and leave, if there is only a few people playing an AT, a portion of possible market share is being lost.
[This is one reason I believe that tanks have a shot at being buffed around I8 or so.]
Mr. Lithuania
Jessica to Nathan in bed: "I'm not really bad, Isaac just drew me that way."
If we can all get a time on the test server with some notice I can get a Ice / Ice tanker there at 35 to test with, AoE debuffs, good crowd control, and defense with a few ok attacks. I think it might be a good test with a diverse tank as well as a few that have less slot options, cause that is one of the things they where pushing for it seems.
[ QUOTE ]
[ QUOTE ]
I have to ask then, what are you expecting out of these tests? What outcomes lead to what conclusions?
If we do nothing but rehash the old roles, how does that address the issue of "redefinition"? (Not that I am saying we shouldn't test the old roles, I'm just trying to get a clear picture of what we are trying to test)
It seems to me that whatever new role a tanker now has, it must be defined in terms of the team dynamic, rather than the team dynamic being defined in terms of it.
I agree that testing such a thing would be impractical. But I question the utility of a test that is limited in scope to just testing tanks and other ATs performing the roles you described. That only answers the question: What makes the best meatsheild, scranker or partial meatshield?
[/ QUOTE ]
I am of the opinion that we have a chance to get the DEVs attention if and only if we can prove one assertion:
In the test cases (and it needs to be cases that cannot have holes shot in them with charges of bias) the following was true:
The team will suffer fewer defeats and gain xp faster if another AT is substituted for a given type of tank.
If we can present case after case where this is true using the same team setup and the same missions, I think we have a shot.
For the purpose of this kind of testing, the *role* of the tank is important from an anecdotal standpoint, but not from a statistical standpoint. What is important to the DEVs, I think, is the hard numbers.
[/ QUOTE ]
Being a hard numbers kinda guy, I actually get the feeling that what Statesman is more swayed by is what he percieves as honest, unbiased philisophical points, rather than numbers. I don't think he's that good with numbers frankly.
I agree 100% with what you have written above. This is why I think designing the tests around the anectodal roles, will bias them in an unhealthy way. The teams tested must be free to adapt beyond the rigid 'roles' to provide any data on how a team will interact with other ATs compared to tanks. This will be different for different team configurations.
[ QUOTE ]
In the test cases (and it needs to be cases that cannot have holes shot in them with charges of bias) the following was true:
The team will suffer fewer defeats and gain xp faster if another AT is substituted for a given type of tank.
If we can present case after case where this is true using the same team setup and the same missions, I think we have a shot.
[/ QUOTE ]
I think you overestimate the devs. After all the tests before and after I5 about Invinc and they basically blew it all off. But for our own edification I'm up for it.
So the question isn't really what roll a tank can play its really whether he can help the team as much or more then any other AT slipped into his place on the team.
Why not start simple? Take the five basic ATs and build a team with one of each. Make the level like 32-39 range so we avoid Epics skewing things but were the ATs have most of their powers. We build a team with each then we start swapping out the tank for others and see what changes. It might give us an idea of what else we need to look at later.
Yes, Da5id. You are right.
If we put together a team and just vary the tanks and other ATs for a given set of missions without worrying about the tank's or other AT's roles (in other words, assemble the team and just let it happen) the optimum balance for the team should be achieved. That data point is what I think we're looking for.
Mr. Lithuania
Jessica to Nathan in bed: "I'm not really bad, Isaac just drew me that way."
Yep. It will just take a little longer to feel out the ATs and players to find that optimal strategies.
(You know this might be fun as a regular event and messing around with just bizzare team combinations)
A good starting place would be what D_O suggests. we can expand from there, seeing as how testing all combinations would be impossible.
I think this is a good idea, DO as far as it goes, but I think the pentad team is a bit too small to get the most out of it.
Now, we might consider a six-person team to start with one of each AT plus a Peacebringer. This will kick up the mission difficulty by +1 and provide even more flexibility.
One of the more important things to look at from an anecdotal standpoint is how the roles of the other ATs shift depending on who the sixth AT is. Does the team suddenly get safer in some cases? Is so much damage being handed out that the team walks through the mission? Is aggro not a problem anymore with one AT?
I also like the level 32-37 range for a starting area. Each AT has it's level 9 Primary but no Epics nor level 9 secondary.
Mr. Lithuania
Jessica to Nathan in bed: "I'm not really bad, Isaac just drew me that way."
What was the dates for this again? I will be on Vacation from the 17th to the 28th. I will be glad to help, and I have a 20 something verison of Guardian on the test server already plus I can copy Guardian now and I expect him to hit lvl 38 or 39 before the vacation. I can post my build right now if any wants to see if it will be beneficial.
And if anyone does want to see it, and I don't see the request for it, just PM me. 
Also, my style of tanking is to grab as much aggro as I can. From whatever is there. Bosses, AV's, lts, whatever. And then let the team take care of them once I have them stickied. It will not be unusual to see me with a full 17 mobs stuck to me.
Prof
p.s. I can also still get the Mark IV store mission.
[ QUOTE ]
One thing I've only heard mentioned once or twice (and one of those times by Statesman) is the role of the Monitor.
What does any kind of healing defender, controller or corrupter do? They monitor the condition of their teammates at all times and make sure that their health does not drop below a certain point.
They additionally, and rightly so, bring more to the table than that. Certainly, they are much more than healers, even if they're mistakenly seen that way.
Perhaps one possible new role of the tank is to provide a constant stream of damage while monitoring the condition of the rest of the team and handing out the taunts that are required to keep them safe?
[/ QUOTE ]
My Empathy/Rad is one of my busiest and most demanding characters to play. The role of "monitor" can be entertaining, but currently tank powersets simply don't lend themselves to that. If you see that someone is in trouble (squishy or not), the only tool available to assist is Taunt (which is a secondary for that matter). Gauntlet is also based off of secondaries.
If you want to be a monitor given the tank sets, your only options are to try and target through another player, or hit the enemies that are causing the other player grief to invoke gauntlet. Given the way combat actually works, that is can be very hit and miss, making the monitor role tenuous at best for a tanker.
Defensive primary, melee secondary. Primaries don't contribute more than a defenders to defensive capabilities, and secondaries damage output is generally so much lower than a melee primary that it is marginal, especially in a game that is all about doing damage.
Finding a role for the character here outside of meatshield is going to be difficult at best. If the Statesman example is the expectation of contribution, then I expect that the AT will be marginalized once the player base can observe tha actual contribution made by tankers at higher levels. However, since the game itself is such a moving target, each issue seems to bring these sorts of questions up for somebody anyway.
The testing should be informative if nothing else. We'll see if Cryptic pays much attention. We did get Jack's attention once here.
When you do your tests with multiple tankers look for the ability to play ping-pong with the mobs. One tactic is to vary AOE knockdown attacks and taunt. Good for argo control bad for damage.
I've not tested it on the new CoH but a tela-tank might do the same thing by bouncing around, taunting and knocking people around.
As for my own characters: I can copy to test the following charactes (most in heavy need of respec)
50 Inv/EM/En tank
37 Grav/Kin controller
36 PB
46 BS/Regen Scrapper
24ish Fire/Fire tank
all sorts of low-twenties to high teen heroes.
[ QUOTE ]
If we put together a team and just vary the tanks and other ATs for a given set of missions without worrying about the tank's or other AT's roles (in other words, assemble the team and just let it happen) the optimum balance for the team should be achieved. That data point is what I think we're looking for.
[/ QUOTE ]
You're going to need a lot more than one weekend, then. There's millions of possible combinations of 6 characters, when you consider different powersets, different power choices, even different pool power picks. And that's not even considering playstyle - two sets of players with the exact same character builds might approach things in a completely different way.
Now, I understand that we don't need to test them all. But say we find 50 different team compositions that are better without a tank. So what? There's millions of other team combinations that didn't get tested and might possibly be good with that tank. That same team might even be better with the tank if the other characters on the team are played differently. And even getting 50 would take much longer than a single weekend.
I don't mind testing, but we need some better criteria than just 'we'll do what feels right on the team'. If that's the only criteria we can come up with, all we're testing is our own playstyles.
Skip
My Char. List and Market Transactions
HeroStats Developer
Legion of Valor
Iron Eagles
[ QUOTE ]
As for my own characters: I can copy to test the following charactes (most in heavy need of respec)
50 Inv/EM/En tank
37 Grav/Kin controller
36 PB
46 BS/Regen Scrapper
24ish Fire/Fire tank
all sorts of low-twenties to high teen heroes.
[/ QUOTE ]
Happy to help out with what I have listed under my signature.
[ QUOTE ]
[ QUOTE ]
If we put together a team and just vary the tanks and other ATs for a given set of missions without worrying about the tank's or other AT's roles (in other words, assemble the team and just let it happen) the optimum balance for the team should be achieved. That data point is what I think we're looking for.
[/ QUOTE ]
You're going to need a lot more than one weekend, then. There's millions of possible combinations of 6 characters, when you consider different powersets, different power choices, even different pool power picks. And that's not even considering playstyle - two sets of players with the exact same character builds might approach things in a completely different way.
Now, I understand that we don't need to test them all. But say we find 50 different team compositions that are better without a tank. So what? There's millions of other team combinations that didn't get tested and might possibly be good with that tank. That same team might even be better with the tank if the other characters on the team are played differently. And even getting 50 would take much longer than a single weekend.
I don't mind testing, but we need some better criteria than just 'we'll do what feels right on the team'. If that's the only criteria we can come up with, all we're testing is our own playstyles.
[/ QUOTE ]
The DEVs already have a million different combinations going in the game. The thing about those combinations is that there's so much noise that the signal can get lost. That's why needed changes can take as much as a year to a game and why, for example, they couldn't tell that Invincibility was broken.
We start simple with one set of information under controlled circumstances--same mission, five of the six players the same every time, rotate the sixth.
This gives us one starting data point. We examine it, draw any conclusions from it we can and then design the criteria for the next test.
It's possible that the test will tell us nothing. It's possible that it might enlighten us on the future of tanking. It's possible that we're all wrong and the team with the tank always do better with this particular combination of other ATs. What we need to look for are things that are consistent across the test.
What we don't have is any reliable data under controlled circumstances. (The Tanks and AV study being a possible exception, but the variables are so many in those tests that it's hard to draw definite conclusions, although some interesting team dynamics have already been revealed. Folks should read that thread for those dynamics.)
Bob, I like your idea for differing team makeups and I think that it will be a good idea to reference it when we make the sextad. Your notes also have convinced me that choosing the substituted AT should be more complicated than simply "popping in a random Defender."
The current testing date is November 26th with active planning beginning around the 16th. We're trying to bang out the "boundary conditions" of the first set of tests right now.
Mr. Lithuania
Jessica to Nathan in bed: "I'm not really bad, Isaac just drew me that way."
My tanks role now is two fold, take the alpha strike, and keep the boss busy. With my fire/axe tank, I can handle just about any boss one on one. The blasters need to snipe out the leuts, scrappers can help keep a few busy.
It's the whole role of the team that has changed, not just the tank. It used to be, tank herds, everyone pounces on the boss, then mops up the stragglers. It's about the dead opposite of that now, and much more realistic to a superhero battle. I mean, as winpy as they are, if we all pounced on a group, regardless of aggro, some of the minions are not going to focus on the tank, but are going to focus on who's attacking them!
Each person in the team has a job to do. Mobs need to be taken out with strategy, not brute force. The tank takes the alpha strike as the controller drops an area hold. This takes the minions (most of them) out of the fight for a minute. The scrappers now need to pounce on the leuts (1 each, not at the same time) to take agro off the tank. The blasters lay down supressing fire and pick off runners when they're not sniping at the boss. The tank keeps the boss off the snipers. Defenders keep the team buffed and mobs debuffed as allways and the controllers heal the tank (mostly) and the rest of the team as needed.
This strategy works for everything except some AVs. Certain types of mobs may be the focus of snipers and controllers before the boss are (SAPPERS!). A team that plays well together, and often, will succeed. Untrained random people need a strong leader or the team will fail.
Many "younger folk" playing this game don't have the patience or experience to play a team strategy. They should save us all the trouble and quit now.
...and soloing as a tank? You're a tank... You don't solo any more thanh a blaster does (accepting that at high level blasters probably solo better than any other class except scrappers and MMs).
[ QUOTE ]
I think this is a good idea, DO as far as it goes, but I think the pentad team is a bit too small to get the most out of it.
Now, we might consider a six-person team to start with one of each AT plus a Peacebringer. This will kick up the mission difficulty by +1 and provide even more flexibility.
One of the more important things to look at from an anecdotal standpoint is how the roles of the other ATs shift depending on who the sixth AT is. Does the team suddenly get safer in some cases? Is so much damage being handed out that the team walks through the mission? Is aggro not a problem anymore with one AT?
I also like the level 32-37 range for a starting area. Each AT has it's level 9 Primary but no Epics nor level 9 secondary.
[/ QUOTE ]
The testing sounds like it might be interesting, and depending on the timing etc, add me to the list of possibles to help out.
The very thing that is so attractive about CoH is the variety of builds, and that is going to make it very difficult to get any meaningful results unless your goals are very specific and well defined.
This is a post where Statesman affirms the roles of each AT:
http://boards.cityofheroes.com/showf...art=7&vc=1
Note that it is dated 5/25/05, which means that is the breakdown of the ATs with I5 and ED in mind, because of this post which says ED was in mind since March 2005:
http://boards.cityofheroes.com/showf...o=&fpart=1
So if both of those posts are true and correct, then the role of the tank is still to be meatshield. It also says that scrappers specialize in soloing. I think you should keep the initial testing as simple as possible. Namely, put together a 4 man team:
1 Blaster
1 Controller
1 Defender
1 Tank
Use an Invul/SS tank and use a build that is as defense oriented as possible. Then use the same team but build the tank as offensive oriented as possible. Hopefully, this will give some idea of whether it is the tanks primary or secondary which is providing usefullness to the team. Next, use the same tank and build it for soloing and do the test again. Lastly, replace the tank with a MA/Invul scrapper built for soloing. I suggest Invul/SS and MA/Invul because they are about as similar as you're likely to get.
A comparison of those results should give you an idea of where to branch off in your testing.
skeelos, we can't use an INV because of the bug in Invincibility, I'm afraid.
The mid-size team idea is an interesting one, but the reason for dropping a scrapper isn't clear to me. Just because the scrapper *can* solo doesn't mean that he will be doing that.
One of the problems with making first a Defensive tanker and then an Offensive one is that puts them right back in the roles that they were in prior to I5--meatshield and skranker. I think we decided to move past that a couple pages ago.
Thank you, though for the offer of your help. I'll put your name down on the list I am compiling (10 plus myself at the present time.)
Mr. Lithuania
Jessica to Nathan in bed: "I'm not really bad, Isaac just drew me that way."
So we're looking 32 to 37 no Inv, neither tanks nor Scrappers.
Well that drops me to two.
Smoke'n Jack Fire/Fire Tank (needs to be respec'd) level 32
Dr. Serber Rad/Rad Defender (also needs respec'ing) level 32
I would think we're looking for whatever build is your normal playing build.
I have a 31 Ill/Kin Controller that should hit 32 soon. He can help out.
[ QUOTE ]
So we're looking 32 to 37 no Inv, neither tanks nor Scrappers.
Well that drops me to two.
Smoke'n Jack Fire/Fire Tank (needs to be respec'd) level 32
Dr. Serber Rad/Rad Defender (also needs respec'ing) level 32
I would think we're looking for whatever build is your normal playing build.
[/ QUOTE ]
Yeah. I'd like the expertise to be approximately even across the ATs and powersets. As has been mentioned before, an INV can be used for the first run, because that won't be counted anyway, since it will take longer than any subsequent run due to familiarity with the map.
I think that the rotating ATs should all be at the same level, no matter how we construct the rest of the team.
If they change Invincibility prior to the test, we can toss the INVs back into the mixture.
Mr. Lithuania
Jessica to Nathan in bed: "I'm not really bad, Isaac just drew me that way."
[ QUOTE ]
[ QUOTE ]
I didn't want to go there.
The thing is, that seems to be what States' expects to be the role of the AT. If you are doing more, then the AT may need to be taken down some more.
How much is enough, and what should you expect. Apparently the contribution that Statesman's Tank made was sufficient to meet expectations.
If that is the case, then the question becomes, is it worth playing a character that contributes that much (little)?
[/ QUOTE ]
He "tanked" 1/3 of a spawn for 7. He took up a little more than twice his share. On a team with any sort of non-melee AT i'd expect any scrapper to at LEAST do this much. I think it's pretty clear that states' grand vision for tanks is nothing more than scrankers who draw a little more aggro than the vastly more offensively effective scrappers.
[/ QUOTE ]
Exactly how does a Scrapper with less defense than a tank do do at LEAST as much defensively? That makes no sense. Since other than Regen and Dark they are using the same numbers but lower than Tanks no Scrapper is going to be able to deal with damage better.
Statements like this are why nothing comes of these threads. There is no basis for them. They are little better than the obvious flames.
The Devs respond to facts not opinions.
----------------------------
You can't please everyone, so lets concentrate on me.
[ QUOTE ]
I'm a Scranker that tanks aggressively through violence. This must be a truly offensive standard to the true tanks. 
[/ QUOTE ]
What an excellen turn of phrase "Tanks aggressively through violence." I love it. Bravo to you sir.
I personally was a classic meat shield up until ED. Now I am squarely in the Scranker camp. So much so that it seemed to me that it would be a better idea to take flurry instead of RPD. I tried it on test and damn if I wasn't right. (For me anyway)
Speaking of the scranker vs. scrapper - I was on a team last night - me lvl 42 inv/ss : 43 katana/regen scrapper : a bubbler : an ice controller : and a kinetics controller. We were figthing Nemesis - mostly fakes with a pile of jagers around them. Due to buffs my survivability was fine and with hasten combined with the speed bufs I was putting out a pretty respectable DPS - not as good as the scrapper however. BUT on a couple occasions, the scrapper got antsy and ran in before me taking the full brunt of the combined enemy attack. This would usually result is his death an me running in and taking up all the aggro (using a combination of footstomp and taunt and hitting the boss ) before the mob turned on the fragile controllers. Usually it worked, sometimes a troller would go down since Tanks can no longer grab aggro the way they used to.
This is the first team I have played with other than my usual SG mates - and here is what I walked away thinking:
The Tank is no longer responsible for commanding all aggro (which I personally think is unfortunate). The main purpose of the tank seems to be that of "wedge buster", that is to say, go in first as usual, take the alpha strike and absorb all the initial damage, then start swinging. The other members of the team then do their jobs which is lock down and defeat the mob while you pound away at the most fitting target - i.e. Boss/Sapper/illusionist/whatever...
So westill do more than a buffed scrapper, but only in the initial moments at the start of a battle. After that we can just stand around an flex and watch the rest of the team do their thing.
As A side note - I solo very well using a combination of secondary powers (attacks), hasten, conserve power and rage to defeat mobs before they can do too much damage to me.
wow.long post.
[ QUOTE ]
I think we really need to have an idea beyound the extremes of Scranker and Meatshield.
[/ QUOTE ]
The big problem is that there's not much of anything between the extremes. That's because the Tanker's capacity to operate at the meatshield extreme has been gutted, and the meatshield role is simply not needed in the game philosophy. So if he's not doing all that much good at that point, it's not going to get any better as he reduces his aggro control role in favor of doing more damage. He should therefore max out his damage as that is the bulk of his contribution, and spreading the damage around to himself is a bit of an afterthought.
In the post-ED world, anyway, everyone above a certain level is effectively a Skranker because there just isn't the spots to put the slots in defense. Post-40 or so your offense is going to be maxed out, or so close to maxed it won't matter much, simply for lack of better spots to put slots in.
I think the philosophy for any remaining Tankers in I6 should be to go in and start walloping the biggest guy you can find to draw some attention, and let the rest of the aggro fall where it may. You're going to be getting a lot of aggro from Gauntlet anyhow, quite possibly more than you can afford to take, and it should be obvious to everyone on your team you can't expected to take on all, most, or even all that much of the aggro anymore. Cold, but sadly, you *are* doing your best; this is as much tanking as you're going to be able to provide. At least this way you're using your defenses to the limit that way, providing an useful service by taking on the alpha, and applying your mediocre damage to help out. Taunt is what it's always been since Gauntlet and auravoke were introduced -- a nice save-the-squishy power, but hardly a mainstay or a must-have.
In this philosophy your role is basically to stand in front of the big guy and do as much damage as you can to him, pray you don't draw so much attention that you get pasted, and keep what you can busy while the real heroes do the work. And don't touch an Archvillain without a pair of pocket Defenders.
This approach is not going to be terribly effective, and I don't think it does much to justify the Tanker in a team. You're still putting out mediocre damage and not doing much to help out with team defense. But it's about the best you can do as it beats the alternatives, such as they are.