Guild Wars 2

Crinn.7864:

Lordrosicky.5813:

Wrong. They have to let the meta settle and people enjoy playing their builds and fine tuning them. Nobody wants regular changes. 6 months is fine for an mmo

6 is a bit much. 3-4 month cycles. Anything less that 3 means the meta may not have fully settled.

1 month is enough time to let the meta settle, look at what happened this season. People played DH up the kitten , they realized they get kitten on by everything else and in the first week of pro leagues bunk mes was played, then second week there were 2 bunks mes’s on each team.

People know well enough the state of the meta after a month passes by, and even to an extent Anet should still be throwing out monthly hotfixes for different things. Look at major popular games that have a HUGE competitive ESports scene, LoL and DoTA both do bi monthly patches DURING their tournament seasons.

While they may not be doing anything drastic during the larger tournies its still small tweaks, and thats what Anet lacks. Is a dedicated pvp balance team.

Anets balance team consists of people from WvW, PvE and PvP… Nothing should be balanced like that, you should have separate teams working on separate things that aren’t as tightly integrated as they currently are.

/grammar sucks balls, idc you understand my point.

Except that ANET has never actually done fixes or balancing on a regular and frequent basis. I see no reason to believe that they will start now.

Actually yes.. I like bunker a kitten of a lot better than zerker. That’s where we are going. With no bunker and the level of damage and cc, look for 2 second fights.

Henceforth, “skilled gameplay” will consist of pushing buttons really really fast.

A.net,

HOT balancing is obviously still in its infant stages. The proleagues are your greatest opportunity to make weekly balance patches and get immediate feedback. Since money is on the line, you can be confident that the proleague teams are all trying to discover the most optimal builds/comps to run in high-level play, and that they test these builds/comps extensively in scrims. Part of being a pro player is understanding the game well enough to quickly adapt to new balance patches.

There’s no need to wait for 80% of the proleague to finish before making substantive changes. For example, after nerfing shieldblock and precog, it was pretty clear by next game that bunker mesmers were still crazy effective. So the next step could have been getting rid of quickness/slow stomps/resses, or even getting rid of the bunker amulets.

Waiting until the proleagues are 80% over just resulted in a bunch of incredibly boring games that the viewers didn’t enjoy watching, and a lot of missed opportunities to playtest more incremental balance changes. Even the proleague players didn’t seem to be enjoying the experience (as evidenced by one team’s ragequit).

It’s too early to tell what the next meta will look like at this stage, which is part of the problem. The current balancing strategy is like throwing a bunch of paint at a wall, which the playerbase is then stuck with for the next 6 months.

You’re already paying over $200k in prize winnings to make the proleagues happen. You might as well get the most bang for your buck by using each week’s proleague game to make frequent targeted balancing patches. If one tweak doesn’t work out, you can just reverse it for next week’s game.

PLZ for the low of god add this amulet combination:

+1050 power
+1050 precision
+560 toughness
+560 ferocity

This will be highly desirable for rangers, mesmers, warriors, and necromancers who wish to play a power build. Currently thees classes can pick berzerker (an go full glass) or marauder to get some defence. The stat combination i proposed would be a competitor to the marauder amulet since the above mention classes have medium and high hp pools and getting more hp is somewhat redundant (not on all builds but still), but they all would greatly benefit from some added armor.

Back on topic i love all the item changes. The removal of celestial and introduction of these new stat combinations will force ppl to make more of a choice when making a build.

Honestly I think the 20% damage buff on unload was stupid, yes it’s nice but does it make a difference does it change the set at all? No. It just makes thief like longbow ranger which is something no class should strive for imo. I don’t like the idea of spamming 3 and doing massive damage at all, it’s not skillful in the slightest.

However, I think one thing that could happen that would make p/p decently viable (at least twice as viable as it was before) is the blast finisher dodge on Daredevil needs to go back to being a leap finisher. This along with Pulmonary Impact to use in combination with skillfully timed headshots would make P/P stronger than it ever was before. As you’re gaining stealth on demand that can’t be interrupted, as well as an amazing damage proc that can easily hit for 3k+ which takes skill to use. Both of which making Headshot as well as Black Powder on P/P very strong skills.

Greatsaw!

Greatsaw!

Greatsaw!

Greatsaw!

I don’t care about anything else. Maybe the scythe. But GREATSAW!

I don’t care if it’s a recolored version of it or something silly like that. PLEASE ANET!

I was there for Halloween. I tried my hardest to get one from a chest. RNG never gave it to me. It’s literally the only weapon I have ever wanted in the game and it’s too expensive to actually buy on the trading post. The Scythe Skin isn’t even available anymore.

These skins were under a stupid RNG system before and need to come back. Please!

I say, it’s a beta, treat it as a beta, not like Preview.

Next time, give it 5 days. During those 5 days, test couple versions. On Monday, let’s say that it’s regular what we’ve got. On Tuesday, make some significant change to Archers for example and see how people react. Maybe on Thursday, bump the player cap to 6 or 8 per team and see how people like it.

Run similar rating/opinion system like in GW2 betas. After the match, give an option to leave feedback in-game.

As I said previously, I get that you want to time your previews and limit them, so people stay excited. It’s okay for majority of PvE, features etc.

But when you want to get feedback on competitive game modes, treat it like a beta and give it a proper shot.

As of my concerns – I would slightly reduce the amount of HP on Doorbreakers and bump their armor instead. They’re too easy to kill as Zerk and take too much time as Condition.

Also, make that Supply Depot more important. After first 5 minutes it was usually that people were not even spending or getting any supply anymore, just waiting for Hero channels. Once the inner gates are down, getting supply is straight useless (because of archers, mostly). Add some variety. After Doorbreakers break inner gate, replace them with another type of mob/buff. Maybe defensive one? Maybe upgraded version of Archers?

Third concern – Guards. Reduce their leash range, make them always prioritize Doorbreakers, make them ranged with cleaving, piercing projectiles.
I bring this up because I was easily able to grab the aggro from all guard packs while escorting Doorbreakers, keep them away from any skritts (just being aware of their bolas killing skritt if they’re in crossfire range), kite to safe spot and finish off by myself while all 5 Skritt easily take down all gates without any focus.

Evan Lesh.3295:

We actually do this during matchmaking to err on the side of inactivity dropping skill level.

This makes sense as long as you don’t mess with the match prediction part of the algorithm – basically, players who have been away for a while are paired down in matches that Glicko would say they are favored to win, but due to unmodeled decay should actually have a 50% chance to win.

Evan Lesh.3295:

One problem we really want to solve for a new ladder system is how easy players can understand their rating.

You hit a trade-off between legibility and accuracy. You can clearly use something like the latest ladder test season, which is pretty transparent in what it’s doing, it just also clearly doesn’t map onto performance well. Glicko and Glicko2 are pretty transparent though require pretty sophisticated statistical training to understand. Not sure how much simpler you can make it without it losing any semblance of accuracy.

Evan Lesh.3295:

Sure, link all the papers.

I presume you’ve read through most of Glickman’s stuff. The most relevant papers for building a ladder are those by Graepel and Herbrich regarding the TrueSkill system.

Going over the basics laid out by H. A. David is worthwhile, or even the original system by Elo; I don’t say it lightly, because all of these systems are founded on certain normality assumptions, and I am a bit worried that your own modifications have started violating those.

If you’re going to take a stab on your own, you should be very familiar with the Kalman filter and Tikhonov regularization, both of which are at the core of the math of any sort of rating system.

Evan Lesh.3295:

Ensign.2189:

Evan – this is a problem with a pretty well understood solution under Glicko – you rank players by their point estimate MMR, minus a multiple of their rating deviation (2 or 3 typically).

We actually do this during matchmaking to err on the side of inactivity dropping skill level. If we had a strict Glicko leaderboard now, I think this is an approach we would take.

One problem we really want to solve for a new ladder system is how easy players can understand their rating. With Glicko, we don’t really get that. This is one benefit of the point system we’ve been testing (Though it needs some in-game info).

Ensign.2189:

I can link relevant papers if you are interested.

Sure, link all the papers.

Ease of understanding the leaderboard shouldn’t take priority over having a respectable leaderboard.

That said if you want to maintain a respectable leaderboard with the points system, you should drop any and all glicko-based matchmaking what so ever and make it a truely random arena. Only then would a points-based system somewhat hold weight.

Evan – this is a problem with a pretty well understood solution under Glicko – you rank players by their point estimate MMR, minus a multiple of their rating deviation (2 or 3 typically). So your ladder rating would be something like

LR = MMR – 2 * RD

This has several convenient properties:

- New players do not immediately shoot up to the top of the board even if their MMR is high, as their RD is also high until they have settled at that level for some time.

- It is a ranking system that, for most players, will have their rating generally move up with games played – as even a 50% W/L ratio at the same skill level will lower their RD, and thus increase their rating. This helps immensely with new players, who get to see some ‘progress’ when they are most likely to be discouraged – brand new players will go up in rating even with losses.

- It handles decay easily, as a player’s RD should increase with inactivity – which consequently makes them fall back down the ladder.

Sure there are some complicated systems you can use that can feel better (various league systems), but unless those are forthcoming a visible MMR – 2 RD is a quick and dirty system that works surprisingly well.

I can link relevant papers if you are interested.

I disagree. Very often I want to play with my guildies, who are completely new to PVP. I hate the hotjoin and I don’t want to lose rank points so we have to resort to unranked games. We lose the majority of our games and we don’t bother with voice chats. And I’m willing to bet these sort of ‘casual’ groups are the majority out there, out of all the premades.

I think the only thing that might remedy this is how much more likely are premades to win against a full solo queue, or a smaller premade? Actual statistical data. I solo queue a decent amount and I gotta be honest, I literally don’t even notice if a team is a premade unless it’s the same guild tag. I’ve done some 500-40, on both sides, as a team queue and solo queue, but they certainly aren’t a norm. It’s hard to not just take the complaining as a hyperbole after a few bad games.

Stronghold has a lot of problems in its current state and if it goes live without some significant changes, it will be skyhammer 2.0 (best thing ever for the first few days then pushed aside to the point of people not wanting it).

I played for about 7 hours almost entirely with a 5man team, so these views are based on coordinated strategies.

PvE Tactics Decide the Match
A group with some support can keep doorbreakers (DBs) alive through player damage until both gates are down. The biggest effect on the DB push is how many are one-shot by the guards. The melee guards are especially potent since they can cleave down three at once.

However, the guards can be pulled off the path of the DBs long enough for the DBs to pass through. But if you slip up even a little, a single attack can take out half your DBs. As a result, the single biggest deciding factor in whether your DBs break the inner gate is execution of PvE tactics.

No Need for More Supply
If you corral the guard NPCs properly, your initial wave of DBs (between 5 and 10) can break through your opponents’ inner gate. Unless you completely fail at PvE, you’ll never need more offensive NPCs. That means no reason to ever visit the supply depot after the first 15 seconds of the game. This negates what was intended to be a crucial feature of the map intended to promote PvP combat.

Race Instead of Competition
With re-supply being unimportant and manipulation of NPC guards being the deciding factor, the game mode turns into a PvE race. Execute poorly and you lose unless your opponents also fail. Defending with players only serve to slightly delay breaking through a gate or putting damage onto the lord.

The most effective thing that a defending player can do is prevent opponents from corralling the NPC guards. The second most effective thing is to instant-res the lord in order to buy time for their team’s offense.

Little Chance for a Comeback
If you happen to have poor execution, there’s no way to recover from it on your own. If you defend your lord, it still eventually dies. Even wiping the opposing team in your lord room doesn’t buy enough time for your team to run across the map and kill their lord. Your opponents will resurrect and just rush your lord again, which is likely at low HP already.

Even in a match where both sides fail the initial DB push, the mist heroes are inconsequential. They can be killed before doing any significant damage to a gate or lord and don’t do anything else along the way.

PvE races that depend on out-smarting a few NPCs won’t go over well in a PvP mode. Once PUG groups settle on some variant of rushing with DBs at the start and have some minimal coordination, games will have become who can PvE best, not who can PvP best. What can be done to change it? Here are some of my ideas:

Lower Doorbreaker HP
Players should be able to kill DBs faster. As a result, there is less need for NPC guards to kill them and gives defending groups an actual chance to stop DB trains.

Guard NPCs Don’t Doorbreakers
It’s already been shown that the current NPCs between the two gates can be easily out-smarted. A small goof in NPC management that should not be the difference between success and failure. The purpose of the guards should be to prevent attacking NPCs from getting through gates quickly without enough help, and they can do that without instantly killing DBs. This change combined with the previous one would put far less emphasis on defending NPCs.

Doorbreakers Attack Once and Die
Suicide Skritt bombers (or to be animal-friendly, they toss their bomb and run away). Their damage would have to be increased (20-25% of gate HP probably compared to current 12.5% or 1/8). The aim here is three-fold. first, You don’t have to stop all the DBs in order to ruin your enemies’ plan or simply slow them down. Second, the push with 5 initial supply of DBs could actually be stopped from getting through the first gate and would be incapable of getting through the second. A slightly delayed 10 supply DB push could be easily stopped from breaching the second gate. This slows down the PvE race and makes the supply depot actually important. Third, it creates a more distinct trade-off with archers. Archers will do more damage if left alone and can hit the next gate or lord, whereas DBs would be to break a heavy defense or to take a quick risk.

Mist Hero’s “Elite” Death-Touches a Gate
The elite would have its current effect on the lord, but also a cooldown to prevent the hero from going through both gates or a gate and hurting the lord before it reasonably would be killed. This would make the mist hero more important and a bigger priority. It could be used to break a turtle team or as a way to leap-frog ahead in the PvE race.

Our Beta Experience:
Already feels like a fun game mode, and refreshing after years of only conquest based maps. We had a guild group (3-5) queue most afternoon, on mumble. Our members are mid-rank, so feel free to take the following feedback with a grain of salt (mix of phoenixes, bears and wolves mostly, but we even had a few rabbits try out the mode and enjoy it).

Thoughts on the map/gameplay:
1) Communication is king. Premade seems like a much stronger advantage here (and would likely need a correspondingly higher mmr penalty than in conquest to keep matches competitive if pugs were to ever be put up against premades). Case in point: we are not exactly amazing, but we felt like we are able to fight above our technical skill level because of the advantages of voip.
2) Simultaneous lord fights feel amazing. Banner. Stability stomps. Boon removal and CC. Really fun moments for us in a couple of games. Unfortunately, they were quite rare. The far more likely scenario was a one-way fight at lord.
3) The map might have a bit of a snowball problem. Although the idea of two-lanes is clearly there to provide a risk-reward to pushing too kitten the enemy, this tended to fall apart. We won a lot of our games simply because we would win the early encounters. Getting down the outer gate early is a very strong change in the map, and even more so if you are able to push in numbers when it goes down.

This might be because of a couple factors:
- Defending NPC’s feel a little weak. Even with active defenders, it was possible to take them down as part of a push, or have one player wreak havoc –because of this archer’s felt considerably weaker and less useful than they probably should be.
- It’s really, really hard to re-establish map presence once you’ve been pushed into a defense. I think out of the all of the matches our guild played, we only lost one match where we got inner gate down first. In part, this is because there is a huge position cost to committing to defense. The enemy will respawn and make it back onto the map too quickly for your defenders to get back out onto the map and take advantage after a successful defense (run supply/reapply pressure on their defenses). The worst part of this was that there was never a real feeling of risk to making a lord push.
- A possible comeback-mechanic/suggestion:As this map already has a ton going on, it may not be wise to add another mechanic to the map, but it feels like it could really use a strong comeback one. The heroes don’t quite do it because of the positional problems from above.
Proposal #1: increasing respawn timers.

Proposal #2 : a one-time movement-speed advantage that only becomes available for use once after inner gate is down. Example: a shrine inside the home base that when channeled gives slick-shoes type speed to the entire defending team for a certain amount of time (usable once per game). This would create a way for a defending team to rapidly reestablish some map presence, without being a mechanic that helps the stronger side, while simultaneously adding a little bit of risk to pushing into their lord room. And yes, portal, but relying on one skill on one profession doesn’t make for an interesting general comeback mechanic.

Thoughts on scoring/win condition/timing:
Overall, the timing of the matches and win condition felt pretty good. There is one caveat – we had two close matches where we won/lost because points from kills even though the opposite lord was at half-health and the other inner gate was still up. That didn’t feel like a good win, and it felt like a horrible loss.

Thoughts on the NPCS/Trebs:
1) Doorbreakers and trebs feel about where they should be. Fix the treb targeting.
2) Archers need some love. Whereas doorbreakers feel like awesome little kamikaze tools, archers feel a bit weak. Maybe consider upping the damage from archers to guards and slightly upping the overall health of guards as well? Also, to our archers: why you run off after non-NPCs?

Nitpicky-stuff:
1) Why have the gate portals close when a gate is destroyed?
2) Is red spawn a few seconds back compared to blue? Seemed a tiny bit more rushed to get to gate and stack swiftness.

Ok, was able to get a good 6 hours (including queue waits) of matches in last night. General impression, liked it. Tried some pug, some groups with friends. Didn’t try any premades where people spec’d for it or brought classes that would compliment the team or were traited for the map.

Was able to run all classes except Guardian. Can see roles for just about any class so that was nice. Could see wanting to adjust some builds for this map. Saw lots of various techniques used and there didn’t seem to be a distinct, do this action which was a positive to the map. Now granted one partial day is not a lot of time to get set into ways but could see a lot of different options that could come into play which means it stays fresh longer.

Saw only one bug over the night where mist essence was displayed on the map but could not be summoned. Power level on the archers and bombers seemed to be pretty good as far as resistance and damage dealt. Treb didn’t come into play as expected as much as compared to Kylo. Pace in match seemed to be fast which was good, you really didn’t have time to waste.

Once again over all liked it and will look forward to playing more of it and seeing other maps with that style of game play.

Finally played enough to get a good understanding.

+Tide and flow of battle here is BIG something Fantastic and nonexistent Conquest.
+The map is beautiful and the NPCs talking makes it feel like a warfeild. Makes me miss the beta days. “You got the keep, keep it!”
+Had amazing success kittenter Mes on Defense. Not sure why, however.
+I can see this game mode being crazy in high level play, with rotations and plays. No surprise seeing its based on a “moba map”. “Siege” no less, my favorite mode outside of “Arena” in Smite.
+Decent to Good balance.

Bad things
-A bit more complex, after two years of Conquest many players were confused, a guidlie explained it several times and we won those games. A TUTORIAL FOR THIS MAP, WOULD IMPROVE PLAYER EXPERIENCE. AND MAKE IT MORE INCLUSIVE
-Archers should do a little bit more damage.
-Heroes need a buff, I feel they should be able to damage players and be a threat to take out.
-Needs a death punishment.

Suggestions
=Tutorial added where the water tutorial was.
=Buff Heroes or..
=Less Heroes. Perhaps where the second set would spawn, instead add a “Planar Anomaly” that summons a Meteor Storm, or grants a buff similar to Ferocity, points on kill. It will add to the tide and flow of the battle to whoever gets it.

I was interested in Stronghold as soon as you mentioned it, I watched the demo streams about it. Was glad to try it today. I’ve done about 10 game overall, almost half and half with my teammates or pugs.

The good :

First thing I wanna say is that it’s good to have some new things! Change is good for time to time to shake the meta and to liven up the scene. We definitely need to continue promoting PvP ; new maps and/or skills changes.

Love the dynamic of the map, it’s kind of overwhelming at first but there is always something to do. Could be also bad at the same time for some player because it kinda feels like you’re not doing enough, but to me it’s not a big deal.

I like the theme of the map : Pirates vs Knights. Could be nice to have some explanation and context about the events ; Is it just a fictional event for the sake of it or does it have some lore related to it? Could use more development on that side.
Heroes are good objective, they are strong enough to have an impact but not too much to break your team if you miss it.

The layers of the maps are well disposed; Lower ground in the lane and higher group on the sides.

The bad :

May be more fun to have group > 5, but you would really need to punish death a lot to do so. (see death punishment later)

Treb don’t have enough impact on the game. Being able to treb your own lord room maybe? How about being able to damage the doors with it?

Most of the time, it’s best to ignore the player when on the defense and just kill the NPC, same thing with initial push when attacking. Need to be more incentive to kill players instead of just focusing NPC.

Don’t have big comeback mechanics or we don’t know how to use it properly because when you get a lead, it’s easier to just rush and forget defense, even if they defend it successfully, lack of death penalties doesn’t do much to the defender.
NPC are too statics it would feel better to have them move around, dodge from time to time.

Timers is not fun. At some point, if the game is close enough and draw long, the winning team in points can just tank and wait for the timer. For me it’s not fun. Consider having unlimited time or puting a higher timer.

The ugly :

Need more maps and separate queue per game type (Conquest/Death match/Stronghold), at least 3.

Hero channel are not fun. Warrior rampage and GG.

Need to punish more for deaths. You can just defend, die spawn right away. A cumulative death timer should do the trick, or add a debuff of some kind when you die multiple times.

Need to replace death guards / repair doors. Supply running is kinda bad because good defender can kill NPC really quick when the offensive team puts a lot of pressure.

Oh and I almost forgot… TEMPLATES. Yes, yes, yes, this is still a high priority for me and that should be shiped with HoT. We need it even more if we end up having the 3 game type in the same map pool for unranked.

I’ve tried to keep it concise, don’t know if I did well! Overall, the map needs some tweek to be satisfying playing it, but good job on it Anet!

Last year when Guild Wars 2 celebrated their first Birthday we were given awesome info pics that should us how many people bought the game, which class was played the most, which race was played the most, which mob killed the players the most etc..

where is year 2’s info pic? I truly want to see the comparison and how well or bad you guys did? If this has been release can someone link it?

There is a major world boss every hour. No matter what hour they chose you would still be here complaining.

They push out the updates as soon as they are available. There is no downtime and the fixes benefit people immediately.

I facepalmed so hard that my hand came out at the back of my head.

This is Roleplaying 1-o-1. You dont know someone’s name unless they are introduced by some source.

It should have gone like this:
“This is the work of Mordremoth!”
“Mordre-what now?”
“M-Mordremoth… The next Elder Dragon?”
“How do you know this?”
“Its..its the vines. They are all over the place.”
“Yeeeesss, i can see that. How do you know it is the direct work of an Elder Dragon?”
“’Cuz Mordremoth is the planty Elder Dragon! It makes perfect sense, dont you see? We heard him roar!”
“We heard something. Thats it. We agreed that it is possibly an Elder Dragon. Now you come and tell me that its name is Mordremoth, and its the ‘planty’ dragon, obviously. I ask again, how do you know this?”
“Well… i… you see, in the Infinity Coil, there was this green area…”
“Uh-huh… go on.”
“Oh! And Subject Alpha was like ‘Roar-roar’ and ‘Blergh-Grawlllrrr’, but he was also like ‘Subject Alpha deals 1560 damage to you by using Teeth of Mordremoth!’ And it was all … earthy… and planty.”
“…Hon, did you hit on the blood whisky again?”

Thank you guys for changing the top-tier reward to be at 1200 rescues instead of 1500. 1500 just really seemed impossible given the overflow situation and time limit, and I was honestly about to give up on the event because rewards below that top-tier are so lackluster.

1200 is still hard (highest I’ve ever been is 1,000, and we really pulled that out last minute), but it feels attainable.

So again, thank you for responding to feedback!

Good change.

1200 is a reasonable number, difficult but not impossible for a bunch of random people thrown together in a random overflow.

TTS I think isn’t a good comparison, as you’re able to have a greater number of people with more coordination, even if we have to manipulate the game mechanics in order to actually get our people into the same overflow. Despite the name Guild Wars, it seems the developers don’t want guilds to be able to work together under the current overflow mechanics.

In the world of Tyria, the death of your main-character loss is mourned/celebrated. A true hero has gone on to the mists. We shall miss him/her.

What will be written on your gravestone in the Shaemoor Graveyard (or wherever your race puts memorials up to heroes)?

gw2lfg.com

You can try and find some cross-server dungeon goers here. Might be your best bet. Also try LFG/LFM in lions arch.

NixZero.7540:

orbs and undermanned buffs should be switched around, they are just some kind of sadistic pun (the side already winning get a power bonus, the side that cant win because is undermanned get a gold/xp bonus to what they cant win).
this said, why the side with less and worse players should win?

I agree with this, that actually makes more sense than a lot of the drivel I’ve seen on here. Orbs would be a nice little perk for those winning, while the outmanned buff would actually be useful for once lol. It could scale with the amount you are outnumbered and such, and you could get like 10% MF per orb or something like that.

orbs and undermanned buffs should be switched around, they are just some kind of sadistic pun (the side already winning get a power bonus, the side that cant win because is undermanned get a gold/xp bonus to what they cant win).
this said, why the side with less and worse players should win?

I was tempted to transfer to a “better” server like many others had. But then I remember their multi-hour queues to get into WvWvW and just smile.

Besides. It is much more fun playing as the under dog and being deep in enemy territory as a solo thief or as a hit and run group instead of a mindless zerg.