A Guide to Group Finder

[Saxxonknight]

Damion, I hope you read this.

 

If you recall the GDC from 2004-2006 I was one of the volunteer staff. Also a veteran of several MMOs, including Shadowbane, over which I used to give you some good natured heckling at the various forums we discussed improving MMO design. One of the comments you made to me in a casual discussion in '06 was I was "swinging for the fence" in the goals of my design theories or proposals.

 

Yes that was and is true. I work in an industry that makes actually moving into the game industry not practical from a financial standpoint to start over carreer wise at my age. But, over 25 years I've studied games, bought hundreds and thought of ways of improving them. I'd like to offer some thoughts for your serious consideration.

 

#1 - Part of the problem with SWTOR (and Rift as well as other psot WOW MMOs) is that they do not have player made UI mods. Now there are some that think its anathema, but follow the thought for why this was a decisive factor in WOWs success and influenced heavily the relative stagnant performance of others.

 

I'll focus this on the role of healers in groups and raids, bearing in mind you have a population of players of varying skill. Ergonomically, playing a healer on the base UI in ANY of these MMOs including WOW is a target switching nightmare. The sheer amount of actions performed above that of a tank or DPS in any fo those games is staggering. In just a 60 second fight vs one boss, you will have switched targets between 20-60 times depending on AEs, GCD lenght, cast times etc. Everyone else won't have had to switch targets at all. Your amount of actions is doubled.

 

For an elite player, so what, they see it as a challenge. However, for casuals its more of a hassle, especially long term. In your game, you need 25% played healers with a group size of 4, this design choice was a vulnerability but its too late to change without major upheaval of other systems. In order to achieve this, you need to make healing practical and enjoyable for a lot of average players. Otherwise, you have a situation where non-healer que times will be obscenely long and further reduce client retention.

 

Why WOW has been so successful in part - because player made mods like Healbot, Grid and others made group and raid healing at a competent level practical, ergonomic and doable for the vast majority of players. WOW only required 20% played healers to function smoothly on a statistical basis for group formation. (their hobble was tanks, and they frankly made a mistake not enabling certain DPS classes (Enh Shaman, Demo Lock, Beast Hunter) to tank 5 mans thus taking the long que group and enabling some of them to bolster ranks of the instant que tanks).

 

The main component of these player healing mods is a click cast group or raid interface. This takes the act of target switching and combines it with the act of casting, thus putting the ergonomic demand on the player on par with a tank or DPS class. It also allows the healer to keep the main target (boss) while casting heals on group mates. In reality what the mods did was make mouse clicks the equivalent of hotkeys. By using keyboard modifiers (SHIFT, ALT, CTRL) you expand the potential actions usable on a 3 button mouse to 12 - easily enough to bind any applicable actions usable on a friendly avatar in this game.

 

By doing so you will have moved the game more toward the forefront of game UI science at this stage. What I mean is, when you release a game in 2012, if you can't move the UI elements and resize them the game is instantly less than jsut about anything else currently out as competition which will have done this within their first few months. Not doing so for a new release is taking a gamble on lost clients (any deficiency, coupled with others, results in a % of loss). This is one factor of game ergonomics that players are less likely to overlook these days. I know personally I find the base locked UI totally gimped with the bars spread across the edges of the screen, making the elements movable was a good move, but its a stolen horse/unlocked barn story - negative opinions were already formed.

 

This factor is really a #1A. As for other player mod UI elements, look at WOW over the years and all the mods they incorporated into their UI. As a game designer you should take that as a clue those features should be standard. I regard games launching without these things (movable UI, click cast) as sub-par because its been obvious for 6-8 years now that its a success bolstering pillar of WOW and in this case a feature that can be copied and implemented in your own way without being "me too." Note that the chat menus in WOW where you can click a name and perform actions with that character was mainstreamed by WOW as well. I thought of that way back when I was playing Shadowbane et al - WOW was the first I've seen that did it - but everyone else since has done it as well. Take heed with the above in this regard.

 

#2 - On your LFG interface, I certainly understand the interserver lockout. I saw the attitudes in WOW with cross server grouping. However, to effectively leverage your populace to keep the game experience vibrant, you really have to mine this resource. So it behoove syou to come up with and implement systems to stamp out the negative aspects of the cross server grouping. Because sooner or later you will have to do it, and frankly sooner you do it and keep the player pool large the better your long term retention is.

 

Just gotta get rid of the greif. This becomes complex because you have to create a system that will tabulate and analyze a wide range of player behaviors to come up with rewards or penalties, and ratings of players. Incoming mass quote here, the player Aethyrprime is right on the money here so a big second to his post:

 

On the vote kick, please require filling in a reason "Why?" before enabling the vote button:

 

Person was disruptive to the team

[ ] Voted Need Constantly (This is the persons fault)

[ ] Was Rude (This is the persons fault)

[ ] Caused team to wipe more than once (This is the persons fault)

--------------------------------------------

Other Reasons

[ ] Too Low of Level (Bug not the persons fault)

[ ] Lacking Sufficient Gear (Can't be helped, some will and some will not carry a team mate)

 

 

In this way people who truly are a disruption to the community can be put on a block list custom for grouping.. The "Other Reasons" do not count against the persons account rather notify the developers that their base group level filtering needs to be tweaked more.

 

Then on the Find group option, the team leader (or solo players) filters would determine which repeat offenders won't show up.

 

[X] No Ninjas

[X] No Jerks

[X] No Jokers

---------------------------------------

[ ] No Lowbies

[ ] No Gimps

---------------------------------------

[X] Legacy Mode Enabled (Overrides legacy settings and puts it to visible as title by default for the duration of this team. This is intended to deter people from using alts with legacy hidden to grief others)

 

 

By putting this on both sides you can minimize your chaos. People who get too many vote kicks for being rude simply will not be teamed with people that have a short patience for childishness. And of course. Any one who gets vote kicked for ninja'ing other class loot can also be prevented from joining most regular players games.

 

Also, Those who don't mind carrying a lowbie or a gimp can allow them in their team, while those who specifically intend to kick that person just seconds after they join anyway can block them from joining by enabling a stricter rule set for match making of player level/gear level averages.

 

If you do not make such a filter system understand the "chaos" will be great as the group finder tries to mash people together for large group ranged content. Lowbies will get kicked, gimps will get kicked and that will leave them with a bad experience since they really did nothing wrong. While people who repeatedly ninja other class loot will still have a new group to prey on coming up real soon. Same with people who get their kicks from upsetting others they will also know that one kick just means another group incoming.. Sure keeping it limited to server does sort of help these a little.. but believe me when I tell you not nearly as much as it should.

 

Now, one thing needs to be added to this for cross server and perhaps even same server is voluntary evaluation. IE

[] Would group with this player again

 

This would essentially provide an ignore list for the grouping so that a known douche doesn't get with the same people who never want to see them again. That is certainly a situation you want to avoid for your clients.

 

Things your system should track:

- times player leaves before completion. Chronic Quitters aren't desirable.

- relative completion times for FPs. This could be a long term measure for competence rating.

- Damage/Healing output (sustained/spiked). The players will figure this out eventually anyway with parsers, but you want to make sure you've got a group that will be capable of completion or you've setup a bad experience for those involved.

 

There are probably a number of other benchmarks you can come up with. The thing is, you want to have this laid out, fleshed out and even ready to deploy when you need it, or even before. Because if you guys start working on it only when you percieve the need, it will be too late, you'lll suffer X% client loss again before plugging the leak in customer satisfaction.

 

 

Anyway man, I hope the position is treating you well. I actually did apply with Bioware for this project back then in '06. However, as you noted, its about impossible to break in at upper levels without paying yoru dues. Mine were already paid in industry and I couldn't take an entry position without taking a wrecking ball to my life. I may see you again at the conference in the next couple years as the more pressing matters of the last 6 years settle out. I hope you have success in turning this franchise around, you guys have a lot of good stuff in here, but the unfinished corners had some bad holes with respect to keeping players involved.

 

-Saxxon

 

(its 1am, you all can read thru spelling and grammar mistakes I trust).

Edited June 14, 2012 by Saxxonknight
Add the sp/grammar caveat

[CJAShadow]

One of the reasons I enjoy healing is because of this challenge, as I am apparently an elite player.

 

Please do not include mods that eliminate the need for target heal selection and the human decision making element of healing. Healing characters have a much easier time in PVP when dps because of how the Ops party health bars work, DPS must pay attention to more of what is on the sceen to succeed in correct target selection. Tanks have an even harder job with this decision making for protection/dps/taunt demands on them.

 

I believe that keybinding eliminates the need for UI mods because of the limitation on skill usage based on cooldowns. Everyone that plays this game has learned how to type at a reasonable level (I am sorry older players, but being able to type just a little is very important as it is just muscle memory). Anyone with a disability or something stopping them from this can use outside hardware such as a mouse with multiple keys and "other devices".

 

The LFG tool selection for "would you group with this person again?" is a really good suggestion.

[Irilith]

I think you underestimate the importance of a proper reward-system. Please take a closer look at how Blizzard is doing it and reconsider your choice of once-a-day rewards.

[Devlonir]

I think you underestimate the importance of a proper reward-system. Please take a closer look at how Blizzard is doing it and reconsider your choice of once-a-day rewards.

 

WoW went with their system because they knew their heroic modes did not give any real reward anymore to the people they wanted to grind them. The phallacy of this earlier decision in Wrath was shown at the start of Cata, where people geared 90% through random dungeon rewards and barely through the actual loot drops.

 

This game still has a huge group of people who want to group but never got a chance and have the best single player gear but barely any group/harmode flashpoint level gear. For now, the act of gearing through the hardmode FP's themselves will be enough incentive. Eventually, a WoW like system might be needed, but for now this implementation looks good enough.

Also.. the game still has daily and weekly FP quests as well. WoW dumped those when they implemented their reward driven Dungeon Finder.

 

Good point you make, but a little too soon to ask for incentives for every run besides the loot drops and 'commendations' you get for most boss kills as well as the daily/weekly quests that are in game.

Edited June 14, 2012 by Devlonir

[RangerRobEU]

Implement cross-server and they'll lose a lot of players too. For every player who wants a faster queue, there is another player who wants a more enjoyable experience when that queue finally pops. Cross-server fuels the farming habits of potty mouthed children and takes the enjoyment out of actually doing the content.

 

I personally hope BW stick by their guns on this because with the server transfers there will be absolutely no need whatsoever for cross-server LFG queues - if you are really that worried about your queue times then roll a tank or healer because it'll help queue times much more than cross-server would anyway.

Edited June 14, 2012 by Parali
bad quote

[Saxx]

Having tested out the group finder so far i'm quite pleased with how it works. There are ways to make it even better, but the current format is quite acceptable.

 

The only part I disagree with is the cross server queing that causes a decline in overall group performance. This is a bit of a misconception that will vary from play session to play session. Played a certain game about war....with craft in it that introduced me to server queing. Truthfully it is always a mixed bag even if you get players from your same server. Only thing that may act as a deterrent from them acting like jerks is the repercussions on the player and as a result they feel that they are "strong armed/bullied" into behaving.

 

Hopefully we will see a cross server quing eventually down the road. I for one would love to see an advanced cross server community group tool itself allowing people to open a window to chat with friends they make on other servers. Have no doubt this kind of thing will take an immense amount of time and effort, but breaking down the barriers of servers within an MMO only enhances the overall experience players have.

[Inzuher]

Perhaps, but the nature of Group Finder queues is that the limitations are usually defined by a shortage of certain roles (healers and tanks, specifically) than by a shortage of potential party members.

 

This might be true, and makes me wounder if you'll be following up on the group finder by adding Dual Specs?

[Moscardo]

Question...

As not Flashpoints and/or Operations are of the same difficulty, or require the same combinations of classes

(for instance Directive 7 is MUCH easier with ranged DPS and not melee ones)...

Will be any way to change/choose, to avoid going to face Ironfist of Esseles with fresh 50... basically ANY way to control the randomess of the LFG tool?

(Please note that I'm asking from the fresh 50 point of view. I HATE when my inexperience or lack of appropriate gear creates difficulty for my team, even if random).

[Rehme]

Im just more distracted by the picture of the rancor that looks like its taking a dump. I dont know. is it just me? lol

[Xiahoufig]

Wouldn't it be better if players were able to access it earlier than level 10 so that they could use it on their origin planets?

 

It's a good point but the heroics on the starter planets (if you can't find anyone to do them for whatever reason) become soloable once you get your companion. (Especially for bounty hunters getting Mako)

[Rehme]

Im just more distracted by the picture of the rancor that looks like its taking a dump. I dont know. is it just me? lol

[Elfindreams]

One of the reasons I enjoy healing is because of this challenge, as I am apparently an elite player.

 

Please do not include mods that eliminate the need for target heal selection and the human decision making element of healing. Healing characters have a much easier time in PVP when dps because of how the Ops party health bars work, DPS must pay attention to more of what is on the sceen to succeed in correct target selection. Tanks have an even harder job with this decision making for protection/dps/taunt demands on them.

 

I believe that keybinding eliminates the need for UI mods because of the limitation on skill usage based on cooldowns. Everyone that plays this game has learned how to type at a reasonable level (I am sorry older players, but being able to type just a little is very important as it is just muscle memory). Anyone with a disability or something stopping them from this can use outside hardware such as a mouse with multiple keys and "other devices".

 

The LFG tool selection for "would you group with this person again?" is a really good suggestion.

 

I would suggest a compromise. There are legitimate needs for custom or variant UI and there are ways to do it that don't create an unfair advantage for those that don't have it (i.e. bots that auto heal are an unfair advantage as then you really aren't playing the game you are just an AI that happens to have a human that can move you around). For example some people need to use on screen keyboards and pointers due to disabilities or health reasons and a way could be come up with to create an addon that could streamline their gameplay such that it is still human interaction which causes the effect, just the selection process is more streamlined for their use case.

 

So basically I am suggesting allow community add-ons but make them undergo a review process (which would in all likelyhood require a small fee) or set strict guidelines on what they can and can't do and start banning if people use ones they shouldn't.

[Elfindreams]

-= wall of text snip =-

 

Now, one thing needs to be added to this for cross server and perhaps even same server is voluntary evaluation. IE

[] Would group with this player again

 

This would essentially provide an ignore list for the grouping so that a known douche doesn't get with the same people who never want to see them again. That is certainly a situation you want to avoid for your clients.

 

-= more wall of text snip =-

 

Out of curiosity, why implement something like that when the existing /ignore functionality already provides that?

[thesadplanet]

The only thing I would change about the group finder would be to add Hard Mode EV and KP to be selectable with the group finder. If Story Mode Denova is going to be selectable, then I see no reason as to why HM EV and KP aren't since they are roughly the same in terms of difficulty and the loot is from the same gear tier.

[Inzuher]

Damion, I hope you read this.

 

If you recall the GDC from 2004-2006 I was one of the volunteer staff. Also a veteran of several MMOs, including Shadowbane, over which I used to give you some good natured heckling at the various forums we discussed improving MMO design. One of the comments you made to me in a casual discussion in '06 was I was "swinging for the fence" in the goals of my design theories or proposals.

 

Yes that was and is true. I work in an industry that makes actually moving into the game industry not practical from a financial standpoint to start over carreer wise at my age. But, over 25 years I've studied games, bought hundreds and thought of ways of improving them. I'd like to offer some thoughts for your serious consideration.

 

#1 - Part of the problem with SWTOR (and Rift as well as other psot WOW MMOs) is that they do not have player made UI mods. Now there are some that think its anathema, but follow the thought for why this was a decisive factor in WOWs success and influenced heavily the relative stagnant performance of others.

 

I'll focus this on the role of healers in groups and raids, bearing in mind you have a population of players of varying skill. Ergonomically, playing a healer on the base UI in ANY of these MMOs including WOW is a target switching nightmare. The sheer amount of actions performed above that of a tank or DPS in any fo those games is staggering. In just a 60 second fight vs one boss, you will have switched targets between 20-60 times depending on AEs, GCD lenght, cast times etc. Everyone else won't have had to switch targets at all. Your amount of actions is doubled.

 

For an elite player, so what, they see it as a challenge. However, for casuals its more of a hassle, especially long term. In your game, you need 25% played healers with a group size of 4, this design choice was a vulnerability but its too late to change without major upheaval of other systems. In order to achieve this, you need to make healing practical and enjoyable for a lot of average players. Otherwise, you have a situation where non-healer que times will be obscenely long and further reduce client retention.

 

Why WOW has been so successful in part - because player made mods like Healbot, Grid and others made group and raid healing at a competent level practical, ergonomic and doable for the vast majority of players. WOW only required 20% played healers to function smoothly on a statistical basis for group formation. (their hobble was tanks, and they frankly made a mistake not enabling certain DPS classes (Enh Shaman, Demo Lock, Beast Hunter) to tank 5 mans thus taking the long que group and enabling some of them to bolster ranks of the instant que tanks).

 

The main component of these player healing mods is a click cast group or raid interface. This takes the act of target switching and combines it with the act of casting, thus putting the ergonomic demand on the player on par with a tank or DPS class. It also allows the healer to keep the main target (boss) while casting heals on group mates. In reality what the mods did was make mouse clicks the equivalent of hotkeys. By using keyboard modifiers (SHIFT, ALT, CTRL) you expand the potential actions usable on a 3 button mouse to 12 - easily enough to bind any applicable actions usable on a friendly avatar in this game.

 

By doing so you will have moved the game more toward the forefront of game UI science at this stage. What I mean is, when you release a game in 2012, if you can't move the UI elements and resize them the game is instantly less than jsut about anything else currently out as competition which will have done this within their first few months. Not doing so for a new release is taking a gamble on lost clients (any deficiency, coupled with others, results in a % of loss). This is one factor of game ergonomics that players are less likely to overlook these days. I know personally I find the base locked UI totally gimped with the bars spread across the edges of the screen, making the elements movable was a good move, but its a stolen horse/unlocked barn story - negative opinions were already formed.

 

This factor is really a #1A. As for other player mod UI elements, look at WOW over the years and all the mods they incorporated into their UI. As a game designer you should take that as a clue those features should be standard. I regard games launching without these things (movable UI, click cast) as sub-par because its been obvious for 6-8 years now that its a success bolstering pillar of WOW and in this case a feature that can be copied and implemented in your own way without being "me too." Note that the chat menus in WOW where you can click a name and perform actions with that character was mainstreamed by WOW as well. I thought of that way back when I was playing Shadowbane et al - WOW was the first I've seen that did it - but everyone else since has done it as well. Take heed with the above in this regard.

 

#2 - On your LFG interface, I certainly understand the interserver lockout. I saw the attitudes in WOW with cross server grouping. However, to effectively leverage your populace to keep the game experience vibrant, you really have to mine this resource. So it behoove syou to come up with and implement systems to stamp out the negative aspects of the cross server grouping. Because sooner or later you will have to do it, and frankly sooner you do it and keep the player pool large the better your long term retention is.

 

Just gotta get rid of the greif. This becomes complex because you have to create a system that will tabulate and analyze a wide range of player behaviors to come up with rewards or penalties, and ratings of players. Incoming mass quote here, the player Aethyrprime is right on the money here so a big second to his post:

 

 

 

Now, one thing needs to be added to this for cross server and perhaps even same server is voluntary evaluation. IE

[] Would group with this player again

 

This would essentially provide an ignore list for the grouping so that a known douche doesn't get with the same people who never want to see them again. That is certainly a situation you want to avoid for your clients.

 

Things your system should track:

- times player leaves before completion. Chronic Quitters aren't desirable.

- relative completion times for FPs. This could be a long term measure for competence rating.

- Damage/Healing output (sustained/spiked). The players will figure this out eventually anyway with parsers, but you want to make sure you've got a group that will be capable of completion or you've setup a bad experience for those involved.

 

There are probably a number of other benchmarks you can come up with. The thing is, you want to have this laid out, fleshed out and even ready to deploy when you need it, or even before. Because if you guys start working on it only when you percieve the need, it will be too late, you'lll suffer X% client loss again before plugging the leak in customer satisfaction.

 

 

Anyway man, I hope the position is treating you well. I actually did apply with Bioware for this project back then in '06. However, as you noted, its about impossible to break in at upper levels without paying yoru dues. Mine were already paid in industry and I couldn't take an entry position without taking a wrecking ball to my life. I may see you again at the conference in the next couple years as the more pressing matters of the last 6 years settle out. I hope you have success in turning this franchise around, you guys have a lot of good stuff in here, but the unfinished corners had some bad holes with respect to keeping players involved.

 

-Saxxon

 

(its 1am, you all can read thru spelling and grammar mistakes I trust).

 

This guy has some valid points

[Elfindreams]

The only thing I would change about the group finder would be to add Hard Mode EV and KP to be selectable with the group finder. If Story Mode Denova is going to be selectable, then I see no reason as to why HM EV and KP aren't since they are roughly the same in terms of difficulty and the loot is from the same gear tier.

 

I believe they are, I am at work so can't log into the PTS to be sure, but I am 75% sure they are there.

[kaydo_]

Very excited this is coming soon. Thanks BioWare! Really enjoyed the play through to 50 then was a little disappointed about having to camp out on Fleet twiddling my thumbs while waiting to find groups. Now I can do my dailies, search for datacrons and check out areas I just leveled past without worrying about missing something.

 

Anyway, this is huge for the game and will really boost the level of enjoyment for all players I think.

[tronot]

I don't understand why you can queue for story mode Denova, but not hard mode EV or KP. Story Denova is much harder than those 2.

[ZenMorph]

Out of curiosity, why implement something like that when the existing /ignore functionality already provides that?

 

The important question, in my mind: will /ignore prevent you from being grouped (via LFG tool) with players on your poo-doo list?

[Tmanarl]

Im just more distracted by the picture of the rancor that looks like its taking a dump. I dont know. is it just me? lol

 

Win. /5char

[NoFishing]

Im just more distracted by the picture of the rancor that looks like its taking a dump. I dont know. is it just me? lol

 

Oh god, now I cannot unsee that! Thanks. :-/

[NoFishing]

The only thing I would change about the group finder would be to add Hard Mode EV and KP to be selectable with the group finder. If Story Mode Denova is going to be selectable, then I see no reason as to why HM EV and KP aren't since they are roughly the same in terms of difficulty and the loot is from the same gear tier.

 

According to the patch notes, story mode Denova has been removed from the options due to the difficulty.

[NoFishing]

The important question, in my mind: will /ignore prevent you from being grouped (via LFG tool) with players on your poo-doo list?

 

From the article:

 

The Group Finder will never group you with someone you’ve placed on your Ignore list

[wyrmac]

I am very disappointed to see that BW, and at least a very vocal part of the Forum community, are against cross server LFG, frankly the reasons given are very weak, if someone is acting in an inappropriate manner you can simply vote to kick them and put them on ignore, If the tool is cross server they will likely be replaced instantly and you will never be grouped with them again.

The same people who everyone seems to be afraid will lessen their LFG experience are players on someones local server, and they are dealt with in the manner I have mentioned above. The main reason for the existence of a LFG tool is because players are having trouble forming groups, this is most likely because of low populations, but also because of the fact that The Community that exists on a server, and the preservation of which seems to be the battle cry of everyone against cross server Ques, already has people that they prefer to group with, keeping it same server only means that those people will all Que together to get an instant teleport to the flashpoint, and receive extra rewards ( which they can still choose do do if the tool is cross server).

Even with the transfers being done now, there will still be times when it will not be feasible for someone to get a LFG to pop, Just because someone plays on a East coast server for example, dose not mean that their scheduled is 9-5 EST, as a subscribing player you should be able to play any aspect of this game that you chose, at any time that you chose, if it is 4 am your local server time, and you want to run a flashpoint, you will have a problem if the LFG tool stays same server, but if it go's cross server, it will draw on every player in the game, and you will be able to play the aspects of the game that you enjoy most at any time that is convenient for you.

This post should also be considered to apply to PvP Ques.

Edited June 14, 2012 by wyrmac

[Elfindreams]

The important question, in my mind: will /ignore prevent you from being grouped (via LFG tool) with players on your poo-doo list?

 

They answer that right in the article this is the discussion for. Yes.