6v6+ Game mode variety

General Discussion
12/16/2016 01:22 PMPosted by NooneSpecial

I'm sure you guys are always experimenting, but I'm always more curious about the ideas that don't work than the ideas that do. Not so much trying to solve them, but more the complications and issues you had with the idea that lead to it not being viable and the 'nail in the coffin' so to speak.

Agreed. Having some tranparwncy about why things are rejected from competitive consideration would be good to know.

Also, why not unlock all these options in customs? Let the community try out all these different modes and see first hand why it might not work, or at least come up with their own way of playing. Maybe there would be some community created game modes that would end up being viable.
12/16/2016 11:37 AMPosted by Apple
Question, but what's the point of not releasing content that's developed to some extent, rather than releasing them at least as options for Custom games? That way so long as you can pull people together you can play such a mode, at least vs AI if you can't find that many people.


This is a really good question. I should have been more clear. When we are prototyping and experimenting, we often use very rogue development tactics to create content quickly -- just to see if it is fun or not. So for example, when we prototype a new hero, we just use an existing hero model and animations to first see if the "rough" version of the hero is going in the right direction. For Ana, we used Widowmaker internally. For Sombra, we used Symmetra's model and animations. For their abilities, we don't even have proper visual effects. So often we'll prototype a hero and when he or she is doing their ultimate ability, we just print text over their head that says "ULT".

It looks really terrible and it takes a lot of imagination and practice in seeing that sort of thing to get over it and assess the core of fun without all of the "dressing" that makes the game not only look good, but play well. As developers, we are very used to rough edges. When we play our maps internally, they are just gray mesh with very basic lighting. When we test new features, they often lack user interface and require us to use console commands to execute. Really, during this type of testing or prototyping, the game is in no shape to be shown or played.

Probably the best example was the first playtest of Overwatch. The entire level was gray and blocky and the only hero was tracer. We had no visual effects or attachment points on her guns so she just shot laser beams out of her eyes...

On top of that, new stuff is inherently buggy and non-optimized. Often times, the way we prototype stuff, it would not even run properly in a live environment, let alone be optimized for 3 platforms (PC, Xbox and PS4).

I have a lot of videos of these types of early testing. Maybe at some point we'll find a way to share some of that with you guys. I think if you saw how truly rough our early tests are, you would have a better understanding of why they never make it into the public.
As an old fart TF2 player, I yearn for the days of 12v12 and 16v16, though OW has grown to become more of its own thing as time goes on. I think they could make it work, but I shudder to think how much balancing that would take, just for a fringe play mode.

What I loved from TF2 that I think could more easily work in OW are payload races (two separate payloads on two symmetrical tracks going to the same point,) Tug of War (one payload pushed by both teams with opposite-facing goals for each,) and 3-point KotH (capturing the center point unlocks a point closer to the enemy spawn, and capturing THAT point wins the game). Linear 5-point capture maps are super fun, but they can go on for a very long time, and that might not be in keeping with the quick turnaround that other game modes seem to stress.

If, god forbid, you guys ever actually want to dabble in 12v12 or more, I'd love to see more open maps with multiple consecutive goals. Consider looking at maps like TF2's Gravel Pit, where attackers need to capture 3 points, but the first two can be captured in either order before unlocking the third.

What I like about OW is that it has a serious competitive side that TF2 never had. To my mind, that should always be the primary focus. But with that said, OW lacks the options for chaotic modes that are less about winning (some of the most popular TF2 community maps are almost impossible to win or lose) and more about having an environment without restrictive time limits to hone skills and experiment without a ton of criticism or blame for throwing a match. A more free-form killing grounds mode would be great for blowing off steam when I'm burnt-out on objective-based play and just want to practice my sniping against other humans for a few uninterrupted hours.

Woof, didn't mean to pile on a bunch of stuff like that, but whenever I see a topic that grabs the team's attention, I get excited.
12/16/2016 01:35 PMPosted by Jeff Kaplan
12/16/2016 11:37 AMPosted by Apple
Question, but what's the point of not releasing content that's developed to some extent, rather than releasing them at least as options for Custom games? That way so long as you can pull people together you can play such a mode, at least vs AI if you can't find that many people.


This is a really good question. I should have been more clear. When we are prototyping and experimenting, we often use very rogue development tactics to create content quickly -- just to see if it is fun or not. So for example, when we prototype a new hero, we just use an existing hero model and animations to first see if the "rough" version of the hero is going in the right direction. For Ana, we used Widowmaker internally. For Sombra, we used Symmetra's model and animations. For their abilities, we don't even have proper visual effects. So often we'll prototype a hero and when he or she is doing their ultimate ability, we just print text over their head that says "ULT".

It looks really terrible and it takes a lot of imagination and practice in seeing that sort of thing to get over it and assess the core of fun without all of the "dressing" that makes the game not only look good, but play well. As developers, we are very used to rough edges. When we play our maps internally, they are just gray mesh with very basic lighting. When we test new features, they often lack user interface and require us to use console commands to execute. Really, during this type of testing or prototyping, the game is in no shape to be shown or played.

Probably the best example was the first playtest of Overwatch. The entire level was gray and blocky and the only hero was tracer. We had no visual effects or attachment points on her guns so she just shot laser beams out of her eyes...

On top of that, new stuff is inherently buggy and non-optimized. Often times, the way we prototype stuff, it would not even run properly in a live environment, let alone be optimized for 3 platforms (PC, Xbox and PS4).

I have a lot of videos of these types of early testing. Maybe at some point we'll find a way to share some of that with you guys. I think if you saw how truly rough our early tests are, you would have a better understanding of why they never make it into the public.


You should post the earliest "footage" and just have it be pong with an arrow at the pellet and the label Tracer.
12/16/2016 01:35 PMPosted by Jeff Kaplan
12/16/2016 11:37 AMPosted by Apple
Question, but what's the point of not releasing content that's developed to some extent, rather than releasing them at least as options for Custom games? That way so long as you can pull people together you can play such a mode, at least vs AI if you can't find that many people.


This is a really good question. I should have been more clear. When we are prototyping and experimenting, we often use very rogue development tactics to create content quickly -- just to see if it is fun or not. So for example, when we prototype a new hero, we just use an existing hero model and animations to first see if the "rough" version of the hero is going in the right direction. For Ana, we used Widowmaker internally. For Sombra, we used Symmetra's model and animations. For their abilities, we don't even have proper visual effects. So often we'll prototype a hero and when he or she is doing their ultimate ability, we just print text over their head that says "ULT".

It looks really terrible and it takes a lot of imagination and practice in seeing that sort of thing to get over it and assess the core of fun without all of the "dressing" that makes the game not only look good, but play well. As developers, we are very used to rough edges. When we play our maps internally, they are just gray mesh with very basic lighting. When we test new features, they often lack user interface and require us to use console commands to execute. Really, during this type of testing or prototyping, the game is in no shape to be shown or played.

Probably the best example was the first playtest of Overwatch. The entire level was gray and blocky and the only hero was tracer. We had no visual effects or attachment points on her guns so she just shot laser beams out of her eyes...

On top of that, new stuff is inherently buggy and non-optimized. Often times, the way we prototype stuff, it would not even run properly in a live environment, let alone be optimized for 3 platforms (PC, Xbox and PS4).

I have a lot of videos of these types of early testing. Maybe at some point we'll find a way to share some of that with you guys. I think if you saw how truly rough our early tests are, you would have a better understanding of why they never make it into the public.

I would SO love a video showing off all the early testing stuff you guys do. Especially the early testing for heroes like Ana or Sombra. That would certainly be an interesting video!
12/16/2016 11:07 AMPosted by Jeff Kaplan
We're always revisiting ideas. We have some fun experiments in the works that are new 6v6 game modes. Some of them will probably never see the light of day. But I think some will make their way to the Arcade.


Is there a reason you haven't tried the suggested "omnic flag" regarding CTF? I know you said normal CTF doesn't work well but when you make it so the "flag" is an omnic that people climb into/become so that it negates any balance issues that certain heroes might bring if they were to carry a regular flag(ex Tracer/Genji/etc). If the omnic flags for both sides have the same stats/abilities/etc, it makes it balanced and it doesn't matter which hero people choose.
12/16/2016 01:35 PMPosted by Jeff Kaplan


This is a really good question. I should have been more clear. When we are prototyping and experimenting, we often use very rogue development tactics to create content quickly -- just to see if it is fun or not. So for example, when we prototype a new hero, we just use an existing hero model and animations to first see if the "rough" version of the hero is going in the right direction. For Ana, we used Widowmaker internally. For Sombra, we used Symmetra's model and animations. For their abilities, we don't even have proper visual effects. So often we'll prototype a hero and when he or she is doing their ultimate ability, we just print text over their head that says "ULT".

It looks really terrible and it takes a lot of imagination and practice in seeing that sort of thing to get over it and assess the core of fun without all of the "dressing" that makes the game not only look good, but play well. As developers, we are very used to rough edges. When we play our maps internally, they are just gray mesh with very basic lighting. When we test new features, they often lack user interface and require us to use console commands to execute. Really, during this type of testing or prototyping, the game is in no shape to be shown or played.

Probably the best example was the first playtest of Overwatch. The entire level was gray and blocky and the only hero was tracer. We had no visual effects or attachment points on her guns so she just shot laser beams out of her eyes...

On top of that, new stuff is inherently buggy and non-optimized. Often times, the way we prototype stuff, it would not even run properly in a live environment, let alone be optimized for 3 platforms (PC, Xbox and PS4).

I have a lot of videos of these types of early testing. Maybe at some point we'll find a way to share some of that with you guys. I think if you saw how truly rough our early tests are, you would have a better understanding of why they never make it into the public.


I hope you share some of those videos, Jeff! I love seeing the behind-the-scenes models and prototypes developers make, and their unique ideas to be able to test things out. It's especially good for people who are interested in being a game developer themselves!
I would think the only way a CTF match would work is if it was a 1 capture the flag mode when one is on attack and one Is on defense. That way the defending team could plan for that tracer or genji. Also they could then put a movement penalty on the flag carrier to focus on the team escorting that flag carrier back.
12/16/2016 01:59 PMPosted by CorgiCyborgi
12/16/2016 11:07 AMPosted by Jeff Kaplan
We're always revisiting ideas. We have some fun experiments in the works that are new 6v6 game modes. Some of them will probably never see the light of day. But I think some will make their way to the Arcade.


Is there a reason you haven't tried the suggested "omnic flag" regarding CTF? I know you said normal CTF doesn't work well but when you make it so the "flag" is an omnic that people climb into/become so that it negates any balance issues that certain heroes might bring if they were to carry a regular flag(ex Tracer/Genji/etc). If the omnic flags for both sides have the same stats/abilities/etc, it makes it balanced and it doesn't matter which hero people choose.


I've wanted this since launch.
Can we get a skin where Tracer shoots lasers out of her eyes
12/16/2016 01:35 PMPosted by Jeff Kaplan
she just shot laser beams out of her eyes...


New hero confirmed.
12/16/2016 01:35 PMPosted by Jeff Kaplan
12/16/2016 11:37 AMPosted by Apple
Question, but what's the point of not releasing content that's developed to some extent, rather than releasing them at least as options for Custom games? That way so long as you can pull people together you can play such a mode, at least vs AI if you can't find that many people.


This is a really good question. I should have been more clear. When we are prototyping and experimenting, we often use very rogue development tactics to create content quickly -- just to see if it is fun or not. So for example, when we prototype a new hero, we just use an existing hero model and animations to first see if the "rough" version of the hero is going in the right direction. For Ana, we used Widowmaker internally. For Sombra, we used Symmetra's model and animations. For their abilities, we don't even have proper visual effects. So often we'll prototype a hero and when he or she is doing their ultimate ability, we just print text over their head that says "ULT".

It looks really terrible and it takes a lot of imagination and practice in seeing that sort of thing to get over it and assess the core of fun without all of the "dressing" that makes the game not only look good, but play well. As developers, we are very used to rough edges. When we play our maps internally, they are just gray mesh with very basic lighting. When we test new features, they often lack user interface and require us to use console commands to execute. Really, during this type of testing or prototyping, the game is in no shape to be shown or played.

Probably the best example was the first playtest of Overwatch. The entire level was gray and blocky and the only hero was tracer. We had no visual effects or attachment points on her guns so she just shot laser beams out of her eyes...

On top of that, new stuff is inherently buggy and non-optimized. Often times, the way we prototype stuff, it would not even run properly in a live environment, let alone be optimized for 3 platforms (PC, Xbox and PS4).

I have a lot of videos of these types of early testing. Maybe at some point we'll find a way to share some of that with you guys. I think if you saw how truly rough our early tests are, you would have a better understanding of why they never make it into the public.


This sounds like something fit for a brawl. I think combining abilities or ultimates with different heroes could be a really fun brawl.

Granted it would look buggy and a bit of work may need to go into it...but having different characters so stuff they're not supposed to sounds fun.
12/16/2016 01:35 PMPosted by Jeff Kaplan
so she just shot laser beams out of her eyes...


And you feel justified depriving us of this!?
12/16/2016 11:08 AMPosted by MacFierce
Otherwise it would just be a rush to see which Tracer or Lucio could get to the flag first.


Why does everyone think those two would rule the CtF game?

Winston is where it's at. HP to tank something more than a sneeze, Bubble when you need a spare couple seconds for Leap that outpaces everyone else, and he counters the only character that *might* be able to keep up.
Plus an Ult that provides the ultimate "Yeah right, try again later losers!" escape.
[li][/li]
No... Tracer would be a trap Carrier. Monkey is where it's at!


What about Genji? Unparalleled vertical mobility and able to easily utilise flank routes nobody else can (with the possible exception of Pharah), very slippery and difficult to hit, and can easily escape with a swift strike if things get too crazy. While Winston has a lot of health, he's fairly slow and very easy to hit, and his only real mobility is a jump that has a several-second cooldown.

If CtF ever does get implemented, I imagine some abilities might get restricted or disabled while a player is carrying the "flag", kind of like how invisible Spies or Soldiers/Demomen with the rocket/sticky jumper can't carry up the intelligence in TF2.

12/16/2016 01:35 PMPosted by Jeff Kaplan
Probably the best example was the first playtest of Overwatch. The entire level was gray and blocky and the only hero was tracer. We had no visual effects or attachment points on her guns so she just shot laser beams out of her eyes...


I have to admit, having Tracer shoot lasers out of her eyes sounds like the most awesome thing ever.
Imagine a mode where you have a normal capture the point but also have a dva on either side with unlimited flight chasing a flying omnic for team power-ups inspired by the seeker in a game of quidditch.
I remember CTF on Team Fortress 2. Most matches ended up feeling like some of the biggest wastes of time in my entire life. They are terribly unfoccused and I hate them in Team Fortress 2, but Overwatch has a very different dynamic that could really work with this mode if done right. It would be better than King of the Hill, that's for sure.

What really bothered me about Jeff's post was the "..in Arcade" part. Does this mean we aren't going to see any new gamemods on Quickplay and Competitive ever? That's kind of sad.
12/16/2016 01:35 PMPosted by Jeff Kaplan
12/16/2016 11:37 AMPosted by Apple
Question, but what's the point of not releasing content that's developed to some extent, rather than releasing them at least as options for Custom games? That way so long as you can pull people together you can play such a mode, at least vs AI if you can't find that many people.


This is a really good question. I should have been more clear. When we are prototyping and experimenting, we often use very rogue development tactics to create content quickly -- just to see if it is fun or not. So for example, when we prototype a new hero, we just use an existing hero model and animations to first see if the "rough" version of the hero is going in the right direction. For Ana, we used Widowmaker internally. For Sombra, we used Symmetra's model and animations. For their abilities, we don't even have proper visual effects. So often we'll prototype a hero and when he or she is doing their ultimate ability, we just print text over their head that says "ULT".

It looks really terrible and it takes a lot of imagination and practice in seeing that sort of thing to get over it and assess the core of fun without all of the "dressing" that makes the game not only look good, but play well. As developers, we are very used to rough edges. When we play our maps internally, they are just gray mesh with very basic lighting. When we test new features, they often lack user interface and require us to use console commands to execute. Really, during this type of testing or prototyping, the game is in no shape to be shown or played.

Probably the best example was the first playtest of Overwatch. The entire level was gray and blocky and the only hero was tracer. We had no visual effects or attachment points on her guns so she just shot laser beams out of her eyes...

On top of that, new stuff is inherently buggy and non-optimized. Often times, the way we prototype stuff, it would not even run properly in a live environment, let alone be optimized for 3 platforms (PC, Xbox and PS4).

I have a lot of videos of these types of early testing. Maybe at some point we'll find a way to share some of that with you guys. I think if you saw how truly rough our early tests are, you would have a better understanding of why they never make it into the public.


Love your post :) I often try to explain this to people but I'm just a random and I was never as eloquent :)

Might I suggest grabbing some of the worse buggiest prototypes of concepts proving unplayable that you've already trashed (that doesn't also contain stuff you might still use, unless you already have) - I suggest this as I expect it to end up linked as the answer to many people's questions about "why can't we have..." or "the devs aren't doing the things I think they should be doing so they must be doing nothing" or whatever in the future :) (probably linked to from other non-blizz games forums too :P)
6v6 control point elimination would work really well in OW (Basically TF2's arena mode.)
12/16/2016 03:56 PMPosted by Ruseman
I remember CTF on Team Fortress 2. Most matches ended up feeling like some of the biggest wastes of time in my entire life. They are terribly unfoccused and I hate them in Team Fortress 2, but Overwatch has a very different dynamic that could really work with this mode if done right. It would be better than King of the Hill, that's for sure.

What really bothered me about Jeff's post was the "..in Arcade" part. Does this mean we aren't going to see any new gamemods on Quickplay and Competitive ever? That's kind of sad.

This might be a "have your cake and eat it too" sort of thing. I've no doubt they could balance and make CTF fun and less of an all-day gridlock affair like a full 2Fort server, but it breaks far enough away from the mechanics of the existing play modes that the pro circuit would likely reject it as a legitimate mode for comp.

Whether we like it or not, Blizzard has a vested interest in keeping OW viable as an esport, so if keeping the challenges fairly homogenous is what makes the pros happy, that's what they'll do. Having play modes that break away from the strategy of a 6-person cohesive unit all pushing for the same objective are, at this point, either going to be arcade modes or not implemented at all.

CTF definitely falls under that category, given that it would break the teams into subgroups of attackers and defenders, totally unlike any other Comp mode. I still want it, though. I'll take it any way I can.
I agree that the pro scene and esports aspects has a large say on what game modes there are. However, they already have a veto system in most tournaments, so why let that stop more options in competitive.

I think the thing that I hope for the most is that some of this stuff gets opened up in customs. Any game with a community that is involved in helping to figure out what works and what doesn't is far more healthy than one that shuts people off from the process. Transparency and collaboration in figuring out what works would be good for the game I think.

Join the Conversation

Return to Forum