Potentially Concerning Madden 26 News (SOFTDRINK Rumor)

[lambo3500]

Regarding #2. 2K Sports hasn't made a simulation football game in 21 years now. You think they could get that going in a year? No shot.

To be fair, I never said in a year. College football took 3+ years and that was with an existing engine and framework that they already had an existing football game in use for. But I still want to see 2k or another big publisher have access to the NFL license.

As I mentioned in the initial post, I said "ALSO", meaning EA can continue to use the license to make Madden, but now they would have competition, albeit maybe in 3-5 years. Sure not as awesome as having a title next year, but the first step is to get approval from the NFL and commitment from a studio. I just think we consumers get no benefit in having one publisher, especially one known to put mediocre annual releases for many of their games as our only NFL football option.

Even if someone (not necessarily you), are a huge EA or Madden fan, or you hate 2k, being a fan of Madden this will likely push EA to put more resources into making Madden better.

[TarHeelPhenom]

To be fair, I never said in a year. College football took 3+ years and that was with an existing engine and framework that they already had an existing football game in use for. But I still want to see 2k or another big publisher have access to the NFL license.

As I mentioned in the initial post, I said "ALSO", meaning EA can continue to use the license to make Madden, but now they would have competition, albeit maybe in 3-5 years. Sure not as awesome as having a title next year, but the first step is to get approval from the NFL and commitment from a studio. I just think we consumers get no benefit in having one publisher, especially one known to put mediocre annual releases for many of their games as our only NFL football option.

Even if someone (not necessarily you), are a huge EA or Madden fan, or you hate 2k, being a fan of Madden this will likely push EA to put more resources into making Madden better.

It would be great to have options in the form of other NFL games. But, the more I think about it, I really don’t know how many companies would jump at the chance to do in 2025 or beyond. Take NBA for example. 2k doesn’t own the exclusive license to the NBA, yet no other companies have jumped at the chance to make an NBA game. EA bowed out of making NBA games. There used to be countless options if MBA games back in the day. Same with MLB. I could be wrong but I don’t think San Diego Studios has the exclusive license to MLB? If they do then I stand corrected. The only other baseball game made with the MLB license in the last decade other than The Show was RBI Baseball and very little people played it.

So realistically, how many companies would jump at the chance to make an NFL game? Also, what has 2k been doing, or what have they shown that leads the NFL to believe that they want the license in the first place? They put out that mobile type game and it was horrible. Have they come out and said that if the license became available that they would make a new game? And yes, I’ve read the thread in the 2k board. I just don’t know what they’ve been up to.

Like I said, I would love to be able to play more than one NFL game. I didn’t buy Madden last year. But, I’ll believe it when I see it.

[lambo3500]

I do agree, it would be an immense battle, but I know 2k is well aware of people wanting another NFL simulation game, and 2k5 is still well received and still not duplicated despite Madden being the only NFL sim game for 20+ years. Furthermore, 2k is also aware that EA at one point was the dominant NBA game, and slowly but shortly 2k took over as the dominant basketball game regarding at least sales, but many will argue also in terms of quality. Lastly, lets be real, EA and 2k are greedy and love MTX, so no better way to bring additional MTX into their revenue streams by having a football game that they can later make a mypark and competitor for MUT.

[SmashMan]

So realistically, how many companies would jump at the chance to make an NFL game? Also, what has 2k been doing, or what have they shown that leads the NFL to believe that they want the license in the first place? They put out that mobile type game and it was horrible. Have they come out and said that if the license became available that they would make a new game? And yes, I’ve read the thread in the 2k board. I just don’t know what they’ve been up to.

Like I said, I would love to be able to play more than one NFL game. I didn’t buy Madden last year. But, I’ll believe it when I see it.

In general, yeah. The current sports games, flaws and all, are so well established that it'd be an uphill climb for anyone just starting out....but man you’ve gotta think 2K has been dreaming of ways to leverage the hype a new NFL2K would have to funnel people into another MyPlayer VC mode.

[jerwoods]

u don't want 2k making football games
1 all they care about is how much $$$ we can milk the person for
2 they funnel u toward my career my team on every screen ea does it too

[lambo3500]

I don't want 2k to be exclusive, but from a consumer perspective it makes sense to want both companies (EA & Take Two/2k) to compete for NFL/Football fans money. My understanding as well is that NBA 2k still has one of the best franchise modes of all sport games in addition to 2k Eras for the most part being pretty well received. Small example, but wear and tear is essentially injury management that was introduced years ago.

[Madden08PCgmr]

.. but man you’ve gotta think 2K has been dreaming of ways to leverage the hype a new NFL2K would have to funnel people into another MyPlayer VC mode.

I really would like to believe that.. but wouldn't they be more aggressive and vocal about getting back into the game?

[extremeskins04]

If player attributes didn't matter, then you wouldn't see top teams make the playoffs literally every year.

[adembroski]

and that playbooks still controlled sim stats.

Neither you nor Softdrink know what this means, so how's it a red flag?

Playbooks control sim stats the same way it controls played game stats. If the playbooks call for a run, the sim stats will generate a rush attempt. If a receiver runs an out, the sim will determine that the receiver based on a coordinate system that will determine what defensive back (again, based on the call produced by the playbook) will be in coverage. It's the same stat gen we've had since... christ, I don't even remember when SuperSim was first made.

I spent several hundred hours into tuning that damn thing in '12 and '13. You can load the whole thing up in Frosty and see all the same variables I used back then (there are a lot more now, since there are a lot more ratings and traits than back then, but it's the same fundamental system.)

What this actually means is that the playcalling logic put into the playbooks (what's simplified and surfaced to the user as Gameplan, with the situations and star ratings [which are weights. IE if you have two plays and you rate on 1 star and the other 2 stars, the second play will be called 66.7% of the time in that situation. Play weight / weight sum.]) What output that produces is determined by the sim engine logic. The playbook tells the sim engine where the players will be at a given point during the simulation of the play (actually, this is a poor way to put it. The playbook will say, for instance, that X is running an Out Route. The designer tuning the engine can tell the sim engine how to resolve an out pattern. Specifically, how far from the line of scrimmage and how far from the center of the field that receiver will be when he makes his cut, and when the ball arrives, given other variables such as time to throw). The playbooks do not actually have any input on what the play will produce. That has a lot more to do with player ratings.

So the only thing the sim stats base playbooks on is the play call. That's it.

[adembroski]

Let's be honest - regardless of how you feel about EA/Madden, competition is a GREAT thing. If you have two or more companies making an NFL Football game or anything for that matter, the final product will undoubtedly be better. When there is no competition, people get away with releasing a sub-par product. We have seen this many times, not just with Madden. Competition between companies meaning they push each other further to 'out-do' each other and actually push the boundaries. Without competition, the world would be a pretty miserable place. Exclusive game licenses should be banned across the board. Set a set price for the license, and whoever pays it can make an NFL licensed sim football game. I am sick of EA having exclusive licenses to sports and the product being nowhere near what it should be. The more competition, the more choice the better. It really is that simple.

I love it when people say, "Lets be honest," or "Let's face it," and follow up with a thing everyone more or less agrees on anyways.

First off, Madden wasn't that great when 2k was around. We tend to think of it as being great through our rose colored nostalgia glasses, but our beloved Franchise Mode as presented by Tony Bruno was a broken mishmash of unrelated systems trying desperately to produce a cohesive whole through special case code, hacked adjustments, and flat out faking it. You routinely had 5-10 2,000 yard rushers a season, we had to reverse engineer the entire playcalling system to get it to produce even close to realistic play calling (see: Madden Manifesto), real life young prospects would never, ever improve and only generated rookies would get any better. The Tony Bruno show would air the same Brandon Lloyd interview 5 times a season, rarely had a thing unique to say to the season at the time. The salary cap was (purposefully) tuned to never actual matter because not being able to sign every 90+ in the free agent pool "isn't fun."

Madden wasn't good back then. Competition did not improve it.

The problem isn't lack of competition, it's a philosophical focus on the wrong thing. There's an obsession in the industry generally with "Fun." Everything's gotta be "Fun."

It doesn't.

It has to be engaging. The pursuit of fun is why baseball didn't have a halfway decent game until the 2000s, and then when everyone realized that engagement meant leaving in the non-fun things like walks and focusing on the engaging things like the pitcher batter dual, we hit a baseball golden age. Want great games from the '04-'06 era? Look not to football, but to High Heat 2004, MVP Baseball 2005, and The Show 2006.

I helped on a number of fixes for that **** show from Madden 04-09 PC [PC stuck with the PS2 generation until 09], starting with the Manifesto, ending with Madden 09's Gommo's Madden Editor (I designed the drills used in the Training Camp add on.). It was a bitch to get that thing playable.

[Kevin McKoy]

Question for you, Adembroski, because you would know this: Regarding what they're doing with Coach DNA, the machine learning, and whatnot—I will say, what they've done with it overall is what I've been waiting for for more than a decade.
Is it true that something of that magnitude couldn't have been done on prior consoles? I'm almost certain it couldn't have been done on the PS1 or PS2.

[Cycloniac]

Neither you nor Softdrink know what this means, so how's it a red flag?

Playbooks control sim stats the same way it controls played game stats. If the playbooks call for a run, the sim stats will generate a rush attempt. If a receiver runs an out, the sim will determine that the receiver based on a coordinate system that will determine what defensive back (again, based on the call produced by the playbook) will be in coverage. It's the same stat gen we've had since... christ, I don't even remember when SuperSim was first made.

I spent several hundred hours into tuning that damn thing in '12 and '13. You can load the whole thing up in Frosty and see all the same variables I used back then (there are a lot more now, since there are a lot more ratings and traits than back then, but it's the same fundamental system.)

What this actually means is that the playcalling logic put into the playbooks (what's simplified and surfaced to the user as Gameplan, with the situations and star ratings [which are weights. IE if you have two plays and you rate on 1 star and the other 2 stars, the second play will be called 66.7% of the time in that situation. Play weight / weight sum.]) What output that produces is determined by the sim engine logic. The playbook tells the sim engine where the players will be at a given point during the simulation of the play (actually, this is a poor way to put it. The playbook will say, for instance, that X is running an Out Route. The designer tuning the engine can tell the sim engine how to resolve an out pattern. Specifically, how far from the line of scrimmage and how far from the center of the field that receiver will be when he makes his cut, and when the ball arrives, given other variables such as time to throw). The playbooks do not actually have any input on what the play will produce. That has a lot more to do with player ratings.

So the only thing the sim stats base playbooks on is the play call. That's it.

So, are you saying it's moreso the sim engine is broken/not working correctly than the playbooks affecting it that's the issue?

For example, I feel like ratings don't matter very much. It doesn't seem to matter which QB you have. They'll pass for 70+ percent completion rate whether they're Mahomes or Will Levis.

My main concern is ratings not mattering much on the sim side of things, so I'm open to how you approach it or if you have a solution for that conundrum. During games, I feel like ratings have more of a noticeable impact.

[adembroski]

So, are you saying it's moreso the sim engine is broken/not working correctly than the playbooks affecting it that's the issue?

For example, I feel like ratings don't matter very much. It doesn't seem to matter which QB you have. They'll pass for 70+ percent completion rate whether they're Mahomes or Will Levis.

My main concern is ratings not mattering much on the sim side of things, so I'm open to how you approach it or if you have a solution for that conundrum. During games, I feel like ratings have more of a noticeable impact.

They water down the ratings by using too many things mishmashed together.

As an example, the last time I popped the hood on Madden (this would have been Madden 24), there were something like 12 different factors determining whether a QB would scramble.

When you have too many attributes, traits, and abilities feeding into one decision, the resulting number tends toward the mean, so everyone ends up scrambling the same amount.

[adembroski]

Question for you, Adembroski, because you would know this: Regarding what they're doing with Coach DNA, the machine learning, and whatnot—I will say, what they've done with it overall is what I've been waiting for for more than a decade.
Is it true that something of that magnitude couldn't have been done on prior consoles? I'm almost certain it couldn't have been done on the PS1 or PS2.

I could go twenty different directions to answer this. Some make Madden look good, some make Madden look poor. I'll just go with Machine Learning was looked on with great skepticism in gaming until very recently. Whether or not it *could* have been done is irrelevant because nobody was going to until it got a lot better generally.

We all know intellectually that AI has been around since the '60s, but we also know LLMs couldn't tell strawberry had 3 Rs in it until sometime in the last year.

A few companies have played with AI in gaming before recently, and it always has wonky results. Oblivion is probably the best example. I think it has turned a corner though. I'm really excited for the AI era in gaming now (even though AI has pretty much hit the wall, it's plenty advanced enough to give a great deal of life to fictional worlds.)

I think in the coming years you're going to have legit staff meetings and franchise mode will go from spreadsheet manager to people manager and it's going to be amazing when they can figure that out.

[Kevin McKoy]

I could go twenty different directions to answer this. Some make Madden look good, some make Madden look poor. I'll just go with Machine Learning was looked on with great skepticism in gaming until very recently. Whether or not it *could* have been done is irrelevant because nobody was going to until it got a lot better generally.

We all know intellectually that AI has been around since the '60s, but we also know LLMs couldn't tell strawberry had 3 Rs in it until sometime in the last year.

A few companies have played with AI in gaming before recently, and it always has wonky results. Oblivion is probably the best example. I think it has turned a corner though. I'm really excited for the AI era in gaming now (even though AI has pretty much hit the wall, it's plenty advanced enough to give a great deal of life to fictional worlds.)

I think in the coming years you're going to have legit staff meetings and franchise mode will go from spreadsheet manager to people manager and it's going to be amazing when they can figure that out.

It's interesting that you say that, because when I spoke to ChatGPT the other day, it essentially said the same thing. It noted that, at best, with prior systems, you might have been able to script code that gave the appearance of machine learning—but that's about it.
It also said, to your point, that companies like Ubisoft only really began giving it serious attention around 2021. Here's the link for anyone who wants to read more:
https://news.ubisoft.com/en-us/artic...2011%2D2012%5D.

In other words—because I can’t speak for anyone else—the football game I’ve always wanted just wasn’t possible until now. I was fortunate enough to get into the beta, and for the first time, this is what I’ve been waiting for.

I’ve said for a while that the biggest issue with the game—on the field—has always been the AI. Whether it's quarterback decision-making or coach play-calling, that was the core problem. Sliders are just the icing on the cake in my view.

So my takeaway, to a degree—and you were in that building, not me—is that there were things the developers wanted to do that simply weren’t technologically possible. But they couldn’t exactly come out and say, “Hey guys, we know what you want. We want it too. But the consoles just aren’t there yet.”

No company is going to let developers publicly admit that. At least, I don’t think so. The closest I ever heard was Rex mentioning something about jersey tangibility and how it just wasn’t doable at the time. I believe it was in reference for pass interference, defensive holding, etc.

[adembroski]

It's interesting that you say that, because when I spoke to ChatGPT the other day, it essentially said the same thing. It noted that, at best, with prior systems, you might have been able to script code that gave the appearance of machine learning—but that's about it.
It also said, to your point, that companies like Ubisoft only really began giving it serious attention around 2021. Here's the link for anyone who wants to read more:
https://news.ubisoft.com/en-us/artic...2011%2D2012%5D.

In other words—because I can’t speak for anyone else—the football game I’ve always wanted just wasn’t possible until now. I was fortunate enough to get into the beta, and for the first time, this is what I’ve been waiting for.

I’ve said for a while that the biggest issue with the game—on the field—has always been the AI. Whether it's quarterback decision-making or coach play-calling, that was the core problem. Sliders are just the icing on the cake in my view.

So my takeaway, to a degree—and you were in that building, not me—is that there were things the developers wanted to do that simply weren’t technologically possible. But they couldn’t exactly come out and say, “Hey guys, we know what you want. We want it too. But the consoles just aren’t there yet.”

No company is going to let developers publicly admit that. At least, I don’t think so. The closest I ever heard was Rex mentioning something about jersey tangibility and how it just wasn’t doable at the time. I believe it was in reference for pass interference, defensive holding, etc.

To be frank, having each player on the field have their own individual "brain" (so to speak) wasn't possible until this generation.

American football is the most cognitively demanding sport in the world, and it's not close. The number of possibilities each player has to adjust too on the fly, especially on defense, is virtually infinite. As for play calling, I mean, the thing is it could have been better. QB decision making could have been better. The old system didn't really have a progression. It looked to the #1 first, but it had no concept of alerts, hots, or different primaries based on what the defense offers. To be honest, when I was there, only a few people had better football knowledge than the average fan, and this was back when the average fan wasn't nearly as informed as he is today.

Another problem is that individual player AI would have had to be built pretty much in data. I can't really say why except that's what I was told. We'd have to use LUA or something similar and it'd be competing with graphics for space. Basically, the way Madden was made back then, as much as possible, the engineers provided designers with tools, and we designers would build the feature in the tools. Most of the franchise logic lives in XML and various scripting systems (a node based one that looks like a mind map that I don't remember the name of, and the Need for Speed scripting engine which is mostly data packets pointing to data packets. It's what we used to tune FranTK. I'm oversimplifying a ton because it's been over a decade and my memory is fuzzy.)

I think it was Brett Kollman that set off the modern revolution in football coverage, but now there's a ton of creators just like him, some better than him, and the run of the mill fan is far more educated on the game than he ever has been. This results in Madden being even more unsatisfying than it already was. To be honest, I think Madden has made huge strides over the past 4 years or so, but it doesn't get a ton of recognition because the understanding of the game is outpacing its improvement by a wide margin.

However, it seems like turnover in the building has really brought the mentality around to satisfying that desire for more. It's not enough to look like football, now we're all watching All-22 on demand and having former coaches dive into NFL film every week, and I think the devs are starting to appreciate that. I have some very serious questions regarding a few people on the team and whether they have the mentality to adjust to the new age, but looking past the usual marketing speak, between the lines I see a lot of appreciation for the sport, not just the game.

[Cycloniac]

So, my main point in this thread was that I think the sim engine should be a touch less reliant on playbooks. I think today's franchise blog was a fantastic response to that pain point. Can't wait to get my hands on the game next month and get to work on a custom roster!