• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

EA Details How ML & AI Bolstered Development of Latest Madden & College Football Titles

T0@st

News Editor
Joined
Mar 7, 2023
Messages
2,555 (3.51/day)
Location
South East, UK
System Name The TPU Typewriter
Processor AMD Ryzen 5 5600 (non-X)
Motherboard GIGABYTE B550M DS3H Micro ATX
Cooling DeepCool AS500
Memory Kingston Fury Renegade RGB 32 GB (2 x 16 GB) DDR4-3600 CL16
Video Card(s) PowerColor Radeon RX 7800 XT 16 GB Hellhound OC
Storage Samsung 980 Pro 1 TB M.2-2280 PCIe 4.0 X4 NVME SSD
Display(s) Lenovo Legion Y27q-20 27" QHD IPS monitor
Case GameMax Spark M-ATX (re-badged Jonsbo D30)
Audio Device(s) FiiO K7 Desktop DAC/Amp + Philips Fidelio X3 headphones, or ARTTI T10 Planar IEMs
Power Supply ADATA XPG CORE Reactor 650 W 80+ Gold ATX
Mouse Roccat Kone Pro Air
Keyboard Cooler Master MasterKeys Pro L
Software Windows 10 64-bit Home Edition
On June 1, 1988, the very first Madden video game was released to the world. Players needed to load up either a Commodore 64/Commodore 128, Apple II, or MS-DOS to launch the game. When they did, they were greeted with 8-bit animations of the NFL's most popular teams and found themselves controlling their favorite players to try and win themselves a Super Bowl. And at that time, it was amazing. Thirty-seven years later and EA SPORTS hasn't stopped advancing Madden and our American Football games.

Most recently, we launched EA SPORTS Madden NFL 25 and College Football 25, which are tentpoles of our beloved American Football Ecosystem. Yet our football games are no longer blocky pixels and four-directional controls. They're among the most realistic sports simulation titles on the planet. We even celebrated the recent Super Bowl weekend with these titles and our very own Madden Bowl, featuring championship games and incredible music all in the heart of New Orleans. This is in no small part due to the incredible teams and their mission to make our games better every single year. And technology plays a critical role in making this happen.




Technology has allowed for graphical advancements, taking players from a bird's eye view of the field to so close you can see drops of sweat. Technology has introduced dynamic and real-time weather events, affecting player movements and ball handling. And BOOM Tech in Madden NFL 25 introduced crushing dynamic tackles.

And for our developers, technology has allowed for the development of 150+ unique stadiums and thousands of unique players-a core part of the unmatched authenticity and immersion our players and fans have come to love-allowing for the launch of titles like College Football 25.


Empowering developers with new tech
Richard Burgess-Dawson, a 20-year veteran at EA was the franchise art director for College Football 25. To Richard, 2025's edition held some of the most exciting technological updates ever that allowed developers to put out content that otherwise would have been impossible, thanks in large to machine learning and AI.

"From a development perspective, there were four incredible technological innovations: how we built stadiums, how we built uniforms, GIBS lighting tech and doing run-time lighting instead of pre-baked, and our character creation," says Richard. "And, of course, machine learning and AI played a key role."

Machine learning and AI for player generation
For both creating players and stadiums, it was all about assets. For College Football 25, there are more than 11,000 players our developers needed to insert into the game. Historically, this would have meant a team of artists going through more than 11,000 player photos and building the players out one by one.

With machine learning and AI, it sped the process up remarkably, so artists could spend time on other parts of the game. "What we wanted to do was figure out what ingredients we needed to make. In order to accurately and realistically represent the 11,000 athletes in the game, we used machine learning to identify 240 head shapes we could make that would best fit each player," says Richard. "AI could then help with pattern recognition. We looked for patterns of people and made a head for them. This created the base for all player heads."

With a batch of 240 heads that could ostensibly represent all 11,000 players in some shape or form, the next thing was connecting the dots between a player's submitted photo and the database of heads.

"If I gave you a flipbook of 240 heads, you'd be stumped," says Richard. "By 58, you're starting to lose your mind. So what we did was use machine learning to come back to the artist with 10 options and then let the artist decide. The AI is an assistant to a human. It would give its best guess and we would go in and tweak." With those options in hand, the artist had everything they needed to create the player-recommended head shapes, skin tones, hairstyles, beards, etc. Then it was about applying their expertise, experience and craft to make the best representation possible for these college athletes.

Focusing on the unique
Yet this ingredients-based technique wasn't only used for in-game athletes. It also helped developers and environment artists create the 150+ unique stadiums found across the United States in College Football 25. "In the past, the artist would go in and make a soup of content-railings, tunnels, lights, and so on," says Richard. "If anyone made a mistake or a change needed to be made, it took a lot of time and effort to pick through what each artist did just to make an adjustment. Let alone the time it took to create the stadium in the first place. With our newly developed Stadium Toolkit, this eliminated so many challenging steps."

The toolkit worked similarly to the player development in that it pulled from the developer's assets of common areas and parts you would find in any stadium-stairs, entrances, tunnels, fire escapes. "Those are just taken care of procedurally," says Richard. "And then the artist spends time on the stuff that's truly unique. Almost every stadium has a jumbotron and a press box that's unique to that stadium; it's their signature. This meant we could hone in on the unique, iconic pieces of each stadium without trudging through the mundane aspects."


That means instead of spending time building each step one by one, artists could instead focus on the unique-Michigan Stadium's iconic "M" logo in the end zones, the Rose Bowl Stadium, UCLA/USC's blue and gold color scheme and its surrounding mountains, or Notre Dame's brick exterior and "Touchdown Jesus" mural. All of these are what fans and players remember, and to Richard, that's what the artists should focus on.

And the best part? This technology will not only be used for future installments of EA SPORTS Madden NFL and College Football, but for any other EA SPORTS game if the developers so choose. "The cool thing is that this isn't just for College Football 25. It can be baked into every iteration from here on out, including Madden, and hopefully, other teams across our EA SPORTS catalog also see its powerful utility," says Richard.

Looking to the future
Technological advancements are a win-win for players, fans, and developers alike. By empowering, accelerating and expanding historically labor-intensive projects such as player likeness and environment modeling, our world-class developers are able to focus on areas that matter most-player experience, gameplay innovations, updates, and more.

EA continues to drive innovation forward, including the addition of new tools and technology. Recently, EA agreed to acquire TRACAB Technologies, enabling teams to deliver enhanced realism and more immersive EA SPORTS experiences Our Searching for Extraordinary Experiences Division (SEED) continues to explore, build, and help determine the future of interactive entertainment. Frostbite, whose engine is shaping the future of games. And many, many more.

View at TechPowerUp Main Site | Source
 
Joined
Apr 18, 2019
Messages
2,558 (1.19/day)
Location
Olympia, WA
System Name Sleepy Painter
Processor AMD Ryzen 5 3600
Motherboard Asus TuF Gaming X570-PLUS/WIFI
Cooling FSP Windale 6 - Passive
Memory 2x16GB F4-3600C16-16GVKC @ 16-19-21-36-58-1T
Video Card(s) MSI RX580 8GB
Storage 2x Samsung PM963 960GB nVME RAID0, Crucial BX500 1TB SATA, WD Blue 3D 2TB SATA
Display(s) Microboard 32" Curved 1080P 144hz VA w/ Freesync
Case NZXT Gamma Classic Black
Audio Device(s) Asus Xonar D1
Power Supply Rosewill 1KW on 240V@60hz
Mouse Logitech MX518 Legend
Keyboard Red Dragon K552
Software Windows 10 Enterprise 2019 LTSC 1809 17763.1757
Not great timing, IMO.
Let people 'warm up' more to AI-ML tools running on their own systems, before trying to butter us up (esp. EA.) :laugh:
 
Joined
Aug 20, 2007
Messages
21,874 (3.41/day)
Location
Olympia, WA
System Name Pioneer
Processor Ryzen 9 9950X
Motherboard MSI MAG X670E Tomahawk Wifi
Cooling Noctua NH-D15 + A whole lotta Sunon, Phanteks and Corsair Maglev blower fans...
Memory 128GB (4x 32GB) G.Skill Flare X5 @ DDR5-4000(Running 1:1:1 w/ FCLK)
Video Card(s) XFX RX 7900 XTX Speedster Merc 310
Storage Intel 5800X Optane 800GB boot, +2x Crucial P5 Plus 2TB PCIe 4.0 NVMe SSDs, 1x 2TB Seagate Exos 3.5"
Display(s) 55" LG 55" B9 OLED 4K Display
Case Thermaltake Core X31
Audio Device(s) TOSLINK->Schiit Modi MB->Asgard 2 DAC Amp->AKG Pro K712 Headphones or HDMI->B9 OLED
Power Supply FSP Hydro Ti Pro 850W
Mouse Logitech G305 Lightspeed Wireless
Keyboard WASD Code v3 with Cherry Green keyswitches + PBT DS keycaps
Software Gentoo Linux x64
Not great timing, IMO.
Let people 'warm up' more to AI-ML tools running on their own systems, before trying to butter us up (esp. EA.) :laugh:
Yeah this is rather tone-deaf atm.
 
Top