logo80lv
Articlesclick_arrow
Research
Talentsclick_arrow
Events
Workshops
Aboutclick_arrow
profile_loginLogIn

Green Hawk Platoon: Switching From UE4 to UE5

RetroPixel Digital's Nick Romick has returned to 80 Level to tell us how the studio's Green Hawk Platoon, an upcoming battle royale where you play as a tiny plastic soldier, has been moved from Unreal Engine 4 to Unreal Engine 5.

In case you missed it

You may find these articles interesting

Intro

My name is Nick Romick, and I’m currently now the Creative Producer at RetroPixel Digital in Vancouver, BC. I’m in charge of all projects coming out of the studio, including short films, web series, and video games. I also develop Unreal Engine courses for Epic Games as a consultant.

I originally studied 3D Modeling, Animation, and Digital Compositing at Dev Studios (now Pixel Blue College) in Edmonton back in 2006. In 2014, I took Houdini courses at CG Masters Academy in New Westminster and have taken about 100 courses and upgrades in between. I love to learn!

Previously I’ve worked for companies like Method Studios (working on VFX for Marvel movies, etc.) and more recently, worked on the set of Warner Brothers productions like ‘The 100’, DC Legends, Nancy Drew, and other local productions. In 2018, I was a producer of a fun indie fan film Galactic Battles, which featured Star Wars, Star Trek, Mass Effect, and Halo, and got over 6 million views on YouTube.

Since we last spoke, we’ve just been working hard on developing new features for Green Hawk Platoon. We decided to change the name slightly, too. There were a lot of really good feedback items from the community last time 80 Level did an article and we’ve really worked a lot of these ideas into the game. We love to hear what the community thinks… after all, they will be the ones playing our game!

The Milestones

Recently, we’ve introduced a lot of new features into the game, which means a lot of the old mechanics were changed significantly or scrapped completely. The biggest milestones we had were trying to develop a fun, exciting game style and mechanics. For example, when you die, your character goes rigid and the plastic stand under his feet returns, signifying that he’s still just a toy. It also removes any unnecessary gore and violence, giving you the impression that this is all occurring in a child’s imagination.

Trying to set this game apart from all the others is our primary focus. We don’t want to copy any other games, and we don’t want this game to feel like a bonus level for some other major franchise. We want it to be unique, fun, and visually stunning.

Switching to Unreal Engine 5

We really felt comfortable designing and building the game in Unreal Engine 4 when we started. Shortly after Unreal Engine 5 dropped though, we really couldn’t argue that the quality and abilities of UE5 really would help us launch our game into the next generation of games; both visually and performance-wise.

We decided to port over as much as we could from the original game builds, but there were a lot of mechanics that had to be overhauled a lot – so we started from the ground up in some areas. Honestly, it wasn’t all that bad. The migration from UE4 to UE5 wasn’t all that difficult that early in the project.

Thoughts on UE5's Key Features

Both Nanite and Lumen are game changers. Adding more visually stunning effects using UE5 features like Lumen really helped amp up the visuals of the game significantly. Things like god-rays and interactive (non-baked) lighting really make the game feel so much more realistic, and it becomes so much more immersive.

Being able to use features like Nanite also really helped us be able to use higher quality models and meshes without impacting performance. Of course, we’re keeping the poly count low, so we don’t have a 250GB game build! So far, it’s been very nice to have a game that looks this good without sacrificing precious FPS.

Organizing the Switching Process

Priority right now is going to game mechanics, network replication, and optimization. The level design is really the last thing on the list since it all needs to perform well before throwing everything into a level.

If you watch the alpha gameplay online, you’ll notice we’re mostly playing around on one level. This is the Alpha Tech Demo, which is really meant to be a vertical slice of the game, showing all mechanics and visuals. It’s less about flow right now.

Some of our developers are working on building and optimizing levels right now – making them look pretty and run well. They are also working on flow, so we don’t have any issues with gameplay. Nothing worse than having a level where snipers can rule the court, or where people can spawn-kill you over and over.

Optimization

Optimization has been a very important aspect of this game! We have multiple machines that we test on, including some older hardware like GTX 970 and 1080’s. Due to Unreal Engine's new Lumen technology, though, those cards won’t look as good since the new features require RX and RTX technology.

The game looks amazing. You’d immediately expect that it would be horrible for FPS (especially at this stage), but it’s been running very well so far on both AMD RX 6600 and NVIDIA 3060 GPUs (which are the lower-end current generation ray-tracing cards).

We’ve made sure to keep our file structures clean and remove any unnecessary geometry and complex collisions that aren’t going to be accessible by players. Network replication is a huge factor, too. Setting blueprints to be replicated the right way is extremely important for games with this many players in them. Both FPS and network latency must be taken into consideration.

We’re developing the game on an AMD system using an RX 6900XT GPU, which is quite powerful. We’re making sure to test out the alpha game builds on lower-end hardware to ensure a nice gaming experience for everyone. Since the new AMD and NVIDIA GPUs are set to be released soon, this makes our job even easier.

The Freezing Mechanic in UE5

It was my 9-year-old son who said that it would be cool (no pun intended) to be able to freeze if you’re in a freezer for too long. So, we decided to add this mechanic to the game, where if a character is hanging around too long in a freezer, they’ll slowly ice over and slow down. If they remain there too long, they’ll start to take damage and eventually freeze solid.

We pulled this trick off by animating a material instance and assigning it to a duplicate player mesh of the character. When it’s activated, it slowly animates from 0 to 100 until the character looks like an icicle. If he moves out of the zone, the animation is then played backward from 100 to 0 until there is no more visual ice on him, and his mobility returns. It really looks neat!

Balancing the Verticality Aspect

Well, since the players are so small, a regular bedroom or kitchen is huge. This presents a unique challenge when getting around maps. Using ladders to climb up to get to another level just doesn’t seem fast enough. This game is meant to be fast-paced with up to 36 players on each map, so climbing slowly can turn out to be hazardous.

We really wanted to focus on getting around fast. We wanted to give players the ability to get up and down easily and keep the flow of the game going. This included introducing items such as jump pads, parachutes, grappling hooks and even reducing fall damage. Using a combination of these also makes it super fun too!

Jump pads are fun. They’ll get you up to another level by jumping onto them and being shot into the air. You might wonder why we added parachutes if we reduced fall damage. You can jump straight down off a surface, or you can use your parachute to glide down and get farther across the map than if you just let yourself fall.

Grappling hooks might just be one of my favorite things in the game so far! Using darts or fridge magnets to hook your grappling hook to, rappelling you through the air really makes you feel like Spiderman! You can also shoot and throw grenades while flying in the middle of a grapple, so it can be deadly too!

We just really wanted to add as much mobility to the characters as possible, increasing the pace of the entire game! It just seemed like the only way to go.

Working With the Community

We’ve been working hard to engage with our community as best as we can. We launched a Steam Store page so people can wishlist the game. We created our own Discord server, and we started a small pre-Alpha testing group, where we play regularly; ironing out any major bugs and taking suggestions on how to make this game the best. We will be rolling out the Steam Playtest beta very shortly, too.

We also started a Twitch channel so we could show a bit of the pre-alpha gameplay for anyone who might be interested in the game at this point. It’s exciting to see the number of people excited about this game right now and it’s hard not to get excited about where this is going!

Funding is very difficult to do as we’re such a small studio. We’ve all stepped back from other projects right now so we can focus on this game and all the marketing and community building that comes with it. I’m personally averaging about 75+ hours/week on this game so there isn’t much time to spend doing side projects (or sleeping). We’ve decided to launch a Kickstarter campaign in order to keep the project going at this rate. We’re really committed to getting this done on time and I’ll be the first to admit that our timeline for launching this to the public is very aggressive. Also, things like server hosting, anti-cheat development, and keeping the lights on and the people fed really add up!

We’ve started introducing in-game advertisements on some of our levels. Before you sigh or roll your eyes, let me explain. It’s all going to be done in product placement that’s non-intrusive. For example, one of the levels that’s coming up is a convenience store. We all expect there to be 100 different products on the shelves anyway, so why not rent these products out to existing companies? This doesn’t look out of place but enriches the experience! Since the characters are only 6 inches tall, it makes every chocolate bar or energy drink into a 15-foot billboard.

The Roadmap

We’re just getting ready for our Steam Playtest (Beta) to go live. Up until now, we’ve been hosting the games peer-to-peer during the Alpha Testing. We can’t really do that for the beta release, so we’re getting ready to spin up some brand-new Amazon AWS GameLift servers to give everyone involved a smooth experience.

Once the beta has run for a bit, we’ll be releasing the game to the public as Free-to-Play Early Access on Steam by the end of October or November (depending on how everything goes with funding and development). Early Access will go on for another couple more months at most, then we’ll be into the full release of the game. This will include a few new levels for everyone.

We’re looking to launch some optional DLC (like skins and special cosmetic items) later in December (around Christmas) and some more new levels weapons, abilities, vehicles, etc.

We chose the free-to-play model so we could build up our player base and rely on making a great game in order to measure our success. If we don’t make an amazing game that people love, we aren’t going to make money. Simply put.

You can also support the development of the game by visiting Green Hawk Platoon's Kickstarter page.

Nick Romick, Creative Director and Virtual Production Manager

Interview conducted by Arti Burton

This content is brought to you by 80 Level in collaboration with Unreal Engine. We strive to highlight the best stories in the gamedev and art industries. You can read more Unreal Engine interviews with developers here.

Join discussion

Comments 0

    You might also like

    We need your consent

    We use cookies on this website to make your browsing experience better. By using the site you agree to our use of cookies.Learn more