End of November Blog
Hey Survivors,
Anton here! I’ve got some exciting news to share with you today. Our next Closed Alpha Update will release next week. This Closed Alpha update is a massive update for our team as it is a culmination of the company overhaul we made at the beginning of 2022. We’ve greatly expanded our team and narrowed our focus during this time, so we could make a better survival game and one worth developing and supporting for many years. It has been a tricky process and not without its share of trials and tribulations, but our newly formed team has brought new life and energy to the project. Closed Alpha update 0.10 represents the dedication and talent our team has put their heart and souls into making. So before we talk about 0.10 in more detail, I want to shout out our development team for being such a great group of individuals who are making a positive impact with Dead Matter.
Let’s address the most asked question about Dead Matter: When is it coming out?
The short answer is that Early Access is the next update after 0.10. We’re planning on announcing a new release date in Early 2023. This new release date will be a part of our first game trailer. We also intend on participating in Steam NextFest in February, where we will have a short playable demo for everyone to try Dead Matter out for themselves before Early Access release.
In August of this year, I mentioned three elements that needed to be addressed in Dead Matter to make it a more successful product and one worthy of an Early Access release.
To refresh your memory, this is what I wrote back then:
One: We need the Infected to be an intriguing and challenging opponent for the player. Our Closed Alpha testers know well that our Infected are not there yet, but we have a dedicated programmer working on improving their spawning and AI. Moreover, our technical artists are in the process of redesigning our Infected to make them more distinguishable and have higher fidelity character models. These implementations will begin to appear in 0.10.
Two: We need to improve our looting system. In 0.9, we are taking the first steps of many to improve this system. The looting experience will continue to be tweaked by our designers and is something we're focusing heavily on in our playtesting. We need this to be an engaging loop for Dead Matter to be a successful survival game.
Three: We need to improve combat. Combat has several components that we're thinking about improving, including guns, unarmed combat/melee, and performance. We have started making changes to combat in 0.9, but like looting, combat will also see continued changes as we progress into 0.10.
In our Closed Alpha 0.10 update, these three aspects have received lots of attention. You may have read about some of our team's work addressing these three points in our September and October blog posts. Today I will break down how we tackled these three goals for 0.10 with the help of some of our team members directly working on these goals.
ONE (INFECTED)
I’m going to invite our AI developer to explain the changes he is making to improve our Infected:
Hey everyone, as the AI developer for QI studios, there are a couple of different areas regarding the infected in Dead Matter for update 0.10 that I would like to discuss. The first is the spawning system, which I've briefly mentioned in previous blogs. In update 0.10, infected will now spawn in different-sized groups that will roam together in and outdoors of various locations. Whether it's a few infected in a hotel, or a horde in the forest, this new spawn system gives the infected more of a pack of wild animals feel and offers a new way for us to challenge players as they explore more and more challenging areas.
Although you will now find groups of infected generally moving together, instead of wandering idly or immediately attacking, infected can now enter one of two additional intermediary states that help provide players more options for interacting with infected. Depending on the distance of a seen survivor, the infected may approach out of curiosity or investigate the last seen location. Similarly, if the infected hear a nearby sound, they will investigate and check out the source location of that noise. With these new states, an aggression level has been added to the infected, meaning that if survivors repeatedly make an infected curious or investigate an area, they will become more aggressive, eventually immediately targeting survivors seen at any distance within visual range. Some of the new things these additions open to survivors are the ability to lure a few infected away from a horde in the forest by bashing a tree with a bat and making noise or triggering a fire alarm during PvP to draw nearby infected in to maybe help you out.
-AI Developer
Thanks!
We will show off some demos of our new Infected AI in our next Vlog.
TWO (LOOT)
As mentioned in previous blogs and our first Road to Early Access Vlog, loot has been a primary focus of the 0.10 update. In comparison to 0.9, it is fair to say that the loot gameplay loop is drastically different.
As discussed, loot now spawns exclusively in loot containers. This change allows us to randomise loot to keep areas fresh and designate where certain types and tiers of loot spawn. To further improve this container system, we altered the Eagle Eye perk to be a baseline perk for everyone and to highlight nearby containers within a short distance. This change makes distinguishing loot containers much more intuitive.
If you want a more detailed breakdown of this new system, check out our previous blogs and vlog.
I’m going to invite our Level Designers so they can discuss some of the loot passes they made to the game world:
Hey Survivors,
During this past month, we have been implementing the new loot container system throughout all the existing locations in DMF. The concept of this new looting system is to have interactions be more intuitive and seamless for the player and have a more defined 'loot pool' in terms of finding items from specific container types.
Alongside the loot overhaul, we have been going through the entire playable map and cleaning up stuff, which means that we are going through and removing props that have converted to loot containers, removing the old containers from the map and replacing them with the new ones. The final step was ensuring that each location has the basic interaction setups, for example, the I/O System. With that done, we also made some changes to the set dressing in most areas.
The creative process for this new loot initiative adheres to systematically going through a location and breaking it down into chunks. Whether the chunks are separate structures in one large space or an individual building with its own contained rooms. This categorization provides better focus in making the lootable containers have a purpose and be appropriate for the area they are occupying.
Let's use a residential house as an example. Most loot falls under a more residential theming for this particular location. That said, each room within a house has its own identity of findable items that provide a subtle feel of variance for the player. This semblance of variety can further intrigue the player and their interactions, as containers may look similar on the exterior but will offer a different order of loot on the interior. It adds a bit of a surprise element to the interaction, making the act of looting have its sense of unpredictability, but at the same time maintaining consistency in the overarching loot pool and its appropriate inclusion for a location.
-Level Design Team
Thanks, guys!
There are a couple of things we added since we last spoke about looting in one of our blogs that I’d like to highlight. First, Infected corpses are now lootable, which means there is a reward for taking down the Infected themselves beyond mere survival.
Additionally, in multiplayer, you can loot enemy players after you kill them. This loot change has a tremendous impact on player behaviour as it incentivizes you or your opponents to actively pursue other players. Killing a rival player is a great way of finding precious loot items like weapons, medical supplies, food, and water.
I also want to highlight something we’ve loosely discussed and that is the ability to loot car trunks. Looting is a massive component of Dead Matter, and you should frankly be finding loot beyond houses and buildings. Dead Matter features a large open world, so limiting loot to just those spots makes exploration a lot less fun. That said, you don’t want all of the loot around the open world to be in hard-to-find bags sitting at the base of a tree or fence. Stationary cars on the other hand are easy to spot, are common alongside roads and pathways, and are easy to engage with, so it’s a brainer to make them into a useful element for gameplay.
I’ll also mention that when you loot a dresser or cabinet with multiple drawers you will now see an expanded UI showcasing each drawer, meaning that you do not need to click every single drawer one at a time to loot a dresser or wardrobe. Again, a change like this is to improve the feel of loot, as we want it to be fun, not annoying.
We’re still working on loot, so I’m going to invite our game designer to share some points of improvement we’re still making:
Hello Survivors!
A little note on the looting initiative that we continue to improve over the weeks:
Create or Adjust loot tables
Better chances to find a backpack
Better chances to find weapons
Add new containers for the Level Design team
Some meshes are now containers that you can interact with
Creating looting volumes with tags for DMF areas
New Loot tables applied to DMF areas
If you're a Closed Alpha tester, please share your thoughts on loot drops and interacting with loot. Your feedback goes a long way in how we design this system.
Game Designer
Thanks!
Loot might not be the flashiest component of gameplay, but it is a principal element in any survival game, so prioritizing it and making it feel worthwhile and fun is why we decided to rebuild and focus on it for Early Access.
THREE (COMBAT)
The last goal we looked at improving for 0.10 was combat. 0.10 is an early iteration of combat improvements, but we believe this step forward is an important representation of our ultimate goal and direction. Combat is at the center of the Dead Matter experience as it affects how you engage with AI enemies and other players.
Combat hinges on a couple of things. The first point it hinges on is performance. Having a consistent and high frame rate is vital to making your combat experience feel engaging. Additionally, stutters and frame rate dips can ruin the flow of the action. When we began reworking Dead Matter earlier this year, this was the main concern of the technical debt we faced as a team.
As fun as new content is, it is not worth it if the game struggles to run at a playable framerate. There are so many aspects to the game's performance that it rarely hinges on a single element in the game. As we continued development, we noted several aspects of Dead Matter that have hampered performance. For 0.10, we made several changes to improve our performance metrics. We converted many objects to use nanite, implemented object culling, and once we migrate to UE5.1 we will implement HLODs. In addition, our newly hired technical animator has begun work on simplifying our metahuman character and Infected models to make them less CPU-intensive (although this likely won't appear in/or will be WIP for Closed Alpha 0.10). Our Code Team has also used performance tracking to target pain points in the game world and diagnose how these problems create hardware bottlenecks. Performance is top of mind and will continue to be so throughout development.
I’m going to invite our automation developer to discuss some of the tools he is making to help us clamp down on performance testing:
Nazdar vsichni,
Warm greetings from automation land. One of my tasks this month was to make a performance capture tool for our map; over time, this tool will allow us to notice performance changes on our map build to build.
The tool works with Sequences, which we place throughout our map in areas of interest (underperforming areas, commonly walked paths, performance-heavy areas with a lot of stuff going on).
The tool can target Sequences based on a tag. These tags are listed in a tag hierarchy, from which we can target a single specific sequence (single location) or various groups of sequences (multiple locations). This system is beneficial to either pinpoint a problematic area with a local test or target the whole town with all its sequences in one big performance test.
The tool proceeds to play the sequences one after another, captures their performance data and exports it into sequence-specific CSV files; if needed, the tool can also capture UnrealInsights profiling.
The tool is operational, and I am currently setting up a website-based app backend that will receive and store the data, which we will later be able to display in a frontend app in any way we find necessary (Who loves graphs?).
I'm planning to run performance capture tests on every development build, so we can see how performance changes daily. Should there be a drastic change, this tool will make it easy to narrow down the day our culprit change has happened.
That's about all I have to share; take care of yourselves a zatim zdarec.
-Tools and Automations Developer
I’m going to invite our lead Environment Artist so she can explain some of the changes she made to the game world to help boost performance:
For 0.10, we’ve done a huge amount of work, much of it very obvious and some a little more subtle, but all to improve the overall experience of exploring the world.
For starters, we did a polish pass on our existing locations to improve gameplay, visuals, and performance. This pass also includes getting them to work with Lumen, so interiors will now be lit correctly and look overall much nicer with improved shadows and GI – so make sure to have those flashlights ready! We also created a handful of new locations from the ground up for you to seek out and explore including the Kananaskis Gun & Archery Range, The LEDA Cold Storage facility, Hunting Camp, and Gonzos Gas Station, which are spread out throughout the map.
The wilderness also received an upgrade in the form of more varied and interesting terrain and foliage, as well as a brand-new water system that really brings our rivers up a notch. And finally, we’ve done a ton of work optimising, including implementing Nanite for many of our assets, so things not only look better but perform better too.
-Environmental Art Lead
Thanks team!
The other aspect of combat that needs improvement is our ballistic system. From end to end, there are many areas for improvement we need to think about. In 0.10, we began to tweak our weapons to make sure they are applying damage to players and Infected. Additionally, when you fire a bullet, it should feel responsive and accurate. While this system is far from final and frankly very early in its development, it will begin phasing out our older ballistic system. I won’t speak with too much more detail about our plans for ballistics, but it will be a primary focus for our technical director, and I will invite him to write an update on his changes later.
I’ve invited our game designer to add his own words to the improvements he is making on weapons:
Hello, again, survivors!
I want to talk about design changes you can experiment with in 0.10!
As the combat is core to the experience of Dead Matter, we want it to feel great. So I did a balance pass on every weapon of the game. I'm still tweaking these stats (work in progress), but here are the changes you can find in 0.10:
Parameters for melee weapons (1-handed & 2-handed)
Damage
Range
Attack speed
Stamina cost
Two-handed melee weapons are now slower and cost more stamina but have more range and do more damage. (Also, fun fact, you can now get an arm fracture preventing you from using two-handed weapons!)
Weapons are now properly ranked by tier and have damage tied to their tier level.
Firearms
Projectiles (damage, velocity, mass, diameter)
Weapon (RPM, Recoil, Time to aim, Sway, Magazines sizes)
Firearms parameters and projectiles are more realistic.
Weapons were tweaked with realistic values to make each type of firearm fill a different role. Shotguns or SMG for close combat, Rifles for mid to long-range combat, etc…
We still have some tests and refinements, but these weapons should be more fun to experiment with in 0.10, so if you are an alpha tester, please share your weapon feedback.
Combat will continue to see further improvements post 0.10 for the Early Access build.
-Game Designer
OTHER CHANGES
There’s a whole host of other changes coming to 0.10 as well. A more technical changelog will come out with 0.10, but for now, here are some of the other changes coming to 0.10.
Gameplay
Permadeath (Each character you create has three lives, once they’re spent, your character gets deleted)
Added Unarmed Combat
Added Auto-Run
Various Audio Fixes For Open World
Various Audio Fixes For Weapons
Medical System Fixes
Medical System Changes
Added Status Effects
Fixed Inventory Persistence Issues
Various UI Changes
Map
Added Lumberyard
Added Slaughterhouse
Added Gonzo’s Gas Station
Added Survivor Camp
Added Hunting Camp
Landscape changes around Dead Man’s Flats
Loot, Nanite, and Lumen pass for Dead Man’s Flats Main Street
Loot, Nanite, and Lumen pass for Dead Man’s Flats Residential
Loot, Nanite, and Lumen pass for Dead Man’s Flats Industrial
Replaced Bow River with Updated River Asset
Art
Mountain texture improvements
A note on MetaHuman models in 0.10:
We’ve been working with metahumans in 0.10 since September, but we’ve noticed that the MetaHuman models needed a performance optimization pass to be used on the larger scale that we intend to use them on. Fortunately, in November, we hired a Technical Animator to optimise our new MetaHuman Infected and Character models so they are less hardware intensive. In their current state, using the Unreal MetaHumans tool, these models are designed primarily for low character counts or cinematics. Therefore, an additional pass that streamlines the character components for gameplay needs to be accomplished. As we wait for these models to be optimised, we’ve temporarily reverted to our older models as they aren’t as performance heavy and that allows us to spawn in more Infected for the 0.10 playtest. Our new Metahumans will return before Early Access launch, but they need a couple more weeks of optimization.
CONCLUSION
That’s all for the November blog.
A changelog will release alongside 0.10.
A Vlog will upload around the 0.10 launch as well.
If you haven’t already, wishlist Dead Matter on Steam, your support goes a long way.
Have a great week.
