As per usual, anything said during the show is subject to change by CIG and may not always be accurate at the time of posting. Also any mistakes you see that I may have missed, please let me know so I can correct them. Enjoy the show!
The focus has been on getting 3.01 patch out as well as incomplete present features and performance optimizations
The development team went to a more encapsulated and complete model to prevent ping-ponging to better see the product from start to finish more efficiently
Quarterly releases will help speed up backer feedback as well as build up a cadence within the dev teams and prevent hold-ups to future build development
Besides working on the 3.0.1 patch the Live Team has been resolving about 25 issues including serious ones like the “white ball issue” along with issues with the rental system and ship shields, and pushing them out to the Evocati
Currently in the white box stage common elements for the hangar are being worked into the truckstop to be later used modularly through the PU (space stations and planets) in various sizes and via various forms of entry (straight in and via roofs)
The Weapons Team is currently busy working on more manufacturers, more different weapon types, and more different damage types for player selective use as can be seen with the new Gemini R97 shotgun
The UK Team has also concentrated on the Vehicle Manager (VMA) and Personal Manager (PMA) fixing bugs and improving features using both internal and community feedback and should improve the UI and user experience
The Gameplay Features Team has been working on the character customizer to allow for changes in the choices of head, skin tone, eye color, hairstyle and color within the first iteration with persistence
Engineering optimizations in the raw data system are being converted to the DATAForce system to take advantage of the Item 2.0 System optimizations
This optimization takes advantage of batch updates, update calling and optimizations over the last few months and with the Item 2.0 System it spins off logic threads and thereby speeding everything up
Future iterations will hopefully have optimizations in network calling and range calling as well as those that are unknown at this time
Weapon balancing is just one part of the whole balance cycle: shields, health, armour, item components, weapons, flight
To ensure that no weapon is overpowered or underpowered in all the key parameters and each weapon type is relatively balanced
Weapon balance is needed to make the game fun and competitive and to avoid the game becoming "samey"
They are looking to do a small shield and weapon balance patch for 3.0 based on community feedback
Shields were reworked for Item 2.0
They are working to being able to present various parameters to the player to enable easier comparisons
3.1 fix more known issues and community feedback for weapons, shields, armour, health, etc.
They've implemented armour: ship of the same size will now have the same health and armour will determine their "toughness"
Armour acts as a multiplier to reduced the amount of damage a ship takes, e.g. 0.9x or 0.7x damage
Armour is a more efficient way to make stronger (or weaker) ships as it doesn't require a different ship template
Ships are now in categories - civilian, racer, cargo, military - which determines their armour type
Armour strength can still vary within an armour type e.g. Super Hornet has stronger armour than Hornet
Armour also has an attribute that affects missile lock e.g. Hornet Ghost has "stealth" armour now
When an issue with a weapon is reported they first start at the weapon itself, and then work through shields, health, armor, and then other game systems to find the issue.
DataForge has helped limit errors that have previously been introduced by simple typos.
Balance is never-ending - always a way to make it better.
The current 3 types of weapon damage (kinetic/physical, energy, distortion) will be expanded to 4 or 5. That will take a fair amount of balancing.
Desire is to give plays the desire to mix and match weapons and not have one loadout that everyone uses on a given ship.
Chris Roberts (CR): Hello and welcome to another episode of Around the Verse. I’m Chris Roberts.
Dave Haddock (DH): And I’m Dave Haddock.
CR: Good to have you on the show Dave.
DH: Thank you.
CR: This week we have our Star Citizen project update for February. Much like the monthly studio updates these project updates will include a quick a report for the devs as well as deeper dives into certain aspects of Star Citizen.
DH: Yup. And in this month’s update we’re actually going to take a look at weapon balancing. But first let’s check in with the developers for quick update on some of the other things they’ve been working on for the PU.
Todd Papy (TP): I'm Todd Papy, Star Citizen Live Game Director. What we've been focusing on since we released 3.0 and since the start of the year is basically getting the 3.01 patch out as well as focusing on getting all the features that we felt were not up to snuff in 3.0, working on those to complete those, as well well as performance optimizations and everything like that.
So with 3.0 going out the door we wanted to focus on what we felt was right and wrong in our development process for that, so in many ways we've switched our process. Before we had a lot of kind of … we called it ping-ponging which was a feature was done in isolation, let's say in LA, and then they needed the UI, so then it was ping-ponged over the UK, and then from there the UK UI Team might be working on something else for a couple of months, and then it would get thrown back to LA, and so it just wasn't working as a cohesive team as well as a development process. So, we focused on adjusting teams and making sure that our teams now are encapsulated and they're seeing a product done from start to finish, and then also at the same time we looked at 3.0 and said, “Hey we don't think these things are right. Let's fix these. Let's focus on these, including performance.”
So with our adjustments to quarterly releases I think this will help the company. It will help basically get feedback from the backers sooner than later on features as well as it will build up a cadence within our development teams as far as getting these releases going forward which will help us only in the future and then at the same time if a feature is not ready in the past it was … we would actually hold up the build. In this case if the feature is not ready it will just go into the next release or the following release after that. We'll just continue to work on the feature, make it better and allow it the time that it needs to actually be built up.
So with Star Citizen across our five different studios we have a lot of work going on, however we'll give you brief updates on 3.01 patch as well as common elements and how those go into building out our environments. Then from there you'll … we'll talk about the VMA and also character customization being done in LA as well as the Item 2.0 performances.
Steven Brennon (SB): I'm Steve. I'm a production assistant on the Live Team, and I've been helping with the creation of a 3.0.1 patch. It came to our attention that there were some very severe issues affecting the public at the moment including a white ball issue where the players will become invisible and then after re-logging they would no longer be able to interact with anything, and they would need a Customer Service ticket to resolve it. So, we really needed to get a fix in for that as fast as possible. When we realized we were going to be doing a patch we also realized this was our opportunity to fix up quite a few of the other issues that the community has been talking about and wanting addressed such as problems with the rental system and balance passes on the ship shields, 'cause they were too difficult to destroy. We actually ended up having around 25 issues in total and most of those have been fixed now. We were able to push to Evocati at the beginning of this week and we're just working on finishing a few more of the issues including the white ball that turned out to be a little bit more difficult than we were expecting, but that one is almost complete now and we're hoping to go this week.
Adam Sanders (AS): In the UK we're working on the common elements for the hangar at the moment. In this particular case we're trying to implement this into the truckstop. Further down the line these common elements will be used throughout the PU … universe. That's the whole point of them. They're modular, and they can be reused depending on location. This point in the process we're working on the white box stage, and the white box stage is used for establishing metrics, so basically the measurements that were required in order to make this modular in this case. So they can also be placed inside of space stations. They … we were trying to intend … they used to be on planets as well. So, respectively for space stations we could have front loaders, so the door … you just fly straight into the hangar. For like planets we're planning on maybe landing through the top, and hangars will come in several sizes, so we're going to have like quite small hangars. We're also going to have giant like extra-large hangars.
Jonathan Jacevicius (JJ): So the Weapons Team has been very busy at the moment working a lot of different stuff. One of the weapons we've been working on is the new Gemini R97 shotgun, and this is working towards what we're trying to accomplish at the moment where we we're really trying to fill out the weapons selection that people have got at the moment, because it's currently reasonably limited. We want to just keep getting more manufacturers out there, more different weapon types out there, more different damage types out there to provide people with a nice selection of weapons to pick from. They can pick their favorites and use the ones that they feel are most effective. So that's … can be seen quite easily in the R97 shotgun cause we've got the Devastator at the moment that fires big bursts of plasma damage. You can pick between your normal shotgun blast, or you can pick between your charge mode, but that very much concentrates on raw damage. It's a very rough and ready looking weapon as well, whereas the new Gemini shotgun is very clean looking, it's more traditional, it fires normal ballistic rounds, and it's very, very versatile. You can pick between a tight spread or a looser spread so you can pick a fire mode that is more suited towards longer corridors or you can pick one that is much more suited to closer … close encounters.
Will Maiden (WM): The UK Team over here have been working on the VMA and PMA, which are the Vehicle Manager and the Personal Manager. Ever since 3.0 we've still got a few bugs that need fixing and a few features that need improving, so our focus is largely on that. We spoke to the community earlier in the month and we got a lot of their feedback and we've added that to our own internal feedback of what did and didn't work. So, we've got a roadmap now of where we want to take that system. So for the first sprint we're focusing on the vehicle side of things bringing that all in line with how the personal system works as well, so any fixes we've put in from the vehicles system will also knock-on in effect the personal system. So, it's mainly bug fixes at the moment, but how our UI side, Zane and Art Director Ian Leland are working on a new front side for how it looks, so it's … the user interface and user experience should improve, and then hopefully we'll be able to get some more features in as those systems come online.
Kirk Tome (KT): The Gameplay Features Team has been working on the character customizer, which allow players to customize their characters, so that they can manage the way that they look as they go around the PU. The first iteration of the character customizer is going to allow the players to change their head, the skin tone, their eye color, their hair style and their hair color. We're still working on the details about exactly when they're allowed to and how many times they are allowed to customize their character, and … but those changes will be persistent, so when you make those changes every time you login to Star Citizen you'll look the way you want. We're also working on the deciding exactly which of the different items that you can change are going to be entitled to the players, but we'll iron out those details before we release 3.1.
Stephen Hosmer (SH): Mostly my side, I've been just converting like taking the work that the engineers did to optimize it, and then I have to convert all of the items like the gimbals, the turrets, the missile racks, the EMP and just converting it to our DATAForge system, because it used to be in a raw data system, and now we have to actually convert it into DATAForge, so it can take advantage of the optimizations and the Item 2.0 System.
Max Hung (MH): I've been helping Hosmer with optimizing the game. So, basically what I do is I converted all of the old items into or some of the old items into the new Item 2.0 System, which takes advantage of batch updates and a few update calling and some of the optimization that we've put in over the last couple of months. The idea is that … the reason why we did this was because a lot of the items that we used to have updated on the main thread of which it is. . . which does networking, which does the UI, which does some gameplay logic, and if all of these … and then if all of these things happen on the main thread everything crawls really slowly, 'cause you have … we have planets to deal with, we have ships to deal with, we have players to deal with and a whole bunch of … every … all of them have their own hundreds if not thousands of items, and so now with the new Item 2.0 System we get to have … we get to spin off other threads to deal with these logics and then let the main thread only deal with the essentials like networking or with visual UI updates and stuff like that. In the future we hope to have additional optimizations like network calling or in range calling where we might … we won't update an item if they're too far away from the player or if the network is being really slow we can probably skip some of the messages that get sent along, and then just use the latest information that we got from either the client or the server, and then who knows. Maybe we'll have some more optimizations that we can't even think about.
CR: Thanks guys.
DH: Now we're going to take a look at how the devs have been balancing the various weapons whether they're ships or FPS ones that you'll find in the game.
CR: Yeah, it's always a bit of a challenge, and I know it's ongoing debate on the forums all the time, and I'm sure it'll be like that until we're done and even after we're done.
DH: [Chuckles] Right.
CR: But in this feature we'll get some details on the evolving combat experience in Star Citizen, so let's take a look.
John Crewe (JC): So weapon balancing is, as the name implies, the balancing of the weapons both ship and FP because the game is one fluid scene between both they do impact on each other a little. But it’s just part of the whole balance cycle we have - shields, health, armour, item components, weapons, flight - and they all blend together into the whole balance scene. And they all impact on each other - problems from one knock onto the other so it’s not always immediately clear where issues are but generally we start with weapons as that’s the thing that people pick up on first as it’s what they do the most of is shooting each other and shooting at things. So that’s where the problems first arise.
Andrew Nicholson (AN): Ship weapon balancing I suppose is the work I do to make sure that no weapon is overpowered, and no weapon is underpowered in all the parameters that we decide. We want to make sure that each weapon type - say the scatter gun - is relatively balanced towards a cannon weapon type or a hypothetical beam cannon type. So the scatter gun will do more damage that a regular cannon but obviously it’s rate of fire is slower. And we just make sure that all these parameters fit in the correct range that we give them on a per size basis, and that nothing is too strong or too weak.
The reason we need ship weapon balance is to make the game fun and competitive. If we allow situations to persist where we’ve got a gun that’s too strong you’ll find that the player base will all use that weapon and it’ll just … it’ll become a very samey game: everyone’s using the same weapons, same tactics. We want to prevent that by having as much variation but without allowing it to be one sort of “king” weapon that rules every other weapon. It’s really in the interests of gameplay that we strive to keep things as balanced as possible.
At the moment I’m working on updates to the 3.0 stream. We’re looking to do a short patch that does some balance work based on feedback from what’s currently live. Mainly I was looking at shield and weapons but these things are all connected and all determine each other. So when I’m looking at the shield health I want to make sure that it’s not too strong or too weak based on the weapons that we have. So weapon damage is a big part of what I’m doing at the moment.
Because the shields have been reworked for Item 2.0 and there’s now different characteristics involved. So with that change we have to make sure that, you know, repeater weapons have a rate of fire that isn’t too high to knock these shields down too quickly.
I’ve been working towards taking what we’ve got - parameters that we’ve got in game that determine rate of fire, damage, weapon spread - boil those into certain numbers that we can then show maybe as bars or just as an overall score to allow players to pick the best weapon for their situation much, much easier. Because right now you don’t get that. You just have to pick a weapon, play with it a bit, and then work it out for yourself.
So to try and get these numbers - these stats - I’m going all the way back to … to rework how we scale weapon damage - and all the parameters in fact. When we make changes to ship movement - the effects on weapons, the effect on targeting - it all ripples down so we see these problems and we can say “that’s going to be a problem soon: we need the time to look at it”.
Sometimes things go out a bit before we’re fully happy with them but that’s the nature of what we’re doing development wise: you have to predict these problems and when we get time to fix them we can. So for 3.1 it will be looking at those problems that we’ve seen and the community have obviously encountered and that we get time to fix them. When that’s weapons, when that’s ship movement: it’s all connected. So I’m hoping to do weapons, we review shields, armour, ship health - they’re all in flux.
Adam Parker (AP): For the longest time - pretty much since the game first came out - all the ships had the same armour values and their toughness was determined in their health. We decided that was incorrect so what we’ve done is we’ve changed it so the health of the ship is only determined by how big the ship is. And the armour is now how tough the ship is. So we scaled the armour with civilian ships, and racers, and cargo, and military ships. There’s an actual armour scale to it now rather than just bigger ships have more - exponentially more - health which was getting unmanageable. So we can control the ship health a lot easier now.
We started doing it at the end after doing the health balancing. What we did was we took the health down and put the armour on so even though there’s less health it still takes roughly the same amount of time to kill the ship. And then shields on top of that. It’s near the end of the ship pipeline once we’ve … because we can’t really playtest the ships when they’re half built so we have to wait for them to be done and then get all the actual tech balance working.
The way armour works it’s a multiplier on the damage the ships taking. We gave the civilian armours a multiplier of 0.9 so that times the damage by 0.9 so that lessens it a bit. But military ships would have 0.7 or 0.6 so the would take less damage overall. So even if the ships ... the Gladius and, say, the 300 they’re roughly the same size with the same amount of health but the Gladius will take more hits because it’s got better armour on. That’s … we prefer to have that rather than just giving Gladius an extra 3000 health because it’s tidier at the end of the day.
It also lets us swap armour around if we need to. There’s pirates in the PU that are supposed to be weaker than player ships. So instead of giving them less health - which requires an entirely separate ship - we can have the same ship the player has and just fit a weaker version of the armour to it. So that saves space and when we need to update the ship we don’t have to update each version of the ship as well: we can just update the ship and then the amour will be fine for all the different varieties of it.
Going by what we’ve got on the website we separated each ship out into its category. So you’ve got civilian ships like the Auroras and the 300s. Then you’ve got the racers like M50. combat ships like Gladius and other things. And cargo ships like the Freelancer. We’ll .. they were put into groups and they were given base armour values which were then modified by the size of the ship in question. So the M50 now has quite weak armour compared to the other ships because it’s a racer it would have thin armour to allow it to go faster.
That’s just how the … that was were we got the baseline for each armour type for. And then further iterated it from there because while the Hornet and the Super Hornet are the same size and the same military combat ship the Super Hornet does have tougher armour than the regular one. So there are separate armour values for each ship: they just belong to a group of armour type now.
The other reason we did this was to have armour also determine missile locks on ships. So for … this is only really done so far with the Hornet Ghost - which has … which is more difficult for missiles to lock on to - we’ve put those values in its armour. So it has stealth prop … actual stealth armour now rather than it just being something we say it has. So we can do that with other ships. So we can now give proper variant armours. So if you want your regular Hornet to be stealthy you can swap it’s good armour off, put some stealth armour on: it might be slightly weaker but it will have better attributes for dealing with missiles.
AN: Sometimes there will be unpredictable things - say like there's an AI ship in the PU that we've not really been playing close attention to but they've used it in a story or a mission and its time-to-kill is way too high but it's like, “Yeah, you've picked a super powerful ship,” sometimes that happens, you know, sometimes people want certain characteristics from certain things that maybe they weren't designed for. It does happen. We have to react to that and there are ways to like just change the items on the ship or to make modifiers present to make the ship weaker or to make the ship stronger - all things you have to react to.
JC: So, when we get reports of issues or things we find when we're testing ourselves and playing internally, we have to sort of go through a few steps to work out whether it's actually a valid issue - whether it's valid for weapon balance or whether it's something from another area or in some cases it's something that players have been exploiting for a long time that we never intended them to do and we fixed it thinking that no one's been doing that and immediately people go, “Well, this doesn't work how it used to work before - something’s broken,” when in fact it's actually behaving how it's supposed to behave.
So, yeah we go through it all and see where everything blends together - like a common one people say, “The weapons don’t feel like they're doing enough damage,” and aside from the weapons perhaps not doing enough damage, that is quite often from network desync - your pips are on something as a client and you feel you should be hitting them but somewhere between the client and the server it goes slightly amiss and then suddenly you're not hitting anything. You feel as a player that you should be hitting them but you're not. Your first instinct is to go, “Well, the weapons are wrong” and there's tons and tons of other scenarios like that where it's something else that's actually causing the issue rather than the weapons but it always comes to us first to sort of identify where that problem is. So, we investigate the issues - first course is always to look at the weapons because that's something we can very quickly work out. It's all in spreadsheets so we know how accurate things should be at certain ranges. how much damage they put out, how long the time to kill on various combinations of ships with shields up and shields down - so we have the perfect paper number and if it’s not close to that, then we know something else is wrong. That could be as simple as someone's just put the wrong number in the XML; put a decimal point in the wrong place. If it's not that simple, then it's starting to go through all the other areas of the game to work out, “Where is it not lining up?” Is it the shields are too strong? The shields themselves have so many values that it could be any number of them. If it's not the shields, is it the armor? If it's not the armor, is it the health? If it's not the health, is it perhaps the way ships are flying? Perhaps something has happened in the flight handling that they’re now a bit twitchier that, as the person doing the damage, you don't notice at that distance? Then we go through all the the various code disciplines to work out, perhaps there's just some underlying bug in the code that only appears in multiplayer, perhaps it's the network again, perhaps it's the UI, we've had times where the pips - they look like they're doing the right thing but they're in fact completely off. All these sort of things we have to go through and isolate and work out where the problem lies.
At the end of the day we want everyone to have a good fight with the weapons. It's got to feel, it's got to sound, it's got to look good shooting people and some of that is not from design numbers that we're tweaking - it could be the VFX of the hits just aren't rendering at a certain distance, so again you think you’re not hitting but you are. The audio might not be working and as designers and working on the balance, it's about bringing the entire experience together from all of those disciplines. Much like the ship setup pipeline, the designers - at the end of the day they're the ones that bring it all together and then continue to monitor it and work on it well after release.
AN: I think over time the process has changed because we've done a lot of work to change how weapons and items all work from the code level, so now that we've transported things or ported things over to use DataForge a lot more, that does make my life easier. You've got parameters there in a tool that doesn't let me typo things - that does happen, you know I've been told quite a few typos they do creep into the game but that's because as designers we’ll have to just hand-edit stuff. The problems that creates sometimes are tiny, sometimes they're quite major, nothing super massive, but they are noticeable so having things locked into records that we edit in a tool makes life a lot safer, a lot less stressful but like fundamentally with how you think about balancing weapons - I would say that hasn't changed, only that maybe I've changed because I'm new to it. So, it's been done before but I'll come to it maybe with fresh ideas, maybe when you talk to other people they've got certain ideas - especially when you talk to system designers that have certain wants and needs for what you're trying to do all kind of influences. We've been doing this a long time, so it's difficult to not be influenced by other people's opinions on how that should work. I'd say the process hasn't altered that much but you're always listening to new ideas that might affect how this works.
JC: Balance is never-ending, we're always looking for more ways for it to feel better, to work in synergy with the other areas of the game. We've got more weapon types coming that do specific features, so in 3.0 it's pretty much just ballistic or physical, energy, and distortion are the three main types but in the future we're going to be expanding that to about four or five so then there will be the whole play with those different energy types - how they work on different ships, how they work on different items - just to try and get people mixing stuff up a little. We don't want to have that one great loadout for each ship that is the best at everything. We want people to constantly be mixing and matching what they're doing for their encounters that they're expecting to have and sort of putting more thought into how they play rather than so and so plays with four of these and he gets everything with it. We don't want that to necessarily be the end game, we want it to suit how you play.
It's never finished, never satisfied, there's always more we can do.
CR: Some really cool stuff there. Remember you can follow Star Citizen’s development with our new Roadmap to track updates on new features and up-and-coming content for Star Citizen’s Persistent Universe. Check out the link in the video description below if you want to go check out our Roadmap.
DH: And that’s it for this week. Reverse the ‘Verse Live airs earlier again tomorrow at 9AM Pacific with guests Andy Nicholson and John Crewe from the Design team. I assume talking about …
CR: Yeah, I’m pretty sure they will. So you have a whole hour to grill them about the questions that were probably raised by the piece you just saw. And make sure to check out the new episodes of Calling All Devs - which is, I think, going over very well so Jared’s being doing a great job on that - and the old stalwart Loremaker’s Guide To The Galaxy - by Dave and the rest of the writing team - that aired earlier this week.
DH: A big thank you of course to our subscribers for sponsoring shows like this and all of our other ones.
CR: Yes definitely thank you. And of course thank you to all our backers: your support let’s us continue to make Star Citizen and Squadron 42 the best games they can be. So thank you very much. Until next week, we’ll see you …
Both: Around the ‘verse!