Not sure if I can answer that question since it's pretty broad, but usually they will do a lot of visual concepts to find the design they want for their characters and environments that would fit best their current project. They will also test some animations and do colorboards to figure out the moods they need before setting down on their chosen art direction. Ideally you want to be consistent throughout all visual elements, from the characters to the way they move to the UI.
-
Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
Ask a professional Game Dev any development or design questions that bother you
- Thread starter Farlander
- Start date
OP
OP
I would say that it doesn't really affect the challenges of a day-to-day work, because there are no easy games to make, just different challenges depending on the type of game. A one-button hypercasual has its own set of challenges that an open-world AAA game doesn't have, and vice versa, despite the scope differences and everything.
I must say though that if we take a look at everything holistically in terms of scope and team priorities, I would prefer if graphical fidelity would have a smaller importance in AAA games. But players expect that to be pushed further and further, so it's sort of another catch-22 I guess? I feel that at some point some graphical details provide diminishing returns for the amount of effort they need/resources they use.
Why is it when your avatar is loaded into a level they are often placed just above the ground and then fall into place? I've noticed this in multiple games where sometimes you load just fast enough to see it happening. I assume it's to prevent clipping/getting stuck in geometry, but not sure.
Thank you very much for the thread and all the questions you can answer (and other devs too!)
I'm going to ask likely a very broad question so bear with me, because as someone who doesn't know a lick of game design I would love to be better educated on the subject here.
Quite simply, what is a Game Engine and what can it/can't it do?
I ask because I typically see a lot of threads blaming the game engine for issues and a lot of back and forth as if that's true or not. So I figure I'd try and ask someone who has worked with one to give some insight and can hopefully tell me what gets commonly blamed as the engines fault really isn't.
I'm going to ask likely a very broad question so bear with me, because as someone who doesn't know a lick of game design I would love to be better educated on the subject here.
Quite simply, what is a Game Engine and what can it/can't it do?
I ask because I typically see a lot of threads blaming the game engine for issues and a lot of back and forth as if that's true or not. So I figure I'd try and ask someone who has worked with one to give some insight and can hopefully tell me what gets commonly blamed as the engines fault really isn't.
Do you have any thoughts on intrinsic and extrinsic motivation and how they relate to modern playerbase retention methods, such as battle passes?
OP
OP
Because performance is not defined by visuals alone, there are a lot of other things that come into play - AI calculations, object streaming, file placement on disc/drive (related to how it streams), dynamic lighting if such exists, physics.... there's a LOT to take into account for proper performance.
Usually designers/artists/sound engineers have different limits defined together - no more than a certain amount of NPCs within a certain radius, limited amount of sound points, etc. etc., but even then final rounds of optimization usually come into play close to the end of development once there's a clearer picture, and for various reasons those are not always successful - lack of time/require to make big changes/etc.
Invisible walls. In the first Horizon as one example. MGSV is another I have tried to shoot through the gap of a 3D object which looks like the arrow or bullet can easily gets through but it gets stuck there on that invisible wall. I don't like this. I've gotten a bullet through a similar situation in a game like say, Crysis 3.
OP
OP
Since PC is PC, there's no particular standards for this sort of thing (unless the launcher you're releasing for has one defined for it, like we had for Uplay - all saves were stored in a particular location), so the answer to this can range from what developers feel is more comfortable for users, or a bit more secure, or... well, quite frankly, sometimes it's just a matter of 'well this is what we have set during development and changing it to someplace else wasn't important enough in comparison to all other shit that needed fixing' :D
Forcing the player into any sort of tutorial/ first mission without being able to change basic settings (key mapping, audio, video et cetera) first.
The new battlefield games are very 'good' at that.
The new battlefield games are very 'good' at that.
Not saying it applies to you, but how does it feel when a developer has to release a game in a state when it clearly needs more work to be fit for release?
Like not in terms of the quality of the game itself, but when you get games that are, or are bordering on being, absolutely unacceptable in terms of performance.
Like not in terms of the quality of the game itself, but when you get games that are, or are bordering on being, absolutely unacceptable in terms of performance.
Because there are 2 different teams/disciplines in charge of each. Player speed is normally set by a gameplay programmer or a designer, and character speed could be set by any number of different people.
Also the questions "What speed feels good to move at" and "how fast should this NPC move" are 2 fundamentally different questions.
OP
OP
This is a very complex question to answer within a single forum post, but will try to go point by point...
- Scope is important, i.e. make sure to not scope more content than your mechanics can sustain.
- Make sure that gameplay has depth which would allow for a greater amount of choices/replayability (depth can come from skill mastery, amount of progression mechanics/skills, etc.)
- Keep a balance between varying more systemic and 'special memorable moment' goals.
- Stuff like mini-games doesn't really solve the problem of variety at its core, but it can help a bit.
- Reuse various gameplay elements at hand in different variations. For a very basic abstract example - a combat encounter with just melee enemies will feel different from a combat encounter with just ranged enemies and that will feel different from a combat encounter with a mix. Obviously there's a lot more to that than that, but to give an idea.
- Try not to do stealth missions if your game doesn't have dedicated stealth mechanics :p Those usually are very annoying, lol.
OP
OP
Playtime, monetization success, player churn. How much all that matters for early planning depends on the people at the top and their mandates.
Regarding DLCs, VERY rarely is a DLC something that was cut from the base game to be sold separately. That does happen, but it's an extreme case to get a project out of the door, usually.
DLCs are usually conceptualized in the middle of production as part of post-launch plan, so that active work on them could start once the main game goes Gold.
Makes sense, so how are games budgeted/staffed generally? I'm guessing it is on a per project per year basis with a ramp up, IE year 1 conceptual phase, year 2 full work phase and you gain 20-50 new employees, year 3 should be release with 3+ months of post launch support and DLC wrap up?
oh cool! I always wanted to ask a few regarding time since my main field is PM.
1. How do you review the timeline based on the project? AzureDevops? you keep notes?
2. Do delays happen a lot?
3. Do you use Sprints and see the game advance every 2 weeks or something like that?
1. How do you review the timeline based on the project? AzureDevops? you keep notes?
2. Do delays happen a lot?
3. Do you use Sprints and see the game advance every 2 weeks or something like that?
Since we are in "E3" season, how early before the presentation do you start working on a vertical slice demo or gameplay trailer. Does this process take a large amount of Development recourses. If you never had to show the game before it was basically done, how much dev time would that save?
I don't know if you'll know this. But how much work is porting to another platform, including something like a PS5 version of a PS4 game.
Oh and thanks for this thread! I love these kind of myth busting perspective pieces.
Oh and thanks for this thread! I love these kind of myth busting perspective pieces.
How much impact is Cross Gen actually having in terms of how modern game design is impacted? Like if say, Horizon Forbidden West was a PS5 only game would it suddenly have looked/played/felt far different than what we got?
Thanks for the thread!
- actually I've always wondered how game development process done (not talking about pre production, planning, beta alpha etc.), more like do you start by developing the tools and assets needed, and then starts with making the game from A to Z, or just random parts where X part assets are finished, and keep same process until all the parts done?
- does the game engine really matters? Especially when it comes to different genres?
- actually I've always wondered how game development process done (not talking about pre production, planning, beta alpha etc.), more like do you start by developing the tools and assets needed, and then starts with making the game from A to Z, or just random parts where X part assets are finished, and keep same process until all the parts done?
- does the game engine really matters? Especially when it comes to different genres?
Thank you for this thread!
Does having to scale games to different consoles SKUs (Xbox One -> Xbox One X, PS4 -> PS4 Pro, Series S -> Series X) add a lot of time/headache to development? And is there any truth to the assertion that something like a Series S could "hold back" what can be accomplished in a modern game?
Does having to scale games to different consoles SKUs (Xbox One -> Xbox One X, PS4 -> PS4 Pro, Series S -> Series X) add a lot of time/headache to development? And is there any truth to the assertion that something like a Series S could "hold back" what can be accomplished in a modern game?
OP
OP
Ah, this is a very good question! Honestly, misunderstanding of what a Game Engine is met very often, because it happens in the backend and people see the game itself so there's this opinion that 'game mechanics = engine', which wouldn't really be true.
An engine is a framework that hosts various features. A good way to look at it would be to look at it as a house, I guess? For example each house has a kitchen which serves a specific purpose, but how that kitchen looks will vary from house to house (i.e. game to game). And then there are pipes that lead water to the kitchen, that make the kitchen function.
So, let's say we have nothing. What do we need for a game? Ok we need to visualize things. That means an engine needs to have a render code that will visualize the 2D and 3D objects that we have. Now that is sort of a standard feature set that every game using this engine can do - it can show things on screen. Memory management, animation systems, sound systems, disk management, all these become part of the engine.
For example Assassin's Creed 1 didn't have a day-night cycle, it had static timeline. Dynamic day/night cycle and dynamic lighting wasn't part of the Anvil engine at that point. When Assassin's Creed II was developed, it added dynamic lighting which gives the possibility to do a day/night cycle - that became part of the engine. However it wasn't optimized yet so a lot of concessions had to be done for overall visuals. But then for Brotherhood the dynamic lighting system was optimized, so it wasn't as resource intensive. And now that dynamic lighting is part of the Anvil feature set - every game working on it can use it.
This is why it's an 'engine'. How that dynamic lighting system is used now it's up to the game - it might not be used for a day/night cycle at all, and if it's used for day/night an engine doesn't force you to make each day/night cycle the same in each game that uses it.
Now there's also tools. Editor and its features. Like Unity Editor, Unreal Editor, or a propriteary engine editor. Technically an editor is NOT an engine, neither is the toolset related to it - it's an interface that allows you to interact with the engine and do things for it.
So, to put it simply.... there's really no limits to what a Game Engine can do, it's up to developers to decide what they have the time/budget to add for it so it could do.
This is why Unity/Unreal and other licensable engines are so popular - they already have a huge set of features readily available which means that people working on those engines don't have to do those feature from scratch, i.e. let's say dynamic lighting.
But this doesn't mean that having proprietary engines is bad - some can be finetuned for certain purposes. For example Trials had its own engine built specifically for physics bike gameplay. And it was very good at it because every feature was built with the purpose of having 60fps physics-based gameplay. But then for Trials of the Blood Dragon there was a decision to add platforming sessions, and in Trials engine we didn't actually have a character controller or an animation system, in previous games character was just a physics object attached to the bike at various points (and when the character moved to shift weight they were just pushed as a physics object), so for Trials of the Blood Dragon we had to implement whole new systems in the engine for platforming sections to exist (and due to pretty short development time it didn't work out THAT great, lol, but I think the team did a great job at implementing things in a short period of time). Trials engine also didn't have any AI systems, all logic was hand-scripted in editor, but for enemies we needed AI so that had to be implemented, with decision trees and everything.
But if we'd spend more time on it, we would be able to transform the Trials engine into an engine supporting good platforming as well - because again, it's just that it didn't have all the features needed for that... but it could, in time.
Hope this clears things up!
TL:DR - Engine is a framework of features that can be reused from game to game.
Oh, and funny note! Some people think that some Ubisoft games reused certain mechanics like drones from game to game because it was easy to reuse them... but it's not, because let's say Watch Dogs and Rainbow Six Siege use different engines, so you can't just copy/paste drone mechanics from one to another, it has to be implemented separately.
Horizon: Forbidden West had a release date that was snuggling up against Elden Ring. Most of the universe regarded this as a Bad Idea. And yet here we are. Same with Titanfall 2 being sandwiched between Battlefield and Call of Duty. My question is; is there *any* veto power from *anybody* that can stop... this? Schedules are a thing, but to so brazenly volunteer to snatch defeat from the hands of even a chance at victory; does marketing have that much power against devs?
Because the mesh you see is not the mesh your character or anything else physically collides with. There is an invisible collision mesh on everything, and often for performance reasons this mesh is much less detailed and therefore has less budget for holes and that sort of thing.
Some games will have those sorts of areas accounted for, but often (especially in older titles) its one of the easiest places to reduce detail and get performance gains
Cool thread bro!
Have you worked in non-game software development and if so, what do you think is the biggest difference in your day-to-day??
Have you worked in non-game software development and if so, what do you think is the biggest difference in your day-to-day??
How much thought/attention is UI/UX design given normally? Not to call anyone out, but it seems like as opposed to things like 3D camera control which is largely seen as a solved problem in modern games, many modern games seems to make the same sorts of mistakes over and over again (things like unclear menu options, illegible or "style-over-function" font choices, inconsistent or overly complex menu structures, and especially on PC things like a lack of quit to desktop and other quality-of-life options). Anecdotally i've always heard that UI tends to be one of the last things added in a development cycle, so it makes sense that it might not get the attention it deserves, but are there any standard frameworks or engine plugins that are available, and if so, why might a developer choose to roll their own instead of just using those? I get that there is probably a desire to make every game bespoke, but UI/UX seems like the type of thing that would benefit from much more consistency.
On a related note, why is it that so many devs seems resistant to providing control customization. Is this just another example of "more work than they have time for", or are there conscious decisions made to lock users into a specific control scheme? If the former, why do we still see many examples of PC ports that open up the controls for keyboard/mouse but keep the controller options locked down?
On a related note, why is it that so many devs seems resistant to providing control customization. Is this just another example of "more work than they have time for", or are there conscious decisions made to lock users into a specific control scheme? If the former, why do we still see many examples of PC ports that open up the controls for keyboard/mouse but keep the controller options locked down?
OP
OP
So a bit earlier I tried to explain what a game engine is, to have more context on that, if you want to read.
Regarding your particular questions - sometimes developers are indeed mandated to use engines that do not have everything the game needs, but also usually there's a high level strategic goal to that - to develop an internal proprietary engine that would be used for all your games. This is what EA has been trying to do with Frostbite.
The real problem is for the production timeline/budget not accounting for adding new engine features. For example, if you don't mandate Bioware to use Frostbite for its RPGs and it is going to be kept for MP FPSes only, then it is never going to gain a feature set required to make this kind of games - robust save systems, facial animations, etc. etc.
And after a game implements those features in the engine, then other projects will be able to more easily reuse it.
But when you do it like that, you have to account for the time it will take to actually IMPLEMENT all that stuff, and not giving enough time is usually the biggest issue when stuff like that happens.
And it's easy to say 'just use Unreal/Unity', but outside of licensing costs which the publisher might not want to pay there are other considerations to be taken into account why it wouldn't be beneficial to do that. It's a pretty complex question but I hope I answered it clear enough!
It feels like some games nowadays have a bunch of systems that feel really disconnected from each other and so I'm wondering how often this is a result of games with huge development teams where work is split into smaller teams that may not always communicate much with each other
This is a good answer. Gamers misunderstand engines so much. I tend to tell people that the engine matters a great deal, but still a great deal less than most gamers think they do.
I've been involved in engine selection for multiple projects and the decisions have always came down to how well the toolchain supports the individual needs of the game, middleware support and the skills of the team (ie what they feel comfortable or familiar with). Its rarely came down to things like the renderer or some headline feature (although it can play a role) like a lot of gamers talk about.
This often isn't clear cut, even if an engine is "better" in some area. There are cases where Unity is better for world creation, and others where Unreal is better. Or, more efficient.
Thank you for your response, I had a suspicion it was something about meshes like that but I would never be able to put it into words like that. 🤝
OP
OP
1. Varies from company to company/publisher to publisher. Generally speaking larger companies have a whole process of pitching/greenlighting/having different "gates" so to speak, while in smaller studios it can be more freeform/organic.
2. I think nowadays there's a lot more possibilities on how you can choose a project you would rather work on. Even in a larger studio.
4. The funny thing about this is that a lot of high-level design principles between all the different types of games - from fighting to puzzle to open-world, are sort of very similar. It's all about loops, pacing, game feel, etc. etc. (broadly speaking), it's just that you have to apply some different details depending on the genre. But I never had big issues transitioning because of that.
5. Not really for one reason: even though platforms have very similar capabilities now, they have very different uses. People on PC play differently (different session lengths, different preferences, different priorities) than on mobile and those play differently than on Switch and those play differently than on PS/Xbox, the latter two being the most similar platforms nowadays due to both being high fidelity consoles.
OP
OP
I worked on photo mode before so I can tell :D
The basics are pretty straightforward, you pause the game and put in a different camera.
Problems come from unexpected things:
- How do you actually share the photos? Where do you store them?
- How do you prevent players from tampering with photos, as some may do so for malicious means?
- How does the photo mode work in multiplayer? You can't pause there and there can be various issues with that, like for example if you're driving in a car and you put photo mode on, suddenly the car keeps driving forward with you while the camera for photomode stays in place.
-- Related to that, do you cut the photomode in MP or make changes specific to it?
And then of course the scope issues - is there enough budget for different character poses, facial expressions, filters (those usually don't take a long time of course), basically everything that's in addition to the core feature set of 'enter a mode to take pictures in'
OP
OP
All true.
It's just that sometimes the 'CRT filters' are literally just scanlines with nothing else, and at those moments one wishes that more priority would be given to properly scope it beyond that and add emulation of at least some other CRT features like a bit of blur and bloom, the reason why those are important is because a lot of times they're key to making CRT visual tricks work.
OP
OP
Well, this happens ever since the dawn of time of making games :D In my personal experience, a project that had a much shorter development time than the scope it required was Trials of the Blood Dragon.
We all knew the flaws it had, but some of the responses were EXTREMELY harsh, from the journalists and from the players too.
I actually stopped reading Rock, Paper, Shotgun before of that. That publication's review of Trials of the Blood Dragon got REALLY personal in saying that the dev team is essentially incompetent and 'who knows what they were thinking', and that review alone brought me to a really depressed state for several months.
This is why when I review games for my personal YouTube channel I make sure to only criticize the product and not the teams behind it or their competency or skills.
OP
OP
Sometimes people are very focused on making sure stuff is done in time. And they work on their PCs. Time pressure can lead to people forgetting to check how it would actually look on TVs from a distance rather than from a monitor close by.
So, I suppose the answer technically is 'sometimes, yeah', but that's usually due to scope/time reasons (a lot of work with little time).
Interesting thread !
I have one question about UI and Menus :
Why is the framerate unlocked in menu screens in some games ? I had to stop playing some games from fear the console / PC would overheat while browsing menu items (ex : metal gear 5 on Ps4 in the chopper)
It's weird to me that the hardware is having a harder time on the menu that in the gameplay sections.
Is there a technical reason to not lock the framerate in menu navigation ?
Thanks for your answers.
I have one question about UI and Menus :
Why is the framerate unlocked in menu screens in some games ? I had to stop playing some games from fear the console / PC would overheat while browsing menu items (ex : metal gear 5 on Ps4 in the chopper)
It's weird to me that the hardware is having a harder time on the menu that in the gameplay sections.
Is there a technical reason to not lock the framerate in menu navigation ?
Thanks for your answers.
Very cool thread.
If you could define game development in fases which fases would they be maybe with some explanation of those fases and for a typical AAA game how much percentage wise roughly would those fases be of the total development.
If you could define game development in fases which fases would they be maybe with some explanation of those fases and for a typical AAA game how much percentage wise roughly would those fases be of the total development.
OP
OP
Best teams are when feedback/suggestions are not limited to just one job family. I've given some good suggestions to programmers how to handle certain technical issues, had a bunch of programmers and artists and testers come up with very nice solutions for design issues we were having.
Main job of a designer is not to come up with ideas, but to 'solve problems', i.e. make sure that the concept is implemented properly - and doing that with input from others is so much more efficient than swimming in your own lake alone and making it look like you're the most important person.
We do, and we have very strict budgets for visuals (memory, CPU/GPU performance, even per thread performance). But developers expectations may be different from players expectations, some sacrifices may be made or there may be time/budget constraints that prevent developers from reaching the targets.
OP
OP
Imagine you're playing a game like Assassin's Creed or Hitman where you can walk up behind NPCs to stealth kill them (something that happens in many games). This means for proper flow your walk speed has to be faster than NPC walk speed, so that's how it gets implemented.
And then you might get a situation where you have to follow an NPC while talking, well making speed adjustments for that essentially becomes a totally different feature on its own that's unrelated to the needs of core gameplay and it might not always get prioritized over other things.
True, I would imagine the bare minimal approach is simply down to it not being seen as a priority though, or that the person tasked with it doesn't really know how to achieve the effect.
It could potentially be performance related too, depending upon how and where its being used.
I do agree that if the effect is important then people should at least try.
OP
OP
Holy shit the most difficult question of this thread.
Uhm.
Uhm.
Well, I don't know, but when I played the latest (or, well, almost latest) Sword and Shield my MVP was Dubwool who was the only one who survived against Champion's Charizard and he won me the Pokemon League essentially :D And I barely used him before that because it was a normal type that wasn't effective against anything, but kept in the team just because I liked the design. And then he saved everybody! Not favorite Pokemon probably, but certainly goes for a memorable story :)