How Close Can Indies Get To AAA Games?

They had the tech , but it took 5 years to develop the game , i am not sure they were 86 employees from start ?

 

https://en.wikipedia.org/wiki/Warhorse_Studios

 

Linear doesn't always automatically equate to easier / simpler / less time consuming.. One advantage of an openworld game is filler, roaming / exploring as opposed to me having to construct set pieces linearly one by one which can take far more time than creating a prod placement alg or even slapping in foliage / rocks by hand.

The only time an openworld game becomes an issue is when the engine isn't in some way actively made for it.. Or if it's just simply massive then of course it takes a lot of time to fill it with something.

A char creation system in an openworld game is the same as one in a linear game, no real advantage. It's ideally about simplifying the process as a whole, the more artwork needed the more time consuming it's going to be.. Not just in terms of art but code too.. The more interactions / dialogue / clothing changes / enviro changes etc. etc. the more animations / scripts etc. etc.

In short one thing leads to another..

Ultimatley I am aiming for a "semi-openworld" RPG that can take advantage of both.. Linear and small enough not to boost scale and implementation efforts to unachievable massive proportions but retain some openworld engagement to again balance the amount of effort required. It's about that "fine line", I think Skyrim is a beautifly simple game which hits that fine line perfectly..

 

What I like about openworld games is that potentially you can have it functioning and 'playable' almost continuously while you add stuff to it. So it lends very well to a situation where you want to put something out there and build on it. Whereas with linear games, everything depends on what came before and it's hard to make an addition without upsetting the structure.

 

Frankly 86 people is impressively small for the amount of content and quality bar they aim for. 200 is a realistic sum, so they're probably using the same time saving techniques indies use.

It's possible for people to suffer mental illness trying to reach things they can't, so I'm not going to pretend or be part of a club that hypes up dreams. I'm out.

But you can still make something you can be proud of, or a top quality smaller experience.

 

It is open world game, i see anything linear in it, perhaps dungeons are linear, but whole world is exploration.

You need to buy plugins that speed up creation, there is many ones that have procedural placement and generation (Gaia, Gena , Octave etc ...) , some that allows you to quickly build geometry in editor like Probuilder, some plugins for particles giving you a AAA grade effects in some clicks.
For characters, you can propose only few characters models base and limit the customisation.

I indies can still get some AAA touch and achieve some game reducing a lot the number of textures, shaders, models, animations, effects, but pushing the quality of the few ones to the maximum.

Or it can become some never ending hobby trying to make Witcher 3 for some people.
I think it can be shocking for those people if they finish their game and next Skyrim or Witcher 4 comes out blowing again any open world game made.

Zelda Breath of The Wild changed open world and gameplay interactions vision, it's a level above any other open world made until now.
I wish next coming open worlds will be as dynamic and interactive or i they will look borring.

 

C'mon Zen give me some credit, you think after 3 years and multiple attempts I've ignored the asset store completely? No I've either combined tools, dismissed them and / or built my own..

I've got to ask just out of curiosity more than anything else, you're giving advice which is fine but where's this coming from? Have you actually ever worked on a large game of these types before and have you got anything to show for it? How do you actually know?

I promise it's not a dig, I'm not invalidating your opinion or having a go.. It's just unless you've done it I'm not sure you can really understand whilst the asset store is a decent crutch it ain't going to save you, like @frosted is working on a kinda openworld tactical (I suppose you could say RPG).. I've seen it in action, a long time ago played a portion of it and he can give me trustworthy advice on specific issues and even though our games are not one for one he can understand what I'm going through or anyone doing this sort of project.

We should set up an RPG anonymous group

P.S nice UE4 screenshots

P.P.S In regards to Skyrim, I was talking about how they approached the design.. They decided on what really mattered, streamlined their processes and it's excellent how they approached the game. If you've ever used the creation kit (to peak inside the inner workings of the game) it's pretty smart really..

 

I don't see the need to try to mimic a thousand people game project as i won't do as good, niether any indie alone or small team.

I don't want to make Skyrim, i know i can't make something as polished, still if i was able and let's say finish something within 3 or 5 years, Skyrim 2 will be out and most people will prefer to play Skyrim 1 for ten bucks.
If making some open world game was some very personnal goal perhaps i would really want to make whatever it would look and play, perhaps i would be trying to make one, but it's not the case and i don't have some need to demonstrate to other people i could make some open world game.


Sure it's a pleasure to find the good tools , achieve technical stuff or new features yourself and achieve great some great results. For example i'm happy succeding to make my own AI system with EQS and covers for fps characters AI, but it's for allowing me to complete some game not to show other people it can be done and i also kept a limited scope as i know i won't make something as big studios can make.

Anyway, making an interesting open world that does not look copy and paste is huge task, even with procedurals tools big studios employs lot of level designers to create all unique places you find in game. Without new gameplay ideas, new places or involving story and characters yo'ull make another rpg among thousand of other rpg looking similar (open world or not).

This is what i mean by less , but maximum quality, like less vegetation assets but maximum visual quality. Same for modular buildings, or characters , make less ones, but keep the best quality possible with AAA touch.

I am aware of modularity from long time a ago, it was pushed forward for their last game
www.youtube.com/watch?v=QBAM27YbKZg

Another interesting video is optimisation, you'll find incredible things happens in big collaborative games like Witcher 3 lol
www.youtube.com/watch?v=p8CMYD_5gE8

 

I thought we'd gathered several threads ago that it's a ridiculous notion to take on AAA as a small indie, nobody is saying they're trying to mimic a game with a "1000 man team".. I could reply to the rest of this but sorry zen, it's just fluff without any real depth to it.. None of it's specifically helpful.

You have to remember most people who will get anywhere with these projects already have quite a bit of experience in games dev anyway.

As for the rest it's a two minute search which has been posted here several times along with most people into RPG's and are dev's will have watched it anyway..

 

<It's worth noting AAA game's detail are generally limited by the technical limitation of hardware, I guarantee you if they could just plop assets without paying attention the visual would have skyrocketed and the discussion would not even have a chance to start. Because, if an indie ever do something that is perceive as AA it will be by strict planning and thoroughly anticipating problem, ie having a very low margin of maneuvering, you make decision and live with them forever. AAA use brute force to walk around the problem and adjust as they go.

Not really an example, but who played grandia 2? It' was a well received JRPG back in the day. It's not AAA by any standard even at the time. But you could feel the tight planning in the lay out of the game, all village had the same number of fluff and important npc, there was always 4 fluff house, 2 services houses, 1 story area where the trigger to continue was and one cinematic events area where you go after the trigger to get the big story events, then you have a field to traverse to the next town with a set number of enemy. Nobody complained, nobody even seem to notice it, each area had its own distinct visual and architecture and it kept the pace of the story steady with constant revelation the story is tropey as F*** but engaging enough. But also it was coded to the metal using primitive tools. When I say we can do AAA, it would be along these way with modern tools and services. Strict planning, locked decision very early, no wiggle room, no prima dona artistic vision.

 

The question "How Close Can Indies Get To AAA Games?" can be simple answered with 42.

 

There are very experienced people also that make games that are not open world and they succeed completing it and even selling it. I've also seen lot of indies showing some great tech demos, but never completing the game because of their scope was too big or they underestimated the amount of work it requires to make some hight quality production.
For example some small indie game like Ghost of a Tale is already a very long development game for one guy, and it's on a reduced scope , it's not open world and it's focused on specific places and gameplay, but already it's a huge amount of work and it took years.

Indeed anyone is free to try to make some AAA games, whatever they succeed or not , complete it or not, sell it or not is doesn't matter as there is no expectations for some incredible AAA game.
Nothing is wrong with the quest to AAA games lol , i am among those that prefer to keep a smaller scope to keep it as projects that are more realist and that can be completed in a reasonnable time frame.

Anyway, perhaps soon your game will start developping and taking shape when you'll finally decide to stick with some 3D engine and stop swapping ( or perhaps it's the technical evaluating and making tech demos is what you really like also).
Kickstarter and recruiting some talented character and level artist is another option among a lot,
i wish you'll take the right decisions to move forward if completing a whole game is your real goal.

 

Sure, even with experience in can fail miserably.. Even AAA projects have bitten the dust that's just the nature of it. I don't believe any of us are trying to compete with AAA, I don't know why you keep saying it.. What we are trying to do is capture some of the magic of our favourite games.

Today I'd actually have to put in zero effort or go out of my way to downgrade graphics for it to look like the original Dragon age.. It's not extremely difficult to make a game look good with modern engines, what we're essentially arguing about is scale and content which sure I can't compete but scale and content does not automatically equal good game.. Most realistic graphics does not equal good game.

Essentially RPG's are difficult to make, especially interesting ones and that's the path I want to follow.. Also with your little jab Fallen Spirit isn't the only project I've worked on over the last so many years.. I've been involved in a few and some have been released, they are far from small as well (MMO's etc.)

@GarBenjamin said it looks like they'd bit off more than they could chew, which is a fair comment.. Here's one of the main reasons, do you know what's worse than biting off more than you can chew? Doing it twice from scratch in another engine because the first one was a bust. If I really do pour my all into my own "pet" project I'm going to learn by the mistakes I've already encountered.

I don't engage in other large projects for free just to feel warm and fuzzy, I do it to learn..

 

Nothing i said was related to your actual game or you personnaly, it was general about also about many other people going AAA trying to keep almost the same scope as some big games, i seen many atempts until they realized they won't reach such content and quality without lot of money or vey big team. They abandonned making the mmo for example or made something that was finally lot more limited very different from their first annoucements and tech demos.
It happens also on big studios like Everquest Next that simply failed, they had unrealistic expectations.


Indeed there is very experienced devs like you able to pull out more stuff than many common users like me and others.
https://www.youtube.com/channel/UCB43TEVl2hmLNDb4PumQVpQ

Again i don't know if it is only a hobby for you making your game, or a tech developement fun, or if your goal is to complete a game, it looks like you don't need or want any timeline or schedules, if it must be completed it will be done when it's done lol .

Indeed, there is lot of games without top graphics but lot more deep and interesting. Even some game like FFXII remaster from PS2 that is far away from Witcher 3 graphics has better reviews than some other AAA games.

 

@ShadowK Must say at first I was thinking... what is he talking about? But oh yeah I do remember now this was in private messages.

Well yeah clearly they had some good ambitions I just think if they scaled back they probably could have really pulled off something very solid. Would have been smaller but a lot better IMO.

Here's the thing... all of these discussions here (in general not specifically you) on graphics and Skyrim / Witcher 3 type of games and so forth. I often wonder so why aren't people just making these games?

I think it comes down to either too much focus on the raw graphics quality which causes continual iteration on that one aspect alone and / or the sheer scope of work involved in making something like a Skyrim is just too much. Although we did see @bakershah do it while in college with the game Hero. Well actually I think he said there were two others helping part-time over the 3.5 years of development.

That's why I thought it was very cool when after making the little side scrolling shmup demo your next project was a month or so on a very small rpg area.

Because once a person has the actual first game done... doesn't matter if it is tiny... to have one really truly complete then it becomes a scenario like Witcher series or games leading up to Skyrim, etc. I sometimes think people (not saying you) forget just how much work has really gone into these AAA games.

It's not just the work for the final product released in the form of Skyrim or Witcher 3. Skyrim was able to build and improve upon Oblivion which in turn was able to build on Morrowind. Witcher 3 of course benefited from work done on Witcher 2 which would have built on work done in Witcher 1. I think looking at it this way reveals more of the true scope of work involved.

I think looking at making a modern AAA game (just going by the thread title here) even those large developmemt teams probably couldn't create the games they created. I know people may think that makes no sense but it makes perfect sense.

If they had started back in Morrowind or Oblivion days with the goal of making Skyrim even if they had all the needed tech (and maybe they did) I doubt they had the experience and foundation to pull it off. Same for Witcher 3. If there hadn't been a Witcher 1 and a Witcher 2 then Witcher 3 would have been much closer to W1 or W2.

That's why it boggles my mind people especially solo devs thinking of making something like Skyrim from scratch. Just jumping from nothing straight there. lol

So anyway... that is they key I believe. Make an rpg focused on a little hamlet or some other tiny area. Get it all working very solid. Probably need a lot of fetch quests and combat grinding to give it a reasonable amount of play time but that is fine. Because it establishes a foothold so to speak. Then version 2 builds on it and can expand the game not only in world size but also in unique stuff to do and rely less on fetch quests and combat grinding. And so forth.

Truly that is what I so often think when I read these threads around here is why we haven't seen dozens of tiny rpgs come out by now considering the amount of discussion about them over the past 3 and 1/2 years I have been here.

 

I will say things are very different in 2017 than when these people set out to make games such as Skyrim. Tools are very powerful and content is readily available. Well kind of anyway. There is a lot of content available but probably not exactly what a solo developer has in their mind.

And these advances... the tech and content availability help out a huge amount. But still there is all of the actual game design and implementation. What is the story? What are the important locations? Who are the important characters and what are their roles in everything? What is the main focus of the gameplay? Does it favor combat exp grind? Or questathon? Discovery? Or a balance of these things? How many fetch quests in a row are too many? How do we make the fetch quests more interesting? There are just so many things to consider when making such a game.

Anyway look at the history leading up to Skyrim...

I wouldn't mind taking a crack at making something like Arena or Daggerfall but definitely not the scope of those games. The video mentions that Daggerfall had 15,000 different towns. I think the latest games have around 500 to 600 total locations. Which of course is still far too many. I'd be looking at maybe a total world size of 2 miles square. If done right I think that area should be plenty big enough.

I am not sure why there is so much focus on size of these game worlds. If you think about your life how much do you do in an area only a few miles square? I think a lot. I do take road trips and enjoy driving around exploring in general but when it comes right down to it when I go over to town if that was in the middle of a 2 miles square boundary there would be a huge amount of things that could be done. Many merchants. Parks. And plenty of wild area as well.

Anyway I did a little experiment tonight on such a game. Obviously nothing much in it yet and no idea what I will do with it if anything. Just playing around really. Notice the awesome ultra low poly trees in the distance that are not yet textured. lol I wanted it to be ultra low poly because tht would greatly reduce the workload and I like the look but there are just so many hd assets available that I used those instead. So all of these models and so forth are made by other people.

Of course, I'd need to do some customization on things. Better AI on the enemies. Customizing the collectible scripts and so forth. And again this isn't in Unity. Right tool for the job and all that. I wouldn't attempt such a thing in Unity due to lack of sufficient time to make good progress. On the other hand this took no more than 30 minutes to do and that was while still learning.

 

Daggerfall was entirely procedural though (the old kind of not really super good)

I admit the things that bog me down when trying to put a paper skeleton of AAA rpg is all the decision about teh world lol. It's such a stress even when going super generic. I have some ready made story accumulate other the years but they only span at best 2 hours.

But I guess it's an aftereffect of burning out on my current project, I swear I'm always on the verge of finishing all the decision beat to chill on pure execution, but the last bit just drag lol.

That's kind of already amazing lol

 

No, they use planning and budgeting, derived from preproduction testing and profiling. And because things do often change, have a well planned (and where possible) automated pipeline (both capital P and lowercase p). Brute force on a massive catalog is outrageously expensive. Though on occasion outsourcing can be a solution.

 

Horizon game on PS4 has been made from scratch in some 5 years and it's a AAA game; it's not based on any previous technology or 3D engine, all started from zero.
https://www.guerrilla-games.com/read/creating-a-tools-pipeline-for-horizon-zero-dawn
Anyway, indeed this shows the amount of work of hundred people, something indies can't do to that level of quality and content.

 

The environments are tough for indies. But now with Shadowmask ligth mode I think we have a shot at getting environments that look more Triple A. Here is a shot from our games most AAAish scene

 

lol
The pillar shadow on ground looks offset, you need contact shadows.

 

Haha, saw that too when posted, now fixed

 

Those aren't mutually exclusive, but okay, let me reformulate my claim.

If you are an indie trying to punch over your weight, you will have to put a designed structure first and fit contents to it, and cut anything that don't fit. Also you won't be able to do everything from scratch, so you would need to build with the affordance you have, ie cut any content that won't mesh, essentially doing a form of kitbashing. Like in the grandia 2 example, I'm sure the basic structure was in before the story.

AAA have the army to co evolve the structure and the content, that's why you have tons of specialist to solve problems, that's the brute force. You can create things from scratch. You still have to reduce risk by making similar decision but you have a chance to test them first. The type of planning and decisions would be different in nature and practice (I mean beyond the broad phases of pre prod, prod, etc..).

The thing that separate AAA and mock AAA by indie is the relative potential ROI.

 

While true, "brute force" typically refers work by hand, not tools and planning. Certainly large / experienced studios will avoid that by creating tools and up front preparation, but that is not exclusive to large studios. Small studios benefit heavily from planning and tools, they don't usually have the resources to deal with large pivots, and if they are externally funded, or have very little or no access to additional funds, it can be the end of the project. If you have very little resources, you have to plan/tool very carefully.

 

Ah that makes a big difference. Create one or more basic town "templates" with a list of properties that support random ranges of customization. Still be impressive though because I imagine each town would have random quests as well. Doable just saying as usual the more you look at it the the more scope is revealed.

I do minimal design up front and kind of let the games design themselves. I have an overall concept / direction in mind that kind of governs everything (and certainly for an rpg would have more than usual) and then implement, playtest and refine again and again. I think trying to design everything (all details) up front can work against you.

Lol well it is basically just a bunch of terrain and stuff so far. Not much game in there at all. It has been fun to experiment with though.

I knew I'd want the game to center around a small hamlet but no idea on any details. After adding the shack and the 2 barbarians in my first work... today I thought aha! These are barbarian brothers who come down from the hill every week and rough-up the people in the hamlet taking whatever they want. First quest... put an end to that nonsense.

So today I spent a good hour or more building out more of the game world, creating the hamlet and adding a couple of other terrain touches including texturing that low poly tree I made just so I could see what it would look like in game.

There is something weird about this thing in that it kind of makes things look much better than they normally would. Like if I stuck the low poly tree in a Unity scene I don't think it would look like this at all. I mean certainly others could make it look as good and even much better. I am saying that for me I wouldn't be able to and would require me spending a mass of time (that I don't have) tweaking stuff trying to understand what I even needed to do. Of course, I wouldn't do that because as long as it shows up I am happy. lol

The fact that stuff looks good to me is just an added benefit. The sky is another example of that. I noticed there is some subtle cloud movement going on. I didn't do that. If it was up to me to do such things it wouldn't be in here. lol

The sky just does it. I've spent no time on anything except just building terrain and populating the area (both of which are super easy to do). I guess there is some kind of auto-magical lighting & other FX happening or something which kind of helps things to look better and even objects that don't fit it to not stand out nearly as much as I'd expect. Or something. I don't know.

Anyway... I did that work and neglected to update the scripts so my hamlet of women instead of being helpless and friendly are bad ass enemies. lol Ah well might work on it some more when i get in a game dev mood again.

 

@zombiegorilla
Well that's my entire point, I apologize if I expressed that badly.

There is also historical precedent in other artforms about small studio relying on tight planning, imposed structures and cheating to get higher results. In fact all artforms evolve by innovation cycle within production too. But any closed production gap will be absorbed by big studio anyway. Be a pioneer is often key in this lol.

I think we are a such a key turn around points, will be too late later to fully capitalize on it. But if it happen it will also mean production value will go up overall, and we could have a very mature production for "good enough" works of medium value relative to the new normal. I mean like TV series have reach a bar where they don't looks like scrappy child version of movies. I remember growing up and noticing there was a small period where there was almost no difference in quality between the two.

They did learn from it, they use templates now to get faster works with less people and fill an entire open world. Which we all can do. There is also some more options:

- Have an initial procedural pass that put the template down, then hand corrected them to add specificity.
- Design an additional specificity pcg algorithm with interesting metrics to generate the specificity directly in the generation.

The former is easier, it gives you something to improvise on.
The latter is harder it mean you already thought out what the composition of the experience in details enough to be a set of rules. I think the riskier one but with the bigger ROI overall. It will eventually mature and we will move to the next production cycle.
Specificity is really the problem to solve with this techniques.

That's actually historically accurate lol, when left to themselves women become fiercer warrior, the amazon are not the only occurrence, In dahomey it happen too, it happen recently in the middle east etc ... lol

 

Never heard of it before. Just watched a video on YT. That is very impressive. All around. A very interesting story and world. Cool mechanics. Actually seems innovative.

Although this may not be a sequel I think surely the team had to be experienced in making AAA FPS / RPG games? And how many were on that team?

 

They are the killzone team, aka top tier dev on the visual aspects.

 

Okay I found their website. The company started in 2000 and now has over 200 professionals. No idea how many actually worked on HZD though. I'd guess they relied a lot on the experience gained and foundations built from the Killzone games.

 

Yes and no, they had zero knowledge in open world design and their story tend to be rated average, their linear fps engine don't translate into open world, they had to start from scratch. They were essentially the fps tech demo team of sony. They weren't slack either.

1. Starting a new big budget franchise
2. in the middle of the console generation, (publisher avoid this, it usually don't work)
3. in a new genre they don't master,
4. with new technical requirement they hadn't tried before,
5. with a woman protagonist,
6. with an original story not relying on tried and true tropes (debatable),
7. with the ridiculous idea of dino robot taken seriously,
8. in a visibly multicultural matriarchal world,
9. in the context of video games' culture war during the last 6 years,
was total suicide.

They not only to manage to pull it off, but sell very very well to establish a new franchise, be the best open world critically and set the new standard for 3 straight whole days, until the new zelda was released blew it out in the genre and set a new higher critical standard (see the comparison video between the two to know why).

And they used procedural generation. That thing tanked mass effect Andromeda.

They still took 6 years, a lot of them in pre prod by a small team. It's the lightning in a bottle that happen once in a while. It was a huge gamble and the team was super nervous lol, but their publisher was 200% behind them and gave them freedom. It's also the total antithesis of the advice I have been given, but we tried to be optimist and realist at the same time lol.

 

They were experienced with linear fps game and multiplayer, they had no experience about open world.
Indeed only experienced people was working on the engine or game content, but they made a new engine and new tools from scratch to make the open world game.

 

This touches on an interesting idea: how different are different genres?

Would a 3D artist really be doing anything differently? Would an animator really be doing anything differently? Would a story writer really be doing anything differently? Would a GUI designer really be doing anything differently? The main difference I see is in the fundamental gameplay structure, but it seems like there's a lot of overlap otherwise.

 

5 years with thousand team people working on game content,engine and tools i don't think a lot overlapped lol
Games like Far Cry or AC indeed evolve on graphics and tools but they keep their base structure made in previous episodes, it's not the case for Horizon.

 

Yup we seem to forget a lot of these four year AAA projects have had another four or more years spent upon them via previous projects. With a lot of the artwork etc. they will save the high poly models and re-topo / re-texture as technology gets better, a lot of the heavy work will have been done and lot of it will be modifications / improvements.

I was watching a Unite video on a project where they needed seventy characters, they said it takes a month / to a month and a half per char, personally I think that's a bit much you'd be looking between two / three weeks for a rigged / animated character and you wouldn't start from scratch every time.. It's a matter of binding clothing meshes / using sockets for weps and using a bunch of morph targets for variation..

Can't remember off the top of my head but I think Skyrim only used one skeleton, the rest was just a limited set of morphs. So spend three weeks to a month on the base char and then the rest should be relatively straight forward..

Here's the thing though, it's still a mountain to climb.. You can easily spend 3 - 4 months getting the char / enemy artwork all together and getting the combat system tested out. It's simple in terms of "how to" but it's just one more thing about these types of games that are ridiculously time consuming.

@neoshaman said it's impressive for you to put that together in 30 mins, since this ain't our first rodeo it's kind of expected really (although I do appreciate you have the skill to move that quickly).. In UE that's roughly how long it takes me to put it all together once the artwork has been done.. A lot of it's to do with the game framework / BP's, as it would take me a week in Unity to as opposed to 10 minutes with UE.

What you really should do is a very small concept at the beginning, once the core is done (from UI, sound, chars, combat, quests, dialogue, interaction system, databases, addon's etc.) it's just a matter of more.. It's a 20 minute job with world machine and a procedural placement algorithm to extend the world (as in terrain) exponentially if you're using the right tools.

This is where I believe many get confused as to why these projects are difficult.. From a small prototype perspective / core setup RPG's are dead simple, then when you start scaling the difficulty just warps out of control quickly.. There's a thousand reasons which I could probably write a book on at this point.

 

Can you describe how the work of a 3D modeller would be different? Can you describe how the work of an animator would be different? Can you describe how the work of a story writer, or a GUI designer, or a programmer, would be different? The tools may be different, but that's like saying programming in C# and C++ have "no overlap."

You're still doing the same things by and large. The main differences are at the high level design and with building gameplay elements. Otherwise, however, these people literally are doing the exact same things, and thus their previous experience really did help them.

 

Indeed these jobs have not changed.
The main change is the overall 3D engine, tools and how to manage new dialog system, missions, new combat system, new open world rendering and all that stuff.
They have never done an open world game, no TPS character gameplay, it's a totally different approach in terms of 3D engine, procedural exterior generation, streaming, optimisation , visibility etc ...

Even character TPS is new and made from scratch
https://www.guerrilla-games.com/rea...hanics-in-the-vast-world-of-horizon-zero-dawn

It's like you made many linear fps games, and next project they ask you to make a tomb rider game with open world voxel terrain system, water dynamics, flying cars or planets travel and space system.

 

Yeah I just can't get into that kind of thing. 4 to 6 weeks on one character. And translated to very part-time folks like myself that would be what maybe 4 to 6 months per character? Who in the hell has time for that kind of thing? Seriously. I guess if you are a teenager living at home and off from school all summer long sure knock themselves out spend their summer break in front of the computer making 2 to 4 characters.

To me honestly it just seems a bit insane. That's just my opinion. I mean if it was working as your full-time job this is what you are being paid to do then absolutely. While these characters are being created the other pieces of the game are also being built by others. But for a solo dev no way I can see that. I know people do this kind of thing. Just saying I don't understand it. I would focus on how to reduce the scope and try to maximize the bang for the buck out of less work. Going solid-shaded ultra low poly would help a lot and then making different assets programmatically by scaling and coloring these would help too.

Absolutely agree with you man. And this is why I look for tools that work in a way I want to work and that allow me to very quickly knock things together. The basic prototyping stuff. You hit the nail on the head. The entire reason I focus so much on that stuff and talk about development speed is because... well honestly I know this will tick some people off around here but you know as well as I do man probably 80% of the stuff we see here is basically just a prototype. Sure they might look awesome and all but when you really get right down to it beyond the graphics there is hardly anything actually done.

You're right. What I threw together this past weekend to me is nothing but just a little experiment. Hell there is literally nothing to it. It's like 1/10 or 1/20th of the work that I put into my 3D Space Invaders demo game. That game had actual real gameplay going on.

 

Considering what AAA was during the PSX and PS2 era, yeah I would assume indie could reach AAA status with a small team. But with the current gen as of this generation, the distance greatens from game to game. Most AAA are open world, big universes, motion capture CGI, etc. Indie may not breach AAA in that standard, but I believe indie are the go to future AAA titles when it comes new ideas, new concepts, new art directions, new gameplay experiences/ways to enjoy a game. Since AAA titles tend to fall in the era of sequels, trilogies and so on. Sometimes a story doesn't need a sequel, but let the players imagine what the world is after a game has reach its ending. We don't see much of the companies starting new franchises until a trilogy hits or a new gen console, which I feel give indies the chance to make a name for themselves/ourselves.

If you think about it. You'd really would only know the companies for 2-3 big franchises. Which if you look at their catalogue, they boast way more titles that you probably wouldn't even know was them.

So to say that. You can be indie, and have an AAA IP. But not be a big team yourself. All comes down to what is being sold and what is not/popular IPs.


But I digress...I'm just spewing newbie knowledge...

 

There is some overlap, as much as going from basic to C, they choose something diametrically opposite in term of requirements. Suddenly you aren't modeling for very control case, where scenes are done in a very specific order with a very specific timing, you are modeling for more interactive characters at the same time appearing on screen, all must have good lipsync, which they don't as some side characters are noticeably bad.

A TPS have way greater traversal options and full body animation with more interaction on screen that must be believable. And the load of open world is very different thatn a straight shooter in terms of programming the streaming aspects. Writing is different too, you aren't writing for fixed events, but for parallel narrative that can be done in many order, share the space, and also must direct the player to the right place.

They also established a lot of first, like using procedural generation, having a new cloud rendering system, a scalable performance capture system. Basically scope management is making everything news. It's like going from basic programming and then jumping to C and having to manage pointer and memory allocations, that's risky. You are still programming, surely, but the added complexity can destroy you, and it did for many studio (see mass effect andromeda).

They did knew how to pick their battle too, there is a video comparing them to zelda, and they have less interactive elements on a more powerful hardware, some which are staple to the genre (their physics is kind of super simplified, more so than what people do with unity at a basic level). But let's be frank most player didn't care, and what they did was super polished to a T. I'm more confident I can clone Horizon, without the polish of scope, than I can do anything close to zelda lol, zelda scare me at the programming level, just the basic physics scare me, it's so solid I just don't know.

 

It depends how the company works or what game you are working on, but most of the time, you don't start from scratch , they have characters library, or character generator, the main modeling you'll do will be on clothes and stuff and character tweaking.For animation you got many available ready to use motion capture motion you can modify if needed.
There is also kitbash for characters for Zbrush, you can pick up pre made sub models and drop them on your character to modify any part of the character or pick up kitbash for clothes and stuff.
Someone with no character experience could get unique characters in some hour (without retopology and rigging). From an indie side if you don't have made or bought your own character library, there is MakeHuman tool for example to create your base character.

I can put 5 days of work to complete a character character usable in game, and some more time can be spend later with small tweaking iterations modifying some parts, some textures or tweaking some animation.
Indeed if you go very realistic and very detailled with specific textures you can spend a lot more what matters is to get some first iteration in some days usable in game.

Companies or small studios also choose tools like SpeedTree before for example and make their own tools to speed up their workflow
www.youtube.com/watch?v=QBAM27YbKZg
Or use procedural level generation ( before tweaking by hand to make specific places )
www.guerrilla-games.com/read/gpu-based-procedural-placement-in-horizon-zero-dawn
We got already such generation tools, for example Unity has plugins like Gaia, Gena, Terrain Composer 2, CTS or Dungeon generators, while some other indies will write their own procedural generation tools.

You can get quickly enought some characters and level world with clever tools and workflow or buying some quality model and texture packs, but it can look very generic and borring lol
Next is to make your game interesting :
- game world background and enviroments
- main characters , their story , their specific look, making them appealing and interesting
- the game story with many plots
- the level design to make levels a bit attractive and get some unique outstanding places
- make the gameplay appealing, something new and a little different from other games.

Going with non realistic art direction can decrease a lot the amount of work. For example Zelda Breath of Fire game, it's cartoony, but it has some amazing level art and one of the most interactive and dynamic gameplay ever made.

 

We indies can't compete on enviroments, it's just impossible, what we can do is compete on thinking outside of the box. For example our VR game has very hight level of realism when it comes to weapon interaction, while the bigger studios are more arcady and let magazines automatically slip into the mag well etc.

Here is our new first person hands WIP video, a bit proud, almost triple A on the trigger finger IK sorry for the tracking issues, htc vive does not like the windows in my home office

 

I agree completely. Actually not even sure why I am in this thread. Part of me says yeah we (even very part-time devs) can do it and I think this is what trips people up and keeps this discussion going.

It can be done but why? It would take so long it would be insane to do so. And actually releasing a state-of-the-art AAA title no way.... that is truly impossible. There is no way a person can do it efficient enough fast enough to release it and have it be competitive state of the art.

What could be done is start now and in 10 to 15 years or more have a Witcher 3 clone. But by that time W3 will no longer be modern AAA and will be "old news". A person would have to focus on what is not even seen yet... targeting maybe 5 to 10 years in the future and using a canned engine.... ANY canned engine... how will you do that when the engine itself will not have that tech til at least 5 to 10 years down the road?

This is why I strongly believe even if a person's goal is to make a AAA style game they focus on ultra low poly, cubes even through the bulk of development. Get all of the game done everything done except for the final presentation aspects. Then at that time years down the road you can replace all of the content with whatever is cutting edge at that time. For example it would have been replacing all your ultra low poly / primitives with HD pbr stuff.

The game itself must come first. Otherwise you will spend all of your time redoing the "final" graphics as tech progresses and filters down.

Of course I also see it like I would much prefer using primitives / ultra low poly for the final presentation and focusing on making a very interesting game itself. One that is so good people would play it if it was black and white cube-based.

 

I don't think there's anything wrong with chasing the AAA thing if one truly wants to. Now, if one is under some kind of impression that that's the only right way to make a game or something, then obviously there's a problem, but there's nothing inherently wrong with tryign to go for AAA quality, if one is willing to put in the time. People can do what they want with their time.

Something like this does seem like the smart move, but for a small team that has a couple 3D artists - what should they do while the rest of the game is being made? Most applicable for single-dev indies, or indies without a dedicated 3D artist.

 

I agree with your point.

The gun manipulation is excellent. How big is your team? My concern with this level of detail is that it adds so much extra time for each content element and places restrictions. Imagine how much work a rocket launcher would be!

That said, as long as you have either adequate resources or limited scope for weapon types, that's some pretty sick functionality. If it feels good in play, that's a big step forward.

Players will love futzing with weapons in VR if the hand interface is fluid.

 

Absolutely! Of course people can do whatever they want. lol If someone gets enjoyment solely from working on graphics or attempting to match AAA then yea that is what they should do. I think it is important they recognize that is what they are doing. Meaning recognize they are not so much trying to make a great game as they are trying to make great graphics or trying to make a AAA-like game.

There is nothing inheritantly wrong with it but it may not be the best way to go if someone wants to build an Indie game dev business either. Or maybe it is. I only am sharing my thoughts on it. And those aren't the end all be all. ha ha.

 

I somehow doubt that going from low poly to full HD with facial animation will be an easy upgrade. I think drawing a box around what AAA means to your team and having at least one full tech and quality test scene (that may be updated if AAA changes) as early in the development as possible is a good idea. Giving everyone as much time to iterate as the coders seems like a must do for AAA imho.

 

It could be. I guess i am thinking the artists can certainly spend their time locking down exactly how the game should look. Including lot of experiments to keep up to date as time passes and tech evolves. But hold off on actually going into full scale production until the rest of the game is nearing completion.

And I say that only in the interest of reducing a large amount of rework throwing out tons of content that a lot of time was poured into to make previously. Instead they would have streamlined workflows and stay up to date with the latest and greatest focusing on some small set of test pieces until the time to strike comes.

 

I want to nuance that, with PCG it's perfectly possible to do organic and man made environment that looks good with few effort.

HOWEVER, the more specific spot won't scale well, Manufacture goods like furniture however need a bit of investment that might not be worth the ROI on the scale on a small project, while you can do them using pcg, they will take as much time as hand made (assuming you have to generate both the algo and the tools) so the ROI won't be great unless you are using to generate a whole planet full of these.

For example: a single house like Edith Finch, which all elements are personal and tell a specific story, and everything has a purpose, would be kind of hard, you would need a general pcg model that is able to generate character, cultures and intersect the two to create a place that is itself embedded in a complex environment.

Now I say that regarding current understanding of PCG in dev cultures, most people treat pcg just like an automatic placement tools, not as a language generator, you would need investment in story generation, interior generation, intentional level design generation, all intersecting together. That require a level of expertise people don't invest in so far (outside people like Emily short or the creator of Tracery).

Character on the other hand are almost a solved problem, but specific addition, character design and acting aren't. But I recommend watching the second part of this video at 27:00

This is research and looks complex, but it will be simplified and reduce to a few parameter once you it become mature. Just like using bone in a character don't mean you model exact anatomy, you use a functional simplified set up that does good enough jobs.

 

Indeed you won't make Witcher 3.

Why not smaller level maps like Inquisition , with less assets, smaller city, less villages , less places, using characters kits, only few unique NPC, few side quests, and less content ?
Some small scope Witcher like game not open world, could be done in Unity or other 3D engine using many plugin tools, plugins generators or quick placement tools.
This game could use non realistic style to decrease the amount of work ( Overwatch, Zelda Breath of the wild ).
And it would still have the AAA touch.

The challenge is again about your ability to make interesting level design and places ? unique appealing characters personnality and design, appealing story and some new gameplay very involving ? This is long reasearch, work and lot of iterations.

 

Thanks for your kind words! We are only two devs! We have no artists employed but outsource all artwork. A new type of weapon do take time. But some weapons that you think are a big difference turn out to be pretty straight forward, for example our first shotgun was alot easier than I first imagine. We reused the magazine code we had, but extracted the functionally that was the logical representation of a magazine and slapped that on the shotgun to simulate the internal tube. In the end if we exclude the rewrite from one type of mags (external) to external/internal agnostics mag code it was just a few lines of code

Rocket launcher? Piece of cake! (If we ignore world destruction). I fear the first light machine gun with physx enabled belt ammo But you are right, our game is progressing a bit too slow, also we both have day jobs. Its hard in VR, its such a small market so you cant take the plunge and work full time.

Yeah, we as seasoned devs me and my businesspartner can do great stuff when it comes to the code/domain (basically the core mechanics), but environments are hard. We make due with asset store environments. But we have outsourced to add custom content to existing asset store environments which as proven to work well for us, for example we added a building with internals to a urban asset





Modifying and adding to existing asset store stuff is a really good way for indies to create interesting environments, still not triple A though