Is unity a good engine for an open world survival game? Or would I be beter off with UE4?

I had this awesome idea in my head for an open world survival game, but after doing some research I quickly saw that I am absolutely not ready to partake such a large project. So I want to start smaller with other games that I can use to familiarize myself with the engine and developing. I will make a few games that I can incorporate into an open world survival but before I do that, I want to make sure the engine I use will be able to do it. So years from now, when I feel comfortable enough to start it and I have a group of people I met along the way to help out, I don't start the project and realize that this engine can't handle a game like this.

So again, just want to know if unity is a good option for this. I've played around with it quite a bit so I'm not too bad with it but still learning a lot. Also, I would like to reiterate that I understand it is a large project and people with much more experience than me wouldn't touch it. I am not going to be starting this project for years to come. Just want to know if the engine can handle it. I've seen how ruthless you guys can be to new people with ideas like this lol.

 

This question came up a lot already and there are dozens of threads about it.

Unity or Unreal?
https://forum.unity.com/threads/open-world-and-unity-or-unreal.497623/
https://forum.unity.com/search/8948929/?q=open+world&t=post&o=date&c[title_only]=1

...and then there is the question:

Unreal or CryEngine?
https://forums.unrealengine.com/com...between-cryengine-or-ue4-for-open-world-games

 

Rust, 7 days to die, The Forest, etc. are all using Unity I think. It mainly depends more of the teams skills and experience to be capable of finishing such a huge game.

All engines have their pros and cons and workflows differ a bit. You could use both for a while and see which one you like more.

 

The long dark, and Subnautica as well, think it has more to do with what the dev puts in then it does the engine.

 

We're mostly ruthless towards the people who try to jump into an absolutely absurd project before they've taken their very first steps as a game developer. It'd basically be like an infant wanting to run a marathon. You don't seem to have that problem though.

Yes. With enough effort there really isn't anything Unity can't do. That applies to just about every major game engine on the market too. You will have to work around certain limitations like floating point numbers having a limit to how high they can go before the physics system starts to fall apart but that's a problem with other engines too.

 

its coming from a well meaning place, just programmers are often very explicit and blunt in their communications.

 

I think the game engine only matters when you know enough that you don't have to ask what the difference is.

Kind of like paying for an expensive wine when you you'd never be able to tell the difference unless you really became an expert in wine tasting.

 

Unity can do it. Unreal can do it. I'd weigh the pros and cons of each engine and decide what works best for you.

 

Honestly Unity can but it is way behind in the out of the box Large Terrain Engine stakes compared to Unreal's push for large open worlds.

Also there is Cryengine and Lumberyard (Amazons version of the same) used ot build and run FarCry.

So it depends on how large an open world your wanting to build, if you want to go large check out Unreal if you plan on a modest island/region <10 km* then Unity should be good.

However you need to check out the in built tools and asset stores and workflow for all of the engines then compare and contrast the results.

As you will need to fill the landscape with stuff that makes it interesting enough to be fun to survive in.

Unity:
Advantages: Quick and Easy to Learn.
Disadvantages: Poor Out of the Box Terrain System Toolkit.

Note: Unity have recently added the Pro Builder tools so hard/flat/building type elements should be a lot easier to work with in Unity.

* PUBG maps are < 10km in size and plenty large.

 

You're smart.

Yes, it is. Rust and Subnautica are prime success examples. To do this you need to learn how it is abstractly done, then apply that in your engine of choice. In the end it matters very little which engine you use if your motivation to succeed is high enough.

 

Learn some developer skills, engine is not super important, but you could start learning more than one.

Then at the time when you feel somewhat ready to start your big project, then you'll be able to make an informed choice.

Otherwise, I don't see what point there is in asking anyone which engine is going to be the better one years from now.

 

One of my earlier Unity games 2011...

After I had a play with the systems in Unity.

Take advantage of the power of game engines like Unity and start learning on your dream projects.

Just expect the game development to be a learning experience.

Half the battle is working on something you want to work on.

Or try maintaining inspiration and drive working on Pong or a Pac Man clone!

 

The engine can do it, but it takes experience with optimising the heck out of things. But that's true for any engine. I've seen UE with previous generation console-style areas struggle, so bad asset management, optimisation, culling etc. can destroy any game.

The major engines are very capable now, so the only determining factors should be:
1.Can you enjoy using the built-in tools?
2.Is the community decent enough?
3.Are there more YouTube tutorials than you think you can get through in a month?

All three definitely apply to UE and Unity. Both also support many popular 3rd-party tools/asset libraries, like the Substance family and SpeedTree, which are advanced subjects which could boost your productivity.

 

Open world 3D game can be made with Unity. However, you have a bigger problem. You say that you have "this awesome idea in your head". This means that you don't even have documentation. Without documentation you can't even choose an engine intelligently. Focus on fleshing out your idea on paper. Don't write any code. Just write documentation like a manager. Do also financial analysis in MS Excel. You will quickly realize that the game you want to make requires a solid 20+ man team to execute properly.

 

Unity, Unreal, CryEngine; can all do it. Can you do it though?

 

Cry Engine, the engine that makes you cry. For real. I'm not nearly 1337 enough to handle that beast.

 

IMO, It is more of a matter of a your budget and manpower and less of a matter of game engine.

 

Don't we all?

 

Just go with Minecraft style graphics it worked for Notch.

​

 

Pointing towards an existing title that was very successful with those visuals sounds great in theory until you realize it's literally the only title that was very successful with them. There have been numerous titles since then with those graphics and just about every single one has never left obscurity.

 

Every time I see a game with such graphics, I immediately skip it.

 

Same.

 

Goes for any low poly stuff for me.

 

I think I'm the only one, who deliberately haven't played Minecraft, ever. I have played only one lowpoly (For the King), but I don't like it at all. Apparently it's not for me.

 

Why though?

 

Aren't all indie games low poly? xd

 

Yeah, tt seems 99% are. We are not though.

 

i have no issues with low poly styles. Sometimes this fits a game very well and can be look gorgeous, if unique.

I do have a issue with that particular voxel-style tho. It screams "lazy" and often "asset flip"

 

Eh, no. Low poly can be very cool and stylish. So, if somebody has a cool low poly style, that's very interesting.

Minecrafty graphics, however, are bland. The image posted is soulless. So that one is automatic "skip".

 

Well I hate it, I dont play any game that does not have realistic graphics. That's why I wrote for me, since thats my opinion

edit: Plus every low poly game looks the same. Its no effort done in the low poly style (Most often)

 

I dislike realistic graphics. They look boring to me.

Basically, every game that goes after realism pretty much looks the same. There are some exceptions, but it is in general this is a great way to make a game look bland.

 

Reality is bland.
Though, we have seen some very nice realistic games over the years. Metro 2033 is a favourite example. I could wander the subway homestations for hours, just look at the detail. And that was pre PBR, now we see even nicer looking games

edit: I rather play ugly realistic graphics than pretty none realistic even. Well to a certain degree, take PUBG for example, not very good looking realistic graphics. But I like it more than Fortnite

 

Ultra-realistic lighting combined with slightly stylized/undetailed textures and cinematic post, produces a delicious result.

It's true that reality is not all that fun to look at. Games are meant to be only the best analogues of it.

 

Another great example of beautiful realistic graphics is Alien Isolation, what a game

 

Thats a good example of what I mean. They used a relatively highpoly workflow (e.g. bevels with weighted normals) and tiling textures. Meaning that textures had to be clean with relatively low, uniform detail. Produces the best kind of result imo.

 

By now the only thing I remember about this game is looking at the closed locker door from the inside. The steam from vents was okay as well, but I only rememebr it because I needed to replicate it and create similar effect.

So, all of the work was pretty much wasted. In my memory it doesn't look any different from, say Doom 3. (Lots of pipes and smoke and stuff. In space)

Ever played ABZU? It is one time game, but it si a good example of what can be done with "non realistic" approach:

Spoiler

The visuals are stylyzed/cartoonish. Looks quite amazing in motion.

The reason why I'm not a fan of realism because it is wasting 98% of your time making 1% of differences.

Visual impact of a scene, in my opinion, is largerly determined by large scale structures, so all the time spent drawing texture grunge essentially goes down the toilet. It makes someone go "wow" maybe once, then it is not noticed.

it is more interesting when somebody tries to remove all detilas that are unnecessary away.

I think the only recent "realistic" game I played recently was conan exiles. Here's character level of detail:

Spoiler

The problem? Well. You never see them this close, so in essence this is a wasted effort.

Also, the game is broken as hell.

Spoiler

Air walking dragon

In the end, visual impact of stylized environment is going to be bigger and more memorable.

 

Only thing I will remember from those games are not buying them

 

I think Halo art is one of the most impressive. Sometimes it's hard to tell whether it's more realistic or more stylized in nature. But it has great lighting and a lot of visual impact.

Spoiler: Halo

 

I still play old games now and then. Games like Quake 1 and HL. Back then those games had "realistic" graphics, not anymore obviously. They "play" very well. I find that what makes a good FPS game is more about feeling and control, than about graphics anyway.

For example, you mention PUBG. That feels janky as hell (probably a combination of controls and netcode), and I basically consider it to be unplayable as an FPS.

You can have realistic graphics set in fantasy locations and scenarios. Then it doesn't look bland, it just looks awe inspiring, because it looks "realistic" but it's not the bland reality we know. That is my favourite aesthetic, very immersive and interesting.

 

I also play those, and they are still realistic. Realistic (for the time) has nothing todo with the quality. No game is realistic since you can tell its fake

 

Thank you all for your comments/suggestions.i have already started working on a game since posting this so I can get familiar with unity.

 

Far Cry 2 and above don't run in latest CryEngine version, Ubisoft uses Dunia engine a heavy modified CryEngine version they improved though new Far Cry episodes.
http://www.moddb.com/engines/dunia

 

Which Far Cry would you say has terrain/foliage systems on par with Unity's out of the box Terrain/Foliage, 1-5?

 

Unity out of the box terrain is a very old system with very basic shaders and very basic vegetation tool.
Vegetation Studio, Speedtree , Gaia, RTP3 or Megasplat, CTS are not out of the box.
And Far Cry games using Dunia have many specific features towards open world creation and streaming, you can't compare a multi purpose 3D engine and a specialized one for a specific game genre.

 

My map is about 10km x 10km, which should be plenty open. For frame of reference, the thing in the box is a giant castle, as seen in the 2nd pic along with an actual player. I would say its a pretty good size. (That's not what she said? Burn)

 

I would use whatever was most comfortable and easier to work with as an individual. Building up worlds with terrains is an artwork thing anyway so irrelevant at the early stages. You need to develop your own code structure to define the game and controller logic rather than just making pretty scenes. That is what matters most. The game logic can easily be ported to other languages or engines etc. later if need be when you have a better understanding of each engine's pros and cons.
Newbies always seem to be mainly concerned with starting on the wrong foot with game engine choice rather than the immense amount of learning that it will take regarding game development in general on any game engine.
Unreal Engine is probably a better choice for handling large worlds but a more sophisticated and higher learning curve to master. Unity is easier to understand for a beginner. I currently have a detailed world in Unity about 50* klm that runs well on a mobile. It could be larger with compromises. All game engines have their limits, strengths and weaknesses much like people do.

 

The advantage of UE is terrain tiles editing is supported out of the box as open world manager and streaming.
Unity can already make very big terrains, but a very big terrain doesn't always mean a very good game.

 

You guys arn't really helping the OP. There's alot of "whatever you want" and "its all on you" and condescending remarks. Just saying that any engine will work is not a very satisfying, and leaves them with making the perplexing decision.

I was researching Unity vs Unreal for survival games as well. I found this thread, and wanted to add better content. So I will add my own.

Based on my research, I am starting to lean toward Unreal. It has a reputation for being used in professional grade games for PC and consoles. Unity is mostly used for mobile games. Not exactly sure why but there must be a good reason. If you want a highly realistic game, Unreal is the best option. I read that Unreal is also better for multiplayer games (if you want your survival game online). That on top of some of the nice features like node based programming.

Unity and Unreal offers some packages in their stores to get started with a multiplayer survival game. Here are the ones I found as of 9/23/18

Unreal packages
https://www.unrealengine.com/marketplace/survival-game-kit
https://www.unrealengine.com/marketplace/survival-game-template
Usurvival
https://assetstore.unity.com/packages/templates/systems/usurvival-95015

If you can learn both engines you can start your project based on the best available templates from the stores. Personally I think the survival game kit from Unreal is the best package as it seems to have the most features like clothing, building construction, and furnaces.

With enough effort and commitment, you can take any game engine and make any kind of game if you modify it enough. In fact, you could argue that you would be just as well off opening up openGl and start typing c++. But that's not the fastest way to get games finished, which is why we use prebuilt game engines. We want the engine that gets us finished the fastest and easiest way, especially if we don't have large teams. so his question is valid.

 

Once upon a time mobile games were a gold mine and the only affordable mobile capable game engine at the time was Unity. Unreal was not affordable, not able to create mobile games, and not easy to use by any stretch of the imagination until UE4 came out which by that point Unity was already very successful.

It's very easy to develop a reputation for professional grade games when that's literally your only audience for more than a decade. Unreal started life as a game engine, costing somewhere in the neighborhood of a million dollars, for AAA studios and it wasn't until recently that they altered their model and only because of Unity.

 

That reason is that Unreal used to cost thousands to license, so of course it has good reputation amongst gamers if nobody else can use it. Unity, while it is used a lot for mobile, also dominates indie PC games. Just take a look at GameJolt, or Steam's indie genre page.

Highly realistic game in Unity (Escape from Tarkov)
Highly realistic graphics in Unity (Book of the Dead)

Multiplayer games I can't argue. Testing multiplayer in Unity is painful as there's no built in "launch 5 clients" button like Unreal does. There's a reason why most indie multiplayer FPS titles in 2018 use Unreal, like World War 3, Insurgency: Sandstorm, and Squad. You're right about this.

Node based programming depends. Some hate it, some love it.

Survival Kit Pro
Survival Props

And here's two free ones:
Survival Kit Lite
Low Poly Survival Essentials

Edit: Also, Rust, the most popular indie survival game on Steam, is made in Unity. The Forest too. Come to think about it, are there any UE4 survival games at all?