Unity

What does everyone here think of it?

It's nice, I like it.

it's a game engine, like UE or whatever else.
the quality of the games is dependent on the quality of the devs, not the engine. only retards will tell you the opposite.

I work in this company so my opinion could be biased but I think it is good and definitely better than people generally think it is

I thought it was all phone trash until I saw that escape from tarkov game

>I thought it was all phone trash

Yeah unfortunately most people think like this

>escape from tarkov
Although being wary is not unreasonable, there IS a lot of trash on unity.

You guys actually support a lot of cutting edge stuff, its a real shame most devs who use it cant bother

because unity makes poorly optimized games no matter how well you code

Republique lags on everything

What types of games would you NOT make in Unity?

Yeah there is A LOT of trash and that is why we have such a bad reputation

There is so much of that trash because it is very easy to make decently looking trash compared to engines that are made for making that trash

Actually good games usually even hide the fact they are made with Unity

Definitely not 1st person shooters

I want to make a game in Unity but the tutorials on you guises sites are outdated. Where can I find decent documentation

Would argue but it's obvious you don't know shit so I won't bother

Seems kinda pointless when UE is a thing.

We have great documentation both on the engine and scripting

As for actual tutorials yeah they are a bit outdated

I could suggest just going on YouTube there is literally a ton of tutorials for all levels

Quill18 comes to mind as one of my favorites

Not him but I don't think I've played a single Unity game that ran well. Even by bigger more seasoned devs.

Thank you

Also not him but have you ever played Wasteland 2? or Kerbal Space Program?

How's the linux version coming along?

There are a bunch of videos/articles on pros and cons of the two

>C#
>.NET
never?

No I didn't
Are these an example of games with bad performance?

>MonoDevelop

Bloated shitty engine whose only pro is easy to use and asset market. It will go the way of Torque in near future.

Yes, close to AAA games too

Don't have to deal with that very much but it is definitely moving forward and has improved a lot compared to what we had last year for example

You mean GODOT

At where I'm looking from UE just seems to have way more pros.

I guess Unity has more documentation and no royalty license, though UE I would argue is kinder to smaller devs that are just staring out since no pro bullshit and less of a focus on expensive third party plugins.

The way I see it is that they are dominating different markets and while Unity is trying to expand to UEs domain UE is just improving

I honestly don't see why any indie developer with a smaller project in mind go with UE

Though I am hardly knowledgeable enough to make such assessments
Just like any of you here

Also in my experience Dreamfall Chapters, Pillars of Eternity, Verdun, Rust, Tabletop Simulator, Umbrella Corps and Rollercoaster Tycoon World are some of the games have all run like absolute shit for me. Despite many of them looking like they came out around 2007.

Well can't comment on those games but if you looked for it I am sure you would find better looking games that are better optimised than those two and run great

Have you tried rust recently? I have heard it has improved a lot
As for others I have heard bad things about some of them too

I am not saying that there is no engine fault in the performance department

As a smaller developer why wouldn't I go with UE?

The times I've tried using Unity I've just ran into features I wanted that was locked behind a paywall, not to mention just how much Blueprints help out. Hell if you have some patience you can make a reasonably effective game without ever touching a single piece of code.

Can someone list out the pros and cons of Unity and UE?

The wordpress of game engines. Not totally bad, but a couple of design decisions towards shallow popularity I wouldn't have made.
Said to be buggy and version incompatible, though.

At some point, Godot might take big parts of its market share since it made a lot of the same decisions while being free and open source.

I remember when the game dev threads were all talking about how UE4 was going to kill Unity and how all Indie devs are switching, it lasted less than a month before everybody went back to posting Unity projects.

Their 3d implementation is still fucking terrible, I'm waiting for that to evolve

Yeah, until 3.x all bets are off.

This

Almost all performance issues with Unity are caused by developers using faulty shaders and sometimes materials, often brought and only lightly modified from the marketplace.

Stick to what you know best eh? With enough work you can make either engine do what the other can either way, what matters is if the starting point is comfortable for you.
Same with any 3D editing package, no one gives a shit whether you model in Max or Maya, just pick the one you feel more connected to.

Then they need to better implement ways for those shaders to not be faulty since seemingly every single devs uses them, big or small.

Once again my opinion is very biased due to my own experience, my work, my lack of experience with UE

I am not a game developer myself and when I started playing around with making a game I tried them both and Unity was my engine of choice after trying both of them

I would disagree with your opinion that UE is easier to work/start with. I say that by looking at preference of developers during game jams and what not

Maybe. All I know I found it way easier to get my ideas through while working with UE. Though I admit I'm a beginner.

Just like with any software I think preference plays a huge part (most anons here don't think like that tho) and I am most definitely not saying that either of the engines is better than the other I myself am definitely lacking in knowledge to make such judgments

>tfw no dark theme

feels batman

Unity user here
Yeah this is a stupid decision in my opinion too the white theme is pure cancer

You can't argue that the garbage collector is not a significant hit on otherwise optimised games

They could have went with a grey because god that white is awful. I wouldn't be surprised if some users were scared off of Unity

It has made great strides in bringing games to Linux, so I can't really complain.

Asset flip city.

Basically this. I've seen great games made using GameMaker and terrible ones made using the Crysis Engine. Better features doesn't mean better game.

Proprietary trash.

My main experience with it comes from playing Kerbal Space Program. Ebin gayme, don't get me wrong, but performance is utter shit because it's ducktaped together by incompetent mexican interns and the gayme's requirement aren't exactly light by design either.

But who knows maybe engine is worth a damn in good hands. I'm impressed by it's Linux support. Every Unity game seems to JustWerk™ under Linux.

install godot

Any game that is not "graphics first":

AI focused games (there exists literally 0 examples of this today)
Networking heavy games (like racing sims or MMOs)

Unity makes it very hard to use the GPU for anything other than graphics (even their physics engine is CPU bounded), so stuff like GPU accelerated networking or AI is harder than it should be.

Also because you are using a garbage collected runtime with .NET games that require real time allocation and deallocation, like open world games are also not suited for Unity.

>decent engine
>no source code access
>game devs using it are lazy as fuck
>game devs using it refuse to update to the latest version
>game devs using it refuse to support Linux
>game devs using it ignoring almost game breaking bugs like accidentally loading environments 4 times over
I'M TALKING TO YOU UNKNOWN WORLDS.

>no source code access
Your just too poor to afford access.

>GPU accelerated networking
Nigga what?

>not having a Ethernet to displayport adapter

There are FOSS alternatives. Use them instead.

In case you're not trolling. Nested prefabs when?

It's alright and not too expensive.
Their C# version is aging and the defaults are often mediocre, could also use some more stability and support. I could go more in details about what's nice and what isn't since i'm using it professionally but i won't tell anything different than what's already known.

>updating the engine
>ever
You update when the version is stable not before and right now we don't have one.
>support the literal 1%
Windows tier hardware diversity and android tier fragmentation with none of the market share, can't blame them for not bothering.

You get source code access if you pay extra. Which is okay.

Retards like you maybe.
Unity performance is horrible. Pick a non retarded game engine if you want anyone to play your stupid shit.

Blizzard managed to make hearthstone perform magnificently and drain much less battery compared to other Unity games. I'm not saying unity is good, but it's obviously not impossible to make an optimised game in it.

>people hating a game engine because they buy a game that was made for streamer click bait.

super bloated

all you need is a window manager and everyone is moving to wayland/xwayland anyways

my mistake i thought op was talking about the desktop not the engine

carry on

>Ai focused games don't exist

The limit isn't hardware, it's user interest - you guys hate AI games.

It turns out most people want autonomy in a game. They don't usually like, outside of idle games, staring at a thing doing stuff without providing any inputs.

Worse, they like winning some of the time. Truly AI focused games have a bad habit of, when players are given a chance to go head to head with AI, either having AI that's too smart so humans get discouraged (see: chess, backgammon, checkers) or is easily gamed and crushed with strategies online (see civ, most 4x games, etc.)

If you look at their absence as absence of consumer demand, it makes total sense.

Why do people keep saying this, the only version that is bloated is the Personal Edition and that is obviously a strategy to make you buy their Premium Version.

They say it because you respond to it, and because it's the standard answer to "what does Sup Forums think of THING?"

How hard is it to include a library of shaders/materials for the most common stuff?

Like a generic concrete material, generic plastic, ect

i want to make a 2d game like hotline miami but with a huge open world map and better ai

Is unity the best studio, or should i use gamemaker or rpgmaker or something

UE is bloody horrible to use, both C++ and blueprints.

Don't use RPG maker. It's terrible. It wouldn't be worth using even if it were free. For open world games Unity or UnrealEngine4 are nice picks. I'm not sure how GM would handle it nowadays since I have yet to use GameMaker2.

its 2d pixel art though,

like only sprites

Unity or Unreal would be my recommendations. I'm helping a friend out with programing right now who's trying to make a game in GameMaker, which uses a made up language called "GameMaker Language" which is incredibly basic, slow, and poorly designed.

It's simple design is what draws people with little experience to it, but you're much better off just learning a real programing language. You'll save yourself a lot of headaches when you try to do something more complex.

using the GPU isn't practical for most AI or physics tasks you would do for games

Unreal engine is better in every way possible.

Yes user, you can use a GPU to deal help with Networking, you can make stream/cuda cores deal with socket connections allowing the processing of hundred of p2p connections in real time by a single GPU, reducing latency and improving overall performance of online games.

I strongly disagree. AI could help balance the Game Flow making the game neither too hard or too easy by auto regulating itself. The games you mention follow prescribed heuristics to oppose the player.

The problem with powerful AI in games is that for commercial reason Publishers prefer to make pretty games than interesting games, a good AI system could easily hoard 90% of the system resources but then you would have to sacrifice visuals.

I've worked with Theano making GPU-accelerated Deep Learning models, good and fast AI is easier to implement with GPUs more than CPUs, yet there exists exactly 0 games that use any kind of GPU accelerated AI technology.

>.NET
wat

That's what Unity would like you to believe :^).

>I've seen great games made using GameMaker and terrible ones made using the Crysis Engine.
Bad example.
>GameMaker: delivers a lot to make simple to mid 2D games
>Crysis: ike Frostbyte, stunning visuals, but little else

It has an excellent logo

you need to process rigidbodies in serial
you can accelerate casts and sweeps with the GPU, but you usually aren't doing enough of that to make it even worth it
same with networking, networking uses next to no processing power, making GPU handle network code is ridiculous

This, let's not forget the thing Sup Forums is actually competent at.

>you need to process rigidbodies in serial
This is false.

You can calculate the Minkowski projection of a set of meshes in parallel (I've done it) and then detecting collision on convex meshes is also GPU-accelerable.

>networking uses next to no processing power, making GPU handle network code is ridiculous
If you have a lot connection it uses a lot, also in games you usually have to deal with interpolation because of package loss and stochastic predictions, this is a clearly GPU-accelerable process.

>You can calculate the Minkowski projection of a set of meshes in parallel
considering the time it takes to upload and download the data to gpu you aren't saving time on this
detecting collisions has to be done in serial because the result of one dynamic object depends on the result of other dynamic objects in the same frame
>also in games you usually have to deal with interpolation because of package loss
also really simple math which isn't even worth the round trip

there's a huge amount of overhead shifting things to and from the graphics card which is fine for stuff like textures and meshes which you upload once and rarely need to access, but makes most non-graphics applications not worth it

>Conversation at HN level

I'm out of my depth boys. Will I ever make it in tech?

I'm more interested in the programming challenge involved in making a game than the creative challenge so I usually start from a relatively barebones framework like love and reinvent the wheel. I'm not a huge fan of Unity because it's so complicated and it's much harder for me to completely wrap my head around the system. However I have an idea for a VR thing (gonna try projecting 4d scenes into 3d space) that I guess I'm going to do in Unity since the support for that seems to be good.

If they added more programming language options like Go, Julia, Ruby, I might use it

check Amazon, there are a ton of books on Unity, also Lynda and Pluralsight have Unity video courses

Absolutely disgusting.

Good engine, not the best.

Best company in the industry though, imo. Their business model is fantastic at promoting growth in the games industry, so good in fact it forced their main competitor to adopt it.

I never use Unity any more but I thank them everyday for what they've done for independent development. Yeah, you get tons of shit piled alongside with it, because they opened the gates to a wider audience and most members of any audience are shitcunts, but its hardly like the millions of shitty app developers or shitty kickstarters make it harder to actually sell a good game. In fact, people making garbage and getting paid for it has highlighted the sheer desperation of most people who are interested in buying and playing games; gamers will part with their money time and time again in the hopes that they get a good experience even 1/100 times, and it doesn't really show signs of fatiguing. I think that's a positive if you're a developer who actually intends on making something good and isn't too much of a brainlet to understand what constitutes something worth selling.

Unknown Worlds have been notoriously shit forever. NS2 was so poorly managed by them and could have been so much more than what it was, but at their core, UW are modders. They seemingly have no aspiration to actually work at the level of game developers, they're still operating at a level that most alliedmodders users do; interesting idea, nice art, poor ability to implement and test, "community-driven" qa, extremely bad curation of their product post-launch.

I would only ever use UE for something open world. But with it being 2d, I guess it depends on your scale as far as you define "open world".

But just by reading your post, if you don't know what engine you want to work with you're probably not ready to handle a project of the sort of scale you're asking for here. I'd use Unity to start with, work with creating some smaller games in concept and prototyping your ideas for bigger projects. Over time you'll learn where your weaknesses are, and you'll then be challenged to learn more about certain areas of game design or get experience managing other people who can fill gaps in your skillset. Its a fun process, and I honestly think that Unity is the best engine to start out with.

will the new maxed out mac book pro be adequate to run unity if i'm making simple 3d shooter?
>inb4 apple