Is Unity the right engine for this game?



I’m seriously pondering why this game has to utilize Unity. It’s obviously a bad choice for a game like this. Do the developers know that if you would release a game utilizing Unreal Engine 4 on Epic Games Store there will not be any costs using the engine? With Unity you can’t avoid the costs.

Unity is not a suitable engine for any type of action game. It works the best in isometric games, turn based games or generally with games without too much 3D animation and slow pace. Action games suffer from Unity and Otherside will once again face this when they start to optimize System Shock 3 at the last phases of development; it’s not going to be easy if even possible, since no other development team has been able to make their Unity game run smoothly at high framerates. I hope Otherside realizes that high framerates in an FPS action game is a must have today.

I understand that workflow, sharing assets and stuff from the Unity store are big benefits to you, but you won’t save any time in the end. What you will face is a very demanding optimization phase, lots of time fixing bugs and other technical issues related to the engine and you will lose much money in comparison when finally your game is sold, because Unity is not a cheap engine.

I hear System Shock 3 is already halfway ready, but could it be possible to ditch those Unity store assets (if there is a legal problem), hire more people to make graphical work like assets and textures and get that money back when you exclusively release the game in Epic Game Store? I know whiners will spot this message and rage, but don’t care about them, we just got news that Epic Store didn’t hurt sales of Metro Exodus.

I wouldn’t mind one bit if the game would be exclusive to Epic Games Store, but I really don’t like buying an FPS game made with Unity. Unreal Engine seems to be the only right way to go beside CryEngine. What you will benefit compared to Unity is superior lighting, texture filtering, fluid movement, good animations, ease of optimization and making large levels load effortlessly. I really don’t see why you wouldn’t use Unreal Engine if you care one bit about the end result, as how the game will feel to play.

I don’t expect any answers from developes or moderators, but do you other forum folks here have any thoughts about the matter?


I was surprised by the choice of Unity, as well.


You say that Unity is an obvious bad choice, but don’t actually give any supporting arguments to your statement; other than mentioning the fact that no other game from this genre has utilized Unity as its engine before. Just because it’s never been done, it doesn’t mean it can’t be. Ultimately, it’s the developers choice.
I also asked one of (if there are multiple) Otherside’s community managers on Discord the same question. The response I got was this: “SS3 used UE4 for a while, but their team was more familiar with Unity and is now using Unity.” I’m pretty confident that switching from UE4 to Unity was a more so a creative decision than anything else.
Also, the whole “Metro Exodus sold 2.5 times better on Epic Store than Last Light”, is a weird argument. Last Light came out 6 years ago. The PC marketplace has grown significantly thanks to games like Dota 2, CSGO, LoL, Overwatch, Warframe, and Fortnite. Not to mention, physical copies on PC were bigger than Steam back then.
Also, 250% sounds pretty low when you think about it. Metro Last Light sales on all platforms Global is 0.68 million.
If we are a generous and say 25% is Steam’s marked share back then it’s 170,000 copies sold, meaning Epic Store have sold around 425,000 copies, that’s really not that impressive. If the game launched alongside both EGS and Steam, it probably could’ve gone over 1000%.


What the actual f*ck? You think they’re gonna release the game with skinned meshes from the asset store? Do you think Unreal Engine doesn’t have a marketplace? “Unity is slow for action games and suitable for turn based games”? Where did you get this from?
I have worked in many games in Unity3D for a decade (not for me as an indie. For big companies), but nevermind me, Cuphead is made in Unity and is freaking fast paced.
Someone in another thread falsely said that Unity has been created for mobile, then ported to PCs. Is Unreal paying you guys to come and talk bad about Unity and maybe promoting Epic Store exclusivity? (??)
Someone in the Steam forums for U.A. also blamed on Unity3D every single problem U.A. had, but at least he gave some technical justifications, such as Unity reverting to sRGB and BlitBit in Windows 10 when set to windowed mode, and a couple other reasons that, while completely not enough, at least did take away a frame or two when rendering.
Sincerely, do you criticize the Unity3D tool from a developer standpoint because you’ve used it or as a end consumer because you’ve recently seen that many games you didn’t like used Unity3D? I do admit that it went somewhat slower since the last decade (but it also added many features) but can’t agree with the rest.


@SKADRIL I had some arguments there, but I can also add Google search results to give you a better picture about Unity’s issues:…0…1…gws-wiz.2XiGc4UtsvQ

This specific Reddit post is a good read:


First of all, don’t you ever suggest that it’s OK for System Shock 3 to be exclusive to Epic games store. I don’t like the Epic Games Store, I don’t want the Epic Games Store, and I don’t like their business practices. I also don’t need yet another storefront. There are only 3 storefronts I actually like. GOG, Steam and Humble. I hate Origin and Uplay, but deal with them because I have no choice. Do I have a problem with new storefronts showing up and giving competition, no. But Exclusivity deals does nothing but segment the market and limit our choices. Don’t ever suggest that that’s a good thing. Just because you’re happy to be forced to use the Epic store, doesn’t mean everyone else is.

As far as Unity being used for System Shock 3. Don’t mistake System Shock for other FPS games. Unreal Engine is great for shooters, but it’s great for action shooters. System Shock was never an action shooter. It was an FPSRPG with elements of survival and horror. With heavy emphasis on suspense, dread, mystery and atmosphere. Not action. It did have action, but the world and it’s story was what was upfront and center. Not the action. Unity can pull that kind of thing off just as well as any other engine, even as good as the Unreal Engine. But if you’re promoting the Unreal engine because you’re looking for an Unreal, Quake, Halo type shooter, then you are looking at and expecting entirely the wrong IP for that kind of game. System Shock has never and never should be, “that” type of game. Unity is more than capable of accomplishing System Shock’s style of gameplay.


As a developer, Unity is an absolute joy to work with. Actually, no, I never say that about any software, so: in some places, it’s a hot mess.

But all other engines on the market make development even more unpleasant. If Unity is a hot mess, most other engines are dumpster fires, and a few are the sucking firey gateways to hell.

Hellish engines destroy game development schedules.

As a modder, I’d argue that since Unity makes no real effort to stop you modding the game, once the game is finished, the community can carry on and give it a whole second life. At least so far as I know, the tools for modding Unity are far above and beyond those for any other engine, and better even than the games where the developers released an editor for just their one game.

And the things that Unity has in the pipe… well, the recent Unity Keynote showed some of that, and it’s hella exciting. If OSE keeps riding this train, there’s no telling what future games will be like. If they don’t, they won’t be able to keep up to speed with it.

And if they fragmented, with Austin going with one engine and Boston with another, they would lose out on skill transfer and cooperation. Which apparently isn’t massive right now, since each team is heads-down focused on developing their own product, but give it a few months.


I generally stay out of religious wars, such as Unity vs. Unreal, or why BRIEF was the best text editor ever created (hint: it was highly moddable).

But this is a good point:

Ultimately the nature of the game needs to dictate the engine. But while engine consistency among projects can become an albatross (see: EA’s Frostbite and Starbreeze’s Valhalla), for a small studio it seems like there’s some real value in maximizing code sharing.


And in that vein, there’s a much better chance that the DX:IW / T:DS engine sharing won’t damage things.


It’s not so much the code sharing, as the skill sharing that’s important.

They apparently aren’t sharing any code between UWA and SS3 (gonna suck a little for modders, have to learn a whole new codebase, but not that hard).

But the important thing is that they can, at crunch mode or if they get in a jam, share developers. They can just ask each other in their company Discord or Slack, “hey, did anyone find a solution to this problem I’ve got with Unity?” and get an answer from twice as many people. They can share tips and tricks and advice of which trees to not try backing up, and which features are cool and should be learned.

This is potentially a pretty huge help with productivity, and even without direct code sharing, means they can stand on each others’ shoulders as they go from project to project.


I think I’d agree, broadly, that it doesn’t hurt to have many people in a studio with similar technical knowledge in their brains. The obvious benefit is studio survivability as people come and go (as happens).

But I’m inclined to persist in thinking that sharing an engine is more of a technical benefit than a people benefit. It’s convenient to be able to ask a teammate, “Hey, how do I do X?”, but to some extent that can be picked up from tool docs, StackOverflow, and some hands-on learning (which has its own value).

By contrast, the value of already doing X – because core architectural code could be shared and the local team were able to move on sooner to building the systems and resources specific to their project – is, I’d say, extremely high. The sooner you have a minimally working game to bang on, the more time you have to figure out the features that game actually needs in order to become polished fun.

In other words, I’m thinking that using one good, working engine across teams has value for both knowledge sharing and code sharing – I just assess the value of sharing code across teams as especially high for a small studio. Knowledge is valuable, but time is precious.

Now, if you’ll excuse me, I’m off to tell another studio how to run their business! VROOM :smile:


I have no problem with people doing smart, efficient things.