Its refreshing to see something different than another unreal engine based game. It must be extremely challenging to have your own inhouse engine. The engine is one thing. The tooling to build levels, animations etc is maybe even harder. I guess it also gives some advantages. More control over low level architecture gives you more opportunities to optimize. Harder with a all purpose engine like Unreal. Cant wait to play this game.
This is why sad to see Cyberpunk moving to Unreal, it's a solid engine, it takes advantage of all CPU cores, unlike unreal.
I'll never understand why have stuttering issues because of shader compilation, and most other in-house engines don't have that problem. And why it hasn't been improved yet, and even if they release an update to improve this tomorrow, it will take years to see games take advantage of it.
It's a huge resource sink to develop and manage your own engine (Cryengine in this case). No one wants to buy it because proper documentation, support, and updates for the engine would cost CDPR even more to produce. Making your own engine is only really feasible if you have a unique use-case that the existing engines don't provide (this is what led Bethesda to create the Gamebryo engine back in the day).
Ultimately, it's more efficient from a cost and productivity system 98% of the time to use something off-the-shelf.
CDPR makes Red Engine for their games, but has been using Unity and UE for other titles.
Bethesda didn't create Gamebryo. Gamebryo was an extremely prolific game engine back in the day made by a now-defunct company, Numerical Design Limited thogh the rights live on. Gamebryo was used by Rockstar, Firaxis, Ubisoft, and others, including some current Korean MMOs. Bethesda did indeed use it as the basis of their Creation engine.
Generally, you should pick tools for the purpose you need. An off the shelf engine may not help tell the story you want to tell—which Remedy has clearly decided with Northlight.
The death of in-house engines is one we should be sad about, because it creates a monoculture of game vision—more games will be more similar then they are different—it's easiest to use defaults when you have other decisions to make.
> The death of in-house engines is one we should be sad about, because it creates a monoculture of game vision—more games will be more similar then they are different—it's easiest to use defaults when you have other decisions to make.
I'm not so sure about that. We've seen incredible innovation on top of Unity and Unreal (and Gamebryo still). There are some tell-tale signs a game might be running on one engine versus the other, but among games on the same engine there is an incredible variety in everything creative done on top of the engines from art styles to gameplay to even indie business models. Unity has plenty of flaws but we've seen so much more diversity in games since Unity has provided a base platform that better lets especially small developers focus on their unique creative visions rather than reinventing low level primitives yet again. A rising tide lifts boats, right?
The sad thing about the death of in-house engines is that the tide doesn't rise more each time one dies. Game companies should open source more of their in-house engines as they retire them. (CDPR should open Red Engine now that they are moving off of it.) Game companies should externalize (if not open source, then open/easy licensing with source access) more of their in-house engines while those engines are still living. I don't expect more engines to be properly productized like Unity or Unreal, but it would still be interesting to see more engines shared in more interesting ways outside of single developer/single publisher silos.
We know Remedy can do very interesting things with Northlight and it seems to have some tricks other engines can't do, so it would be nice to see if developers that aren't Remedy can also do interesting things with it. If Remedy ever retires Northlight it would be nice to see how it did some of its tricks in a way other engines can learn from.
I hope valve doesn't fumble the bag with source 2, and actually makes it a viable competitor in the engine space.
My limited exposure to it (via s&box) is extremely positive, the hammer mapping tool is absolutely amazing to work with, and from what I've seen way ahead of unity / unreal on this front at-least.
