Why not shift NMS to a new game engine?

[NZ.Zero]

All those game-limiting bugs we need to work around - part limits, janky physics, falling thru floors etc etc, you can bet those are all because of the hard limits of the game engine.

The game engine has done great service but it's limits are very obvious, and is clearly buggy as hell. Basically it has the feel of an old plane that's been added to and patched up many times, but at some point you just need to replace that old airframe for something that can meet modern requirements.

So why don't HG update it? Build a new game engine, maybe even a modular one that can be gradually transitioned to?

Of course it would be a lot of work and money, but much like an old airframe, the cost of working on old ones just goes up and up, and over the years they become less and less fit for purpose, so around now is the point when it starts to make total sense to start driving the shift. (Actually it was a few years ago, but just like old airframes, the NMS engine is still limping along way past it's use by date).

I actually think the economics would work out, because NMS is a great asset and could be financially viable for quite some time longer with a decent update of the underlying engine.

[william_es]

Originally posted by NZ.Zero:

So why don't HG update it? Build a new game engine, maybe even a modular one that can be gradually transitioned to?
.

Possibly you could build this new engine for them? You have some great ideas there. And if if works out, you could sell the engine to other games.

[NZ.Zero]

There are plenty around, why build one from scratch?

[Mr. Bufferlow]

I think from HG's perspective, the reward/cost ratio is just too low for the quantity of work involved. It could take them a couple of years (or more) of dedicated effort. They are a very small development team and even a big development house would find doing a year or less turnaround hard.

Our main hope is they do a remaster at 10 years. That would have the advantage of selling all new software to buyers instead of hoping for a few additional buyers to absorb the cost and provide a little profit each update.

[Aurumworks]

People think switching out a game engine is like copy-pasting in some code and done. It's more like removing your skeleton and putting in another, possibly not even a human shaped one.

[Azure Fang]

Two reasons, probably.

1) As per interviews, the team remaining on NMS is small. An engine port would need studio re-focus and, with the still unannounced project in development, that's not feasible.

2) Engine ports are a huge undertaking. More often than not, this requires completely rebuilding the game from scratch. It's never as easy as just "hitting the build button" under the new engine. Even between different iterations of the same engine, such as from Unreal 3 to Unreal 4, programming language support, asset management, even basic API calls can differ and need to be totally reworked.

If you want a snapshot of the ♥♥♥♥♥♥♥♥ that comes with engine porting, look into the Viscera Cleanup Detail status updates as they detail the woes of "porting" from UDK to UE4: https://steamcommunity.com/app/246900/allnews/

[Zymmer]

Originally posted by Aurumworks:

People think switching out a game engine is like copy-pasting in some code and done. It's more like removing your skeleton and putting in another, possibly not even a human shaped one.

I think this guy should do this all on his own - since it is so easy apparently. Like the first commenter said.

[Lindy Bomber]

Different engine, different problems. Non-solution.

[NZ.Zero]

I never said it's easy. Just increasingly necessary. You can't avoid the inevitable forever and the only thing that's certain is that if you don't start, you'll never get there.

Originally posted by Lindy Bomber:

Different engine, different problems. Non-solution.

Yes some problems of course, but an architecture designed for the purpose would obviously be better. (The game is very different now). Not to mention a decade of industry progress and all the knowledge HG has gained.

[Lindy Bomber]

The unnamed big project they have been working on for 3-4 years is probably build on a new engine. Once that title is launched they will probably put NMS to bed. So porting NMS to a new engine is probably not an economical use of their time.


Maybe 10 years from there will be an anniversary addition using what ever game engine they develope over that time.

[Azure Fang]

Originally posted by NZ.Zero:

I never said it's easy. Just increasingly necessary. You can't avoid the inevitable forever and the only thing that's certain is that if you don't start, you'll never get there.

The question then comes down to "is it worth the cost?" The effort to port a game takes money. Will the port draw enough sales to cover the cost of the port or, foregoing that, is it worth it to the company to eat the cost to give their existing customers such a costly update?

NMS isn't an MMO, nor could it be classified as a "live service". It's a complete product but an outlier in that its developer decided to keep releasing content for it post-release. Even these content updates cost money (payroll hours, mostly) and, unless the updates themselves drive sales enough to cover their cost, that's a drain on the company that needs to be covered somewhere otherwise it will lead to the death of the company. That's why the rare few games that DO produce engine ports end up, in most but not all cases, selling them as a new product rather than giving them away to previous owners. That cost needs to be recouped, and it's a risk as previous owners may be less inclined to re-buy a game they already own.

[voidCaster]

Originally posted by Zymmer:

Originally posted by Aurumworks:

People think switching out a game engine is like copy-pasting in some code and done. It's more like removing your skeleton and putting in another, possibly not even a human shaped one.

I think this guy should do this all on his own - since it is so easy apparently. Like the first commenter said.

Not sure what Zymmer read, but I know what they both mean.

There are literally uncountable programming languages. Some programming languages have been used to write other programming languages. When you're talking about YOUR PC -> NEW MAN'S SKY CLIENT -> <???> -> YOUR CPU AND GPU the unknown part is a red-eye trans-Atlantic flight oversold on seating during bad weather.

Absolutely every engine, whether it's Unity or Unreal or PhysX or Havok, is another separate platform. Think about being a computer programmer, and having to port code from Windows to Macintosh. If you don't get that, think about all the problems one platform has that the other doesn't, and how long it takes Microsoft (a billion-dollar publically traded company employing tens of thousands of programmers worldwide) to fix anything.

So the next platform not only has a different syntax, but its IDE has a different interface (and its programming language is either completely different or totally off-the-effing-wall) and that platform has different strengths and weaknesses and inherent glitches...

[Desoroxxx]

The beauty of software is that you don't need to replace everything all at once, and as far as I know NMS doesn't have an engine.

[Azure Fang]

Originally posted by Desoroxxx:

The beauty of software is that you don't need to replace everything all at once, and as far as I know NMS doesn't have an engine.

You would be wrong. All games run on an engine of some sort, be it proprietary, pre-packaged, or a raw language interpreter.

NMS utilizes a proprietary engine built in-house and augmented with some pre-packaged middleware, most notably Havok.

[Norseman]

Originally posted by NZ.Zero:

(...) So why don't HG update it? Build a new game engine, maybe even a modular one that can be gradually transitioned to? (...)

The massive amount of labor to change to a new engine and/ or rewrite all code to "fix" things will not be not profitable for a old game as No man's Sky is. That simple.

[Desoroxxx]

Originally posted by Azure Fang:

Originally posted by Desoroxxx:

The beauty of software is that you don't need to replace everything all at once, and as far as I know NMS doesn't have an engine.

You would be wrong. All games run on an engine of some sort, be it proprietary, pre-packaged, or a raw language interpreter.

NMS utilizes a proprietary engine built in-house and augmented with some pre-packaged middleware, most notably Havok.

No, a proper game engine is a software you can use to built a game, but both can be separated, as far as I know NMS doesn't have that, the "engine" is built specially for the game if you remove NMS from the "engine" there is basically nothing,