Putting that level of procedural flexibility into every system they have is pretty amazing. Hopefully I can achieve even a fraction of that with my own.
Yeah after seeing the tech in No mans sky’s engine produce such nice looking environments that are procedural and then have them be almost infinite (In a multiplayer environment as well) It makes me kind of look at my own project as ‘meh’ now. I wish UE4 could do this but I mean their terrain engine is so customly built and their engine is created specifically for this purpose from the ground up I’m not sure if we would ever have access to this ability.
On the other hand you could say procedural created worlds can get repetitive and are not as good as hand making everything, but the complexity in the random variation and the amount of variation which actually looks appealing in no mans sky makes you think again. For outdoor environments it really is an thing to be able to visit a planet that is generated in a way you have somewhat created as an artist but is different and something you havent seen before at the same time.
For single player we do have the ability to do infinite worlds in UE4, I’ve seen some impressive procedural works with Unity engine that people have done, maybe someone will create a custom terrain system for UE4 that can have this ability, since I’m not sure if we have full control over the existing one tied to blueprints. The LOD system going from space to land is like an technological achievement in itself. We need a more elaborate terrain solution to generate new chunks as you move along based on your position. We also need a better terrain LOD system for smoother transitions and maybe something could be built on top of the material editor tied to blueprints for random variation with a bunch of world machine data
Hopefully someone from the community or an epic dev may implement this some day. I think it would be a fun project, many other games will most likely be using the tech in their games in the future as well, No mans sky is just the beginning, pioneers & the first people to do it properly harnessing the power of current hardware
Honestly - The no mans sky devs are really the only people who can give Epic Games a run for their money if they ever open up their engine for licensing.
The guy asks him: “-Why noone has done this before?”
He don’t know why… … The reason is simple; developers hate math 
complex math is used for what is absolutely necessary in games only, but that should change. Game creators (not just the tech providers) should all be very intimate with math.
But oh boy, math is so hard. I mean the real maths, like the algorithms they used in this game, is pretty complex and game creators always have a more artistic approach trying to avoid ‘science’ as much as possible.
Their engine is probably only useful for what they’re doing. That’s usually the case
Whats to stop them from adding other gameplay elements to it like a shooter where you can exit the vehicles too? Or any other projectile / traces based gameplay? I mean they already have their terrain system working with multiplayer, there is a slight chance you can encounter another player. Theres AI and everything too… Why is it only useful for “what they’re doing” ? The hard stuff is already there, you could take their engine and do way more than just a spaceship sim with it.
I think the tech could be somewhat duplicated in UE4 in a less complicated way, maybe we would have more terrain lod popping & not as complex differentiation in procedural terrain generation, but I would pay for anything procedural related in the marketplace
If the No Man’s Sky devs made an engine it would look alot like Crysis. Great for what its built for AND if you know how to use it. It would take a decade or more for them to reach even a fraction of the flexibility and general game designing power that UE4 has.
Theoretically you could make No Man’s Sky in UE4. You have the source. Good luck with the math though. And probably rewriting the renderer and who knows how many other tools and systems. Tons and tons of work.
I am more interested in the idea of what No Man’s Sky has achieved. Applying procedural concepts to everything. I want procedural music, terrain, meshes, characters, game systems, etc, etc. You can build in tons of complexity in UE4, but it takes a lot of work and time. Pays off in the end though.
For UE4, that means more procedural access to systems in general, preferably from blueprint. Complete blueprint/procedural access to landscape, world browser, and every other system there is.
In the end, there is always C++ of course. With enough work I am pretty sure you can make anything in it.
What are you talking about? How can you compare a game engine with tools like UE4 with a boxy game?
You can do this game in UE4, what stops you?
It’s hard to see how interesting gameplay will fit into this. Still it may turn out to be one of the better walking simulators.
That’s the crux of the problem with open worlds, and especially procedurally generated ones.
As as No Man’s Sky looks for exploration, I’d be more impressed by a dungeon crawler that could generate interesting conflict procedurally (even if it didn’t have an amazing level generator per se).
People are waiting for a very engaging gameplay like what Star Citizen wants to deliver.
I believe ppl will be very mad when they findout that the gameplay is actually flying to a planet and making lists of new discovered animal species then repeat…
My ideas exactly. At first I was really excited about No Man’s Sky but now I think we all get bored in a very little time. I’ll give it a shot of course but I think Star Citizen will be the star.
After No Mans Sky came out I was heartbroken,I’m a 14 year old thinking I was going to create the next AAA Game nobody’s ever heard of or maybe even thought about. So i went into production, just me and only me. I got alot of cool things done with Unity game engine but with what they’ve done I could never outbeat that. I don’t have the manpower or scripts to do my idea. Sadly this is as far as i got. Don’t mind the sunlight that was the last part I was on. Also the rotations were a bit off. Goodbye Space Project. Sorry for my voice for whomever it concerns.[video]- YouTube
From a technical point of view its an amazing piece of kit, as a game not so much.
hoping free DLC will fix that abit
I can’t see how DLC will do anything big, it took 5 years to get to where they are
Atlas is just the entity at the focus of the “story” which doesn’t end up being much really, it’s the path you would go on to get to the center of the galaxy but is a bit of a letdown.
This game is the flop of the year (because hype and production wasn’t in AAA scale); selling at $60 was a huge mistake.
The little story won’t help to fix it. The one man made ffxv-mayCry spinoff has more value than this one.
I’m sorry that I wasn’t wrong about it when I wrote that comment last year.
As soon as the hype vanishes, that “7” will turn to a 4 or 5 score, because hype still adds value to the game.
Great example of powerful marketing used on weak product.
Hello Games would be fine if not for the marketing hype… the costs of marketing forced them to go for the $60 price tag.
Now they have a lot of enemies.
Hehe … they already made so unbelievable much money that they will not invest a single $ into further development of NMS. Instead they either are happy that they dont have to work a single day for the rest of their lives - or they do plan on a new game.
It was about 212.000 simultanious players on day one just on steam.
So if you wanna get rich then create a new game mechanic and hype the **** out of it. Everybody will be buying it out of curiosity.
This might still works for the next 2, 3 games of this kind 
Actually the studio earns from $9 to 18$ from the $60 price tag; average $12 if your studio has good reputation in consoles…
So if they spent, let’s say, $2KK to develop the game plus $1KK for marketing then they have to sell first 50.000 copies to pay Sony for the funding; only then they begin earning something.
Let’s say the game sells 2.000.000 copies across PS4/PC; the studio must keep banked at least 50% of that to develop next game, the other 50% should be split between the 15 devs (proly won’t)… so each of them would earn around $800.000,00; split by 4 years of work, $200k per year.
Taxes in UK I think will take around 40% of that earnings, so… Would be left $120.000 per year of work for each member of the team.
Good monies, but not “rich”
Maybe they manage to build another game without 3rd party funding with that.
Game dev earns that low…? I know it’s low but that amount is like might as well get a normal job or IT industry for same pay. If the money doesn’t split, the boss is rich lol.