Unreal Engine 5.7 is out

winjer

Member

Unreal Engine 5.7 rolls out with new PCG, Nanite Foliage, and MetaHuman tools




Unreal Engine 5.7 (what's new):

  • Production-Ready Procedural Content Generation
  • PCG Editor Mode
  • Procedural Vegetation Editor (Experimental)
  • Quixel Megaplants support
  • Nanite Foliage (Experimental)
  • Substrate material framework (Production-Ready)
  • MegaLights (Beta)
  • MetaHuman Creator plugin for Linux and macOS
  • MetaHuman batch processing via Python/Blueprint
  • Live Link Face external-camera support
  • Refactored Animation Mode
  • Selection Sets
  • IK Retargeter updates
  • Enhanced Skeletal Editor sculpting and weighting
  • Dynamic Constraint Component for props
  • Live Link Broadcast Component
  • Updated Composure real-time compositing
  • In-editor AI Assistant
  • New Unreal Editor Home Panel
 
Nanite foliage, hell yea

Awkward Jim Carrey GIF by Ace Ventura
 
Uh, who cares... no one is even using 5.6 yet for that matter cuz for most UE5 games development started long before any major updates and no one is updating all the work they've done over the years to work with newer UE5 versions.

Like, imagine having to replace all the foliage in your game and use Nanite instead. Not gonna happen, ever. All these major Engine updates will matter for newer games only.
 
Last edited:
I like that UE5 has a ton of options, all an option, nothing is a must.

- You want cutting edge stuff? Have fun!
- You want just UE3 stuff? Have fun!
 
Uh, who cares... no one is even using 5.6 yet for that matter cuz for most UE5 games development started long before any major updates and no one is updating all the work they've done over the years to work with newer UE5 versions.

Like, imagine having to replace all the foliage in your game and use Nanite instead. Not gonna happen, ever. All these major Engine updates will matter for newer games only.
So the engine should've just stagnated and stayed at 1.0? The benefit is games in the future will be more feature rich. Moaning for the sake of moaning
 
Uh, who cares... no one is even using 5.6 yet for that matter cuz for most UE5 games development started long before any major updates and no one is updating all the work they've done over the years to work with newer UE5 versions.

Like, imagine having to replace all the foliage in your game and use Nanite instead. Not gonna happen, ever. All these major Engine updates will matter for newer games only.
You've just answered your own question.
 
If you don't count any others sure.
The others don't look as good even if some of them perform better.

So yeah, if you are a graphic whore and have a capable enough pc to mitigate some of its problems, ue5 is definitely the best, i don't really care if it sucks on console, i play mostly on pc so i'm not gonna speak for the console experience.

Decima can get the second place if it makes you happy, especially when you consider that the best aspect of that engine (the character rendering) is made with metahumans ue tools.

If you leave the characters out, ubisoft is the rightful second place between avatar and shadows (i know they are 2 different engines).
 
Last edited:
So the engine should've just stagnated and stayed at 1.0? The benefit is games in the future will be more feature rich. Moaning for the sake of moaning
You're completely missing the point of my post:messenger_unamused:

Epic should work with the devs on a deeper level and make version transitions faster and effortless with each major Engine update, as well as provide necessary means so that older games with absolute dogshit performance could also benefit from this. Each new version intros not only visual upgrades, but also a lot of updates related to performance, streaming etc. etc. S.T.A.L.K.E.R. 2 for example uses outdated af UE5.1 if iirc and so it causes a lot of problems and performance issues that are no longer present in new versions of the engine.

Visual upgrades are completely different cuz of the reasons I mentioned in my previous post, unless we're talking about Hardware Lumen, GI etc. which only needs some tweaking to work instead of complete rework of environments.
 
Last edited:
You're completely missing the point of my post:messenger_unamused:

Epic should work with the devs on a deeper level and make version transitions faster and effortless with each major Engine update, as well as provide necessary means so that older games with absolute dogshit performance could also benefit from this. Each new version intros not only visual upgrades, but also a lot of updates related to performance, streaming etc. etc. S.T.A.L.K.E.R. 2 for example uses outdated af UE5.1 if iirc and so it causes a lot of problems and performance issues that are no longer present in new versions of the engine.

Visual upgrades are completely different cuz of the reasons I mentioned in my previous post, unless we're talking about Hardware Lumen, GI etc. which only needs some tweaking to work instead of complete rework of environments.
You know I kinda agree. At this point we won't see games using UE5.7 for another four years!

Look I have no idea how any of this works but wouldn't it be nice if you can just carry over all your work to the latest build?

It reminds me of Daikatana that used Quake 2 engine that already looked outdated.
 
Nanite foliage, hell yea

Awkward Jim Carrey GIF by Ace Ventura
Experimental so not ready for anything but limited testing.
VCG headlines are... creative.
5.7 has one big new feature which is now production ready - procedural world generation.
The next one which is going from exp to beta in this version is Megalights - which they don't mention at all for some reason.
And voxelized Nanite foliage is releasing as experimental. Which likely means that we won't see it being used until some 5.10 which in turn means games releasing around 2029 or so?
 
Experimental so not ready for anything but limited testing.
VCG headlines are... creative.
5.7 has one big new feature which is now production ready - procedural world generation.
The next one which is going from exp to beta in this version is Megalights - which they don't mention at all for some reason.
And voxelized Nanite foliage is releasing as experimental. Which likely means that we won't see it being used until some 5.10 which in turn means games releasing around 2029 or so?

It's right there in the article and in the first post, that:
  • Nanite Foliage (Experimental)
  • MegaLights (Beta)
 
It's curious that modern engines like UE increasingly have procedural generators for worlds, assets, animations, character models...

And there are no media outlets like Eurogamer or people on X crying because the work of modelers, animators, and artists is being lost, as is the case with AI-generated voices. It seems that there are first- and second-rate artists, or else AI is evil just because.
 
It's right there in the article and in the first post, that:
  • Nanite Foliage (Experimental)
  • MegaLights (Beta)
There are no mentions of Megalights up until (a copy-pasted, I presume) "What's new" list.
And Megalights is arguably a more interesting feature than anything which got into the headline.
 
There are no mentions of Megalights up until (a copy-pasted, I presume) "What's new" list.
And Megalights is arguably a more interesting feature than anything which got into the headline.

That is because not everything fits in the title. Everybody knows that.
 
"New MetaHuman tools" which are of about zero interest to the consumer public found their way there though.

Are you really making a drama over a tittle, just because it didn't include your favorite tech? Despite it being listed in the main article.
 
So the engine should've just stagnated and stayed at 1.0? The benefit is games in the future will be more feature rich. Moaning for the sake of moaning
Nanite and lumen is going to suck even in future hardware, because it't not an advanced tech, it's simply poor optimized. When they get a stable release probably we will get a new tech making UE5 obsolete.
 
Are you really making a drama over a tittle, just because it didn't include your favorite tech? Despite it being listed in the main article.
I'm not making any drama, it's you who are defending a title on a click bait rumor site. I'm just saying that the title is stupid.
 
It's curious that modern engines like UE increasingly have procedural generators for worlds, assets, animations, character models...

And there are no media outlets like Eurogamer or people on X crying because the work of modelers, animators, and artists is being lost, as is the case with AI-generated voices. It seems that there are first- and second-rate artists, or else AI is evil just because.
I assume you're referring to that EG review of Arc? What a completely moronic review that was, the author has clearly no understanding of the AI usage in that games' development and the features it enables, he didn't even mention the cool feature where it transforms the game's microphone input into a characters voice. They focus on hating AI, whilst ignoring real issues like union busting, extreme crunch and CCP/Saudi funding. I didn't see any points docked for Fifa for predatory MTX, I didn't see TLOU docked points for the devs extreme work hours. And as you say, what's the impact difference between the new AI tech and these procedural tools. I guess those guys are not as important though /s.
AI has grey areas associated with it, especially when it comes to data training/sourcing, but the usage of it in games development is inevitable and Embark's usage looks to be a great and fair use of it. To laser focus on AI and just say AI=Bad is absolute garbage tier "journalism", selective outrage and virtue signalling just so he can wank his own ego off.
 
Last edited:
You know I kinda agree. At this point we won't see games using UE5.7 for another four years!

Look I have no idea how any of this works but wouldn't it be nice if you can just carry over all your work to the latest build?

It reminds me of Daikatana that used Quake 2 engine that already looked outdated.
You can. Some things might break and need fixing up in the process but it's absolutely doable. How do you think all the live service games that keep getting updated work?
 
As gorgeous looking as this engine is at creating incredibly dense realistic looking open worlds, i've yet to see it come close to R* Rage engine in terms of overall imo fidelity & complexity shown off on the GTA6 trailers, how is it that possible and why the hell don't other developers license Rage from R*, is the Rage engine really that good? I've yet to see an actual game look anywhere close to these tech demos even from previous versions
 
The biggest feature I want is better documentation of the API and of the engine and processes in general. And an easier way to install it on Windows (Linux is ironically easier because all the shit you need is bundled with the binary).

The engine is fucking useless if the documentation is shit. And Unreal has one of the worst documentations I have seen for a commercial product.
 
As gorgeous looking as this engine is at creating incredibly dense realistic looking open worlds, i've yet to see it come close to R* Rage engine in terms of overall imo fidelity & complexity shown off on the GTA6 trailers, how is it that possible and why the hell don't other developers license Rage from R*, is the Rage engine really that good? I've yet to see an actual game look anywhere close to these tech demos even from previous versions
Not to do with engine, it's just the massive amount of time R* spend on creating their games. Those systems would be possible in another engine, just very few other devs have the massive budget, manpower and time to implement them to the same quality as R*. Rockstar wouldn't license out their engine and there likely wouldn't be any good documentation for others to make good use of it. GTA 6 really does look a cut above anything else this generation though without a date.
 
Top Bottom