I'm honestly starting to think that Unreal Engine is killing the AAA games industry, leading to lazy development practices and unoptimized games. Of all the UE5 games to have come out lately, I...
I'm honestly starting to think that Unreal Engine is killing the AAA games industry, leading to lazy development practices and unoptimized games. Of all the UE5 games to have come out lately, I can only think of one that doesn't run like complete shit and that is Clair Obscur: Expedition 33.
Early versions of UE5 didn't run all that well, but titles are getting better as the engine improves and studios share their best practices for the engine. This tech demo already demonstrates a...
Early versions of UE5 didn't run all that well, but titles are getting better as the engine improves and studios share their best practices for the engine. This tech demo already demonstrates a pretty notable performance improvement over the first batch of UE5 demos running on PS5.
A big reason developers are pushing to use UE5 is because many of its core technologies have huge potential to streamline development workflows while still allowing them to push for bigger and better design. Expedition 33 actually started as a UE4 project, and they pivoted to UE5 mid-development because they felt the time savings and design opportunities afforded by Lumen and Nanite outweighed the amount of work it would take to switch engines.
Cinematic trailer here: https://www.youtube.com/watch?v=i_aRhNcHryA Of note, this is just a tech demo, iirc CDPR had some cool tech demos for Cyberpunk 2077 and look how that launch went. Still...
Of note, this is just a tech demo, iirc CDPR had some cool tech demos for Cyberpunk 2077 and look how that launch went. Still though, looks pretty impressive!
I'm not too sure, I think the game will end up looking rather close to what we see here, this tech demo is surprisingly tame, and seems to be generally on par graphics wise with games made for...
I'm not too sure, I think the game will end up looking rather close to what we see here, this tech demo is surprisingly tame, and seems to be generally on par graphics wise with games made for last generation running on current generation (which, by now, current generation is almost old enough to be the last generation...).
I'm not really seeing anything visually in the groundbreaking visuals that tech demos usually showcase, the particle scattering, the light rays, the foliage, the movements, they are all pretty tame. Sure, it has some serious art direction and fidelity, but I am not seeing anything that screams 'impossible' or even a step beyond what we have now. In fact, I would say it's a step behind in a lot of ways, take a look at the world (not the cut-scenes that) of recent games like Expedition 33, or even Final Fantasy XVI. These are about on par with each other.
Edit: I would have to say that natural world wise (01:10), it's almost a half-step beyond Ghosts of Tsushima in the visual department, but that would be unfair to compare as that game is a masterpiece.
If anything, they are purposefully being rather tame to either 1) astound us closer to release and/or 2) learned from Cyberpunk 2077. All in all, I am looking forward to this.
Yeah I feel like I'm the only one with a memory when it comes to Tech Demos and video games, and it's not just Cyberpunk/CDPR. I always feel like the tech demos come out look so close to photoreal...
Yeah I feel like I'm the only one with a memory when it comes to Tech Demos and video games, and it's not just Cyberpunk/CDPR. I always feel like the tech demos come out look so close to photoreal and has so many immersive details, but then the final game even with maxed out graphics on a decent PC looks far below what originally came out.
I feel this a lot with Unreal 5. I have seen some AMAZING photoreal tech demos from Unreal 5, yet I have not seen or played a game using Unreal 5 that has looked as good as the demos.
At some point I have to ask myself, what is the point of the demos? "Look at what we could do! But realistically will never do."
And it'll run like absolute shit. Some studios can make Unreal work. Those are the same studios who can make any engine work because they are experts in optimizing. CDPR, Bethesda, and a few...
And it'll run like absolute shit.
Some studios can make Unreal work. Those are the same studios who can make any engine work because they are experts in optimizing.
CDPR, Bethesda, and a few others probably shouldn't be allowed anywhere near this engine. I don't know if it's because it allows for shortcuts or because it requires more optimizing, but we have so many examples of what happens when large, beautiful open-worlds are made in UE5 without the necessary fine-tuning.
CDPR's engine developers are absolute wizards with optimization, they just get thrown under the bus by leadership that pushes to release games before they are ready. A bunch of the work being...
CDPR's engine developers are absolute wizards with optimization, they just get thrown under the bus by leadership that pushes to release games before they are ready.
A bunch of the work being demoed here to make Unreal more performant in open world games was done by CDPR engineers.
I kind of hate the photorealistic look of UE games. Even when it's done well it's just so bland. It seems like the goal is to create games where if you disabled the HUD and took a screenshot you'd...
I kind of hate the photorealistic look of UE games. Even when it's done well it's just so bland. It seems like the goal is to create games where if you disabled the HUD and took a screenshot you'd never be able to figure out what game it is.
Style over fidelity every time. The reason Dark Souls 1 still looks good is because stylistically it's knitted together with excellent insight into what works and what doesn't. Dark Souls 2 looks...
Style over fidelity every time.
The reason Dark Souls 1 still looks good is because stylistically it's knitted together with excellent insight into what works and what doesn't. Dark Souls 2 looks worse because of a focus on contemporary graphic design choices that are now aging.
We can even go back further, Earth Worm Jim still looks great. Yeah you can see it's from a couple of decades ago, but it's pleasing to look at with unilaterally its own style application.
I agree with you, to an extent. Clair Obscur applied their own style to the engine and it did set it apart from the other UE5 games. Not quite as much as to be unrecognizable if you side-by-side it with something like Veilguard, but still a well done application of stylistic vision.
Ultimately it's more the triple-A chase to be as photorealistic as possible without thinking through what style they're looking to implement. UE5 compounds the issue moreso than any other engine did.
I should've been clearer, but yeah I was referring to the default style that comes out of UE games that don't put in the work to differentiate themselves rather than anything the engine forces to...
I should've been clearer, but yeah I was referring to the default style that comes out of UE games that don't put in the work to differentiate themselves rather than anything the engine forces to happen. Studios can control their graphics pipeline to make it look different, just like other tools can go for photorealism. It's also the fault of GPU marketing and such suggesting that the "next gen" everyone should be chasing means ray tracing and complex hair simulation and other stuff focused on making it look more real rather than unique.
For fun I'm also editing in a conspiracy theory regarding game realism down here. As games, especially AR and VR games, become more realistic do we start running the risk of making the "games make people violent" people right? If it looks real enough will it blur the line between fiction and reality to an unsafe degree for some people?
Oh I wasn't disagreeing with you! Even stylistically distinguished games still have that Unreal 5, for lack of a better word, sheen on all textures. Maybe? I think we're still ways off. At this...
Oh I wasn't disagreeing with you!
Even stylistically distinguished games still have that Unreal 5, for lack of a better word, sheen on all textures.
Maybe? I think we're still ways off. At this point we're still at the level that it only happens to someone already at the breaking point and/or mentally unstable enough to blur the lines between fiction and nonfiction anyway regardless of the medium.
Unreal is suffering from the same problem as Unity. They've built an engine that is easy enough to work in that anyone with a modicum of programming ability can cobble together a functional game...
Unreal is suffering from the same problem as Unity. They've built an engine that is easy enough to work in that anyone with a modicum of programming ability can cobble together a functional game using store-bought assets and ship it on Steam in 6 months. It's inevitable that the first games to ship on new versions of the engine are all samey, poorly optimized crap.
Talented developers can still use it to make visually appealing, performant video games. But they will always come later and in smaller quantities. UE5 first came out 5 years ago, and with AA and AAA development cycles being what they are, it makes sense that we are only just now seeing high quality games that properly utilize the engine.
The game has been in full scale production for less than a year, and this is a tech demo built to demonstrate new engine features to other Unreal developers, at a developer conference. I wouldn't...
The game has been in full scale production for less than a year, and this is a tech demo built to demonstrate new engine features to other Unreal developers, at a developer conference. I wouldn't read too much into the art direction outside the cinematic they showed at the start.
I am seeing a lot of Assassin Creed and Horizon (Zero Dawn, Forbidden West) influence on this, well more than I ever anticipated from CDPR. For AC it's the movement, the world, the way the NPCs...
I am seeing a lot of Assassin Creed and Horizon (Zero Dawn, Forbidden West) influence on this, well more than I ever anticipated from CDPR. For AC it's the movement, the world, the way the NPCs are reacting and moving, and how the character is generally sauntering. For Horizon it's more visual identity, like the scene starting at 04:14 is something I would have thought was just a recording of a Forbidden West, or Horizon with a 4k texture pack + lighting addition applied.
Edit (added this in another comment, adding it here too): I would have to say that natural world wise (01:10), it's almost a half-step beyond Ghosts of Tsushima in the visual department, but that would be unfair to compare as that game is a masterpiece.
These aren't necessarily bad things, but it's just incredible where you can see the influence so clearly. I am also feeling a little bit of Fable II/III in the mix there, not sure, might just be the chickens running around needing to be chased.
Those games had a lot of Witcher 3 in them so it’s just the ouroboros of adventure game design. The game looks good and I’m interested to see where Ciri’s story goes, but I’m so fatigued by the...
Those games had a lot of Witcher 3 in them so it’s just the ouroboros of adventure game design.
The game looks good and I’m interested to see where Ciri’s story goes, but I’m so fatigued by the Ubislop gameplay loop that I couldn’t get more than 4 hours into Horizon: Forbidden West—the sequel to a game I loved—and I don’t think I’d have a different reaction to this. If they don’t meaningfully evolve the gameplay, or just depart from the wide-open world format in general, I doubt I’d be able to suffer through it. Witcher 3 itself didn’t have enough depth to its traversal, crafting, or combat systems to hold my interest for the full game length, especially once you start getting into the DLCs. So as much as I enjoyed playing through the stories the last third of my time playing it felt like a major slog.
In contrast, I could play Spider-Man and Miles Morales forever. I fully ran out of things to do and still wished there was more to do. Arkham Knight I also think I just started to get to the point of being sick of it by the time I was done. But some of these 60+ hours games are trying to scrape not-enough-butter over way too much bread.
I can't wait to not be able to run this when it comes out. That's not even just a general industry thing, but a CDPR specific pattern. I remember even Witcher 2 being pretty demanding when that...
I can't wait to not be able to run this when it comes out. That's not even just a general industry thing, but a CDPR specific pattern. I remember even Witcher 2 being pretty demanding when that came out.
It's too early for me to engage with stuff like this and think it's seriously reflective of what's to come. But another UE5 game? Seriously? Was RED engine too much work to maintain?
It's too early for me to engage with stuff like this and think it's seriously reflective of what's to come.
But another UE5 game? Seriously? Was RED engine too much work to maintain?
Yes. They swapped to UE5 with Cyberpunk. https://www.pcgamer.com/cyberpunk-2077-director-says-studios-switch-from-redengine-to-unreal-engine-5-isnt-starting-from-scratch/ I mean, people complain...
Pardon me, in retrospect my tone probably didn't read just as "surprised". I know I'm in the minority here, but I'm more concerned about UE5 monopolizing the engine space. I wouldn't consider...
Pardon me, in retrospect my tone probably didn't read just as "surprised".
I know I'm in the minority here, but I'm more concerned about UE5 monopolizing the engine space. I wouldn't consider keeping a well-maintained engine as bloat, doubly so if you can license that engine out, but I can see how many would disagree.
Having a unique engine makes hiring game developers much harder. In comparison, someone with years of experience in Unreal Engine can hit the ground running if your game is also in Unreal Engine....
Having a unique engine makes hiring game developers much harder. In comparison, someone with years of experience in Unreal Engine can hit the ground running if your game is also in Unreal Engine.
Having a common framework across multiple companies should also, in theory, help everyone whenever the framework is improved. If Unreal Engine had an update tomorrow that improved performance on Intel graphics cards, every company using Unreal Engine would see a performance boost.
Maintaining your own unique special engine is really hard, and there aren't as many benefits as you'd think since hardware architecture has become more standardized. See also: the browser rendering engine unification under Chrome.
While I do understand why they decided against having their own engine, there are also a lot of problems that come with using off-the-shelf engines. They can be outweighed by the pros, but they...
While I do understand why they decided against having their own engine, there are also a lot of problems that come with using off-the-shelf engines. They can be outweighed by the pros, but they very much exist.
One very big con, one that bites a great many game projects, is that doing anything the engine isn't strictly expecting can range from mild headache to hair-pulling insanity. This also means that while it is true that optimizations may come to your project for free, you can also get deoptimizations and breakages if you've strayed from the beaten path. In practice this tends to result in absolutely massive amounts of internal friction to taking upgrades at all.
I think it often gets underestimated how much time can be lost to the complexity of trying to fit every project into the same mold. Some of the most cursed game projects I've worked on professionally had directives to use certain engines that just weren't a great match for the protect. So much time lost trying to fit mismatching parts. In some cases the mismatches would be so extreme that the source of truth would move out of the tools entirely, often into excel or something, with wacky import scripts to glue it back together. Amusingly, one of the better workflows I've used for developing a game was a tool that vaguely resembled a hierarchal spreadsheet editor for a custom engine (the game was an MMO, a genre notorious for having trouble with off-the-shelf engines).
I think the big reason for the centralization of engine technology actually has a bit less to do with things like performance updates or hiring than people think. I think the primary driver is console ports.
Very salient points, I'm inclined to agree with you on all counts. And it isn't as if alternatives like Godot don't exist. I wrote up a long comment then deleted most of it since who cares....
Very salient points, I'm inclined to agree with you on all counts. And it isn't as if alternatives like Godot don't exist.
I wrote up a long comment then deleted most of it since who cares. Basically talking about monopolization and employee treatment concerns while acknowledging my stance is more ideological than practical.
So ultimately, the realization they're dropping red engine altogether and not just on 2077 was more like a "damn this is the way things are going huh" than "they should do what I want because I know best".
Agreed with nearly everything in your comment, except for this part. Blink (Chrome’s engine) can unilaterally implement features in order to “standardize” them, even when the features have serious...
See also: the browser rendering engine unification under Chrome.
Agreed with nearly everything in your comment, except for this part. Blink (Chrome’s engine) can unilaterally implement features in order to “standardize” them, even when the features have serious drawbacks: WebUSB is a great example of this, as are the DRM extensions that Google has forced through. Coupled with how websites will occasionally perform browser checks (as opposed to feature checks), and Gecko’s (Firefox’s engine) difficulty in keeping up feels moreso to me like a deliberate action on Google’s part to destroy standardization via monopolization, rather than some natural market force which encourages it.
I think you're talking about the outcomes of Chrome's monopoly on the browser market. Its stranglehold is terrible for web standards, and they've obviously been behaving in uncompetitive ways now...
I think you're talking about the outcomes of Chrome's monopoly on the browser market. Its stranglehold is terrible for web standards, and they've obviously been behaving in uncompetitive ways now that they have a borderline monopoly.
As for adoption rates, Chrome's popularity was mostly organic. It only comes preinstalled on Android and Chromebooks. Users seek it out, and now Chrome is in a monopolistic position which they're happily abusing.
On the developer side, I was comparing web engines with game engines because both have major companies that have given up on developing their own unique engine, instead using something off-the-shelf. Microsoft is huge and could absolutely maintain a web engine, but they still moved Edge to Blink. Many companies will weigh their options and decide not to reinvent the wheel or come up with their own bespoke solution.
There's a lot of general pessimism/skepticism over this (and, yeah... I get it), but this demo running on a base PS5 at 60fps with ray tracing on is incredibly impressive now matter how you slice...
There's a lot of general pessimism/skepticism over this (and, yeah... I get it), but this demo running on a base PS5 at 60fps with ray tracing on is incredibly impressive now matter how you slice it -- when they cranked the crowd size up to 300, my eyebrows shot up. It's really good to see some major performance leaps in the UE5 engine because yeah, it's been a big problem this generation.
Also just wanted to point out that the old Cyberpunk 2077 gameplay reveal demo was running on a (at the time) ~$2000 PC. This W4 demo running on a standard console feels a lot more... honest, I guess? I think it's reasonable to expect the final game to look and run a lot like this. We will see though.
This got a bit rambly, but I don't have the brain power right now to redact and polish these thoughts: I want to be so excited for a new Witcher Game as a finally picked up the third game again...
This got a bit rambly, but I don't have the brain power right now to redact and polish these thoughts:
I want to be so excited for a new Witcher Game as a finally picked up the third game again recently to finish it. But coming back to it, I can't shake the feeling that a lot of the complexity from different mechanics gets in the way of the game feeling fun and am worried it's going to be even heavier in the fourth. Between oils, potions, and crafting, I often feel like I'm missing out on playing the game "right". There's just enough that I feel like I need to keep track of that I don't feel like I can lose myself in the game as an immersive experience. I feel like this is more an issue with recent game design more than being a CDPR issue specifically. There's just too much for me personally. Like I love the art, the fighting, and sound design but then I feel like the actual game play gets too many little things show horned into it. It probably sounds counter to the idea that sequels should add more than prior games, but I'd honestly love if they cut out a lot of side mechanics and made the fighting and adventuring mechanics more interesting. I feel like I'm probably in a minority for that sentiment given how complex the systems in games have become.
All that said, I am looking forward to picking it up and continuing the story. I thought it would have been amazing during the third game to switch between Geralt and Citi more actively during the game rather than just a few key points (and only when the game forces the POV switch). I feel like having multiple characters you play as and can switch between dynamically in a game gets underutilized. I think you can create really interesting story telling opportunities. I'm a little disappointed that doesn't look like it's going to be an something they do in the fourth game either, rather it looks like we're following Ciri exclusively.
I really wish I could get Witcher 3 to click with me. I love the story and characters, but the gameplay, UI, inventory management, and controls are so relentlessly clunky that I've never really...
I really wish I could get Witcher 3 to click with me. I love the story and characters, but the gameplay, UI, inventory management, and controls are so relentlessly clunky that I've never really had much fun playing it. I've tried so many times over the years to get through it simply for the story, but the gameplay always loses me.
I feel that so much. I've been able to get past how hard I find the inventory management and crafting. At this point I'm content to check if I'm carrying a better weapon or armor so I can fight...
I feel that so much. I've been able to get past how hard I find the inventory management and crafting. At this point I'm content to check if I'm carrying a better weapon or armor so I can fight tougher monsters, but find it too stressful to 'optimize' my equipment because there's so much to keep track of and it takes so many clicks to move between things. That said, it's such an artful game and that draws me to work past some of the clunky mechanics.
I only played Witcher 3, and only until I rage quit when the Gwent tournament ended in a match that was stacked against me instead of random. But I agree with you. The setting of Witcher was very...
I only played Witcher 3, and only until I rage quit when the Gwent tournament ended in a match that was stacked against me instead of random. But I agree with you.
The setting of Witcher was very interesting to me. The mechanics started out interesting but never felt impactful enough to be fun, just impactful enough to be required. Overall I liked the game but when I think back in it (it's been a while) I remember exactly the things you mentioned.
I really enjoy Gwent despite not being very interested in virtual card games. That said, I've avoided playing it because it feels like a few hours of additional gameplay for an entirely different...
I really enjoy Gwent despite not being very interested in virtual card games. That said, I've avoided playing it because it feels like a few hours of additional gameplay for an entirely different game! Like, they created a great game that honestly stands on its own, but it feels like it takes me away from what I'm trying to actually play the game for. So I've played the minimum amount required otherwise I'll definitely never finish the main campaign.
I'm honestly starting to think that Unreal Engine is killing the AAA games industry, leading to lazy development practices and unoptimized games. Of all the UE5 games to have come out lately, I can only think of one that doesn't run like complete shit and that is Clair Obscur: Expedition 33.
Early versions of UE5 didn't run all that well, but titles are getting better as the engine improves and studios share their best practices for the engine. This tech demo already demonstrates a pretty notable performance improvement over the first batch of UE5 demos running on PS5.
A big reason developers are pushing to use UE5 is because many of its core technologies have huge potential to streamline development workflows while still allowing them to push for bigger and better design. Expedition 33 actually started as a UE4 project, and they pivoted to UE5 mid-development because they felt the time savings and design opportunities afforded by Lumen and Nanite outweighed the amount of work it would take to switch engines.
Cinematic trailer here:
https://www.youtube.com/watch?v=i_aRhNcHryA
Of note, this is just a tech demo, iirc CDPR had some cool tech demos for Cyberpunk 2077 and look how that launch went. Still though, looks pretty impressive!
I'm not too sure, I think the game will end up looking rather close to what we see here, this tech demo is surprisingly tame, and seems to be generally on par graphics wise with games made for last generation running on current generation (which, by now, current generation is almost old enough to be the last generation...).
I'm not really seeing anything visually in the groundbreaking visuals that tech demos usually showcase, the particle scattering, the light rays, the foliage, the movements, they are all pretty tame. Sure, it has some serious art direction and fidelity, but I am not seeing anything that screams 'impossible' or even a step beyond what we have now. In fact, I would say it's a step behind in a lot of ways, take a look at the world (not the cut-scenes that) of recent games like Expedition 33, or even Final Fantasy XVI. These are about on par with each other.
Edit: I would have to say that natural world wise (01:10), it's almost a half-step beyond Ghosts of Tsushima in the visual department, but that would be unfair to compare as that game is a masterpiece.
If anything, they are purposefully being rather tame to either 1) astound us closer to release and/or 2) learned from Cyberpunk 2077. All in all, I am looking forward to this.
Yeah I feel like I'm the only one with a memory when it comes to Tech Demos and video games, and it's not just Cyberpunk/CDPR. I always feel like the tech demos come out look so close to photoreal and has so many immersive details, but then the final game even with maxed out graphics on a decent PC looks far below what originally came out.
I feel this a lot with Unreal 5. I have seen some AMAZING photoreal tech demos from Unreal 5, yet I have not seen or played a game using Unreal 5 that has looked as good as the demos.
At some point I have to ask myself, what is the point of the demos? "Look at what we could do! But realistically will never do."
The tech is cool but the art gives me the ick for some reason, it's like... so generic or uncanny valley vibe, not sure.
It's the Unreal 5 guarantee. It all looks the same, which is that it looks good but not quite.
And it'll run like absolute shit.
Some studios can make Unreal work. Those are the same studios who can make any engine work because they are experts in optimizing.
CDPR, Bethesda, and a few others probably shouldn't be allowed anywhere near this engine. I don't know if it's because it allows for shortcuts or because it requires more optimizing, but we have so many examples of what happens when large, beautiful open-worlds are made in UE5 without the necessary fine-tuning.
CDPR's engine developers are absolute wizards with optimization, they just get thrown under the bus by leadership that pushes to release games before they are ready.
A bunch of the work being demoed here to make Unreal more performant in open world games was done by CDPR engineers.
Oh don't get me started. The performance is downright terrible. It siphons frames from even the newest of gear.
I kind of hate the photorealistic look of UE games. Even when it's done well it's just so bland. It seems like the goal is to create games where if you disabled the HUD and took a screenshot you'd never be able to figure out what game it is.
Hire opinionated art directors and let them cook.
Style over fidelity every time.
The reason Dark Souls 1 still looks good is because stylistically it's knitted together with excellent insight into what works and what doesn't. Dark Souls 2 looks worse because of a focus on contemporary graphic design choices that are now aging.
We can even go back further, Earth Worm Jim still looks great. Yeah you can see it's from a couple of decades ago, but it's pleasing to look at with unilaterally its own style application.
I agree with you, to an extent. Clair Obscur applied their own style to the engine and it did set it apart from the other UE5 games. Not quite as much as to be unrecognizable if you side-by-side it with something like Veilguard, but still a well done application of stylistic vision.
Ultimately it's more the triple-A chase to be as photorealistic as possible without thinking through what style they're looking to implement. UE5 compounds the issue moreso than any other engine did.
I should've been clearer, but yeah I was referring to the default style that comes out of UE games that don't put in the work to differentiate themselves rather than anything the engine forces to happen. Studios can control their graphics pipeline to make it look different, just like other tools can go for photorealism. It's also the fault of GPU marketing and such suggesting that the "next gen" everyone should be chasing means ray tracing and complex hair simulation and other stuff focused on making it look more real rather than unique.
For fun I'm also editing in a conspiracy theory regarding game realism down here. As games, especially AR and VR games, become more realistic do we start running the risk of making the "games make people violent" people right? If it looks real enough will it blur the line between fiction and reality to an unsafe degree for some people?
Oh I wasn't disagreeing with you!
Even stylistically distinguished games still have that Unreal 5, for lack of a better word, sheen on all textures.
Maybe? I think we're still ways off. At this point we're still at the level that it only happens to someone already at the breaking point and/or mentally unstable enough to blur the lines between fiction and nonfiction anyway regardless of the medium.
Unreal is suffering from the same problem as Unity. They've built an engine that is easy enough to work in that anyone with a modicum of programming ability can cobble together a functional game using store-bought assets and ship it on Steam in 6 months. It's inevitable that the first games to ship on new versions of the engine are all samey, poorly optimized crap.
Talented developers can still use it to make visually appealing, performant video games. But they will always come later and in smaller quantities. UE5 first came out 5 years ago, and with AA and AAA development cycles being what they are, it makes sense that we are only just now seeing high quality games that properly utilize the engine.
The game has been in full scale production for less than a year, and this is a tech demo built to demonstrate new engine features to other Unreal developers, at a developer conference. I wouldn't read too much into the art direction outside the cinematic they showed at the start.
I am seeing a lot of Assassin Creed and Horizon (Zero Dawn, Forbidden West) influence on this, well more than I ever anticipated from CDPR. For AC it's the movement, the world, the way the NPCs are reacting and moving, and how the character is generally sauntering. For Horizon it's more visual identity, like the scene starting at 04:14 is something I would have thought was just a recording of a Forbidden West, or Horizon with a 4k texture pack + lighting addition applied.
Edit (added this in another comment, adding it here too): I would have to say that natural world wise (01:10), it's almost a half-step beyond Ghosts of Tsushima in the visual department, but that would be unfair to compare as that game is a masterpiece.
These aren't necessarily bad things, but it's just incredible where you can see the influence so clearly. I am also feeling a little bit of Fable II/III in the mix there, not sure, might just be the chickens running around needing to be chased.
Those games had a lot of Witcher 3 in them so it’s just the ouroboros of adventure game design.
The game looks good and I’m interested to see where Ciri’s story goes, but I’m so fatigued by the Ubislop gameplay loop that I couldn’t get more than 4 hours into Horizon: Forbidden West—the sequel to a game I loved—and I don’t think I’d have a different reaction to this. If they don’t meaningfully evolve the gameplay, or just depart from the wide-open world format in general, I doubt I’d be able to suffer through it. Witcher 3 itself didn’t have enough depth to its traversal, crafting, or combat systems to hold my interest for the full game length, especially once you start getting into the DLCs. So as much as I enjoyed playing through the stories the last third of my time playing it felt like a major slog.
In contrast, I could play Spider-Man and Miles Morales forever. I fully ran out of things to do and still wished there was more to do. Arkham Knight I also think I just started to get to the point of being sick of it by the time I was done. But some of these 60+ hours games are trying to scrape not-enough-butter over way too much bread.
I can't wait to not be able to run this when it comes out. That's not even just a general industry thing, but a CDPR specific pattern. I remember even Witcher 2 being pretty demanding when that came out.
You should look into GeForce Now.
Eh, maybe i'll look into GeForce later…
Maybe y’all should Look into the Heart Now…
It's too early for me to engage with stuff like this and think it's seriously reflective of what's to come.
But another UE5 game? Seriously? Was RED engine too much work to maintain?
Yes. They swapped to UE5 with Cyberpunk. https://www.pcgamer.com/cyberpunk-2077-director-says-studios-switch-from-redengine-to-unreal-engine-5-isnt-starting-from-scratch/
I mean, people complain about bloated development teams, making and maintaining your own engine is exactly that kind of bloat.
Small correction: Cyberpunk still uses RED, they switched after.
Pardon me, in retrospect my tone probably didn't read just as "surprised".
I know I'm in the minority here, but I'm more concerned about UE5 monopolizing the engine space. I wouldn't consider keeping a well-maintained engine as bloat, doubly so if you can license that engine out, but I can see how many would disagree.
Having a unique engine makes hiring game developers much harder. In comparison, someone with years of experience in Unreal Engine can hit the ground running if your game is also in Unreal Engine.
Having a common framework across multiple companies should also, in theory, help everyone whenever the framework is improved. If Unreal Engine had an update tomorrow that improved performance on Intel graphics cards, every company using Unreal Engine would see a performance boost.
Maintaining your own unique special engine is really hard, and there aren't as many benefits as you'd think since hardware architecture has become more standardized. See also: the browser rendering engine unification under Chrome.
While I do understand why they decided against having their own engine, there are also a lot of problems that come with using off-the-shelf engines. They can be outweighed by the pros, but they very much exist.
One very big con, one that bites a great many game projects, is that doing anything the engine isn't strictly expecting can range from mild headache to hair-pulling insanity. This also means that while it is true that optimizations may come to your project for free, you can also get deoptimizations and breakages if you've strayed from the beaten path. In practice this tends to result in absolutely massive amounts of internal friction to taking upgrades at all.
I think it often gets underestimated how much time can be lost to the complexity of trying to fit every project into the same mold. Some of the most cursed game projects I've worked on professionally had directives to use certain engines that just weren't a great match for the protect. So much time lost trying to fit mismatching parts. In some cases the mismatches would be so extreme that the source of truth would move out of the tools entirely, often into excel or something, with wacky import scripts to glue it back together. Amusingly, one of the better workflows I've used for developing a game was a tool that vaguely resembled a hierarchal spreadsheet editor for a custom engine (the game was an MMO, a genre notorious for having trouble with off-the-shelf engines).
I think the big reason for the centralization of engine technology actually has a bit less to do with things like performance updates or hiring than people think. I think the primary driver is console ports.
Very salient points, I'm inclined to agree with you on all counts. And it isn't as if alternatives like Godot don't exist.
I wrote up a long comment then deleted most of it since who cares. Basically talking about monopolization and employee treatment concerns while acknowledging my stance is more ideological than practical.
So ultimately, the realization they're dropping red engine altogether and not just on 2077 was more like a "damn this is the way things are going huh" than "they should do what I want because I know best".
Agreed with nearly everything in your comment, except for this part. Blink (Chrome’s engine) can unilaterally implement features in order to “standardize” them, even when the features have serious drawbacks: WebUSB is a great example of this, as are the DRM extensions that Google has forced through. Coupled with how websites will occasionally perform browser checks (as opposed to feature checks), and Gecko’s (Firefox’s engine) difficulty in keeping up feels moreso to me like a deliberate action on Google’s part to destroy standardization via monopolization, rather than some natural market force which encourages it.
I think you're talking about the outcomes of Chrome's monopoly on the browser market. Its stranglehold is terrible for web standards, and they've obviously been behaving in uncompetitive ways now that they have a borderline monopoly.
As for adoption rates, Chrome's popularity was mostly organic. It only comes preinstalled on Android and Chromebooks. Users seek it out, and now Chrome is in a monopolistic position which they're happily abusing.
On the developer side, I was comparing web engines with game engines because both have major companies that have given up on developing their own unique engine, instead using something off-the-shelf. Microsoft is huge and could absolutely maintain a web engine, but they still moved Edge to Blink. Many companies will weigh their options and decide not to reinvent the wheel or come up with their own bespoke solution.
There's a lot of general pessimism/skepticism over this (and, yeah... I get it), but this demo running on a base PS5 at 60fps with ray tracing on is incredibly impressive now matter how you slice it -- when they cranked the crowd size up to 300, my eyebrows shot up. It's really good to see some major performance leaps in the UE5 engine because yeah, it's been a big problem this generation.
Also just wanted to point out that the old Cyberpunk 2077 gameplay reveal demo was running on a (at the time) ~$2000 PC. This W4 demo running on a standard console feels a lot more... honest, I guess? I think it's reasonable to expect the final game to look and run a lot like this. We will see though.
This got a bit rambly, but I don't have the brain power right now to redact and polish these thoughts:
I want to be so excited for a new Witcher Game as a finally picked up the third game again recently to finish it. But coming back to it, I can't shake the feeling that a lot of the complexity from different mechanics gets in the way of the game feeling fun and am worried it's going to be even heavier in the fourth. Between oils, potions, and crafting, I often feel like I'm missing out on playing the game "right". There's just enough that I feel like I need to keep track of that I don't feel like I can lose myself in the game as an immersive experience. I feel like this is more an issue with recent game design more than being a CDPR issue specifically. There's just too much for me personally. Like I love the art, the fighting, and sound design but then I feel like the actual game play gets too many little things show horned into it. It probably sounds counter to the idea that sequels should add more than prior games, but I'd honestly love if they cut out a lot of side mechanics and made the fighting and adventuring mechanics more interesting. I feel like I'm probably in a minority for that sentiment given how complex the systems in games have become.
All that said, I am looking forward to picking it up and continuing the story. I thought it would have been amazing during the third game to switch between Geralt and Citi more actively during the game rather than just a few key points (and only when the game forces the POV switch). I feel like having multiple characters you play as and can switch between dynamically in a game gets underutilized. I think you can create really interesting story telling opportunities. I'm a little disappointed that doesn't look like it's going to be an something they do in the fourth game either, rather it looks like we're following Ciri exclusively.
I really wish I could get Witcher 3 to click with me. I love the story and characters, but the gameplay, UI, inventory management, and controls are so relentlessly clunky that I've never really had much fun playing it. I've tried so many times over the years to get through it simply for the story, but the gameplay always loses me.
I feel that so much. I've been able to get past how hard I find the inventory management and crafting. At this point I'm content to check if I'm carrying a better weapon or armor so I can fight tougher monsters, but find it too stressful to 'optimize' my equipment because there's so much to keep track of and it takes so many clicks to move between things. That said, it's such an artful game and that draws me to work past some of the clunky mechanics.
I only played Witcher 3, and only until I rage quit when the Gwent tournament ended in a match that was stacked against me instead of random. But I agree with you.
The setting of Witcher was very interesting to me. The mechanics started out interesting but never felt impactful enough to be fun, just impactful enough to be required. Overall I liked the game but when I think back in it (it's been a while) I remember exactly the things you mentioned.
I really enjoy Gwent despite not being very interested in virtual card games. That said, I've avoided playing it because it feels like a few hours of additional gameplay for an entirely different game! Like, they created a great game that honestly stands on its own, but it feels like it takes me away from what I'm trying to actually play the game for. So I've played the minimum amount required otherwise I'll definitely never finish the main campaign.