The Matrix Is Unreal
fxguide.com
fxguide.com
Here’s a great deep dive into some of the new technologies that highlights the strengths and weaknesses of some of the new engine tools: https://youtu.be/ib6_c6uliLg
This is extremely exciting this is possible on a year old console and not a 3090.
The price of keeping up with every new enhancement on PC is directly linked to each new engine update, but somehow game releases can be "optimized" for gameplay on ~3 year old consoles... Also why I haven't bought any new PC games since GTA5.
It is a bit of a pain having to monitor stock drops but most online stores have a queue system in place as well as purchase limits to try and make life harder for people that buy up every console they have to just put them on eBay at twice the price.
A colleague of my wife managed to get a PS5 in a fnac store a few weeks ago as well. Just went in randomly and asked and they had them in stock. Only a few and they sell out quick so was just luck in that instance but if you put in a little effort you can get one in a week or two (depending on stock allocation of course) without having to pay over the MRSP.
Purpose built gaming hardware has significant advantages over a typical PC. Modern consoles have SoC designs with incredible bandwidth between CPU/GPU/RAM that you simply can't have on upgradable form factors.
The closest thing in the general purpose computing world are the M1 family of SoC MacBooks. UMA on the M1 in particular could be harnessed to have physics and graphics geometry in sync with each other without the penalty of bus transport - so things like high-detail deformable environments become much more feasible. But again, even these machines are geared towards general purpose computing, not graphics.
So where's the PC platform with these advantages? I mean, most PC players will replace their whole system instead of upgrade individual components (anecdotal / speaking for myself). Where's the PC-like-a-console?
I guess the Steam Deck will be the one to look out for. But I would also expect gaming laptops to have those advantages.
You need the CPU, GPU, motherboard, and RAM to be one product shipped by one company, but instead you've got a CPU from Intel, a GPU from AMD/Nvidia, a motherboard from the PC manufacturer, and off the shelf RAM.
"Gamers" want an Intel CPU because of the branding (although this is shifting towards AMD now), they want a particular graphics card because of branding, they may even care about the specific motherboard and/or RAM.
Apple are able to do this console-like integration and performance because their target market don't care enough about each individual component, the Mac is the product and the brand they get behind. PS5 owners aren't buying it because of the brand of CPU.
I think the speed advantage would be one welcomed by the average PC user, and if Intel aren't already colluding to make an answer to M1, they're heading towards extinction
Would this demo not run on a pc?
I understand that pc hardware is more expensive but it was also my understanding that performance ceiling is actually much higher than consoles, especially after the next gen of pc hardware comes out after a console release.
It’s hard to tell how much more expensive a pc is because of console manufacturers’ loss leader accounting.
With UE5, it's the engine that tunes the rendering quality for the HW available.
about that - screen space ray tracing has been around a while, and even if you miss little things (no diffusion from lights off screen or behind a shade) the resulting eyecandy is often enough to trick people mind even if it's not physically correct... and three generation older hardware is more than enough to run that.
Fwiw this demo is on another level than call of duty. The world feels huge, extremely busy - there’s cars and people everywhere - and so varied.
Indeed, but the truth is it's just a demo, now when you add all the rest that's needed to make a real game you start to have issues that woudn't have with a demo.
7318 people (4132 developers, 3186 thanks)
https://www.mobygames.com/game/playstation-4/red-dead-redemp...
I never had a ps4/xbone, skipped the last console cycle and stayed on PC then recently got a PS5 where every title has an option to play at 60fps. There is no way in hell the consoles are going to be able to push this fidelity at that frame rate. Let alone their promise of 120fps that has been floating around.
And the tech like nanite & lumen. It's not as if you can just "turn it off" and pick up a bunch of performance. It seems like you have to develop your game with these tools in mind.
In general I think the new consoles are going to have some trouble, even today games don't render at native 4k, they render much lower then get upsampled, and dynamic resolution is also in place to pick up the slack. The promise of 4k/60fps on a console is kind of smoke and mirrors.
The 'golden age' arcade games of the early 80s mostly ran at 60fps. And this trend continued in the arcades, with the most impressive early-90s 3D games also choosing framerate over detail, which was absolutely the right decision for Daytona USA, Sega Rally, Ridge Racer, and more. A solid 60fps pretty much defined 'arcade quality'.
It's understandable that early 3D home consoles had to aim a bit lower. But we're now on the 5th generation of 3D home console, and we're still making excuses to run at lower framerates than games of the 80s/90s?
But 4K, that just game too soon. Every recent generation of console has struggled to keep up with the rate of screen resolution increase. And 4K was a huge jump, from the 2 million pixels of 1080p to around 8 million pixels.
We could have been happy with 1080p for a lot longer if we'd had solid 60fps gaming with a bit more antialiasing, and streaming services with higher bitrates.
High-DPI screens are great if you're reading text off them at close range. But not really needed at a normal TV-watching or gaming distance.
I think 30fps is a straight up usability issue. The only times I've ever had motion sickness or a headache from games is when the game is 30fps (though not all 30fps games gave me these issues)
I hope the performance/quality modes become standard this generation.
I think you're wrong about frame rate though. I find 30fps unplayable. Also, people have had a taste for it. They have been playing 60fps titles since launch on the next gen consoles, it makes the frame rate all the more strange when you play something running at half that.
Not sure what you are talking about with games using unreal engine. It's also used for shooters, Gears of war is one of the flagship unreal titles. I'm sure the folks at unreal don't want people to think "the engine isn't for shooters".
Especially considering the engine originated from shooters -- Unreal and Unreal Tournament.
The fact that this runs in real time, at 30fps, on a $400 console is mind blowing. It's even able to downscale to a smooth 1080p/30 on the Xbox S. You would not dream of achieving this level of fidelity 5 years ago, there is nothing out there that compares. Nanite & lumen are the reason it performs so well, not the other way around.
On top of that, the promise of 4k/60 on PCs still hasn't been met. You can spend $3-4k and still barely achieve that with a 3090. This engine might be the first to actually get there.
Sure my AUD $6K PC is faster (with a multi core overclocked Ryzen, and 3070, fast memory, PCIE4 NVME storage, premium case/mobo/power/cooling), but at $750 what you get blows my mind. You'll struggle to match it at 3x the price in a custom built PC, and that was with pre hike GPU prices.
Also while i agree on principle with you, this is still a movie, and one of a trilogy that defined a significant part of culture i might add, so it's a risky move at best.
But yeah, besides facial quality(which i think are vastly underwhelming) the movie looks pretty nice, and the rest doesn't even matter when you consider most of movies/shows today are VFX.I remember seeing the first ever UE5 tech demo years ago(i think it predates ray-tracing if i'm not mistaken) and i was kind of surprised.
"Meh" is the maximum level of emotional response that can be elicited.
The invention of literal time travel, the invention of AGI, or the manufacture of a black hole as an energy source would all be met with mehs. Just a bunch of people dead inside. I still come here for the links and the 5% of comments that are insightful and not pedantic.
Is this impressive? Yes & kudos to the people who built this. Is it life changing in any meaningful measure? Not really...it isn't a scientific breakthrough or complete paradigm shift of understanding that will the lens which you view your existence? Probably not. And expectations of this already happening were built into how I see the steady march of technology development.
This is an incremental instead of step-function change in expectations and was already a pre-ordained outcome. You can argue that my expectations are to high and maybe this came earlier than expected but this isn't some shock to my system.
I suspect that I am not the only one who feels this way.
And if you are part of the team and just feel like people don't appreciate you work - that isn't the case. It is very impressive but it isn't a eureka moment of excitement (at least to me).
If I misunderstand the significance - please guide me to what I am missing here.
Nothin is pre-ordainer for sure, that said there certainly seem to be fairly clear directions in technology. I think what was missing from my understanding is that this is a much more efficient use of resources for a better engine. Sounds like a leap forward in graphics engine (now developers will be captured on the Unreal Engine - great business model for the company).
It still isn't a eureka moment that the other poster was really hoping for. Its very impressive but I am not going to swoon over the achievements. That doesn't make me "dead inside" because I can't get fully amped up for this post and I think that the poster missed how people might view look at this development which is why I added my own comments.
The motion version definitely still exhibits uncanny valley, though—lights that are too hard on flesh, tightness and sharpness where there should be softness. That I have to resort to hand-wavey descriptors is cool; they’re getting close! But crossing uncanny valley is HARD.
What this demo convinces me of is not that we’re there; we are not. Claiming that we are is disingenuous, and makes me distrust the claimant. But it makes apparent that we will get there. And that’s impressive.
I d wait until it s in my hand and I touch the tv trying to open the window for Keanu before I get excited :D
It looks quite good. Not reality but the cityscape is pretty close..
The fun part is the end where they show the some of the secret sauce of how it was rendered.
I wonder if CG models the vertebrae yet? Tough to motion capture, but a huge part of why humans don't move like stick figures.
The first part is a CGI representation of Keanu and Carrie-Anne. I was shocked at the quality; it had “the demoscene feel,” if you know what I mean - “I did not know the hardware could do that.” Artists going full OCD, just because it was the right thing to do. Great work.
Parts of it had that CGI feeling, but most of the first two minutes, I swear I was watching video and not CG.
Part 2: the city tour. I am not sure if there’s a detailed YouTube video of same, but the artists used some clever tricks, as any good 3D artist would. You can fly around and peer into offices and apartments through their windows. There are a limited number of static rooms, and they use frosted glass to blur the details. Interiors are fully modeled and there is parallax through the glass. HVAC units on rooftops are robust, down to the warning messages. Great use of LOD.
I saw Z-fighting and LOD “popping” when surveying the city from on high.
I walked away from this (um, got up from the sofa) thinking Epic’s gambit is to push Unreal Engine 5 as a virtual movie set.
They went all out on the character models in the first part, with Keanu and Carrie-Anne. There were a few uncanny valley shots (“dead eyes”) but very impressive overall. Some of the shots could be taken as a reference, or gold standard.
Even more impressive is that it’s running on a (loss leader) $600 Sony PS5.
The demo runs capped at 24fps? That has to feel almost unplayable in an interactive setting? A dip below 30 in an interactive first person perspective is usually pretty disturbing.
I think they have been doing this for quite some time already:
PS5 is 499 btw and there's even a 399 model (digital edition).
Disappointing to see there's no PC version (yet) although I hear it's in the works. Would love to put my RTX 3080 to the test.
Overall very good graphical quality demo, would love to see this on a maxed out PC without so much scaling/a higher framerate though.
Objects are often absolutely beautiful and realistic, but interactions between objects are ridiculous. Some of this is probably just "not enough processing power", but it is also due to where the rendering bumps up against the primitive controls. For example, you have a very realistic human. But then the human walks against an object, and she does this standard video game 0 mph duck-walk. Part of the problem here is that real humans don't bash into walls. So how do you realistically represent that, when in fact it's just Johnny pressing forward on his controller?
That’s far from an unsolvable problem though, and this is a graphical tech demo, not a full game with the associated staff, budget, time, etc.
The Demon’s Souls remake (which I think actually looks nicer) has tons of destructible boxes, tables, stone work, etc (the original had much of this as well). I’m more interested high graphical fidelity being used in these smaller, authored spaces than in massive, often empty open worlds.
Filling up a small map with “real” content by hand is hard, doing it on a huge scale is impossible.
I suppose the tech can’t get across the valley without first passing through it
There's a lot of uncanny CGI in modern movies. There's also a lot of CGI that is so perfectly realistic you have no idea it isn't real. This is usually more environmental and less characters.
If the tech gets past the uncanny valley then you'll never know. This is basically true by definition. :)
No they’ll tell me
(And I’ll probably not agree, but if I do they’re still going to tell me, trust me they haven’t stopped telling me)
Similar to ray-tracing, where you get great realism, but it is just an absolute resource hog. The number of vertices in soft-body physics animations come to mind. The real world has curves and deformations, and hair, with millions of strands, that each need tens or hundreds of physics-enabled vertices to look ultra realistic. Walking through a city, you might have 100 NPCs on the screen at once, who all have hair.
You simply need so much freaking compute power to pull this stuff off, I can't wait to see what it looks like in 10-20 years.
This is why I'm most excited for high-core counts leading to game development that allows these kinds of things to run in their own hardware threads. Imagine a world where each of those 100 NPCs has hair simulated on one thread, clothing simulated in another, world interaction on another, even AI generation on its own thread.
It's crazy just looking 10-20 years back and seeing where we are today. Compare this tech demo to Skyrim (2011) or GTA III (2001) and it's just incredible that we're looking at photorealistic, ray-traced, real-time "gameplay" like this on a home video game console. Or just compare GPU specs, the GeForce3 (2001), GTX 580 (2011), and RTX 3090 (2021) all have texture fill rates of 1.92, 49.41, and 556.0 GTexel/s respectively, with similar increases in raw floating-point performance. The future is exciting.
Oh, wait...
Also the acceleration of people and cars feels still too artificial. It is not the looks, but the physics that are still full on the uncanny valley.
I think as graphics get more and more realistic looks wise, game designers need to slow down the game, and make it feel more real.
One of my favorite car chases is from the movie Ronin. It feels extremely real, like you are there, as they didn't use any outlandish effects, but kept it grounded on how a real car chase would feel.
Super impressive demo though.
You could animate that as hugging the wall, leaning against it or something, instead of a 0mph walk.
It is hard to stress how good and impressing that game was at the time.
That doesn't seem like much of a theoretical obstacle to me. The primitive controls only signal intent - you don't have to animate the human shuffling against the wall, you can just not move. The controller should be like whispering into the avatar's ear, not QWOP.
Or if you do want to grant the player lower-level physical control, you can have the human actually collide realistically with the wall. After all, people do bash into walls, just not commonly. You can even make them bump their head, and take damage.
I spent a lot of time on both and read a lot about the issue, and most end up skipping physics on the PC, because it just feels worse controlling it realistically. This might be because we're used to that though. Anyway after all, you're making a game and games have to be fun to play.
E.g. the goal was probably to show off the new Nanite and Lumen rendering features, and probably MetaHuman for the main character faces and facial animations. At some point you have to say "ok, we focus on this stuff, and accept compromises for other things". If the team's goal would have been to create the best crowd simulation or environment interaction ever, they would probably have achieved this, but at the cost of other things.
Carrie Anne Moss, who on-camera looks stunning for her age....looks like she aged at least twenty years and her head was stretched, almost like they got the aspect ratio settings wrong. She's almost unrecognizable.
When they're in the car and the camera pans back to "IO" in the back seat, I felt like I was watching Who Framed Neo Rabbit.
I seem to recall the game "Bound", where you guided a low-poly ballerina through fairly abstract levels, had a really nice solution to "people don't bash into walls": if you tried to guide your character into a wall, she'd just start doing some stretches against it, which feels pretty accurate given how I tend to behave when I'm attending dance classes regularly. :)
Other characters might do different things; a mopey teen might slouch against the wall, a gun-haver might press against it and peer to the side, looking for cover, etc.
Obvious reasons not to do this start with "oh great, one more thing to add to the character state machine and debug" and "oh great, a few more animations to fit into the budget". Maybe also "it stops feeling like you're controlling the character". Someone whose day job is video games probably has at least a dozen more reasons, possibly even stories from when they tried it and abandoned it.
It's definitely possible to "not bash into walls"
You don't always need to have bespoke animations- sometimes keyframes and realistic constraints can get you there. I think Unity is working on a framework for natural human animation.
David Rosen pulled off quite a few little masterpieces -- his gamedev competition entries always had fantastically unique aspects. I cannot believe "Black Shades" was almost 20 years ago! (I have no affiliation, but I am glad to see https://www.wolfire.com/games is still alive and kicking!)
Lugaru and later Overgrowth have been the outcomes of what can only be called a decades long obsession with getting the player controls just right, allowing very natural and responsive control and movement, through creating emergent behaviors from first principles, resulting in a rich and natural interaction with the environment (and other characters, see the fighting mojo), in contrast to the many, many AAA titles that just constantly break the immersion through unnatural, jarring behaviors and frustrating limitations.
It's significant too: nothing frustrates players quicker then controls feeling "unresponsive" by doing something that's different to what they instructed.
Which is what some game does fundamentally bad. When using skills in some game. Your action got locked completely and won't respond to your request until animation finishes. (And when the mob hit you in the middle, you can't do anything because animation is playing.)
But some games makes more efforts into these and allow you to transition to another action in the middle of one, blend them seamlessly or even play them at same time. That makes the game feels much more responsiveness and enjoyable. Feels you are actually moving your body instead a online character with high action latency.
Is that really part of what uncanny valley represents now? I feel like we're moving the goal posts. Fifteen years ago the characters really could be uncanny and disturbing. I don't get that anymore. I feel like it's very close to real, but not real. However, it's in no way off-putting or weirdly creepy.
Watching the demo, I find it very impressive. Still in the valley, they key word being "still" - I think they managed to get past the bottom and are now climbing up.
Aside from the motion though they look incredible.
"Even now people are confused, having watched the footage several times. “The first time you see the older Keanu that was a video shoot that we (Epic) filmed of him,” Pete explains. “After they come back after Morpheus that was all the CG versions. It was color corrected to match and look the same.” There is also one point where we see footage in the mirror this was real film footage that was comped in UE5. The first shot of Morpheus is also not CG, but then becomes CG when the young Thomas Anderson (Neo) appears. Even the desk scene at the beginning when Anderson is asleep was a real-time recreation of a scene from the original film."
Still impressive technically and visually. Let's not forget that these things are running in realtime, they aren't a video. the whole "uncanny valley" thing is exaggerated. in the next 5 years you won't be able to tell the difference. In fact, you're brain will be constantly asking itself whether what you see is real or a 3D model. And I bet it did if you played the demo. That's a serious leap forward.
Now most videogames will still featured heavily stylized/cartoonish models, because some people are afraid of models being too realistic.
When watching it I realized what you just wrote.
This is really impressive.
I get this in movies. Sometimes I even get this in scenes where they're using real actors and not CGI. Harsh lighting with make up can make humans look somewhat fake to me. One of the Mission Impossible movies gave me this effect.
At this stage I feel like the engine is very good and sure processing power is still a factor, but I feel like the biggest factor is cost/time. Small details that make something look natural are sure to go when there is budget or deadline.
Edit: here is a short video about the full size animatronic T. rex. https://youtu.be/fFTsYGgdR9k
Who cares? It's a video game, that's how they work. I'd rather have developers work on making things fun than trying to make a creepy realistic Keanu not walk into walls
This is what blew my mind the most with this demo, and I think will enable some crazy cool evolution in open world gaming.
Like in that video where you had to count how many times people passed the ball, but failed to notice a gorilla walked through the scene.
Let me first state that its amazing overall, very impressive.
Having said that, it exhibits pretty noticeable pop-in, also quite a few instances of what looks like garden variety z-fighting of textures/shaders in some places.
In a way these actually jump out at you a bit more in this demo than they do in less photorealistic rendered environments probably for reasons similar to the uncanny valley phenomenon being discussed throughout this thread.
The car chase sequence is mind blowing and then you get to free roam the city by foot, car or even drone. Nanite and lumen are something special.
The number of cars around the city is very impressive, until you realize a lot of them are non-interactive static props. It's a tech demo, and I doubt the game itself will reflect much of what you see here.
Even the free roaming is impressive, the whole scene is lit by the sun. They recognized there are still things to work out like the car collisions that you mentioned, all the cars are rendered with nanite until you collide with them, at that point they are being swapped to "traditional" models.
I mean for a tech demo it even has way more consistent pedestrian and vehicle simulation consistency even a way long distances than AAA games like Cyberpunk 2077.
That is some amazing tech! It actually feels like Unreal Engine would be really well suited for other genres of games, not just the historically prevalent FPSes, such as creating RTS games or maybe even something like Cities: Skylines, in which being able to create large detailed urban environments could be amazing!
The Verge also did an interview with Keanu Reeves and Carrie-Anne Moss on the process of creating the demo which is quite interesting to watch, too [2].
[1] https://www.unrealengine.com/en-US/blog/introducing-the-matr...
[1] I am aware of the fact that "back in the day" they used three 580s in SLI or something crazy to run it, but that performance was already surpassed two years later by single card setups.
For the technology itself, it is incredibly impressive but more than anything and something which is hard to demonstrate with a demo, is how easy it is to actually use. Nanite models are simple to make from 3d objects, this with photogrammetry would be a dream.
Pragmatism dictates that, even if reality is a simulation, who cares, it's real enough for now.
Why would data, which can be replicated, isn't limited to one body, and doesn't care if it lives or dies, be anything like a human on the inside?
I suspect they used this technology to make the movie, or parts of it. Why else would this be made?
For example: UE was used a lot for The Mandalorian (https://arstechnica.com/gaming/2020/02/the-mandalorian-was-s...).
https://en.wikipedia.org/wiki/LazyTown#History_and_productio...
"Uncanny valley" is the usual term - https://spectrum.ieee.org/what-is-the-uncanny-valley
Next I want to see what kind of tech demo they can make run smoothly on a mobile device at 120Hz. I.e. does the tech scale downward too? Or is there no point to it unless you require extreme scale and fidelity.
Like companies such as Tele Atlas did back in the days for map data (when they didn't have much more than that I guess).
https://www.gamesindustry.biz/articles/2019-11-12-epic-acqui...
Edit: well not of entire cities, more about 3D assets. The city itself was built procedurally using Houdini.
But I dunno. I hope advances in videogame engines do more than just help the biggest AAA titles improve their photorealism.
What do we get as the technology improves beyond the point shown in this demo? Just better graphics, or a democratization of the production of photorealistic games? Does this somehow open the door for smaller teams to create games that involve photorealism, or for more budget and staffing resources in the development of photorealistic games to be directed toward other aspects of the game?
People have been saying that for years, and then the following year comes along that blows away expectations yet again. IMO, I won't be satisfied until I literally can't tell an artificial rendering from a real photo/video.
Yes, less dealing with LODs using nanite, less worry about drawcall count, less budgeting tri count. And Epic provides huge photogrammetry scan libraries that work well with it.
Yes, definitely. I kinda wish more resources would be diverted to improved physics, better procedural generation, and especially more realistic AI. When do we get to play a game against AI that is learning from the player and constantly changing its behavior in a truly dynamic way? Or how about a game where you can move and interact with every physical object you see instead of everything being a bolted down prop except for that jarringly different item that you're obviously intended to interact with? That would be orders of magnitude more exciting to me than hyper realistic hair on the player models.
Then we get to the game portion of the demo, and... it's a half-step above a quicktime event. Now, look, yes, I get it, this isn't actually a game, and the game is an afterthought and not the main point. (By that, I mean the game itself, not the cars, not the agents jumping around, but just the underlying mechanics.) Still, consider where we've come from. Think Atari 2600 graphics and the sheer staggering number of orders of magnitude improvement in the graphics since then. Even if they did have tons of people working on this, all those people were working at levels of efficiency undreamed of in the 2600 era, by almost any objective measure you can imagine. (Even though we'd never be able to agree on one.)
Yet the game itself used in this demo is basically something you could fit on the 2600. I mean, the game itself, the mechanics. Obviously not the graphics, or even a facsimile thereof, but just the game's mechanics. It's still "point at the thing you want to blow up". No generalized conversation engine. No generalized music engine. If you'd like to be generous, call it an order of magnitude better than the 2600 could handle.
The gulf between the graphics and the games we have with them keep getting larger and larger, and it doesn't help that in a lot of ways the AAA games have been boxed into what the engines do. It's not just that "open world sandboxy combat" is the only game they want to make... it's that these huge amazing engines almost require that sort of game, because that's what the engines do. The sameyness of AAA games is forced by this graphics focus because trying to do provide anything but the simplest game mechanics with this level of graphics is impossible.
Which is why I've basically abandoned AAA and go indie now. The lower graphics capability turns out to be a positive advantage, because it means they can make games that aren't The AAA Engine Game.
Yeah, was my impression of the gameplay we saw, and my worry about whether it will encourage that kind of design.