See the incredible photorealism in Ninja Theory's upcoming Project: Mara

midian182

Posts: 6,417   +56
Staff member
Something to look forward to: Those of a certain age may remember when the idea of photorealistic games was just a fantasy. Today, that vision has become close to reality, and Project: Mara, the upcoming horror from Hellblade developer Ninja Theory, could be the most realistic-looking title to date.

It's been almost exactly a year since Microsoft-owned Ninja Theory announced Project Mara, which it describes as a "real-world and grounded representation of true mental terror." The company wants to recreate the "horrors of the mind as accurately and realistically as possible."

In the latest developer diary video (top), Ninja Theory's creative director Tameem Antoniades reveals the incredible attention to detail that went into recreating the game's spacious city apartment, based on a real-world location.

In its efforts to create an exact digital replica, the team first analyzed and scanned materials from the apartment—a process that took several months. The challenge was then to ensure the dimensions of the in-game version are precise. This was achieved through a firm called Clear Angle Studios, which used LiDAR scanners to create a "point cloud" that allows "some of the best references we've ever seen."

Ninja Theory hasn't just recreated the obvious elements; it even created procedural tools to generate the floor's lint and debris. "Artists aren't there just to create an object, they're there to create systems that can create an object and infinite variations of that object," said Antoniades.

Hopefully, all this detail will translate into a good game. No word yet on a release date, but we should hear something soon.

Permalink to story.

 

Mister_K

Posts: 2,028   +713
Pretty neat! Future of cinematography is most likely digital, specially if you can recreate sets completely in VR in photorealism and manipulate conditions to your liking (Mandalorian for instance is getting closer to it). I know, I know, green screen but its not the same as in-engine.

Guess it will become like film stock, where most productions are using digital cameras some still use 65mm.
 

Uncle Al

Posts: 7,965   +6,731
Great article. Simple and to the point. And I believe these folks are certainly setting the new "standard" for the industry much like comparing the first editions of DOOM and what we see today. Great stuff!
 

Athlonite

Posts: 202   +63
Blah blah I've heard this same thing over and over again since the early nineties where the game dev industry has made claims of producing photo realistic textures and it has never come to fruition so I wish them luck but I highly doubt it will be used in anything within 10 years
 

Markoni35

Posts: 905   +365
Even our "real" world is not very realistic. I've found tons of problems in the physics engine of the universe. For example:

1. Everything is quantized, from particles to energy. Probably even time is quantized. That means it's made of finite number of levels or states. It's a clear indication of gaming engine. Real world would be smooth to infinity.

2. Quantum tunneling is obviously the result of collision detection errors. Probably the time slices are too long, resulting in particles passing through each other between consecutive time slices. They should probably shorten Planck time.

3. All subatomic particles of the same type are identical. Obviously they used the same predefined templates, instead of allowing them to be customizable, or made of smaller bits. But this definitely reduced the memory usage.

4. Implementing asymptotic limitation of the maximum speed (so called "speed of light") is a pretty rude way of limiting the map size and required LOD. Resulting in significant compute and memory savings, but at the same time ruining the illusion of living in a real world. In a real world adding energy would increase the speed beyond any artificial limit.

5. Replacing particles with statistical values is a nice way to reduce compute load, but unfortunately it ruins the illusion of the real world, at least for the physicists.

6. Superconductivity is yet another side-effect of simulation. In a world with infinite precision, some kind of resistance would always exist, and there would be no superconductivity. But with a finite-level quantization, when the energy levels are below the quantization limits, the resistance just disappears.

So, I'd like to ask for an upgrade of our "real world". This must be some early preview version. Let's upgrade the hardware, because with the development of science, the players are becoming more and more demanding.
 

havok585

Posts: 255   +97
Even our "real" world is not very realistic. I've found tons of problems in the physics engine of the universe. For example:

1. Everything is quantized, from particles to energy. Probably even time is quantized. That means it's made of finite number of levels or states. It's a clear indication of gaming engine. Real world would be smooth to infinity.

2. Quantum tunneling is obviously the result of collision detection errors. Probably the time slices are too long, resulting in particles passing through each other between consecutive time slices. They should probably shorten Planck time.

3. All subatomic particles of the same type are identical. Obviously they used the same predefined templates, instead of allowing them to be customizable, or made of smaller bits. But this definitely reduced the memory usage.

4. Implementing asymptotic limitation of the maximum speed (so called "speed of light") is a pretty rude way of limiting the map size and required LOD. Resulting in significant compute and memory savings, but at the same time ruining the illusion of living in a real world. In a real world adding energy would increase the speed beyond any artificial limit.

5. Replacing particles with statistical values is a nice way to reduce compute load, but unfortunately it ruins the illusion of the real world, at least for the physicists.

6. Superconductivity is yet another side-effect of simulation. In a world with infinite precision, some kind of resistance would always exist, and there would be no superconductivity. But with a finite-level quantization, when the energy levels are below the quantization limits, the resistance just disappears.

So, I'd like to ask for an upgrade of our "real world". This must be some early preview version. Let's upgrade the hardware, because with the development of science, the players are becoming more and more demanding.
lay off the shrooms, bud.
 

Avro Arrow

Posts: 794   +906
TechSpot Elite
That level of detail is amazing. I can just imagine the GPU horsepower that will be needed to render this environment. Photorealistic environments would put a crippling load on the GPUs of today.
 

Avro Arrow

Posts: 794   +906
TechSpot Elite
Even our "real" world is not very realistic. I've found tons of problems in the physics engine of the universe. For example:

1. Everything is quantized, from particles to energy. Probably even time is quantized. That means it's made of finite number of levels or states. It's a clear indication of gaming engine. Real world would be smooth to infinity.
Who says that real world would be smooth to infinity? How would you know this if you've never experienced it? Maybe you just have digital eyes in an analogue world. Or at least, that's what they're hoping that you'll think. :D
2. Quantum tunneling is obviously the result of collision detection errors. Probably the time slices are too long, resulting in particles passing through each other between consecutive time slices. They should probably shorten Planck time.
Yes, obviously. And déjà vu is just a glitch in the matrix. :D
3. All subatomic particles of the same type are identical. Obviously they used the same predefined templates, instead of allowing them to be customizable, or made of smaller bits. But this definitely reduced the memory usage.
Well sure, but is that reduction in memory usage worth it if people are able to tell the difference just by wearing a tin-foil hat? Those developers were pretty sloppy and lazy if they were actually trying to create a convincing environment. It just would have required a little bit of imagination. I would've hoped that anyone who could create this place had a measurable amount of that. :D
4. Implementing asymptotic limitation of the maximum speed (so called "speed of light") is a pretty rude way of limiting the map size and required LOD. Resulting in significant compute and memory savings, but at the same time ruining the illusion of living in a real world. In a real world adding energy would increase the speed beyond any artificial limit.
Yes, especially when some of their subatomic and quantum particles seem to violate this law with real effects on temporal causality that we don't yet comprehend. It looks like someone forgot to "carry the two" when they were setting up the existential calculations for them. They probably thought "Bah, who cares? It's not like the subjects can see these particles anyway!" which is typical of a developer who has been given an unreasonable amount of time with which to push out a properly-done simulation. Those damn capitalists! Only they could prioritise profitability over doing it right the first time! :D
5. Replacing particles with statistical values is a nice way to reduce compute load, but unfortunately it ruins the illusion of the real world, at least for the physicists.
Yeah, but that's a small part of the population and they've modified American culture to distrust anyone with a university education, let alone actual PhD degree holders like physicists. They just spout more and more insane theories on InfoWars for the more dim-witted masses to consume, and they do.
6. Superconductivity is yet another side-effect of simulation. In a world with infinite precision, some kind of resistance would always exist, and there would be no superconductivity. But with a finite-level quantization, when the energy levels are below the quantization limits, the resistance just disappears.
Well, even superconductors have some resistance. I mean, the only superconductor element that I can think of in which the level of resistance is statistically zero would be metallic hydrogen. Unfortunately, metallic hydrogen can only form at extreme pressures that would be fatal to life as we know it. It could only occur naturally in our solar system on gas giants. I think that the most likely would be Uranus or Neptune because I think that Saturn and/or Jupiter might be too hot to allow hydrogen to properly crystalise.
So, I'd like to ask for an upgrade of our "real world". This must be some early preview version. Let's upgrade the hardware, because with the development of science, the players are becoming more and more demanding.
I know, eh? I saw that the Blue Jays just signed George Springer to a 6-year deal for $25,000,000USD per year. The players really ARE getting demanding! That's just nutz!!! :D
 

toooooot

Posts: 1,418   +693
We ve been through this before. 98% of people have low or very old middle end graphics cards.
It will also take several more generations of consoles at the very least to produce games with realistic graphics.
Whats the point doing these showcases when at the end they always do the same: adjust/lower graphics for current hardware to handle it
 

Markoni35

Posts: 905   +365
Who says that real world would be smooth to infinity? How would you know this if you've never experienced it? Maybe you just have digital eyes in an analogue world. Or at least, that's what they're hoping that you'll think. :D
In the docs that leaked from their computers into our virtual world it's evident their world has a lot more details. Which is logical, since our world is just a simulation. Unfortunately for them, a few years ago they've discovered something horrible. Their world is quantized too. That made them worried.

Yes, obviously. And déjà vu is just a glitch in the matrix. :D
Obviously.

Well sure, but is that reduction in memory usage worth it if people are able to tell the difference just by wearing a tin-foil hat? Those developers were pretty sloppy and lazy if they were actually trying to create a convincing environment.
They've used some of the most expensive quantum computers to run our world, and this was the maximum detail they could squeeze out. It did work pretty well for a few billions of simulated years. But then those damn physicists started to dig deeper.

One of the solution was to just kill all the physicists. That way there would be no need to upgrade the hardware. Another solution was to "stupidize" the young generations. Once the older generations die, there would be no smart physicists anymore. From what I can see, seems that's the plan they accepted.

Yes, especially when some of their subatomic and quantum particles seem to violate this law with real effects on temporal causality that we don't yet comprehend.
Oh, you mean quantum entanglement? Or something like delayed-choice quantum eraser experiment? Well, the leaked document explains it was a debugging tool for particle real-time telemetry. But they forgot to remove it in one of the releases (kinda like surgeons forget scissors inside patients).

After we've discovered the weird quantum effects, they couldn't remove them anymore, because we'd become suspicious. I mean, theoretically they could, but then they'd have to erase our brains, our books, and all traces of their existence. Which would be a lot of work. Fortunately, the "stupidization" plan will solve that problem too.

They probably thought "Bah, who cares? It's not like the subjects can see these particles anyway!"
Yeah, and they were right for a few billions of Earth years. They didn't believe that artificial intelligence (that's us) would eventually become so smart.

Yeah, but that's a small part of the population and they've modified American culture to distrust anyone with a university education, let alone actual PhD degree holders like physicists. They just spout more and more insane theories on InfoWars for the more dim-witted masses to consume, and they do.
Yeah, but some of those insane theories are actually true. They are just mixed in the same pot with insane theories. And nobody tells you which are which.

Well, even superconductors have some resistance. I mean, the only superconductor element that I can think of in which the level of resistance is statistically zero would be metallic hydrogen. Unfortunately, metallic hydrogen can only form at extreme pressures that would be fatal to life as we know it. It could only occur naturally in our solar system on gas giants. I think that the most likely would be Uranus or Neptune because I think that Saturn and/or Jupiter might be too hot to allow hydrogen to properly crystalise.
This is supposed to be a family-friendly site. Let's not talk about Uranus.

I know, eh? I saw that the Blue Jays just signed George Springer to a 6-year deal for $25,000,000USD per year. The players really ARE getting demanding! That's just nutz!!! :D
Don't know about that, but I've heard that Lewis Hamilton for his 2021 contract wants:
- 200 millions euros for 4 seasons
- 10% of the team championship winnings
- limited production AMG One sports car
- additional role that "goes beyond being a driver and more than a mere testimonial" (whatever that meant).

If Mercedes give him that, the world has really gone nuts.