Ask Slashdot: How Did Real-Time Ray Tracing Become Possible With Today's Technology?

dryriver writes: There are occasions where multiple big tech manufacturers all announce the exact same innovation at the same time -- e.g. 4K UHD TVs. Everybody in broadcasting and audiovisual content creation knew that 4K/8K UHD and high dynamic range (HDR) were coming years in advance, and that all the big TV and screen manufacturers were preparing 4K UHD HDR product lines because FHD was beginning to bore consumers. It came as no surprise when everybody had a 4K UHD product announcement and demo ready at the same time. Something very unusual happened this year at GDC 2018 however. Multiple graphics and GPU companies, like Microsoft, Nvidia, and AMD, as well as other game developers and game engine makers, all announced that real-time ray tracing is coming to their mass-market products, and by extension, to computer games, VR content and other realtime 3D applications.

Why is this odd? Because for many years any mention of 30+ FPS real-time ray tracing was thought to be utterly impossible with today's hardware technology. It was deemed far too computationally intensive for today's GPU technology and far too expensive for anything mass market. Gamers weren't screaming for the technology. Technologists didn't think it was doable at this point in time. Raster 3D graphics -- what we have in DirectX, OpenGL and game consoles today -- was very, very profitable and could easily have evolved further the way it has for another 7 to 8 years. And suddenly there it was: everybody announced at the same time that real-time ray tracing is not only technically possible, but also coming to your home gaming PC much sooner than anybody thought. Working tech demos were shown. What happened? How did real-time ray tracing, which only a few 3D graphics nerds and researchers in the field talked about until recently, suddenly become so technically possible, economically feasible, and so guaranteed-to-be-profitable that everybody announced this year that they are doing it?

Comments

Sinple

[Lunix Nutcase]

Simple: Hardware got more powerful.

Re:Sinple

[KingMotley]

I would guess they figured with graphics cards having 3500+ cores and ample memory for massive lookup tables, suddenly it seems feasible. That or a patent just expired, or both.

Re:Sinple

[Anonymous Coward]

They have been working on it for 10+ years

now.

Re:Sinple

[mikael]

They started using neural networks and machine learning to look at optimizing algorithms. Things like solving anti-aliasing problems. Normally they had to super-sample every pixel hundreds of times. Most cases, the result is the same. An ML algorithm let them figure out the most important cases and reduce the number of samples needed. Instant speedup.

Re: Sinple

[Anonymous Coward]

We kill the Batnan.

Re:Sinple

[Anonymous Coward]

Slightly more complex: NVidia uses both rasterization and ray tracing hybrid to render the video content. It's not pure ray tracing.

Re:Sinple

[arglebargle_xiv]

Simple: Hardware got more powerful.

Simple: Graphics card vendors required some new buzzword to sell their most expensive hardware.

Re:Sinple

[Type44Q]

Don't feed the trolls. ;)

Re: Sinple

[Anonymous Coward]

Actually the answer is not that simple. Hardware didn'tmake a magic leap, de-noising did. After a paper published at Siggraph about denoising the algorythms used for path tracing almost every raytrace renderer started including it. This allows fewer rays to be cast for comparable results, which now allows GPU's to also get close enough. It is a software jump, not a hardware one.

Aliens

[Anonymous Coward]

#import

Smaller transistors

[Anonymous Coward]

What else has changed in the last half century? We don't even have the Concorde anymore. Airplanes still look, act, and sound the same as in 1969 when the 747 first flew. Cars, houses, clothes, buildings, roads, food, all more or less look the same.

Only processing information has gotten better by orders of magnitude, because information is massless.

So while you whittle your life away playing stupid games, just remember than none of the Space Age fantasies are ever going to become real.

But you have nice games and cordless vacuum cleaners.

Re:Smaller transistors

[Anonymous Coward]

You're probably not paying enough attention to planes. They crash less frequently than they did and airports are switching to systems where the planes make a smooth approach without the stair-step approach pattern that was previously used.

Other than that, there's not going to be much difference because by the '80s planes were just that good. Any other changes are ones that are mostly only going to be noticed by people actually working with them.

The computing technology stuff tends to be more accessible and more present in our day to day life than it was, so we tend to notice it more. But, there again, we're getting close to the point where additional improvements aren't going to be as obvious as they were previously. I've had this computer now for a few years and the previous one was quite a few years old when I replaced it. That would have been inconceivable to me 20 years ago as the technology was improving that quickly.

Re:Smaller transistors

[Anonymous Coward]

and cordless vacuum cleaners.

You can keep your fuckin cordless vacuum cleaners.

Damn brain washed fucks.

Re:Smaller transistors

[Locke2005]

I disagree; flying cars are hitting the market any day real soon now -- but you will be required to have a pilot's license for most of them.

Re:Smaller transistors

[OneHundredAndTen]

Other than that, there's not going to be much difference because by the '80s planes were just that good.

What has been touted as big passenger-visible advancements in commercial airlines in the last few years is slightly larger windows, slightly higher pressure in the cabin, and slightly higher moisture. That's it. As for safety, while it is true that airplanes are safer than they have ever been, the truth is that they were already extremely safe in the 70s. The sad, depressing truth is that it takes exactly the same time to fly from London to San Francisco as it did fifty years ago. That's half a century ago. It is indeed cheaper than it used to be, but the experience is, if anything, more miserable than it has ever been. And, bearing in mind that the airline industry has become a race to the bottom in the last 10 to 15 years, the experience is going to become, if anything, worse.

Re:Smaller transistors

[Anonymous Coward]

They're not flying cars, they are roadable aircraft. Difference!

Re: Smaller transistors

[dunkelfalke]

Aircraft safety sucked in the seventies. Planes that were knowingly defective were released and killed people.

Re:Smaller transistors

[Anonymous Coward]

You are ignoring the relative drop in price, which has been a huge advancement that has seen working class families travel the world. That, safety, and the sheer capacity (the amount of people flying today is far greater) are three huge advancements in the field which are shocking in breadth and depth. Also, don't ignore the wi-fi, USB ports, in flight entertainment system, and so on.

You can have flights that are *far* more comfortable for the same relative price that was a flight 50 years ago; fly first class and it's still cheaper. You have better food, better service, seats which can become flat for some really comfortable sleep, a great selection of free booze, and so on.

The experience is going to become worse for people who pay less. That is not a bad thing, that is how "entry level" works. The price/performance ratio is where the advancement is, and that has gotten better.

Re:Smaller transistors

[Joce640k]

The sad, depressing truth is that it takes exactly the same time to fly from London to San Francisco as it did fifty years ago.

It's unfortunate that we can't change the laws of physics, yes.

It is indeed cheaper than it used to be

Ticket price is linked to fuel consumption. The price of air travel would rise sharply if we tried to fly any faster.

Re:Smaller transistors

[Anonymous Coward]

It's unfortunate that we can't change the laws of physics, yes.

The laws of physics don't prevent faster planes. In fact, a well know, now retired plane had an emergency procedure called TAL - Trans-oceanic Abort Landing. Though never used, the idea was that if a problem occurred, the plane in question could make an emergency landing in e.g. Spain 20 minutes after departing from Florida.

It's not physics that sets the limitations, it's the state of technology. The plane mentioned above had five huge engines, two of which could only be used once, and the other three needed major refurbishing after every flight, so it wasn't useful for regular scheduled flights.

Re:Smaller transistors

[Opportunist]

You have discovered the "good enough" problem. At some point in development of pretty much anything, you hit the 80/20 barrier where improvement becomes prohibitively expensive, to the point where it is economically unfeasible. The Concorde was one of those 20% that are unfeasible. Hence it vanished.

You will notice that most of the technology you mention has been around for much longer than computers. Cars, houses, clothing, buildings, even planes. The computer, with integrated circuits, is less than 50 years old. The first halfway affordable graphical interface came 40 years ago. Windows 95 was just over 20 years ago. And hardware supported 3D processing in mainstream computing is barely 20 years old.

And remember that it always takes at least 3-5 years for a technology to take over and replace its predecessor to a point where development focusing on the new technology becomes economically feasible.

Now think back to when the technologies you mentioned were 20 years old. Was there room for improvement? Hell, there was even a lot of room for improvement 50 years after their first appearance.

Re: Smaller transistors

[Anonymous Coward]

Cars dont have carburetors and run for 100,000 miles with essentially no maintenance beyond wear components. My guess is you never owned a car before the 80s to have any idea how much things have changed.

Re:Smaller transistors

[hipp5]

We don't even have the Concorde anymore. Airplanes still look, act, and sound the same as in 1969 when the 747 first flew. Cars, houses, clothes, buildings, roads, food, all more or less look the same.

They still look roughly the same because that's a pretty good geometric design. But that doesn't mean they are the same. There have been huge advances in aviation technology over the past 50 years. For example, the average fuel burn of new commercial jets fell 45% between 1968 and 2014. Modern jets have better engines and are made of modern materials (carbon fibre is becoming more and more common). This means cheaper prices and longer distances for direct flights.

Buildings certainly do not look the same. New homes in the '60s were usually 1100 SF or smaller (at least in my area). Compare that to new homes now that are 2000 SF, much more efficient, and have a lot more modern conveniences. Now they're not filled with asbestos and lead paint.

And I could go on and on with food, clothing, and the rest.

Re:Smaller transistors

[Noah Haders]

I think price is a huge revolution. I can go from LA to Iceland for $400 RT. The shape of planes is the same, but the shape of air travel is completely revolutionized.

Re: Smaller transistors

[Anonymous Coward]

Having owned cars from the 80s, 90s, oughts, and 10s, the driving experience is extraordinarily different, for the better.

I went from a carbeurated rear wheel drive v6 with about 180hp that will kill you dead in an accident to an electronic fuel injected front wheel drive i4 with over 200hp that has airbags and antilock brakes and 4 season tires and electronic stability control that will detect if you're about to spin out and steer you with the back brakes.

It's hard to quantify the difference in driving experience. Winter driving went from terrifying to relaxing.

Re:Smaller transistors

[Anonymous Coward]

"the average fuel burn of new commercial jets fell 45% between 1968 and 2014"

Imagine if computers improved 45% in the same time frame.

Thanks for making my point for me.

Re: Smaller transistors

[Anonymous Coward]

You apparently haven't looked our the window. The stakes, takes and winglets are new. You apparently can't hear, because they sound a lot different. High bypass fans sound a lot different than old turbojets. In fact, the newest fans have a very different outlet shape for stage 3 nose compliance. You also can't count, as transoceanic aircraft have gone from 4 engines to 2, except of course the new 747s and -380 which are more than twice as big as the largest aircraft of the 70's. You also failed to notice that the panel fill if gauges in front of the flight engineer are gone, as is he.

Re: Smaller transistors

[Anonymous Coward]

Well, let's see. It takes about the same time to boot windows 10 on s new laptop as it did MS-DOS, so by irrelevant metrics, computers haven't improved.

Re:Smaller transistors

[Anonymous Coward]

The sad, depressing truth is that it takes exactly the same time to fly from London to San Francisco as it did fifty years ago. That's half a century ago.

You forgot that the limitations are abode by the law of physics. As a result, improving certain features could significantly impact others that are already stable. In this case, if you want to significantly increase the speed of traveling (reducing travel time), you would have to scratch everything all over again. There are many things involved in safety to do such a thing. New material and layout/design may need to be redone. The cost would go up a lot due to R&D. It is not worth it at all unless there is a new significant technology emerged that can be applied to this application.

Re:Smaller transistors

[Anonymous Coward]

And if airplanes doubled in speed every 18 months like moore's law, then they should be traveling faster than the speed of light by now.

Re: Smaller transistors

[macmurph]

It actually takes longer to go from SF to London today. In the past, we had the Concorde, no TSA, and a lot less traffic on the way to the airport. That could easily shave off hours from the trip. I don't recall if Concorde landed on the west coast, but I think it was in Dallas, Tampa, DC, and NYC.

Re:Smaller transistors

[Anonymous Coward]

Imagine if computers improved 45% in the same time frame.

Thanks for making my point for me.

Err, you are the one who is missing the point. You conflate real life mechanical with electrical applications. Don't you see? How fast a car travel and how fast an electron travel? If you can't even understand that in the first place, you are way off the point.

Re: Smaller transistors

[Anonymous Coward]

Technically the solid rocket boosters could be reused as well.

Re: Smaller transistors

[Anonymous Coward]

Alas, information is not massless. There would be no way to consume or store or if that were true.

Even electrons have mass.

Re: Smaller transistors

[Anonymous Coward]

Don't confuse the medium with the message.

A signal wire can transfer unlimited amounts of data without changing its mass.

Re: Smaller transistors

[david_thornley]

IIRC, the Concorde was not allowed at supersonic speeds over land areas, which means that it was no faster than an ordinary airliner going from the East Coast to San Francisco, and much less efficient.

It isn't out of the blue

[kiminator]

The short answer is that it isn't brand-new. As the article mentions, nVidia has been doing real-time ray-tracing demos for about a decade. Various tricks and approximations are used to make this a reality. Game developers have largely shied away from it because similar results can be achieved with typically better performance using other methods.

This announcement, particularly with the involvement of Microsoft, indicates that the companies finally feel that the technology is mature enough to actually be used in a game. My guess is that it will likely still be some time before it is put to use. The fact that it was announced at the same time likely indicates that the three companies have been working together for some time behind closed-doors to agree upon the DirectX Raytracing API.

Re:It isn't out of the blue

[Anonymous Coward]

Indeed, they could do this many, many years ago, the problem was always having enough rays to make it worth while. Ray tracing a couple rays in real time is relatively easy, but tracing enough of them to get results that are better than the current system was hard.

I remember messing around with ray tracing programs back in the '90s and they were great, but it took huge amounts of time to render a scene. Far too long for even simple board games, let alone a typical 3D game.

At some point, it was inevitable that ray tracing would be preferable to the standard raster graphics that we've been using. The file sizes necessary to get anywhere near what raytracing can do would have made it impossible to keep up. Having better algorithms and specialized computation units focused on it just moves that point closer to the present.

The thing I'm really looking forward to though is that they should be able to scale with new advances in monitor technology. Both in terms of pixel count and in terms of aspect ratio in a way that wasn't realistic previously.

Re:It isn't out of the blue

[kiminator]

Also, apparently their current demos are running on some severely beefy hardware (e.g. here), and they're only using raytracing for a portion of the scene. This may play a part in gaming eventually, and it's good that the APIs are getting out there, but it probably will be a while before it makes it to the next Call of Duty game.

Re:It isn't out of the blue

[Anonymous Coward]

today's beefy machine is next years commodity parts.

Re:It isn't out of the blue

[Shinobi]

Even if ray-tracing isn't used for graphics, you can use it for the sound engine, with the benefit of making it hardware accelerated. Would make a whole lot of games more interesting, for example Stealthers like Thief, horror games etc etc.

Re: It isn't out of the blue

[Anonymous Coward]

Mate, that's a big plus one, especially for immersive FP, AR and VR games. Anyone doing ray traced binaural?

Re:It isn't out of the blue

[allcoolnameswheretak]

Game developers have largely shied away from it because similar results can be achieved with typically better performance using other methods.

True, but it was also a self-reinforcing trend. Raster-based graphics was how it all started, so chip-makers and API providers where optimizing for that, which in turn entrenched this way of rendering further, 25 years on, graphics chips and API's are an amalgamation of super-optimized functions, tricks and hacks to squeeze every ounce of performance from that rendering methodology. And the fact is that this way of rendering has very little to do with the real world.
I would welcome the trend towards ray-traced rendering because it reflects how the physical processes work in the real world, which is vastly more logical and intuitive than the OpenGL or DirectX graphics specification. Surely ray-traced rendering will also use hacks and shortcuts, but I'm sure it will be nowhere near the bag of tricks that that current graphics engines work with.

On the other hand, this is bad-news for some of the heavily invested graphics-gurus, since all of their accumulated knowledge could become moot if ray-tracing becomes the new standard.

Re:It isn't out of the blue

[Shinobi]

So apparently there's a mod out there that doesn't know that you can actually do ray-traced audio propagation.

Here's a bit of light viewing on how it's been used in one field since the 80's: https://youtu.be/ZY1Kiih8sTU

Re: It isn't out of the blue

[Anonymous Coward]

A great many of the raster graphics tricks are essentially bulk, special case raytracing with no conditional branching to suit SIMD GPU architectures. Is the branching in simple raytracing that makes applying current GPU tech so hard.

Re:It isn't out of the blue

[Megol]

AFAIK Aureal did a limited form of this for their 3D soundcards. IIRC tracing a few "rays" in a simplified world model to simulate reflections etc.

Re:It isn't out of the blue

[Anonymous Coward]

It's a shame they wound up going out of business, that was the superior technology. It's a shame that Creative didn't buy the technology when they went out of business.

Re:It isn't out of the blue

[Impy the Impiuos Imp]

I recall a Quake demo done with ray tracing in the 90s. They estimated it would take well over a 30,000 Hz Pentium whatever to do it in real time. It looked great, witb light projecting through stained glass windows.

Re:It isn't out of the blue

[Anonymous Coward]

I recall a Quake demo done with ray tracing in the 90s. They estimated it would take well over a 30,000 Hz Pentium whatever to do it in real time. It looked great, with light projecting through stained glass windows.

Have you ever implemented a small program to do Ray tracing? It could take a second just to calculate a still image depending on how deep you want and your machine power. A true Ray tracing, you would need to calculate each pixel. If you want to do it in real time, in a video of 30 FPS and at 4k UHD graphic, you would need some sort of a mixed/adaptive method to do that. The Quake demo is a video which can be done before hand (not dynamic). Also, what size was the video screen?

Someone's cheating

[Anonymous Coward]

More specifically, Nvidia made custom hardware to do the calculations.

Re: Someone's cheating

[sound vision]

Nvidia makes custom hardware to play raster games too. They call them "video cards". I hope they can cheat the ray tracing right into my computer.

Terminology

[B.Stolk]

Be careful with your terminology.

Real Time Ray Tracing with one primary ray and one shadow ray for each pixel, was viable last year as well (at 1080p.)
But this will not render indirect light, thus no Global Illumination.

You may be referring to Real Time Path Tracing, where you need to shoot a lot of rays for every pixel.
This is currently not possible, and also not possible in this year's GDC demos: I think most of the demos were hybrids (rasterizing+tracing), and definitely not full Global Illumination.

Re:Terminology

[ganv]

You are touching on an important issue here. 'Real-time' and 'Ray-Tracing' are both open to definition: At what screen resolution, frame rate, and number of rays? And is it a hybrid or full ray tracing solution? That asked, I am very interested to know the answer in the original post: Is this a tipping point where they finally decided the hardware is good enough to market the ray tracing they have been working on or is there some substantial improvement in algorithms or in custom hardware dedicated to ray tracking?

Re:Terminology

[sanosuke001]

Yeah, this is the answer. They showed off hybrid systems so they'll do SOME things using ray tracing where most of the heavy lifting is still going to be rasterized for the foreseeable future.

Re: Terminology

[Anonymous Coward]

Yes.
A little bit.
Sort of.

Re:Terminology

[natex84]

I think the current demos were also done on some insanely high-end Quad GPU configured workstation, that costs around $60k ( https://arstechnica.com/gaming... and https://www.nvidia.com/en-us/d... ).

And even with all of the above hardware, I think they are just running at 24fps?

It looks like this is a long ways off, for a reasonably priced high-end home gaming machine (ie. $4k budget).

Re: Terminology

[Anonymous Coward]

Yeah. A long ways off. Like 2 years at least.

math?

[Anonymous Coward]

Just saying, maybe some algorithm cycled down, seems it would explain the lack of a single patent holder.

Ummm, this isn't too shocking

[Anonymous Coward]

Real time ray tracing was a topic since I was in college 20 years ago. A lot of PhD students have done their research. Algorithms optimized. Hardware advancing. Futurists at major corporations (fancy name for people responsible for monitoring tech) saw hardware evolving.

So Microsoft probably saw that it was going to be viable in 5 years (or whenever) in 2013. They probably starting developing an API in conjunction with major hardware manufacturers. So they all worked together to bring ray tracing to the masses eventually.

Now all this work is paying off and consumers get the end product.

This is no different than major corporations using augmented reality to validate construction designs. Ya, they are doing that atleast in labs. Take the 3D model as built and validate that it is as built. Or quickly look at something in the real world and figure out where the 3D model data is and figure out in a second what vendor supplied the broken part. Ya, this stuff is happening!

Re:Ummm, this isn't too shocking

[belthize]

Or even 30 years ago ... well ray tracing was anyway. Real time ray tracing was a topic in '87 much like the holo deck on ST:NG was; a, "won't it be cool when ?", kind of topic.

It's been a long time coming and it's still not quite here as posters further up have laid out, but it's getting closer.

Well, not all of a sudden...

[Junta]

Demos of real time raytracing have dated back to 2009 or earlier, albeit with various limitations. Raster based rendering is faster and going to raytracing means much better lighting, at the expense of some resolution/geometric detail/hardware requirements.

I think what is being seen is that we've been well beyond the point of diminishing returns as far as raster can reasonably get in terms of better quality. Sure, we can cram more and more polygons and sure we can raise the bar to 4k resolution, but the bang for the back is small. Given that situation, video card market has an issue, they need to do demand generation.

So pushing 4k for gaming helps, and 8k would be nice, but it's just really hard to tell the difference at this point.

If that's a hard sell, then VR certainly can knock things back if it gets traction. With wider FOV, stereoscopic rendering, and optimal experience being at least 90Hz, that would certainly deliver. However, as much as I am a fan of it, it's far from a given that VR is ever going to be large enough to drive adoption. at volumes that can sate the business needs of the GPU vendors.

So raytracing is a third option to make best of breed graphics card noticeably struggle with something that's very visually apparent. Suddenly the market content with status quo of ever refined raster graphics simply must make the leap to have some marketable advancement.

The people pessimistic about any such advancement just continuously have their expectations calibrated to how fast it can perform raster graphics. Raytracing does mean having to step back, but we have enough headroom to take the hit.

Re:Well, not all of a sudden...

[Anonymous Coward]

here are two I remember
* heaven seven by Exceed (april 2000)
* Still Sucking Nature by Federation against nature (april 2003)
Federation against nature did a ray-tracing benchmark called realstorm

Re:Well, not all of a sudden...

[Anonymous Coward]

Actually, VR is probably the reason these guys are switching to ray tracing. Here is a very interesting article on how they manage to reduce latency by using ray tracing instead of rasterizing: https://www.roadtovr.com/exclusive-nvidia-research-reinventing-display-pipeline-future-vr-part-2/
In short, with ray tracing you have the lens warping at no cost, and they can throw more rays at the center of the image (where it really matters because the retina has a rather small high resolution zone), and less rays in the periphery. They can even throw rays more often at the center of the image, they don't have to render the whole frame each time!

Re:Well, not all of a sudden...

[butzwonker]

I agree that it's about demand generation to sell more hardware, but I disagree about the quality of rasterization. I've bought an AMD 1800x with Geoforce GTX 1080, which is not the latest highest high end but still fairly high end, and the high end games I've looked at since then still smear everything with post-shaders and have not enough level of detail at a distance (with crap like "depth of field" turned off, of course). I don't know, maybe game designers are blind but when I look out of my windows I see trees with may more branches and leaves and much more detail very sharply until the horizon.

Consumer CPU/GPU combos have by far not reached the tipping point yet where landscapes really look realistic, and as it seems due to technical limitations.

For that reason I'm sceptical about introducing ray-tracing now. A rasterization-based game will be way more performant than a ray-tracing game, so I find it hard to imagine that ray-tracing game engines can compete with traditional rendering techniques in the near future.

Re:Well, not all of a sudden...

[Anonymous Coward]

Actually one of the first OpenGL API extensions for VR allowed you to render different parts of the view in different resolutions. Mapping that to the screen could be combined with the distortion correction at no additional cost. Since that is handled on the geometry level (viewport multicast) you could minimize cost of pixel operations with I think just a 5x overhead for vertex computations for screen space operations( high res center, 4x low res borders ).

In short, with ray tracing you have the lens warping at no cost

You still have to do a time warp on the whole image to mask the remaining latency, which you could just combine with the distortion correction anyway ...

Re:Well, not all of a sudden...

[Anonymous Coward]

Will be interesting to see if VR adoption increases much after the new Spielberg movie (Ready Player One) which is based on a fantastic book.

Re:Well, not all of a sudden...

[Junta]

I think in that scenario, there may be differences to be had, but there's much less chance a consumer will look at far-away scenery and think a marginal increase in that is worth a couple hundred more dollars.

Contrasted with lighting/reflection/refraction improvements, which can be pretty dramatic.

That was simply my point, that going further may be possible without pulling in some raytracing, but the subjective impression isn't going to differ nearly as much.

Atoms instead of triangles

[technosaurus]

This story reminded me of https://www.euclideon.com/ ... They compose everything as "atoms" instead of triangles. If I ever make it to Oz, i'd like to check it out.

Re:Atoms instead of triangles

[Anonymous Coward]

I thought that was a scam to get government funding

Re:Atoms instead of triangles

[Anonymous Coward]

Not really. The graphics and gaming industry scoffed at the idea you could treat a scene as a database - getting only the point you actually need to render.

This is the rest of the world reinventing their wheel.

Re:Atoms instead of triangles

[Megol]

No those that know the stuff pointed out the problems with such types of engines, a kind that was known and tried but not used as polygon engines were simply put superior. Sampling artifacts, memory use, inflexibility due to extensive pre-processing...

Early demonstrations of the engine show some limitations (even though demonstrations tend to hide problems as much as possible), repeated objects aligned the same way to decrease memory needs, no moving sprites etc. Later demonstrations showed sprite type objects but with limitations (IIRC axis aligned only), more variation in models etc. but then computers were faster and had more memory. Later still I saw it was to be used as a point cloud renderer, possibly the best use of something like that.

But again the problems with such rendering techniques are well known, it isn't used as it isn't good for most things.

It is HYBRID, not full Realtime ray tracing

[Anonymous Coward]

It is using rasterization for the scene, and then hybrid ray tracing for the reflective, shadow, and opacity components were it applies. This is not true ray-tracing, but rather another rendering "trick" to optimize for speed. Most of our rendering for games today is approximations, because it is good enough. Same thing here.

Re:It is HYBRID, not full Realtime ray tracing

[AmiMoJo]

That's how most ray-tracing works these days. Ray-tracing plus the same kind of hacks for things like sub-surface scattering, radiosity, motion blur, air/heat effects, material simulation without having to simulate every fibre of clothing etc.

On top of all that you have artistic considerations. Few games want to look like reality, they want to look stylized and hyper-real. They want to use tricks that focus the player's attention, just like films and TV shows do.

It was in the EULA

[Anonymous Coward]

It was in the EULA, on page 2,763 paragraph 12 subsection 23 A. What do you mean, what's a EULA. It's that thing you're supposed to read before installing the software. You know like the book your about the Spanish kid your car came with. You know, Manual.

Advancing tech and bored consumers

[Anonymous Coward]

The short answer, as many have already pointed out, is none of this was unexpected. Also, saying that "bored" consumers pushed the increase in display resolution is simplistic. Whether FHD or 4k, no current video display system compares to the sensory inputs of a majority of humans - the bandwidth required is just too great. Manufactures continue to strain for that brass ring of true representation, whether for mass markets or professionals doing Real Work (tm).

Denoising

[Anonymous Coward]

Admitted I don't know anything about rendering, but they say that they're using a relatively low number of sample rays per pixel and then running an AI-based denoising algorithm to clean it up. Also it's just for lighting and such.

Money.

[Anonymous Coward]

How Did Real-Time Ray Tracing Become Possible With Today's Technology?

And time; this would certainly have been possible years ago, at considerably greater expense.

Was doing real time RT in the 80s!

[Anonymous Coward]

Ok, ok, it was on 2x2 resolution...

Seriously though, the big problem with mapping real time RT to today's graphics hardware is ray decoherence. Graphics hardware likes branches (like if/else or loops) to go the same way, because it's SIMD: single instruction multiple data. It CAN support each lane having a different control path, but it loses efficiency the more of that starts going on. In the worst case it can be a severe loss.

So, primary rays are not too bad but the more they reflect around and have different lifetimes interacting with different things (think reflects from a narrow object where the reflected rays go in different directions) the worse it gets. People have been working on algorithms to mitigate this problem, which has helped, and of course graphics HW has been getting faster and faster.

Not feasible for gaming

[PhrostyMcByte]

Ray tracing is an impressive technical feat, but the argument against it for gaming still stands:

However fast you get at ray tracing, you can instead use that power for rasterization and do far far more.

The day may come where that gap doesn't matter anymore, or where we find a way to overcome it... but for now, I think this technology will primarily be used to help accelerate very simple environments, and more complex for offline rendering.

Re:Not feasible for gaming

[shaitand]

"However fast you get at ray tracing, you can instead use that power for rasterization and do far far more."

Not when you are real time rendering the graphics to be integrated on the fly along side real time light vectors flying at the retina. Think pokemon go, without the phone, or the screen, and with the pokemon actually sitting on your counter with it's legs hanging down.

Re:Not feasible for gaming

[Anonymous Coward]

Right now that's true, but it won't be true forever.

But, it will only change if somebody is actually making use of hardware advances to move us past the point where it isn't better. Games like the Assassin's Creed series already have entire environments built out of models, going the next step to ray tracing won't be that difficult once the engine is there that can render it in real time on the player's computer.

This approach where the ray tracing is mostly just handling shadows and water effects is a huge step in the right direction, but definitely won't be the last step. The rate at which hardware is improving is significantly faster than the rate at which monitors are adding more pixels and colors. Meaning that at some point, and probably in the next few years, it will be reasonable to ray trace a FPS, perhaps not at full resolution, but at high enough resolution to make it worthwhile.

Re:Not feasible for gaming

[squiggleslash]

The major innovation here is that it'll soon become possible to render, in real time, silver balls, cones, and cubes floating above a checkerboard, something that's just not possible with regular 3D graphics.

Re:Not feasible for gaming

[91degrees]

The main benefit is we get realistic refractive and reflective effects - which tend to be pretty slow with rasterisation as well. Not sure if we'll be seeing some sort of hybrid technique soon.

Raster 3D Graphics seem to be tapped out

[rsilvergun]

Remember the jump from the SNES/Genesis to Playstation/Saturn? How about the first time you saw a Dreamcast game? Those were big leaps. But PS2 to PS3? If you'd been paying PC games you'd seen stuff on par with PS3. And PS3 to 4 was hardly a leap at all.

The trouble is modern graphics have gotten _hard_ to make. Pixel shaders are a bitch. They're too labor intensive. What's needed is something that lets you do great graphics with less man hours and fewer bugs. If ray tracing isn't gonna do that then it might as well be PC's answer to 3D TVs. Especially if it's only kicking out 30 FPS.

Re:Raster 3D Graphics seem to be tapped out

[Kjella]

The trouble is modern graphics have gotten _hard_ to make. Pixel shaders are a bitch. They're too labor intensive. What's needed is something that lets you do great graphics with less man hours and fewer bugs. If ray tracing isn't gonna do that then it might as well be PC's answer to 3D TVs.

I think that problem extends far beyond graphics, if you're going for realistic graphics it also has to behave and interact with everything like it's real. Like if the wind is blowing everything has to be fluttering in the wind. If you're making a dog it can't just look like a dog, it has to move like an actual dog with bones and muscles dragged down by gravity and leave paw prints in the mud. To say nothing of humans, uncanny valley here we come. Heck, I think it'd be difficult just to properly simulate taking a baseball bat to inanimate objects. All the dents and chips and scrapes and smears and splinters are very difficult to get right unless you bring in a ton of material science and deformation models. Usually we just fake it with an animation or two, replacing the "whole" model with a "broken" model and skipping all the hard parts. But it only looks half real...

Where have you been?

[amoeba1911]

You misuse the term "raster graphics", as that refers to bitmaps, not vector based real-time 3D rendering. Real-time ray-tracing has been around for almost two decades, but it wasn't mainstream until recently. As the parallel computational power of GPU increases, ray-tracing becomes the inevitable future.

Re:Where have you been?

[Megol]

Well current solutions do use raster graphics - but I'd be very surprised if realtime raytracing didn't use it too. Representation -> Renderer -> Framebuffer -> Screen.
You are of course right about the misuse of the term.

There have been at least one polygon renderer that didn't use a framebuffer (don't remember the name/company) but with severe limitations: for every rendered pixel a hit test had to be done with every polygon which strongly reduced the numbers of renderable polygons. Add the problem of "racing the beam" and even with pipelining polygons had to be stored in expensive SRAM memory as variable latency couldn't be hidden. I'd prefer a framebuffer.

Let us wait and see

[140Mandak262Jamuna]

We dont know how many of them are real real-time ray tracing and how many are me-too announcement for PR.

Second just because they call it real time ray tracing does not mean it really is. It could be a technique to create a low res ray tracing to calculate the radiosity, reflectivity etc and then use the standard raster graphics to use these to adjust the texture maps of raster graphics.

So let us wait and see if what they call real time ray tracing and what we call real time ray tracing are one and the same.

It's not 100% Ray Tracing.

[Berkyjay]

All the companies that announced this and the media that is reporting on it all gloss over the fact that the Ray Tracing that is being talked about is supplementary the standard rasterized shading engines. We aren't going to be getting 100% ray traced scenes. We'll get rasterized scenes with things like reflections and shadows being ray traced. This will fix the issue that rasterization has where objects outside of the viewing frustum are computationally ignored. 100% ray tracing is still a pipe dream with the current hardware standards that most consumers have.

...and clever Algorithms

[Roger W Moore]

According to this video it is not just more powerful hardware but also that they came up with the idea to use only a fraction of the rays normally required and to then used a power denoise algorithm to generate the final image.

Re: ...and clever Algorithms

[Anonymous Coward]

Exactly. Moreover, this only works for relatively moderate scenes, where you don't have to decouple multiple independent light sources from complex scenery.

Re: ...and clever Algorithms

[Anonymous Coward]

This pegs it. Ray tracing is being used selectively as part of the standard raster pipeline we have today. It's still not feasible for real time raytracing for every pixel and handling reflections perfectly may never be doable.

Re: ...and clever Algorithms

[Pseudonym]

Just to be clear, it's not "reflections" (as in mirrors) that's the problem necessarily, it's the fuzzy effects: diffuse, glossy, translucent, aerosols, etc.

It may never be doable because of Blinn's Law.

Re: ...and clever Algorithms

[Anonymous Coward]

50k a month says yes we can, Opus Architects, and im only limited to architectural vis ... yes the software is costing like 10k-50k a year, but the results cleanup 10k work a week.... soo uhh ............. take heed?

Re:...and clever Algorithms

[Anonymous Coward]

An Anonymous Coward below pointed out that for VR applications you get the lens warping for free if you raytrace and if you want to you can render the center of the picture more often.
That made me wonder if this denoise algorithm can handle getting rays from multiple camera angles.
Would it be possible to trace as many rays as you can at 60fps and merge the rays from the last n frames to get enough data to denoise?
Yes, it will probably look a bit wonky if the camera moves fast, but the question is if it is more or less jarring than stuttering from low framerate.

Ideally I guess you would like to take into consideration that the rays are from different positions and try to get some form of motion blur from it, but that is probably not an easy feat.

Updating the center of the screen more often than the peripheral part is likely easier.

Re: ...and clever Algorithms

[Anonymous Coward]

Handling reflections perfectly is trivially easy with raytracers, and there have been plenty of real-time raytracers even for low-end consumer hardware in the 90's.

People tend to have a misunderstanding on what raytracing is. There are many different rendering techniques with different advantages and drawbacks. Many have the idea that raytracing is a holy grail that simulates the physics of light to produce photorealistic images but is too expensive to run in real-time on consumer hardware. It's weird because that's the exact opposite of the case - raytracers are actually quick, cheap non-photorealistic renderers that surprisingly similar to rasterisers (but far simpler to implement, good/bad at vastly different types of geometry, and more sensitive to resolution). Maybe popular imagination got the concept of raytracing mixed up with some GI algorithm such as photon-mapping?

Re: ...and clever Algorithms

[Pseudonym]

Blinn's Law is never wrong.

Re: ...and clever Algorithms

[Anonymous Coward]

When you hear "ray tracing", 90% of the time it means "path tracing"

https://en.wikipedia.org/wiki/Path_tracing

Re: ...and clever Algorithms

[Anonymous Coward]

Anything that alters light's behavior makes it difficult, "reflections" is just a generic term for light (photon) interactions with other particles. Perhaps with quantum computing it may be more reasonable, but even ray tracing is a flawed model of light because it basically discards quantum effects by generalizing behavior at a macroscopic scale. Ray tracing assumes a deterministic behavior of light exists when bases in our best models of light today, we know it's behaves very undeterministically.

Re: ...and clever Algorithms

[cellocgw]

Blinn's Law is never wrong.

... wondering what Muphry's [sic] Law has to say about that.

Cellocgw's Law: Cellocgw's Law is always wrong.

beat that!

uninformed author

[gravewax]

What the fuck is the poster on about. Ray Tracing has been a known researched thing that was on the horizon for years, it hasn't suddenly miraculously appeared and was always expected once the hardware crossed the threshold of performance for it to be released for more general consumption.

Magicleap

[shaitand]

As long as magic leap has been vaporware, the graphics vendors have known real time ray tracing was going to be required in a big way and have been working on it behind the scenes.

Re:Magicleap

[Anonymous Coward]

Real time ray tracing is not required for Magic Leap to work.

Re:Magicleap

[Anonymous Coward]

yep, Magic leap just requires more of your investment dollars and a few more years and we promise it will be amazing.....honest word. They are just as good an investment as cold fusion in box or Nigerian princes.

Re:Magicleap

[JustNiz]

Magic leap has zero dependency on raytracing.

Re:Bitcoin.

[Anonymous Coward]

Cryptomining has advanced the demand and the state-of-the-art for GPU's

No, no it didn't. Pretty much zero innovation came about from nerd-coin mining. A bunch of people lost a crapton of money though.

As a result, we have real-time ray tracing

No we don't, we have some half-assed hybrid of ray tracing, most of it is still raster.

the year of Linux on the desktop

I hate to break it to you, nobody actually gives a shit about desktops anymore. We did get the decade of the Linux smartphone. You know all of those Android devices out there that run the Linux kernel.

What the hell is this buffoon blathering about...

[Anonymous Coward]

It all depends on what you want to classify as raytracing.

If the cut off is to just trace rays, it’s been done for a goddamn quarter century.
Look at demo scene stuff, chrome 4kb, Heaven Seven and others.

If you want results at 4K with Renderman quality with all the bells and whistles on, well it’s not going to happen at 1 frame a second, much less 60. (For a visually complicated scene)

Also the dumdum that wrote this tripe wonders why we can do today with current technology what we couldn’t do with the current technology of yester-year. That is donkey-brained thinking at its finest.

I don't know.

[Anonymous Coward]

Many of us were raymarching distance fields 5+ years ago, I'm not surprised they figured out raytracing - many great minds in graphics development.

Because it's not really ray tracing.

[Anonymous Coward]

It's ray group estimating.

Saves a crapload of time.

paah, was doing this years ago

[Anonymous Coward]

My friend managed to do realtime raytracing years ago. He did have a large Cray at his disposal, but anyway that's not important, he still was able to do it.

So first off, the Nvidia/MS thing is crap

[locater16]

The first thing to understand is that the Nvidia and MS press releases are complete crap. Oh it "works". The UE4 Star Wars thing is real enough. It's also running on $12,000 worth of GPUs at 30fps in 1080p, and is only partially raytraced.

Now does this mean realtime raytracing isn't here? Well the answer is no, it is here. The game Claybook is entirely raytraced, unlike the DirectX demos. It's in early access and runs on nothing more than an Xbox One or PS4, and does so at 60 frames per second. Here's the trailer: https://www.youtube.com/watch?... Now THIS is possible because graphics programmers have gotten quite clever over the years. The cleverest bit is called signed distance fields. This can be thought of a volume of points, or boxes, that all store the nearest distance to a solid "surface". Going through this structure allows you to raytrace very very quickly, as you know how much empty space you can skip each time without hitting anything. And since this data is relatively small for each point it doesn't use up a lot of memory either. It's so fast and low memory you can run a demo in your browser here: https://www.shadertoy.com/view...

Obviously there's a bunch of other clever programming going on in Claybook and other titles. But SDF's are the biggest thing to understand. That and that the MS Raytracing API is totally uninteresting from a performance perspective. In fact it's rather awful.

Re:So first off, the Nvidia/MS thing is crap

[UnknownSoldier]

> The UE4 Star Wars thing is real enough. It's also running on $12,000 worth of GPUs at 30fps in 1080p

Uhm, try $50K for the Nvidia DGX Station running on four Volta GPUs.

> Well the answer is no, it is here.

Outcast, back in 1999, did real-time ray tracing and voxel rendering

The only difference is 20 years later we can do it hardware.

Re:So first off, the Nvidia/MS thing is crap

[Anonymous Coward]

Obviously there's a bunch of other clever programming going on in Claybook and other titles.

Looks more like clever design: not a single translucent or reflective surface in the game, that simplifies the raytracer quite a bit and cuts down on the recursion.

Re:So first off, the Nvidia/MS thing is crap

[Bobtree]

The game Claybook is entirely raytraced

No, it isn't. Here are their GDC 2018 slides: https://www.dropbox.com/s/s9tz...

They are using Unreal Engine 4 for shadow cascades, ambient occlusion, lighting, motion blur, and presumably the background scene. The SDFs are raytraced for first-hit surface intersections, soft shadows, and extra ambient occlusion. The visual giveaway is that there are no reflective surfaces.

Re:So first off, the Nvidia/MS thing is crap

[Bobtree]

Outcast, back in 1999

That summary is crap. Outcast was only raycasting a 2d heightfield. No raytracing, no voxels.

Re:So first off, the Nvidia/MS thing is crap

[Anonymous Coward]

I don't know what operations are required for ray tracing, but if it can be done with 16-bit FP matrix multiplies, those Voltas are fucking crazy!

Each Tensor core can do a 16-bit FP (half-precision) 4x4 matrix multiply+add in a single cycle, for close to 125 TFLOPS per GPU. If they're using 4 of those, they could be harnessing 0.5 PetaFLOPS!

Of course maybe ray tracing isn't amenable to Tensor cores, and then they'll have 50 TFLOPS of single-precision computing.

dom

Re:So first off, the Nvidia/MS thing is crap

[UnknownSoldier]

The original Art Directoror disagrees.

He used the word voxel 11 times and refers to the 2D heightfield as "voxel tiles" but technically that's correct. A 2D heightfield is just one way to represent a sub-set of 3D voxels.

http://francksauer.com/index.p...

You are right about the ray casting though.

Was Quake Wars: Ray Traced the first ray-traced game?

Real Time Ray Tracing by what definition?

[Darkling-MHCN]

Full realistic ray tracing to provide real global illumination be it real time or otherwise has always been impossible. Ray tracing is only a simulation and has always employed artificial cutoffs and other hacks. Why? Well light is analogue, to simulate it properly requires a literally infinite amount of processing power. I doubt there'll be anything different about this tech, it's just a new api.

Re:Real Time Ray Tracing by what definition?

[Anonymous Coward]

Well light is analogue

No. it is discrete.

Photon mapping is a thing.
The algorithm described in the wikipedia link is however a simplified version used to assist raytracing but it is theoretically possible to do a full photonmap render.

Note than one lumen gives you about 4 * 10^15 photons per second, so a 800 lumen light bulb in a 60 fps frame gives you about 5 * 10^15 photons to trace.

If you trace that many photons from a light source you will get a realistic view of a room illuminated by a 60W incandescent bulb.
You will probably need to denoise the image. Your eyes/brain normally does it for you, but it doesn't work when displayed as a static image on a screen.

Re:Real Time Ray Tracing by what definition?

[Anonymous Coward]

Well, the reason it's impossible for ray tracing to provide global illumination is because ray tracing is necessarily a strictly local illumination algorithm. That's certainly not to say that it's impossible to use a ray tracer to render a scene with global illumination, but rather that to do so you must either modify the algorithm to the extent that it doesn't count as a ray tracer anymore or else have global illumination handled by a completely different algorithm and use that global illumination engine as a LOCAL illumination light source (either as a "light" or a surface property) for the ray tracer itself.

It's like trying to print a colour mage using a black-and-white printer. It's impossible as defined. Maybe you can modify the printer to be able to print colour images, but then it technically isn't a black-and-white printer anymore. Of course you can take a hybrid approach and print outlines then colour the picture in with crayons.

No, It didnt.

[Joviex]

You fundamentally misunderstood the technology; partial solutions for partial rendering with interpolation is not 100% raytracing.

Re:No, It didnt.

[Anonymous Coward]

Spoken like a true scotsman.

Even simpler.

[gl4ss]

it's just used for effects not for full scene. they are not shooting a ray per pixel(or multiple). and following it where it goes. for most of the scene they don't need to do that and for the reflective surfaces they only need to do it once sfor every so many pixels.

people we're doing partial limited realtime raytracing to make some effects on pentium 100mhz's.. that you use raytracing as part of an effect is just.. well, it's not the same as raytracing with povray - not at all.

two words.

[Anonymous Coward]

Magic Leap

Work smarter not harder

[Anonymous Coward]

The number of rays that a GPU can trace per scene has increased at pretty much the same rate as any other scene element. The difference is that with machine learning it has finally become practical to determine which rays within a scene make enough of a contribution to merit the cost of tracing them, and bog standard rasterization takes care of the rest of the lighting.

It sounds like you're asking a different question

[Pezbian]

It sounds like OP is asking "Why are GPUs so damn powerful so soon?"

In a word: Cryptocurrency mining

People were willing to pay for faster and faster GPUs for their mining farms and their profits allowed it. Look at the overwhelming demand for dedicated space heaters, er, miners. If there was no money in it, it wouldn't have happened.

It has been said that porn built the internet. Based on traffic share, that's a safe bet. There was money in being a 1 frame per second camgirl and there was money in video, which was difficult to get over a dial-up modem. A few fibers and cables later, even T1 lines were eclipsed.

yeah well..

[SuperDre]

It's still not possible (in high resolution/high framerate) on current consumer GPU's.. The star wars demo for instance was done on a set of very VERY expensive (not available to the public) GPU's which will not be available for another few years.

Different Constraints

[briester]

Raytracing and rasterizing behave in memory differently, making the tradeoff between them less about processing and more about memory size. Rasterization and raytracing perform the same functions, but in a different order. Rasterization iterates over every triangle to see if it covers each pixel, and repeats that loop over every triangle for every pixel in the image. Raytracing iterates over every pixel, testing for intersection with each scene primitive.

The big difference is how memory is used in this process. Rasterizers can stream triangles in order very efficiently, while raytracers must have the entire scene in memory all at once. We've had more than enough computational capacity in GPUs to do reverse raytracing (perfect shadows and refractions, no support for global illumination, caustics, etcetera,) for more than a decade. Memory *size,* however, has only recently become great enough to support the same quality scene geometry that we're used to seeing from rasterized titles.

It was me...

[Anonymous Coward]

... I told them, smooth the picture bit... like divx smears/blurs the picture to compress it better.

This idea and the subsequent smoothers have made it a reality.

Far less rays needed now to produce acceptable quality pictures.

Amazing how one little idea can change the world ! ;)

It didn't....

[Anonymous Coward]

Is this a trick question?

Complexity

[fyngyrz]

Handling reflections perfectly is trivially easy with raytracers, and there have been plenty of real-time raytracers even for low-end consumer hardware in the 90's.

Even with a simplistic ray model, the more objects there are in the scene, the more complex the trace becomes. Until you hit an object, you don't know at what angle the ray diverges from that object; it may, for instance, go through several partially transparent objects before it reaches a light source, or it may reflect off of several items that are colored, or metallic (which, if done properly, introduces further complexities.)

The classic ray tracer has to examine every object N for every pixel generated M:. So the more objects and pixels there are, the longer it takes to render the scene. Complexity is straight-up NxM. This is without even thinking about things like soft shadows.

That is the bottleneck; and you're not going to solve it for all scenes just by throwing hardware at it.

You can solve it by simplifying the scene until the hardware can keep up, and as near as I can tell, that's what's being described here.

Or IOW, there's more hype here than truth.

Re:Complexity

[guruevi]

It's not just simplifying the scene, it's also simplifying the ray tracer model. Instead of tracing every pixel, it takes a group of pixels and renders them all the same and then applies some filters after the fact to make it look better. So instead of tracing the source of 1920x1080 pixels it's tracing perhaps a few hundred 32x32 or 64x64 sections and then applying a denoiser.

Re:Complexity

[fyngyrz]

Which is another (rather, a further) way of saying "it's not really a ray tracer at all" and therefore no, we don't have realtime ray tracing. :)

You ask why?

[Anonymous Coward]

because you're clueless...Have you been reading the graphics roadmaps?

This was all layed out in the roadmaps in the early 2000's and so far they're right on schedule for 8k everything.

Everything you know is wrong.

[Anonymous Coward]

In 2003, I was shown a video during governmental / military training where the claim was made that "in 10 years, we will be producing real-time rendered video that is indistinguishable from actual real-world video" This was a civilian made video highlighting some awesome (for the day) work that was happening at some university.

The copyrights at the end of that video made it clear that it was already 10 years old. Don't believe everything you see.

So there's that but then there's also the obvious cache of civilian tech that is warehoused just waiting to be released at the right time. When I was a child, they told me about cameras being everywhere, debit cards, e-ink and many other forms of future tech that just sorta trickled into mainstream over the next 25-30 years.

Got kids in school? ask them about what they are being groomed to accept, perhaps they'll have a glimpse into whats coming in the next 25-30. Some will probably be very cool - some will probably be very scary.

The simple answer.

[Anonymous Coward]

Moores law is not dead.

Re:It sounds like you're asking a different questi

[Anonymous Coward]

No, graphics hardware has been steadily marching this direction for years. CUDA enabled crypto, not the other way around. GPUs moving to general purpose cores was driven by games needing different amounts of processing power at each stage in the graphics pipeline.

Blinn's Law Does Not Say This

[Roger W Moore]

It may never be doable because of Blinn's Law.

Blinn's law says that the time to do something remains constant with the usual result that improvements go into the quality of the result. This is pretty self-evident with real-time video since your time budget is fixed by needed to maintain the frame rate and for any given frame rate you might as well use all the time available. Hence, it says nothing about whether "true" real-time ray tracing will ever be possible. The real question is not whether it will become possible but whether it will become practical. This is determined by whether or not there are alternative, less computationally intensive algorithms that produce visually similar (or better) results for that frame rate. If there are not then real-time ray tracing will the be the algorithm of choice.

I Don't Care

[Anonymous Coward]

To me, this is progress and it's exciting. I've been asking for R-T ray tracing ever since I heard about ray tracing. We gotta start somewhere with consumer access and this is it. It might be crap at first but it will get better and the tech gets all the advantages of mass market, mass adoption. Hopefully the games designers just use it in limited ways initially so it always looks good.

We've had so many iterations of GPU tech now. I've benefited from the larger screen sizes and better resolution, certainly. However I've noticed that in game titles, even game remasters/remakes, the reviews will tout how much better the game looks. I look at the comparative shots and yes it's better, but I often find myself a bit underwhelmed. There's no, "holy crap, that's awesome!" moment. Maybe the designers of the earlier version pulled off miracles?

R-T ray tracing has the potential for "holy crap, that's awesome!" moments.

It was definitely . . .

[Anonymous Coward]

Aliens.

the uneven future, since forever

[epine]

Or even 30 years ago ... well ray tracing was anyway.

Back in 1980, the University of Waterloo mathematics and computer science building had a locked public display case featuring artifacts from the senior-level computer graphics course, most of which involved ray tracing (standard chessboard-reflected-in-shiny-sphere kinds of things, but with the scenes aggressively simplified—like only three chess pieces of a dozen polygons each and the board reduced to sixteen squares).

What separated the great from the good was treatment of subtleties such as getting the specular highlighting right. I was young and naive and didn't have Wikipedia at my fingertips, so I don't recall much more.

They also had a public information kiosk back in 1980 which used some kind of polygon-fill graphics language to render an interactive dial-up information browser in all the best oversaturated colours. Just like the Internet, if the Internet consisted of exactly one host, and it was dead slow. But shiny! Everyone tried it out—for exactly three minutes (that would be about your third screen rendered).

They also had rooms full of card readers, which they couldn't bother themselves to eliminate until some fancy anniversary shindig a few years later.

In one terminal room you'd have IBM 3270 block-oriented displays, in the next you'd have the god-awful WIDJET terminals, in the next you'd have IBM PCs running APL (for statistics students), in the next you had Commodore SuperPETs, custom tweaked by the Computer Systems Group or related ecosystem (these people were later responsible for the Watcom C++ compiler).

WIDJET (Waterloo interactive direct job-entry terminals) was basically a JCL front-end with the ability to store about six whole files on non-card storage, where mostly you sat twiddling your thumbs waiting for your runs to pass through the job submission queue.

One such room was more advanced and you could *edit* your next assignment with your previous run still in the queue, but you had to sign up way ahead of time to actually get a seat in this room.

In the other room, you might submit a compile, sit there for ten minutes watching it bounce up and down the queue (upper-year jobs had priority) and then decide "oh, I did something wrong in the previous assignment" and you'd click "edit" on some previous source file, only to discover the message "job submission cancelled" popping up on the status line of your display and since you'd already spent years with the immense power of the TRS-80 or Apple II at your fingertips, you'd bash your head into the desk and wonder how you'd become mired in this technological institution of hazing, abuse, and mediocrity.

The SuperPETs were okay, but the custom Pascal did some kind of partial compile to catch syntax errors before running the interpreter. The error reporting was beyond horrible. 50% of all syntax errors were the same message: "syntax error near or before end of file." It was much like Donald Trump tweaking "WRONG!" at you if you got a single brace out of alignment.

"Sorry! I'll hunt through my entire program looking for the one I messed up." This worked for localized changes. But generally what you actually did after a big edit was added or removed braces at random until you found the problem through a tedious process of bisection (which, however, was an order of magnitude less tedious than dealing with WIDJET, even in the good room).

My first year at Waterloo was the biggest computer science mindfuck of my entire life.

Eventually I discovered you could kind of get into a groove with the IBM 3270 terminals—as archaic as they seemed—if you blinked longer than normal after each press of the ENTER key. Plus, these had chat, so you could send "nice sunrise" messages to all your friends at 05:00.

Fro

People in the know

[Bengie]

People who are good at what they do are typically monitoring or communicating with other people who are good at the same thing.

Or they're at least all watching the same thing and great minds think alike and all that. Kind of like telescopes. No one had even thought of a telescope, yet when optics became good enough, the telescope was independently invented by several different people within a year of each other, but all far enough apart that communication among them would have taken longer than a year, so it was impossible for them to have known what was up with the others. Most ideas that seem novel are really obvious to masters of the domain as long as they're presented with the same problem. This has been a sore point of our patent system for a long time. Just because something has never been done before doesn't mean the person who made it is special or put a lot of effort into it. They were just lucky with timing. Yay, lottery!

My guess is a combination of masters being masters and all saw the tipping point around the same time, and keeping their ears to the ground.

2004 /. article on Beowulf cluster Quake iii rays

[RalfM]

I remember back in 2004, when Beowulf clusters were still a cool thing, Quake iii getting the real time raytracing treatment. Great to see it in individual machines now, but I also wonder what cluster tech has replaced the old Beowulf...

https://games.slashdot.org/story/04/06/07/2350243/quake-iii-gets-real-time-ray-tracing-treatment