I've been curious about muscle simulation lately. Found a pdf that describes a way to procedurally animate extra bones that represent muscles. Linked it in this thread here:
But I guess nobody has any experience with it yet.
the simple muscle system in euphoria isn't as sexy as sliding under skin and bulging/stretching but the ragdoll fidelity from it along with the "self preservation" ai is pretty inspiring next gen.
the simple muscle system in euphoria isn't as sexy as sliding under skin and bulging/stretching but the ragdoll fidelity from it along with the "self preservation" ai is pretty inspiring next gen.
That's cool.
In the pdf I linked they're describing a system where, for instance, you'd place a bone for the biceps. The animators won't control that bone, so there's no extra animation-time required, but instead the bone will translate based on the rotation of the lower-arm bone. So the more you bend the elbow, the more the bicep-bone will move, the more the bicep will look like it's flexing.
I was thinking this may be possible with a SkelControl in unreal, but haven't tried it. Was just wondering if anyone else try it before, or if it has been used in a game?
With all the realtime raytracing software being developled (iray and such) I'm very excited for the future of gaming. Especially with Dx11's dynamic tessellation. I can't wait until my in game assets look exactly like they do in my sculpting package! The thing that scares the most about next gen is the amount of time it's going to take to put all those details into the models and textures.
The art Pipeline is going to have to change drastically to keep up with everything. We need more tools like Weta's Mari but better optimized for games. The game I'm working on now requires a lot of detail on large spaceships(these things are supposed to be about 1/4th the size of a small planet). Took me over a week and a half to paint/texture these ships in photoshop. I downloaded the trial edition of mari and I did an entire ship in 3 days... and the first day was spent learning how to use the tools.
The next hurdle to overcome is all the time it takes to fix things when you have make changes. A non destructive iteration method. So that changing a characters geo wont break the rig, etc, etc. I'm sure it's possible. It's just going to take some creativity.
The newest Game Developer has Bungie's senior graphics engineer Hao Chen in an interview discussing next gen 2.0 strategies and their top priorities for such challenges.
As Bungie prepares not only for its leap onto non-Microsoft programs, but also for a transition into the next generation, we spoke in-depth with Chen about the future of graphics on consoles.
He raises a good point concerning our failures even with the current generation. Namely good shadows!
I've certainly seen how even today people are having a lot of trouble rendering shadows without a lot of blockiness or dithering.
HC: That's kind of the problem with computer game graphics these days. A lot of things people consider solved problems are actually quite far from being solved, and shadows are one of them. After all these years we don't have a very satisfactory shadow solution. They're improving; every generation of games they're improving, but they're nowhere near the perfect solution that people thought we already have. What do you think might be the answer? Your potential megatexture solution, or something else?
HC: We are still far from seeing perfect shadows. Shadows are a byproduct of lighting. All frequency shadows (shadows that are hard and soft in all the right places) are a byproduct of global illumination, and these things are notoriously hard in real time.
There's just not enough machine power, even in today's generation or the next generation, to be able to capture that kind of fidelity. There are also inherent limitations to the current techniques, such as shadow maps, for example. When the light is near the glancing angle of a shadow receiver, then it is impossible to do the correct thing.
With the current state of the art shadow techniques we can manage the resolution much better, and we can do high quality filtering, but we still have long ways to go to get where we need to be, even if we just talk about hard shadows from direct illumination.
I think megatextures could help, but still fundamentally there are things you cannot solve with our current shadow meshes. And until the performance supports real-time ray tracing and global illumination, we're going to continue seeing hack after hack for rendering shadows. Every year we see a few new hacks of current techniques, and with each hack, we see a little bit of improvement on the quality.
Kind of depressing outlook... coming from a traditional background I dream of a day when I can compose more with light and shadows again in what will be the new predominant art form to replace cinema.
I would give up normal maps and tessellation both just to be able to paint like:
Composing equally in this way with light and shadow and next gen surpasses cinema easily as the pre-dominant art form. Would also do more for defeating the uncanny valley or creating what is beautiful in Hyper-Real when realism just isn't good enough.
I would give up normal maps and tessellation both just to be able to paint like:
This is actually a very interesting point. My take on this is that, all the shots you linked to could have been achieved in games made ... back in the PS1 era. Now obviously I am not talking about crazy detailled tech for surface rendition and lighting - just talking art direction and shot design. Such things are why these images are so striking, and games like Vagrant Story, MGS1 and many others show that this has always been possible!
Now of course there would be a lot of trickery involved, to a point where a scene could not work from all angles, for instance. But ironically that's exactly how movie shots are made!
Recent games with powerful engines like Crysis2 or Battlefield show the highest level of fidelity we can hope to have to in the future years, with everything realtime and "next gen". But ironically I think these games are visually quite inferior to what ID achieved with Rage, with everything baked down to diffuse textures. It just looks so much more solid.
My hope for the coming generation of games is more better, striking art direction and less of a race for fancy graphical features. Games look very convincing already - no need for crazy hair sims or super high-polygon displaced meshes ... especially if it slows down production even further than today!
I would have to disagree myself. The best motion Pictures are still more immersive of an experience than my best gaming experiences and what I find striking is how immature and hackneyed a lot of the imagery is in games even with current gen. If I believed that matching cinematic light and shadow fidelity would never be possible someday, I would never have left broadcasting in the first place. If there were PS1 gen dynamic self shadowing shadows with the fidelity to recreate soft high key dramatic compositions like those pictured I would sure like more info on that technology. Otherwise I agree with what the bungie engineer was saying. We have nothing truly revolutionary techwise for dynamic shadows. And if Maxwell technology 1 year away is truly 25x more powerful than fermi it sure would be nice if some of that power was spent on a beautiful shadow solution instead of having to "BAKE" that beauty down in an art form that should be composed with motion in mind.
What those still photos don't show is that same fidelity "in motion" though I am a fan of Jordan Cronenweth's photography in Blade Runner. ( the last photo ) The compositional beauty of light and shadow in a piece of work by Cronenweth or Toland or any other giant is Kinetic. ( Motion Pictures )
I would agree that complex muscle systems and tessellation is super sexy but yes that level of "dynamic" lighting is the holy grail.
Not that Noseferatu's shadow had delicate sensitive shadows in it's day... but dynamic shadows can be very fun when not locked down. And creatively thats where the handicap can be frustrating. Still think it's only a matter of time though.
nosferatu has fun shadows but toland is still king ( also if I remember right this shot was actually animated )
btw: according to wikipedia tha above image and I imagine the film "nosferatu" is public domain at least in the US.
Would also do more for defeating the uncanny valley or creating what is beautiful in Hyper-Real when realism just isn't good enough.
Realism is never good enough, that's why you have to art direct, light and compose stuff properly :P There's a huge gap between looking real and feeling real.
We kind of have basic tools for achieving dramatic feel. Yes, shadows might bleed here and there, but it usually it doesn't do much harm if shot is properly lit and composed. Lots of games have contrasty lighting and filmic colorgrading/tonemapping nowadays. I'd love to see more interactivity/freedom in games. Static worlds and locked shots kind of defeat purpose of games, no matter how carefully tweaked they are.
What about volume shadows? I've read through a few papers on the, and they seem to be capable of hi-fidelity, highly accurate soft shadows. Apart from nobody being able to use them, of course...
Real-time global illumination and lighting is another issue. Frostbite 2 did something similar, didn't it? Real-time radiosity, if I recall correctly, on the PC. Ambient occlusion (other than SSAO) would be nice, too.
synched aural foley landscapes orchestrated as intrinsically as the score.
Got my Game Developer Magazine in today...
I suspect Jessie Harlan who writes the Aural Fixation column might have been snooping around pc.
Either that or there are more folks more hip than myself using next next gen to discuss the things to come? The title of the column this month is... Whats Next for the Next NextGen?
Usually his articles hint at the future state of the art in audio techwise and creatively but this is the first I have read where he flat out predicts a new interactive aural composition for game score and audio effects...
Complex instance culling and
stream management systems are already a
must, and will only continue to become more
of a fundamental need. Sound designers can
expect the usual fidelity increases to both asset
bit and sample rates and, subsequently, the
usual increased impact on both memory and
disc footprints.
But thats the small stuff. The next generation
of game sound is going to be about maturity.
Weve shown how big the worlds we can create
can be. Now we need to show how well we can
get them to sound.
As such, the new frontier of sound design
is mixing and mature implementation. The days
are gone where its acceptable to simply have
static master levels for sound and music that are
occasionally ducked by voice. Nuanced mixes
and intelligent systemic mixing systems are the
next big focus. As we gain the ability to add more
real-time convolution reverbs and more detailed
surround ambiences, were going to need the
ability to deftly sculpt frequency space and create
situational mixes that change depending on player
feedback and myriad shifting game states
He goes on to write how cloud computing will make null the disc space required to make such progress possible.
People get into CG for a variety of reasons but I find that very few get into game cg for different reasons. Usually they grew up loving their bit graphic donkey kong past and as far as they are concerned they are living the dream ( unless in an abusive environment of course ).
To others ( albeit the minority ) as technology evolved it became apparent that realtime graphics is/will be a medium without boundries or restrictions and all things being possible awakens their creative passion for what may be.
If you are that person... then that day has not arrived. yet.
When it does I imagine what that camp works on will not even resemble a game. But just like what is great in the best of cinema... will be as immersive and compelling as any self absorbed button mashing experience.
If you are the type of person who can open up the Stone Giant demo or an area of RAGE that doesn't require any gameplay...
And just be happy to wander for the sake of immersion alone. Than u just might be a redneck... er? I mean u just might belong to that aforementioned camp.
I suspect however that Jessie's vision of a new paradigm for interactive realtime musical soundscapes will not evolve as smoothly as he envisions. Sharing for a long time the same desire for such an interactive experience I have found very little reception for such progress. Not that there is negativity but the same dynamic where traditionaly composed scores and current recording engineering is the end all, be all solution where orchestrated one-off compositions are the gold standard.
And with a legacy of great musical moments how could anyone argue otherwise?
In which case how does evolution happen if percieved as unnecessary or perhaps even threatening to what is comfortable?
I imagine the only way is to just to see it for what it truly is? A different thing. Not necessarily superior but evolution none the less. In which case nintendo old school is just dandy doesn't mean that someone chomping at the bit for technology that opens a door some have been standing in front of for a very long time doesn't have a valid contribution.
I think we simply all look at this from the narrow end of the issue. That's only a natural thing, because we are all so involved with the creation of the assets, sculpting them, baking them, and so on. But all it comes down to is : an AD needs to know exactly what he/she wants in order to have it achieved in the game, regardless of tech level.
If someone had shown an image similar to that Vagrant Story screenshot above before the game came out (say, at the beginning of the PS1 era, when stuff like Ridge Racer was at the high end of the visual fidelity scale), everybody would have said it was impossible to achieve on the current available hardware. Yet these guys knew what they wanted, including that nice rimlight effect that was never seen before. But they had fantastic traditional artists who knew exactly what they were doing, and together with engineers they found solutions. IIRC the thin rim light here is not achieved with lighting, but through a brightened, offset copy of the silhouette, therefore providing more fidelity than a regular highlight would have. How amazing is that!
I think most predictions of graphical enhancements for next-nextgen are just the usual tech dreams we keep hearing all the time. (No more normalmaps! It's gonna be amaaaaazing!!) However that's not how real time graphics tend to evolve! We take the path of optimization, better texture compression algorithms, smarter caching of assets - not the one of wasteful use of ressources with the excuse of more horsepower.
Looking at games today, the few lacking areas are tings like animation blending ; environmental reaction ; draw distance ; and some materials that could need some improvements. All the rest are just problems "by design" : AI behaving poorly, crappy game mechanics, and so on. These don't improve magically with tech cycles
Visually we already reached what we need in terms of visual fidelity. Shadows aint great but they are good enough. Remember how good Resident Evil 5 looks ? The Steve Mc Curry quality of some of the screenshots could easily pass as ps4/nextXBox stuff. Again, the pillars on that Vagrant Story screenshots cast a wonderful soft shadow. Of course it is just a painted billboard in the background. But when you think about there is close to no difference between this, and the way Rage handled environment cast shadows, all "painted in" by the baking process.
Time to really play with all this incredible tech we have and start making beautiful art with it!
Awesome example of excellent design PIOR! I wish I had a ps3. ( nice of square to release on ps3 )
However, knowing that amaaaazing advancement is coming either way... I am mighty glad to have bet on the level of fidelity heading our way. Either way u look at it. It's win win. ( With us nuts craving that level of shadow detail... in the end is still a good design craving? )
I never found denying following a craving for artistic bliss to ever end negatively.
A great ps1 example still does not diminish what will make a huge impact with the lemmings that finally play the game. The audience does like their precious just as much as the artist who craves that brush.
Fer Instance I have yet to meet the musician who makes the point that great music should be possible on toy plastic guitars and in the same spirit offer to give up their Gibson SG. And thankfully for good reason. The Gibson SG does actually offer quality that translates into Bliss.
( cinematic levels of software rendered real time by 2020! Thats only 8 years! ) http://graphics.cs.williams.edu/archive/SweeneyHPG2009/TimHPG2009.pdf
What I never expected was that my phone would some day be in some respects: a super computer compared to the Avid system I began on 14 years ago. I think you may have hit the nail on the head in regards to a narrow focus. However, I think it comes from assumptions held long before switching realtime graphics. ( perhaps some artists are just plain shadow nuts ). To whom the current dynamic shadow solutions "actually implemented" hardly seem to be the promised land envisioned by such a nut... ( or "good enough" )
But whose general purpose/usage seem nothing more than a method to seperate characters from the ground plane.
Again, the pillars on that Vagrant Story screenshots cast a wonderful soft shadow. Of course it is just a painted billboard in the background. But when you think about there is close to no difference between this, and the way Rage handled environment cast shadows, all "painted in" by the baking process.!
It's not always that simple though, which is why a companies such as bethesda that owns Id are still better off from not using idtech5, in fact, skyrim could not have been made in the same shape it has today if using idtech5.
The best looking games of every generation has always been the most static, non-dynamic and locked games you can find in terms of scenery, and they had so much in common with movies, much like you said.
but designing a game is not always like designing a movie-set, it's like designing a theme-park of the highest quality, where you can go everywhere and look behind the houses.
A game like rage can get away with what it does, because the game will know where the player can and cannot go, and bakes the whole mega-texture accordingly. There is no such thing in something like skyrim, there is no place where the player can't go, and there's no way to bake anything down.
In the end vagrant-story looked amazing due to amazing texture-work, small enclosed areas for performance, and locked cameras and very few character in the scene.
What happened was that on the successor of the ps1, the ps2, you could have these kind of visuals in a fully detailed world, without a locked camera, where you have the freedom to move around and not be limited in the amount of characters.
Yup Eld, I totally agree - visual trickery is just only trickery I think what comes out of your examples (Skyrim being all dynamic, and the design process behind VG) is that ... we don't need to wait for tomorrow's tech to do awesome games today!
If anything, I think that the understanding that we reached very high visual fidelity already might (re)open the doors to strongly stylized projects. We know that we can achieve BF3 and Forza quality now ; since this "next gen race" is over, now (that is to say : "next-next-gen") might be the time for more visually original titles, like Journey or The Last Guardian, using tech as a mean and not an end. Going back to simplification, abstraction, and impressionism ...
Yup Eld, I totally agree - visual trickery is just only trickery I think what comes out of your examples (Skyrim, and the design process behind VG) is that ... we don't need to wait for tomorrow's tech to do awesome games today !
If anything, I think that the fact that we reach such high visual fidelity already might (re)open the doors to nice visually stylized projects. We know that we can achieve BF3 and Forza quality now ; since this "next gen race" is over, now might be the time for more original titles, like Journey or The Last Guardian - using tech as a mean and not an end !
BF3 struggles to achieve PC fidelity on the consoles though, and 30fps at that, so a bunch of that fidelity is reached by strangling the console to death
But truly: The artist however is the last outpost of all game art, and even on the ps2 I would find some of the most beautiful games I've ever played, god of war 1 & 2 already looked amazing on the ps2, but then came god of war 3 looking even more amazing.
I don't think we'll ever reach that 'enough is enough' point, and at the point where we'll reach complete photorealism we'll also most likely see some of the most mindblowing artistic creations that moves away from photorealism yet using all that power.
We don't need the coming generation to make fantastic looking games, but it gives us more room to make fantastic looking games in genres that didn't have the resource-budgets to look fantastic.
that was my theory too, pior, and I really, really hope that's the case...
but I doubt it.
I've seen far too many consumers bitch about products because they weren't 'realistic' looking, even if realistic means noisy and fuck ugly textures... there's always going to be companies that simply shoot for that even if they can't achieve it remotely successfully
We don't need the coming generation to make fantastic looking games, but it gives us more room to make fantastic looking games in genres that didn't have the resource-budgets to look fantastic.
I agree. I think there is still potential room for making the fantastic looking games/movies look even better though. I'm saying this because you never know what cool technique(s) could come next, I never imagined that games and movies could look as good as they do now about 20 years ago. Partly because of the quality of TV's now and because they use a combination of different lighting/effects/compositing/etc that wasn't used back then, people weren't able to predict how well it would look. Sure, I was a kid back then but my imagination for games like Sonic and Super Mario Bros went as far as animated motion pictures like the Disney ones.
But disney movies are a great example of how fantastic artistic genius and skill dominated the less significant technical achivements. Art direction defeats all! An open ended game can't necessarily have low level visual fidelity like a linear game does, and that is something you can design around. Where GTA4 used physics, clever use of (exaggerated) particle effects and lighting, and careful direction of the style of the animations and the textures to make the game look really goddamn good all the time even if there aren't giant textures and fine lighting effects.
I think you misunderstood me, Suprore, mainly because I don't think I was being clear on what I meant. I didn't pick Disney feature films as a bad example, they're great for their time. There is a technical inferiority however if you compare it to today's games, not only because of our time's machine capability. Compare these two for example;
[ame="http://www.youtube.com/watch?v=OQXY1WQkhdw"]http://www.youtube.com/watch?v=OQXY1WQkhdw[/ame]
With
[ame="http://www.youtube.com/watch?v=rAx9Q9z4Pdo"]http://www.youtube.com/watch?v=rAx9Q9z4Pdo[/ame]
I believe the little mermaid is a perfect example of how good a feature film could be at that time. It's great but the characters' cels are very flat colored, of course, and the backgrounds are static. Trine 2 looks like an animated painting and my imagination didn't go that far, I'm not sure if anyone could predict something as amazing as this back then. Who had seen a painting come to life this clear and realistically? Also remember that TV's quality wasn't as clear as today's TVs and that didn't make imagining something like this any easier.
Yeah Eld, I totally agree with you on the "breathing room" point. If "next-next-gen" end ups meaning true 720p at 60fps for all games, I'd be happy already!
Goraaz, i feel you now, I think you and me just differ in taste. Both examples are fantastic, but to me the little mermaid holds a LOT more artistic value to learn from and admire. Trine 2 is beautifully realized and full of amazing art, but tlm has more stunningly exemplary artistic skill on display within the constraints they had to work with.
But yeah, regardless, the more games can reconcile an artists ability to design with their ability to realize images onscreen, the better.
Better be true 1080p, I don't get everyone's obsession with 60fps, as long as it doesn't drop below 30 you can't really tell. But we watch movies and tv all the time filmed at 24, and never complain about it. With motion blur its a lot harder to tell the different between 24 and 60.
Better be true 1080p, I don't get everyone's obsession with 60fps, as long as it doesn't drop below 30 you can't really tell. But we watch movies and tv all the time filmed at 24, and never complain about it. With motion blur its a lot harder to tell the different between 24 and 60.
Well, that's like, your opinion man
It might not affect you, but it does affect some people a huge deal. I think the difference is very noticeable, and impacts gameplay a lot. Basically, games running around 30fps hare not only less smooth, they also tend to have a lot of post-processing going on, causing frame delays. It is quite noticeable in the console versions of Crysis2 and Gears Of War 2. Compare these to the tight response time of MW2 or Rage : the difference is huge!
I agree with Pior, Eld and some of the other guys... its all about art direction.
Check this out, these are some screenshot I took from a 12 year old (!!!) PSX game called Breath of Fire 4 (really cool rpg):
These are all full 3D environments usually covering large areas and in most cases you could rotate the camera in steps of 90 degrees as well (notice the icon in the top left). To me these graphics are more impressive and beautiful than the ones Ive seen from recent AAA title rpg's. I studied the graphics by extracting the 3d models and textures with various tools and the developers pushed the console to it's limits!
Surely new hardware would make thing easier, but I think without restrictions art direction tends to get a little sloppier due to the amount of work/assets that needs to be done. I recently finished Fallout: New Vegas, a perfect example of rushed sloppy 'next-gen' art direction if you ask me, it had no soul.
Surely new hardware would make thing easier, but I think without restrictions art direction tends to get a little sloppier due to the amount of work/assets that needs to be done. I recently finished Fallout: New Vegas, a perfect example of rushed sloppy 'next-gen' art direction if you ask me, it had no soul.
That stands true, which is why skyrim, which is just about the same kind of scope looks leagues better than new vegas, even though it runs on exactly the same hardware.
i've been waiting for games to have anti-aliasing since the n64 and i'll keep waiting until we get there. narrowing scope because of artistic vision and because of technical limitations are completely different.
It's totally cool to have upscaled sprite art in a commercial game if you want (for example MBAACC) but it sucks if you are forced to do it because of hardware (i'm looking at you, half-res fx buffers).
I recently finished Fallout: New Vegas, a perfect example of rushed sloppy 'next-gen' art direction if you ask me, it had no soul.
Oh cmon, next-gen? They barely have shadows :P That game have lousy tech, and pretty lousy art aswell. I dont think any amount of art direction would have saved that.
I dont know, I dont think good art direction solves everything, there are tons of cool things that you simply cant do, or that will look like shit on todays hardware. And the problem is that to get even CLOSE to the artistic vision you had in mind, you have to be extremely technically proficient, and you have to invest alot of man hours to optimise everything so you can actually get the result that you wanted. More power = less limits : D
Goraaz, i feel you now, I think you and me just differ in taste. Both examples are fantastic, but to me the little mermaid holds a LOT more artistic value to learn from and admire. Trine 2 is beautifully realized and full of amazing art, but tlm has more stunningly exemplary artistic skill on display within the constraints they had to work with.
But yeah, regardless, the more games can reconcile an artists ability to design with their ability to realize images onscreen, the better.
You're right, perhaps Trine 2 wasn't a good example. My point was that when new technique is introduced artists try out methods without exactly knowing how the results will look together with other cool techniques. I brought up Trine 2 as an example because it's a good looking game. Perhaps the princess and the frog would be a better example if we're comparing it to MLM? It's an animated feature film but the combination of modern day effects and compositing choices make it look in a way we couldn't predict about 20 years ago. Heck, years ago some disney people thought that 2d was over and decided only to go with 3d but Princess and the frog along with other 2d feature films proved them wrong.
What I'm also looking forward to is a better and more real sounding speech synthesizer. We don't use the voice actors' faces for the games a lot and using these synthesizers will eliminate them from the equation completely.
Better tech isn't going to hurt art design, it's only going to give it more breathing room. Sure someone could work in mspaint, one pixel at a time, and approach the same results as in photoshop, but tech has made it so that you don't have to spend a ridiculous amount of time to paint a decent high-resolution painting. And that's just 2d, which is infinitely less dependent on the tech. What about getting to the point where I could sculpt just about anything, drop it right into the engine, and have it stream in with all its detail kept intact? Until then what I make is being reduced, emulated, and approximated to work in real time.
If people want to, they can always go back to old tech and use that, but I don't see many people eager to run back to painting one pixel at a time in mspaint. As tech progresses forward we don't abandon what we were able to do before; we make it so that less "hacks" have to be implemented to approximate what we want, we increase the fidelity and number of factors we have control over.
When I see an older game like Vagrant Story or even a ps2 game like SotC, it makes me wonder how much better the artistic vision could have been realized with today's resources. They were definitely fighting against what they could do technically, and that definitely had an impact on the execution of the artistic vision.
As noble of an idea as it is to think that artistic vision exceeds everything, the tech is what produces and displays the final results. I don't see how people here of all places could be so dismissive of its importance.
Good design or bad design or "soul" has absolutely nothing to do with technological advance. You could 1up the world with lo-fi and work with the constraint of crayons but it will not mean your work has any more a valid claim to "soul" or good design. Fidelity is not the evil enemy of good design. A bad design is a bad design nothing more nothing less. Seeing beauty and what is clever in lo-fi simply makes you a fan of low fi. I am not sure how any creative person believes that our current levels of technology will not surpass current fidelity levels of realtime rendering. If that time has come the world would have stopped software rendering by now. As far as I know that time has not "yet" come. Even then if we wake up the next morning with a box that can render immersive hologram environments, That no one would find value in realizing that potential is a rather scary thought. That some should conserve traditional pipelines is noble... sure. But I really have to wonder what ulterior motives there are to nay saying in a next-gen thread? suggesting Fidelity = no soul? If you want to posture and be hip then go ahead and make a meatboy or a limbo. Such lo-fi work is amazing considering the constraints. OBVIOUSLY good design. Being a luddite on the other hand, does not make you hip. Considering that the current gen has only really matured towards the end of it's lifecycle. I really do believe that we are NOT experiencing the same kind of blind cycle. And that the talent is larger and tech-savvy. I can't disagree that there are plenty examples of bad design work out there. Perhaps leveraging dynamics shadows simply to seperate characters from the ground plane when advanced shadow fidelity will allow for "dynamic" compositions" is one of those bad design choices we should be discussing?
Rewatched "The Long Voyage Home" and after finishing I have to admit, that Pior seemed to be right. That most compositions in the best of work could be recreated with todays restrictions. Even though there doesn't seem to be an equivalent. And for no good reason.
You can actually physically see the difference in framerates up to around 120hz or so. Thats pretty much a fact, the difference only becomes apparent with very rapid camera motion though (FPS games).
I rather think that instead of buffing and improving the graphic and make people upgrade graphic cards all the time, that they should work more on AI, so the worlds can be proper dynamic and not just about a box falling the right way.
Good example is Skyrim and Deus Ex, with buckets on heads and people not noticing that you are hacking their terminal...that they are sitting in front of. Need more work on AI, maybe some kinda of DirectAI library which can be improved all the time, module style. Add this Walk-block, attach this 2-eye-front-view-block, with a biped-animal behavior-block and run it.
I rather think that instead of buffing and improving the graphic and make people upgrade graphic cards all the time, that they should work more on AI, so the worlds can be proper dynamic and not just about a box falling the right way.
Good example is Skyrim and Deus Ex, with buckets on heads and people not noticing that you are hacking their terminal...that they are sitting in front of. Need more work on AI, maybe some kinda of DirectAI library which can be improved all the time, module style. Add this Walk-block, attach this 2-eye-front-view-block, with a biped-animal behavior-block and run it.
Most people get it wrong though when they talk about good and bad AI, the bucket on head is a side-effect of much better AI, where crimes committed are the ones the npc will actually see, and in this case a bucket will actually block his eyesight and preventing him from seeing it.
Proper scripted behavior could've prevented that scenario, but that is countless of more scenarios that would have had to been taken into consideration.
The key word is fun and believable behavior, which can be done with the really simple and basic AI, even the most advanced AI can be blamed as being bad by the layman gamer.
In terms of AI related libraries, path-finding is one of those things that could be made cheaper, it's one of those things that costs a ton, the rest is just properly believable scripted reactions and behaviors.
You can actually physically see the difference in framerates up to around 120hz or so. Thats pretty much a fact, the difference only becomes apparent with very rapid camera motion though (FPS games).
The reason why 30 or even 60 fps aren't enough for games is because of temporal aliasing, ie stuttering between frames. If you have a movie clip of a completely black screen slowly blending into white then 25 fps will be more than enough, but every other frame being black and white might very well need over 100. it gets a lot harder to predict if the images are more complex than just solid colors, but the key factor here is (temporal and spatial) hyper-acuity, or the ability of the visual system to percieve aliasing beyond their "physical capability". Even on a 600dpi screen you can still percieve text aliasing quite clearly. Throwing more frames or higher pixel density on the problem is hella brute force, and doesn't really solve the underlying issue.
Good luck getting shadow quality like that with the current gen hah. Shadows are still so shitty; cause of strobing, small map sizes, and other elements.
The reason why 30 or even 60 fps aren't enough for games is because of temporal aliasing, ie stuttering between frames. If you have a movie clip of a completely black screen slowly blending into white then 25 fps will be more than enough, but every other frame being black and white might very well need over 100. it gets a lot harder to predict if the images are more complex than just solid colors, but the key factor here is (temporal and spatial) hyper-acuity, or the ability of the visual system to percieve aliasing beyond their "physical capability". Even on a 600dpi screen you can still percieve text aliasing quite clearly. Throwing more frames or higher pixel density on the problem is hella brute force, and doesn't really solve the underlying issue.
That's what I was looking for, at 60fps it may feel more responsive, but it probably wont look any different to anyone watching, if it is done correctly.
Trying to render frames twice to four times as fast is a bad solution.
but it probably wont look any different to anyone watching, if it is done correctly.
Trying to render frames twice to four times as fast is a bad solution.
Not trying to be a dick, but do you have even a cursory knowledge of how these things work?
My knowledge stops at right about 'cursory' and it's pretty obvious to me how doing it 'correctly' is probably about a million times as expensive as just rendering it four times as fast. Your 30 fps game with film camera quality blur and depth and anti aliasing and lens effects is a none fps game.
You're suggesting we lower our goal to 30fps so that we can use more expensive rendering, and then when you hit the '30fps looks like shit' obstacle, you're proposing we do something exponentially MORE expensive to make it okay. Why would we want to run a game at 30fps instead of 60fps, and then use all the graphical power we earned just to make it look as smooth as it did when it was running 60?
Or are you proposing we run the game engine at 120 fps or whatever, to respond to player input, render it at 30fps, and write some kind of super predictive magic code to make it feel responsive and intuitive even with the desync between what the player does and when they see it happen?
Replies
Maybe thats what he's talking about, two generations from now... now THAT's a scary thought!!
[ame="
For now it got stuck on the 360/ps3 era, usually any kind of heavily shader-based game, even if it is incorrect.
That's cool.
In the pdf I linked they're describing a system where, for instance, you'd place a bone for the biceps. The animators won't control that bone, so there's no extra animation-time required, but instead the bone will translate based on the rotation of the lower-arm bone. So the more you bend the elbow, the more the bicep-bone will move, the more the bicep will look like it's flexing.
I was thinking this may be possible with a SkelControl in unreal, but haven't tried it. Was just wondering if anyone else try it before, or if it has been used in a game?
The art Pipeline is going to have to change drastically to keep up with everything. We need more tools like Weta's Mari but better optimized for games. The game I'm working on now requires a lot of detail on large spaceships(these things are supposed to be about 1/4th the size of a small planet). Took me over a week and a half to paint/texture these ships in photoshop. I downloaded the trial edition of mari and I did an entire ship in 3 days... and the first day was spent learning how to use the tools.
The next hurdle to overcome is all the time it takes to fix things when you have make changes. A non destructive iteration method. So that changing a characters geo wont break the rig, etc, etc. I'm sure it's possible. It's just going to take some creativity.
http://www.gamasutra.com/view/feature/6579/all_that_glitters_an_interview_.php
He raises a good point concerning our failures even with the current generation. Namely good shadows!
Kind of depressing outlook... coming from a traditional background I dream of a day when I can compose more with light and shadows again in what will be the new predominant art form to replace cinema.
I would give up normal maps and tessellation both just to be able to paint like:
Composing equally in this way with light and shadow and next gen surpasses cinema easily as the pre-dominant art form. Would also do more for defeating the uncanny valley or creating what is beautiful in Hyper-Real when realism just isn't good enough.
it would be sweet to see this become the norm, even cutting edge games seem to be using last gen FX, especially for water sprays.
This is actually a very interesting point. My take on this is that, all the shots you linked to could have been achieved in games made ... back in the PS1 era. Now obviously I am not talking about crazy detailled tech for surface rendition and lighting - just talking art direction and shot design. Such things are why these images are so striking, and games like Vagrant Story, MGS1 and many others show that this has always been possible!
Now of course there would be a lot of trickery involved, to a point where a scene could not work from all angles, for instance. But ironically that's exactly how movie shots are made!
Recent games with powerful engines like Crysis2 or Battlefield show the highest level of fidelity we can hope to have to in the future years, with everything realtime and "next gen". But ironically I think these games are visually quite inferior to what ID achieved with Rage, with everything baked down to diffuse textures. It just looks so much more solid.
My hope for the coming generation of games is more better, striking art direction and less of a race for fancy graphical features. Games look very convincing already - no need for crazy hair sims or super high-polygon displaced meshes ... especially if it slows down production even further than today!
What those still photos don't show is that same fidelity "in motion" though I am a fan of Jordan Cronenweth's photography in Blade Runner. ( the last photo ) The compositional beauty of light and shadow in a piece of work by Cronenweth or Toland or any other giant is Kinetic. ( Motion Pictures )
I would agree that complex muscle systems and tessellation is super sexy but yes that level of "dynamic" lighting is the holy grail.
Not that Noseferatu's shadow had delicate sensitive shadows in it's day... but dynamic shadows can be very fun when not locked down. And creatively thats where the handicap can be frustrating. Still think it's only a matter of time though.
nosferatu has fun shadows but toland is still king ( also if I remember right this shot was actually animated )
btw: according to wikipedia tha above image and I imagine the film "nosferatu" is public domain at least in the US.
We kind of have basic tools for achieving dramatic feel. Yes, shadows might bleed here and there, but it usually it doesn't do much harm if shot is properly lit and composed. Lots of games have contrasty lighting and filmic colorgrading/tonemapping nowadays. I'd love to see more interactivity/freedom in games. Static worlds and locked shots kind of defeat purpose of games, no matter how carefully tweaked they are.
Real-time global illumination and lighting is another issue. Frostbite 2 did something similar, didn't it? Real-time radiosity, if I recall correctly, on the PC. Ambient occlusion (other than SSAO) would be nice, too.
Got my Game Developer Magazine in today...
I suspect Jessie Harlan who writes the Aural Fixation column might have been snooping around pc.
Either that or there are more folks more hip than myself using next next gen to discuss the things to come? The title of the column this month is...
Whats Next
for the Next NextGen?
Usually his articles hint at the future state of the art in audio techwise and creatively but this is the first I have read where he flat out predicts a new interactive aural composition for game score and audio effects...
He goes on to write how cloud computing will make null the disc space required to make such progress possible.
People get into CG for a variety of reasons but I find that very few get into game cg for different reasons. Usually they grew up loving their bit graphic donkey kong past and as far as they are concerned they are living the dream ( unless in an abusive environment of course ).
To others ( albeit the minority ) as technology evolved it became apparent that realtime graphics is/will be a medium without boundries or restrictions and all things being possible awakens their creative passion for what may be.
If you are that person... then that day has not arrived. yet.
When it does I imagine what that camp works on will not even resemble a game. But just like what is great in the best of cinema... will be as immersive and compelling as any self absorbed button mashing experience.
If you are the type of person who can open up the Stone Giant demo or an area of RAGE that doesn't require any gameplay...
And just be happy to wander for the sake of immersion alone. Than u just might be a redneck... er? I mean u just might belong to that aforementioned camp.
I suspect however that Jessie's vision of a new paradigm for interactive realtime musical soundscapes will not evolve as smoothly as he envisions. Sharing for a long time the same desire for such an interactive experience I have found very little reception for such progress. Not that there is negativity but the same dynamic where traditionaly composed scores and current recording engineering is the end all, be all solution where orchestrated one-off compositions are the gold standard.
And with a legacy of great musical moments how could anyone argue otherwise?
In which case how does evolution happen if percieved as unnecessary or perhaps even threatening to what is comfortable?
I imagine the only way is to just to see it for what it truly is? A different thing. Not necessarily superior but evolution none the less. In which case nintendo old school is just dandy doesn't mean that someone chomping at the bit for technology that opens a door some have been standing in front of for a very long time doesn't have a valid contribution.
I think we simply all look at this from the narrow end of the issue. That's only a natural thing, because we are all so involved with the creation of the assets, sculpting them, baking them, and so on. But all it comes down to is : an AD needs to know exactly what he/she wants in order to have it achieved in the game, regardless of tech level.
If someone had shown an image similar to that Vagrant Story screenshot above before the game came out (say, at the beginning of the PS1 era, when stuff like Ridge Racer was at the high end of the visual fidelity scale), everybody would have said it was impossible to achieve on the current available hardware. Yet these guys knew what they wanted, including that nice rimlight effect that was never seen before. But they had fantastic traditional artists who knew exactly what they were doing, and together with engineers they found solutions. IIRC the thin rim light here is not achieved with lighting, but through a brightened, offset copy of the silhouette, therefore providing more fidelity than a regular highlight would have. How amazing is that!
I think most predictions of graphical enhancements for next-nextgen are just the usual tech dreams we keep hearing all the time. (No more normalmaps! It's gonna be amaaaaazing!!) However that's not how real time graphics tend to evolve! We take the path of optimization, better texture compression algorithms, smarter caching of assets - not the one of wasteful use of ressources with the excuse of more horsepower.
Looking at games today, the few lacking areas are tings like animation blending ; environmental reaction ; draw distance ; and some materials that could need some improvements. All the rest are just problems "by design" : AI behaving poorly, crappy game mechanics, and so on. These don't improve magically with tech cycles
Visually we already reached what we need in terms of visual fidelity. Shadows aint great but they are good enough. Remember how good Resident Evil 5 looks ? The Steve Mc Curry quality of some of the screenshots could easily pass as ps4/nextXBox stuff. Again, the pillars on that Vagrant Story screenshots cast a wonderful soft shadow. Of course it is just a painted billboard in the background. But when you think about there is close to no difference between this, and the way Rage handled environment cast shadows, all "painted in" by the baking process.
Time to really play with all this incredible tech we have and start making beautiful art with it!
reminds me of this claydough:)
However, knowing that amaaaazing advancement is coming either way... I am mighty glad to have bet on the level of fidelity heading our way. Either way u look at it. It's win win. ( With us nuts craving that level of shadow detail... in the end is still a good design craving? )
I never found denying following a craving for artistic bliss to ever end negatively.
A great ps1 example still does not diminish what will make a huge impact with the lemmings that finally play the game. The audience does like their precious just as much as the artist who craves that brush.
Fer Instance I have yet to meet the musician who makes the point that great music should be possible on toy plastic guitars and in the same spirit offer to give up their Gibson SG. And thankfully for good reason. The Gibson SG does actually offer quality that translates into Bliss.
( cinematic levels of software rendered real time by 2020! Thats only 8 years!
http://graphics.cs.williams.edu/archive/SweeneyHPG2009/TimHPG2009.pdf
What I never expected was that my phone would some day be in some respects: a super computer compared to the Avid system I began on 14 years ago. I think you may have hit the nail on the head in regards to a narrow focus. However, I think it comes from assumptions held long before switching realtime graphics. ( perhaps some artists are just plain shadow nuts ). To whom the current dynamic shadow solutions "actually implemented" hardly seem to be the promised land envisioned by such a nut... ( or "good enough" )
But whose general purpose/usage seem nothing more than a method to seperate characters from the ground plane.
Oh yeah? well so does yer avatar so PhHhHttttt! :poly124:
It's not always that simple though, which is why a companies such as bethesda that owns Id are still better off from not using idtech5, in fact, skyrim could not have been made in the same shape it has today if using idtech5.
The best looking games of every generation has always been the most static, non-dynamic and locked games you can find in terms of scenery, and they had so much in common with movies, much like you said.
but designing a game is not always like designing a movie-set, it's like designing a theme-park of the highest quality, where you can go everywhere and look behind the houses.
A game like rage can get away with what it does, because the game will know where the player can and cannot go, and bakes the whole mega-texture accordingly. There is no such thing in something like skyrim, there is no place where the player can't go, and there's no way to bake anything down.
In the end vagrant-story looked amazing due to amazing texture-work, small enclosed areas for performance, and locked cameras and very few character in the scene.
What happened was that on the successor of the ps1, the ps2, you could have these kind of visuals in a fully detailed world, without a locked camera, where you have the freedom to move around and not be limited in the amount of characters.
At vagrant storys worst it would look like:
Wait, I started sculpting 3 days ago
If anything, I think that the understanding that we reached very high visual fidelity already might (re)open the doors to strongly stylized projects. We know that we can achieve BF3 and Forza quality now ; since this "next gen race" is over, now (that is to say : "next-next-gen") might be the time for more visually original titles, like Journey or The Last Guardian, using tech as a mean and not an end. Going back to simplification, abstraction, and impressionism ...
BF3 struggles to achieve PC fidelity on the consoles though, and 30fps at that, so a bunch of that fidelity is reached by strangling the console to death
But truly: The artist however is the last outpost of all game art, and even on the ps2 I would find some of the most beautiful games I've ever played, god of war 1 & 2 already looked amazing on the ps2, but then came god of war 3 looking even more amazing.
I don't think we'll ever reach that 'enough is enough' point, and at the point where we'll reach complete photorealism we'll also most likely see some of the most mindblowing artistic creations that moves away from photorealism yet using all that power.
We don't need the coming generation to make fantastic looking games, but it gives us more room to make fantastic looking games in genres that didn't have the resource-budgets to look fantastic.
but I doubt it.
I've seen far too many consumers bitch about products because they weren't 'realistic' looking, even if realistic means noisy and fuck ugly textures... there's always going to be companies that simply shoot for that even if they can't achieve it remotely successfully
I agree. I think there is still potential room for making the fantastic looking games/movies look even better though. I'm saying this because you never know what cool technique(s) could come next, I never imagined that games and movies could look as good as they do now about 20 years ago. Partly because of the quality of TV's now and because they use a combination of different lighting/effects/compositing/etc that wasn't used back then, people weren't able to predict how well it would look. Sure, I was a kid back then but my imagination for games like Sonic and Super Mario Bros went as far as animated motion pictures like the Disney ones.
[ame="
With
[ame="
I believe the little mermaid is a perfect example of how good a feature film could be at that time. It's great but the characters' cels are very flat colored, of course, and the backgrounds are static. Trine 2 looks like an animated painting and my imagination didn't go that far, I'm not sure if anyone could predict something as amazing as this back then. Who had seen a painting come to life this clear and realistically? Also remember that TV's quality wasn't as clear as today's TVs and that didn't make imagining something like this any easier.
But yeah, regardless, the more games can reconcile an artists ability to design with their ability to realize images onscreen, the better.
Well, that's like, your opinion man
It might not affect you, but it does affect some people a huge deal. I think the difference is very noticeable, and impacts gameplay a lot. Basically, games running around 30fps hare not only less smooth, they also tend to have a lot of post-processing going on, causing frame delays. It is quite noticeable in the console versions of Crysis2 and Gears Of War 2. Compare these to the tight response time of MW2 or Rage : the difference is huge!
Playing Rage at a rock solid 60fps made shooting so enjoyable, everything is much more snappier and fluid.
There's an inherit input-latency with lower framerates due to the time between frames, so a higher framerate will mean an actually snappier input.
Check this out, these are some screenshot I took from a 12 year old (!!!) PSX game called Breath of Fire 4 (really cool rpg):
These are all full 3D environments usually covering large areas and in most cases you could rotate the camera in steps of 90 degrees as well (notice the icon in the top left). To me these graphics are more impressive and beautiful than the ones Ive seen from recent AAA title rpg's. I studied the graphics by extracting the 3d models and textures with various tools and the developers pushed the console to it's limits!
Surely new hardware would make thing easier, but I think without restrictions art direction tends to get a little sloppier due to the amount of work/assets that needs to be done. I recently finished Fallout: New Vegas, a perfect example of rushed sloppy 'next-gen' art direction if you ask me, it had no soul.
That stands true, which is why skyrim, which is just about the same kind of scope looks leagues better than new vegas, even though it runs on exactly the same hardware.
It's totally cool to have upscaled sprite art in a commercial game if you want (for example MBAACC) but it sucks if you are forced to do it because of hardware (i'm looking at you, half-res fx buffers).
Oh cmon, next-gen? They barely have shadows :P That game have lousy tech, and pretty lousy art aswell. I dont think any amount of art direction would have saved that.
I dont know, I dont think good art direction solves everything, there are tons of cool things that you simply cant do, or that will look like shit on todays hardware. And the problem is that to get even CLOSE to the artistic vision you had in mind, you have to be extremely technically proficient, and you have to invest alot of man hours to optimise everything so you can actually get the result that you wanted. More power = less limits : D
You're right, perhaps Trine 2 wasn't a good example. My point was that when new technique is introduced artists try out methods without exactly knowing how the results will look together with other cool techniques. I brought up Trine 2 as an example because it's a good looking game. Perhaps the princess and the frog would be a better example if we're comparing it to MLM? It's an animated feature film but the combination of modern day effects and compositing choices make it look in a way we couldn't predict about 20 years ago. Heck, years ago some disney people thought that 2d was over and decided only to go with 3d but Princess and the frog along with other 2d feature films proved them wrong.
What I'm also looking forward to is a better and more real sounding speech synthesizer. We don't use the voice actors' faces for the games a lot and using these synthesizers will eliminate them from the equation completely.
If people want to, they can always go back to old tech and use that, but I don't see many people eager to run back to painting one pixel at a time in mspaint. As tech progresses forward we don't abandon what we were able to do before; we make it so that less "hacks" have to be implemented to approximate what we want, we increase the fidelity and number of factors we have control over.
When I see an older game like Vagrant Story or even a ps2 game like SotC, it makes me wonder how much better the artistic vision could have been realized with today's resources. They were definitely fighting against what they could do technically, and that definitely had an impact on the execution of the artistic vision.
As noble of an idea as it is to think that artistic vision exceeds everything, the tech is what produces and displays the final results. I don't see how people here of all places could be so dismissive of its importance.
Rewatched "The Long Voyage Home" and after finishing I have to admit, that Pior seemed to be right. That most compositions in the best of work could be recreated with todays restrictions. Even though there doesn't seem to be an equivalent. And for no good reason.
http://sixmartinis.blogspot.com/2008/03/to-vector.html
Edit: damm it Gestalt! you beat me to it. I swear I just wasn't copying your post!
Good example is Skyrim and Deus Ex, with buckets on heads and people not noticing that you are hacking their terminal...that they are sitting in front of. Need more work on AI, maybe some kinda of DirectAI library which can be improved all the time, module style. Add this Walk-block, attach this 2-eye-front-view-block, with a biped-animal behavior-block and run it.
Most people get it wrong though when they talk about good and bad AI, the bucket on head is a side-effect of much better AI, where crimes committed are the ones the npc will actually see, and in this case a bucket will actually block his eyesight and preventing him from seeing it.
Proper scripted behavior could've prevented that scenario, but that is countless of more scenarios that would have had to been taken into consideration.
The key word is fun and believable behavior, which can be done with the really simple and basic AI, even the most advanced AI can be blamed as being bad by the layman gamer.
In terms of AI related libraries, path-finding is one of those things that could be made cheaper, it's one of those things that costs a ton, the rest is just properly believable scripted reactions and behaviors.
The reason why 30 or even 60 fps aren't enough for games is because of temporal aliasing, ie stuttering between frames. If you have a movie clip of a completely black screen slowly blending into white then 25 fps will be more than enough, but every other frame being black and white might very well need over 100. it gets a lot harder to predict if the images are more complex than just solid colors, but the key factor here is (temporal and spatial) hyper-acuity, or the ability of the visual system to percieve aliasing beyond their "physical capability". Even on a 600dpi screen you can still percieve text aliasing quite clearly. Throwing more frames or higher pixel density on the problem is hella brute force, and doesn't really solve the underlying issue.
Good luck getting shadow quality like that with the current gen hah. Shadows are still so shitty; cause of strobing, small map sizes, and other elements.
Dark Athena seems to do it well too - still gotta play that one ...
That's what I was looking for, at 60fps it may feel more responsive, but it probably wont look any different to anyone watching, if it is done correctly.
Trying to render frames twice to four times as fast is a bad solution.
Not trying to be a dick, but do you have even a cursory knowledge of how these things work?
My knowledge stops at right about 'cursory' and it's pretty obvious to me how doing it 'correctly' is probably about a million times as expensive as just rendering it four times as fast. Your 30 fps game with film camera quality blur and depth and anti aliasing and lens effects is a none fps game.
You're suggesting we lower our goal to 30fps so that we can use more expensive rendering, and then when you hit the '30fps looks like shit' obstacle, you're proposing we do something exponentially MORE expensive to make it okay. Why would we want to run a game at 30fps instead of 60fps, and then use all the graphical power we earned just to make it look as smooth as it did when it was running 60?
Or are you proposing we run the game engine at 120 fps or whatever, to respond to player input, render it at 30fps, and write some kind of super predictive magic code to make it feel responsive and intuitive even with the desync between what the player does and when they see it happen?