• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

FAST Racing Neo out this week (still a December release)

stop making yourself look clueless, it's embarrassing, here is 18 minute video of the gameplay, shadows are static, they never reacting to the position of the sun.

https://www.youtube.com/watch?v=tiqRE2E-ChM

https://www.youtube.com/watch?v=faVqbbW6BX4

here is a great lighting done right, if you can't notice a difference get your eyes checked.
You can clearly see it being a dynamic shadow lile 6 seconds into this 18 minute video when the vehicle raises from the track dude... What's wrong with you.
 

Glass Joe

Member
Game looks good! F-Zero GX is one of my favorite games ever.

Now, I know this won't play the same, I just hope I can get into that light/dark mechanic. I remember not enjoying the original Wii demo much (I think it forced motion controls too?). So with this game do the shoulder buttons help with sharper turns? The biggest concern I have from watching that 18 minute preview is the player is constantly hitting walls.
 

blu

Wants the largest console games publisher to avoid Nintendo's platforms.
If only we could swap ninjablade's eyes out with some of those! :p
Haha. Well, this old pair of eyes here usually sees more techniques and algorithms in screenshots than the average healthy dose recommended for onlookers - I don't think nb would enojoy that : )
 
Haha. Well, this old pair of eyes here usually sees more techniques and algorithms in screenshots than the average healthy dose recommended for onlookers - I don't think nb would enojoy that : )


I know the feeling. In all honesty, I wish I could go back to simply seeing the overall beauty of a rendered image.

I can still see that now, but after years of computer science and programming theory under my belt, I notice every little detail of every screenshot. Current gen games still seem to be suffering from texture-streaming and LOD issues, and I can't for the life of my understand why current gen games don't use lots of anisotropic filtering. Also, every developer touts PBR these days, but many implementations seem to be lacking in execution, or sometimes the materials aren't rendered properly.

Most of these things go unnoticed by the average gamer, but I can't unsee them even if I wanted to!
 

HTupolev

Member
The 'shadow' that you see under floating objects isn't really the same kind of shadow that objects on the ground have, it's basically a form of ambient occlusion. And we all know that ambient occlusions don't have protrusions like hard shadows tend to have.
A roughly car-shaped object sitting a car width above the ground will tend to have a softened shadow compared with something closer, but the direct sunlight occlusion should still strongly dominate the appearance. The hovercraft shadows in the FRN video are pretty obviously not accounting for environmental light direction, and it's hard to argue that the overall result is "accurate" when the much farther-back environment geometry is leaving hard directional shadows right alongside the hovercraft's blob shadow.

They're probably using a blob shadow because it lands at a convenient crossroads of "dirt cheap to render" and "more useful for gameplay feedback than a correct shadow."
 

Broritos

Member
Game looks good! F-Zero GX is one of my favorite games ever.

Now, I know this won't play the same, I just hope I can get into that light/dark mechanic. I remember not enjoying the original Wii demo much (I think it forced motion controls too?). So with this game do the shoulder buttons help with sharper turns? The biggest concern I have from watching that 18 minute preview is the player is constantly hitting walls.

I wonder about the controls too. Racing games have more depth in their controls than they appear sometimes. Mario Karts drift boost mechanic, F-Zero GX's lean as well as its own drift mechanic that can combined with the lean, realistic and semi-realistic racers and their strict breaking and acceleration physics, they all demand more from the player than simply holding the accelerator and moving left and right.

FRN seems like its betting it all on the light mechanic which seems 'okay'. I really hope there's more to it though.
 

Exile20

Member
Can we all just agree the ignore Ninjablade whenever he post in threads. I mean we know when he will show up. Let's just all agree to ignore no matter what he says. I mean too many threads have turn to crap because of his presence. With that being said game looks amazing and hopefully we let shin'en know by making this their highest selling game to date.

I am definitely doing my part. Without an FZERO on the wii and and wii u, hopefully the Nintendo crowd puts up.

They checked most of the important checkboxes

Slow, fast, faster, too fast for me
60fps
4 player local
8player Online
Play online with friends(I can't believe this is a feature but after Mario Tennis, who the fuck knows?)
16 tracks
Great graphics
Motion/traditional controls
off tv
Leaderboards
 
A roughly car-shaped object sitting a car width above the ground will tend to have a softened shadow compared with something closer, but the direct sunlight occlusion should still strongly dominate the appearance. The hovercraft shadows in the FRN video are pretty obviously not accounting for environmental light direction, and it's hard to argue that the overall result is "accurate" when the much farther-back environment geometry is leaving hard directional shadows right alongside the hovercraft's blob shadow.

They're probably using a blob shadow because it lands at a convenient crossroads of "dirt cheap to render" and "more useful for gameplay feedback than a correct shadow."

You're missing the point. It's not that the craft wouldn't have a shadow like a car on the ground, it's that that kind of shadow would not be visible on the ground, because it would be in mid-air. What WOULD be visible on the ground is some kind of AO type shadow, which is what you see under the craft.

Furthermore, they're still not blob shadows. The shadows still correspond to the shape of the vehicle. And more importantly, this game uses SSAO, which gobbles up performance in any game and is completely unnecessary in a racer when the AO could have been baked into the environment instead. If they were concerned about shadows eating up performance, they would have nixed the SSAO, not give the crafts blob shadows while the environments use something far more technically demanding.
 

jariw

Member
Now, I know this won't play the same, I just hope I can get into that light/dark mechanic. I remember not enjoying the original Wii demo much (I think it forced motion controls too?). So with this game do the shoulder buttons help with sharper turns? The biggest concern I have from watching that 18 minute preview is the player is constantly hitting walls.

At 18:30 of the E3 Treehouse Live, you see the 5 control buttons (Accelerate, Brake, Switch Phase, Boost, Slide):
https://www.youtube.com/watch?v=9XJxqrYllo4

I think that's video is still the best run-through of how the gameplay is designed.
 

blu

Wants the largest console games publisher to avoid Nintendo's platforms.
They're probably using a blob shadow because it lands at a convenient crossroads of "dirt cheap to render" and "more useful for gameplay feedback than a correct shadow."
I don't see the blob. FRN shadows look like good 'ol projected shadows gaussian-blurred into oblivion, ergo the AO appearance. And of course they are decal-painted on top of the track.
 
How about we test this in real life, shall we?

Quick and dirty shadow projection test.

Cup on the table, shadow clearly visible (and some ambient occlusion as well)

iM752WR.jpg



Cup hovering above the table

5fD5lk4.jpg


Uhh...where the fuck is the shadow?!!!?


You see, as I lifted the cup up, the shadow lifted up as well... RIGHT OFF THE EDGE OF THE DAMN TABLE! Did real life fuck up? No, that's just how light occlusion works.

If we apply the same rules to FRN, the craft shadows would fall off the track and they would not be visible. Thankfully, FRN has some occlusion under the vehicles at all times, for gameplay purposes, as well as being somewhat physically accurate.
 

Aostia

El Capitan Todd
I am definitely doing my part. Without an FZERO on the wii and and wii u, hopefully the Nintendo crowd puts up.

They checked most of the important checkboxes

Slow, fast, faster, too fast for me
60fps
4 player local
8player Online
Play online with friends(I can't believe this is a feature but after Mario Tennis, who the fuck knows?)
16 tracks
Great graphics
Motion/traditional controls
off tv
Leaderboards


Totally. Day one
 

jariw

Member
Why is everyone calling this a spiritual successor to F-Zero? It's not the same team is it?

Probably because F-Zero is the most well-known futuristic racing game on Nintendo platforms. Shin'en has made pretty clear that this isn't anything F-Zero (other than the voice actor). FNR is its own thing.
 

HTupolev

Member
You're missing the point. It's not that the craft wouldn't have a shadow like a car on the ground, it's that that kind of shadow would not be visible on the ground, because it would be in mid-air. What WOULD be visible on the ground is some kind of AO type shadow, which is what you see under the craft.
Shadows don't have a mid-air cut-off point. They get dispersed continuously by distance as their penumbra dominates more and more.
In terms of the impact of the overall shadow under an "infinite distance" source, this effect will scale by the size of the object. If a hotwheels car shadow really starts to lose its visibility when the car is 10 inches off the ground due to the sky conditions, a real car's shadow might break up to the same degree when the car is 60 feet off the ground. This happens because bigger shadows need bigger penumbra to be overall dispersed by the same extent.
In your coffee-cup example, you've got a fairly gentle directional light in an environment with a lot of ambient light, and your shot doesn't have wide enough FoV to show the shadow even if the shadow was still coherent (it'll be way off to the side, judging by the angle of the light).

Sun on a clear day is an extremely dominant directional source, and objects can retain a fairly coherent shadow even when they're many times farther off the ground than their size. For instance, the shadow of a cyclist's head can still be quite sharp when it reaches the ground, even though it's many times farther off the ground than its size.

g5wFY9A.png


(The fact that the cyclist's head has a hard connection to the ground doesn't really matter here; if the cyclist was instead flying, the shadow would look basically the same, sans bicycle.)

These hovercraft would have to sit many times higher off the ground for their shadows to be significantly dispersed, in the sunny conditions shown.

Furthermore, they're still not blob shadows. The shadows still correspond to the shape of the vehicle.
True, I was using imprecise language. I just meant that they're not being rasterized from the light source, and were probably instead just a soft image being projected below the hovercraft.

And more importantly, this game uses SSAO, which gobbles up performance in any game and is completely unnecessary in a racer when the AO could have been baked into the environment instead. If they were concerned about shadows eating up performance, they would have nixed the SSAO, not give the crafts blob shadows while the environments use something far more technically demanding.
That's a fairly unrelated point. Performance saved in one area can be used elsewhere.

At any rate, I'm not really arguing that performance was the main reason they went with the method they did, just that it's a contributing added benefit. The drop shadow is a good gameplay choice, and it would be difficult to get an always-aggressive shadow like that to mesh well with an accurate shadow. It's best to just chuck it in and not worry about it.
 
Shadows don't have a cut-off point. They get dispersed by distance as their penumbra dominates more and more.
In terms of the impact of the overall shadow under an "infinite distance" source, this effect will scale by the size of the object. If a hotwheels car shadow really starts to lose its visibility when the car is 10 inches off the ground due to the sky conditions, a real car's shadow might break up to the same degree when the car is 60 feet off the ground. This happens because bigger shadows need bigger penumbra to be overall dispersed by the same extent.
In your coffee-cup example, you've got a fairly gentle directional light in an environment with a lot of ambient light, and your shot doesn't have wide enough FoV to show the shadow even if the shadow was still coherent (it'll be way off to the side, judging by the angle of the light).

Sun on a clear day is an extremely dominant directional source, and objects can retain a fairly coherent shadow even when they're many times farther off the ground than their size. For instance, the shadow of a cyclist's head can still be quite sharp when it reaches the ground, even though it's many times farther off the ground than its size.

g5wFY9A.png


(The fact that the cyclist's head has a hard connection to the ground doesn't really matter here; if the cyclist was instead flying, the shadow would look basically the same, sans bicycle.)

These hovercraft would have to sit many times higher off the ground for their shadows to be significantly dispersed, in the sunny conditions shown.


True, I was using imprecise language. I just meant that they're not being rasterized from the light source, and were probably instead just a soft image being projected below the hovercraft.


That's a fairly unrelated point. Performance saved in one area can be used elsewhere.

At any rate, I'm not really arguing that performance was the main reason they went with the method they did, just that it's a contributing added benefit. The drop shadow is a good gameplay choice, and it would be difficult to get an always-aggressive shadow like that to mesh well with an accurate shadow.

Shadows do have a cutoff point to when they are no longer visible on the ground if that ground is limited in area though, depending on the angle of incidence. At grazing angles, shadows appear to stretch further away from the object casting a shadow. What increases this distance even more is the distance that the object is from the ground. When an object casting a shadow moves away from the ground, its angle of incidence becomes increasingly oblique with respect to the light source, resulting in its shadow actually moving even further away from it.

In your example, the camera is pointed at the shadow. That's not the same as the camera sitting behind the cyclist. Furthermore, if you lifted the entire bicycle off the ground, the shadow of both the bike and cyclist would move out of the FOV of the camera. It doesn't matter how hard the shadow is cast, which brings me to me next point.

The tracks in FRN are pretty narrow, and many of them stand well above their surrounding environment. If the crafts' shadows were to move past the edge of the tracks, you wouldn't see them. Not because they magically disappeared, but they're just not in view of the camera anymore.

But here's a visual example to help you understand my point.

lunar.eclipse01.jpg


Look at the angle of Earth's penumbra with respect to Earth itself. If we were to put a white canvas underneath the Earth in that picture, there would be no shadow on the canvas. It has nothing to do with the shadow being dispersed, it's just that the shadow is pointing in a different direction.

Same thing with my coffee cup demonstration. The shadow didn't actually disappear, it just moved off the table completely out of view.

Now, am I suggesting that FRN hovercrafts actually do have hard cast shadows and that they've just been moved out of view of the camera? No, but considering that that is what would have happen had the crafts actually cast directional shadows with respect to the sun, it would be pointless to use them. Instead, Shin'en opted for the method that they have now, which works well enough, and is still somewhat realistic. Even with grazing shadows, there can be cases of ambient occlusion directly underneath an object.

And finally, I don't think they'd opt to save performance on vehicle shadows instead of environment SSAO. SSAO is unnecessary and offers no benefit to the player other than looking good in screenshots, while accurate shadows of the hovercrafts offer direct and immediate benefits to the player.
 

AlStrong

Member
Somehow, a discussion about projection/planar shadows wasn't what I rendered in mind when I read the list of other things Shin'en mentioned.
 

HTupolev

Member
Shadows do have a cutoff point to when they are no longer visible on the ground though, depending on the angle of incidence.
I edited my post earlier to clarify, I was referring to the fact that the shadow doesn't really "exist in the air" and then "not exist when it gets to the ground" in a binary way.

If the crafts' shadows were to move past the edge of the tracks, you wouldn't see them. Not because they magically disappeared, but they're just not in view of the camera anymore.
Yes, but that would mostly only happen when the sun is in front of the camera and low enough to push the shadow behind the bottom of the FoV. A large majority of the time under daylight conditions, this shouldn't be the case.

Sometimes the track design should arguably help to keep the shadows in view, since it often has up-sloping geometry on the sides.

If this game was mostly played with a driver-seat camera with a low field of view, then the sun shadow of the craft would spend a lot of time off-screen, but we're dealing with a reasonably healthy follow camera pulled back a bit from the hovercraft.

Even with grazing shadows, there can be cases of ambient occlusion directly underneath an object.
That's technically true, but direct sunlight tends to overpower the hell out of AO. In instances where the shadow is off to the side and a clear sun is hitting the road below the hovercraft, you really shouldn't have much visible darkening. It's more justified when the hovercraft is in shadow, although even then it should usually be much fuzzier and less pronounced than it is.

SSAO is unnecessary and offers no benefit to the player other than looking good in screenshots
That's not true. There's a lot of big environment detail that parallaxes slowly on-screen, and without a baked lightmap or some other means of approximating high-frequency GI, contact points in shadow would look ungrounded.

I'm not always a fan of fast screen-space AO solutions as they can often come with nasty artifacts, but they definitely can have an impact on the image that goes beyond what the pixel-peeping screenshot analysts notice.

while accurate shadows on the hovercrafts offer direct and immediate benefits to the player.
Accurate shadows would probably offer far fewer gameplay benefits than inaccurate ones in this case. :)
 

AdanVC

Member
Todas was the day I actually learned how shadows actually works inside an indie videogame for Wii U thread, thanks to brainchild. Impressive explanations!
 
I for one don't mind ninjablade's presence. He's like a cartoon character. One-dimensional, has an amusing schtick, etc. It's like putting on a Bugs Bunny cartoon while you fold laundry. Good for a background chuckle while you engage in things that actually matter.
 
I edited my post earlier to clarify, I was referring to the fact that the shadow doesn't really "exist in the air" and then "not exist when it gets to the ground" in a binary way.

No, it's both. It exists in space (similar to how the Earth's Lunar eclipse works) as well as the ground... until there is no more ground for the shadow to be projected onto ;)

Yes, but that would mostly only happen when the sun is in front of the camera and extremely low, low enough to push the shadow behind the bottom of the FoV. A large majority of the time under daylight conditions, this shouldn't be the case.

Sometimes the track design should arguably help to keep the shadows in view, since it often has up-sloping geometry on the sides.

If this game was mostly played with a driver-seat camera with a low field of view, then the sun shadow of the craft would spend a lot of time off-screen, but we're dealing with a reasonably healthy follow camera pulled back a bit from the hovercraft.

I don't really want to get into an argument about the precise distances needed for the shadows to move out of view. You're not necessarily wrong, but without actual measurements in-game, we really won't know. However, considering the potential problems with dynamic shadows of floating objects in a racing game, I think Shin'en was wise to go with the implementation that they did.


That's technically true, but direct sunlight tends to overpower the hell out of AO. In instances where the shadow is off to the side and a clear sun is hitting the road below the hovercraft, you really shouldn't have much visible darkening. It's more justified when the hovercraft is in shadow, although even then it should usually be much fuzzier and less pronounced than it is.

True, it's not entirely realistic, but it does happen in the real world. My point here isn't so much that it's a common occurrence in the real world, but that it doesn't look out of place compared to the real world, especially since you can barely see the shadow anyway.


That's not true. There's a lot of big environment detail that parallaxes slowly on-screen, and without a baked lightmap or some other means of approximating high-frequency GI, contact points in shadow would look ungrounded.

I'm not always a fan of fast screen-space AO solutions as they can often come with nasty artifacts, but they definitely can have an impact on the image that goes beyond what the pixel-peeping screenshot analysts notice.

Shin'en themselves said it was unnecessary. They could have baked in the AO if they wanted to. It didn't need to be in real time. It's just a bonus feature, which is pretty crazy, considering how demanding it is.


Accurate shadows would probably offer far fewer gameplay benefits than inaccurate ones in this case. :)


But either method would be a better trade off than SSAO :)
 
How about we test this in real life, shall we?

Quick and dirty shadow projection test.

Cup on the table, shadow clearly visible (and some ambient occlusion as well)

iM752WR.jpg



Cup hovering above the table

http://i.imgur.com/5fD5lk4.jpg

Uhh...where the fuck is the shadow?!!!?


You see, as I lifted the cup up, the shadow lifted up as well... RIGHT OFF THE EDGE OF THE DAMN TABLE! Did real life fuck up? No, that's just how light occlusion works.

If we apply the same rules to FRN, the craft shadows would fall off the track and they would not be visible. Thankfully, FRN has some occlusion under the vehicles at all times, for gameplay purposes, as well as being somewhat physically accurate.


you seriously should change your user name, you don't even know how simple shadows work, just because the car is hovering doesn't mean the shadow disappears.

http://postimg.org/image/tl2mp2ib1/

http://postimg.org/image/qoc0upjcd/

if you watched the wipe out vid, thats exactly how the shadow looks the fact that in FRN the shadow never changes shape or move tells, its not dynamic, and doesn't react to lighting.
 

HTupolev

Member
You're not necessarily wrong, but without actual measurements in-game, we really won't know.
We can make informed guesses. For instance:

In steady travel, the hovercraft are clearly less far off the ground than their width, based on the width of the shadow versus the height between shadow and craft.
Let's assume, to be conservative, that they're twice as long as they are wide, and that their distance from the ground is the same as their width, and that the height of the craft is negligible.

We can see the tail of them"straight down" shadow on the ground, so if the Sun was directly ahead and inclined a bit above the horizon...

plLnPL0.png


...It would have to have a run-over-rise of at least 2 to knock the shadow off the screen to the bottom. That implies an inclination of about 28 degrees or lower.

Now, this is the worst-case scenario for the Sun pushing the shadow off-screen. Much of the time the camera is pulled back, and at other angles the shadow would usually be visible at much lower inclinations (for instance, if the Sun was instead behind the camera, it would basically have to go down to the horizon for the shadow to disappear).

If you trace a cone straight up into the sky so that its edges are at a 28-degree inclination, that cone will subtend 3.33 steradians. A hemisphere is 2pi steradians, so even in this cartoonishly conservative estimate, a correctly-projected shadow would be visible more than half the time during the day.

True, it's not entirely realistic, but it does happen in the real world.
Under direct sunlight it borderline doesn't as far as the naked eye is concerned. I'd snap a picture under true sunlight, but since it's night, here's an example from the Ultra Sophisticated Closet World Simulator With White Walls To Produce Lots Of Ambient Light From a Strong Directional Source Such as a Lamp Sitting On The Floor™:

20151119_183318_resized_zpslyudrojx.jpg


AO is clearly visible in places like the shadowed box crease on the right side, but what little dimming you get from the small fraction of the hemisphere blocked by the battery is utterly tiny compared to the intensity of the direct light source.

but that it doesn't look out of place compared to the real world, especially since you can barely see the shadow anyway.
It's quite clear in the video, and must be in order to aid the player's sense of location.

The reason it doesn't look that out of place is that FRN isn't dead set on strict realism.

(And also probably that our senses are somewhat dulled, as the number of games that have mostly correct-looking shadows over large environments without aliasing or bizarre artifacts is basically zero.)

Shin'en themselves said it was unnecessary. They could have baked in the AO if they wanted to.
Hmm, link to what exactly they said? There are reasons to use SSAO instead of a baked map even if dynamic objects don't interact with it, like storage/memory considerations.

Shin'en doesn't seem like a group that would do something "just because" when they could get equally effective results in an entirely cheaper way.
 
you seriously should change your user name, you don't even know how simple shadows work, just because the car is hovering doesn't mean the shadow disappears.

2j14hso.jpg


2j14hso.jpg


It doesn't disappear, it just moves from underneath the vehicle.

Oh, and your links don't work. Whatever site is hosting them is banned from this site.
 
It doesn't disappear, it just moves from underneath the vehicle.

Oh, and your links don't work. Whatever site is hosting them is banned from this site.

fixed the link, but the fact your still arguing this is mind boggling, it really shows you have no idea about your talking about, just because the cars hover 5-10 feet doesn't mean shadows don't react to light, look at wiper hd and especially wipeout 2048 a more advanced engine, shadows change form and position, depending where the light is coming from, like in real life and gasp the cars are hovering.
 

?oe?oe

Member
I looked forward to the reviewers going in depth of how the shadows are in this game. 7.9 from IGN. Too little shadows.
 
We can make informed guesses. For instance:

In steady travel, the hovercraft are clearly less far off the ground than their width, based on the width of the shadow versus the height between shadow and craft.
Let's assume, to be conservative, that they're twice as long as they are wide, and that their distance from the ground is the same as their width, and that the height of the craft is negligible.

We can see the tail of the shadow on the ground, so if the Sun was directly ahead and inclined a bit above the horizon...

plLnPL0.png


...It would have to have a run-over-rise of at least 2 to knock the shadow off the screen to the bottom. That implies an inclination of about 28 degrees or lower.

Now, this is the worst-case scenario for the Sun pushing the shadow off-screen. Much of the time the camera is pulled back, and at other angles the shadow would usually be visible at much lower inclinations (for instance, if the Sun was instead behind the camera, it would basically have to go down for the shadow to disappear).

If you trace a cone straight up into the sky so that its edges are at a 28-degree inclination, that cone will subtend 3.33 steradians. A hemisphere is 2pi steradians, so even in this cartoonishly conservative estimate, a correctly-projected shadow would be visible more than half the time during the day.


Under direct sunlight it borderline doesn't as far as the naked eye is concerned. I'd snap a picture under true sunlight, but since it's night, here's an example from the Ultra Sophisticated Closet World Simulator With White Walls To Produce Lots Of Ambient Light From a Strong Directional Source Such as a Lamp Sitting On The Floor™:

20151119_183318_resized_zpslyudrojx.jpg


AO is clearly visible in places like the shadowed box crease on the right side, but what little dimming you get from the small fraction of the hemisphere blocked by the battery is utterly tiny compared to the intensity of the direct light source.


It's quite clear in the video, and must be in order to aid the player's sense of location.

The reason it doesn't look that out of place is that FRN isn't dead set on strict realism.

(And also probably that our senses are somewhat dulled, as the number of games that have mostly correct-looking shadows over large environments without aliasing or bizarre artifacts is basically zero.)


Hmm, link to what exactly they said? There are reasons to use SSAO instead of a baked map even if dynamic objects don't interact with it, like storage/memory considerations.

Shin'en doesn't seem like a group that would do something "just because" when they could bet equally effective results in an entirely cheaper way.


First, let me say that I seriously admire efforts to demonstrate why you believe that the shadows should still be visible on the track. There's only one problem with your analysis; your assumptions would only hold true under certain conditions, and not necessarily common conditions either. So while what you say might be true, the results would be far too inconsistent as the craft races along the track.

This game will have lots of banking and drifting around corners. In many videos thus far, we see many players crashing into the rails of the track; it is a common occurrence. If the shadows of the hovercraft are displaced to the sides of the vehicle, then when the player drives along the edge of the track, the shadows would no longer be on the ground; they would either cast on the rails, or in the case of tracks without rails, the shadow would completely fall off of the track.

There really is no getting around this. If the shadows were completely accurate, there would be lots of times were you wouldn't see them on the ground, depending on how you're driving. It just doesn't make sense to use them when you consider how inconsistent they'd be.

And again, the AO-like shadows underneath the FRN vehicles isn't really AO. It just has a similar appearance to it, albeit more pronounced than in real life, for gameplay reasons.

As for Shin'en's statements, it was a while back, so it'll take me some time to find it, but basically they were questioning if adding SSAO was worth it when in the end no one would really notice the effect when actually racing at blistering speeds, but they kept it in to really push their limits.
 
fixed the link, but the fact your still arguing this is mind boggling, it really shows you have no idea about your talking about, just because the cars hover 5-10 feet doesn't mean shadows don't react to light, look at wiper hd and especially wipeout 2048 a more advanced engine, shadows change form and position, depending where the light is coming from, like in real life and gasp the cars are hovering.


You mean these games?

1863751-19021wipeout2048gamescom003.jpg


wipeout-2048-vita_ghep.jpg


What, pray tell, do you suppose happens to these shadows when you race along the edge of the track and the shadow is no longer on the track? Are they suppose to just visibly float in mid-air?

More importantly, those are not the kind of shadows that the vehicles in FRN have. FRN's vehicles have dynamic shadows, they just aren't cast by the sun's light.
 

HTupolev

Member
This game will have lots of banking and drifting around corners. In many videos thus far, we see many players crashing into the rails of the track; it is a common occurrence. If the shadows of the hovercraft are displaced to the sides of the vehicle, then when the player drives along the edge of the track, the shadows would no longer be on the ground; they would either cast on the rails, or in the case of tracks without rails, the shadow would completely fall off of the track.

There really is no getting around this. If the shadows were completely accurate, there would be lots of times were you wouldn't see them on the ground, depending on how you're driving. It just doesn't make sense to use them when you consider how inconsistent they'd be.
I think those objections miss the point. It's true that the shadow boundaries can technically disappear across short spans of time for those reasons, but those are high-frequency behaviors that won't make the shadow less noticeable any more than causing a light to flash will make it less noticeable than a steady light.

Similarly, in games with consistent hard shadows, the hard shadow of an object vanishes when it heads into a shadowed area. Does this characteristic make it less important that the objects are included in the shadow casting? Perhaps in games which spend almost no time in direct light, but not if you're constantly hopping in and out of shadow, even if the small object's shadow is only expressed on half of the frames.

And again, the AO-like shadows underneath the FRN vehicles isn't really AO. It just has a similar appearance to it, albeit more pronounced than in real life, for gameplay reasons.
That's essentially what I said, I thought.
 
I think those objections miss the point. It's true that the shadow boundaries can technically disappear across short spans of time for those reasons, but those are high-frequency behaviors that won't make the shadow less noticeable any more than causing a light to flash will make it less noticeable than a steady light.

Similarly, in games with consistent hard shadows, the hard shadow of an object vanishes when it heads into a shadowed area. Does this characteristic make it less important that the objects are included in the shadow casting? Perhaps in games which spend almost no time in direct light, but not if you're constantly hopping in and out of shadow, even if the small object's shadow is only expressed on half of the frames.

But the problem is that the vehicles float above the ground. Without a shadow directly underneath them, there would be an issue with depth perception, making the vehicles more difficult for some people to drive properly.


That's essentially what I said, I thought

Right. I was just clarifying that I wasn't arguing that it was actual AO.
 

HTupolev

Member
But the problem is that the vehicles float above the ground. Without a shadow directly underneath them, there would be an issue with depth perception, making the vehicles more difficult for some people to drive properly.
I agree; I've noted a couple times that the decision makes sense when taking gameplay feedback into account.
 
I agree; I've noted a couple times that the decision makes sense when taking gameplay feedback into account.

Yeah, perhaps I wasn't clear enough before, so I apologize about that.

If the inconsistency of the shadows posed no problems in terms of gameplay, I wouldn't have a problem with it, but since it does, I'm glad that Shin'en isn't using vehicle shadows where the sun is their light source.
 
But the problem is that the vehicles float above the ground. Without a shadow directly underneath them, there would be an issue with depth perception, making the vehicles more difficult for some people to drive properly.




Right. I was just clarifying that I wasn't arguing that it was actual AO.

You're really reaching here, first you went from that's how shadows cast in real life, to now having shadows reacting to lighting on the cars would ruin gameplay, almost every racing game has cars casting somewhat proper shadows this gen, this game doesn't, end of story. Wipe out 2048 even has them

https://youtu.be/e97FgfJh-xY

Final proof this game has blob cheap shadows, at about 320 seconds in there is a crash, shadow of car is none existent. also at 4:56, when the helicopter robots with red lights come so they can give you the green light, there shadows are laughable, and are not reacting to lighting at all.
 
BTW, the multiplayer graphics don't appear to be significantly downgraded at all (besides the frame rate for 4 players), which is reassuring.

screenshot23.jpg


screenshot24.jpg
 
Top Bottom