LAIR Thread - * Rules: post #469

[rounin]

Physics based doesn't mean accurate. Unless they are doing a calculation of every single water molecule and taking into account the riverbed's geometry there is definately room for error. I don't know just how they are doing the water, but I have seen a few shots that didn't look quite right...

There is always "room for error" because it is a game/simulation so its not saying anything. However, I stated that it is reasonable because it is supposedly physics based (with the pretense that I remembered right), and thus most of the odd shots I remember people talking about do have real-world examples that look like that.

In any case, what I'm saying is that you may try to find the screenshots you were talking about and I'm quite certain there are going to be people who can post real-life examples that look like that under possibly similar situations as in game. No whether or not it looks artistic or pleasing or not "odd" may be a different issue. As far as some of the people criticizing the "noisy" water is concerned, the real-life examples that one posted are "weird" or "odd", but its not like you can criticize the real world.

 

[Neb]

Some of the new shots on Playsyde feature improved lighting, better terrain textures and a few other advancements.
But for each of these, you get something like this, too...
lowpoly dragon
super noisy water

The good news is that it does get better looks over time. Must be a crazy race against the clock if they don't delay it a bit more; it would certainly help the graphics a lot.

They must've scrapped everything we've seen in the first techdemo, though, that's the only explanation for showing such rushed content that we've seen in the past months...

Yeah the water looks noisy, like setting the mipmap level to minus effect on textures in PC games. But for this game it looks good neverthless. The first one with the dragon does not look good. It's not the dragon that stands out but the sky and the effects. Perhaps from an old build?

 

[Shifty Geezer]

Physics based doesn't mean accurate. Unless they are doing a calculation of every single water molecule and taking into account the riverbed's geometry there is definately room for error. I don't know just how they are doing the water, but I have seen a few shots that didn't look quite right...

The waves have looked fine to me, apart from the tiling. The physics calculations of the water seem to be happening in need gridded tiles. The noise I guess is a lack of sampling/filtering in the physics calcs. If it's shimmering on screen, it could look good with a degree of realism, or really bad like random fuzz. Think FFX texture shimmer of grass!

 

[3roxor]

Imo there is nothing wrong with the water in that pic.. In real life when you look at the water with the sun direct above it, you will see exacty that 'noisy' glittering on the water. it's accurate.

 

[deepbrown]

Imo there is nothing wrong with the water in that pic.. In real life when you look at the water with the sun direct above it, you will see exacty that 'noisy' glittering on the water. it's accurate.

This debate reminds me of the GTHD mountain backdrop debate. It's 2d because mountains don't look like that when you're moving past them...etc etc. Whereas when you've lived in Switzerland etc. with picturesque snowy mountains in the background...they look very much like non-moving paintings even when driving past them.

Yup water CAN look like that...and I'm sure as you're playing through the game the waters physics and look will change in real-time so it won't be a problem of it looking like that all the time.

 

[betan]

De gustibus non est disputandum.

I can totally understand if you think Lair is ugly overall.
But you seem to be nitpicking regarding technical aspects. That's not to say it is perfect. Low poly dragon and their progressive mesh code can probably be improved. But the bridge shot shows much more than noisy water (which animates good enough). I certainly don't know any game that cares to render character models from that distance.

And what's up with Latin? Does it loose meaning in English?

The waves have looked fine to me, apart from the tiling. The physics calculations of the water seem to be happening in need gridded tiles. The noise I guess is a lack of sampling/filtering in the physics calcs. If it's shimmering on screen, it could look good with a degree of realism, or really bad like random fuzz. Think FFX texture shimmer of grass!

I thought only geometry of the water was procedural. Is it shaded by CELL as well?

 

[Shifty Geezer]

They must be creating normal maps or something alongside displaced waves. There can't be that much geometry in the water! Small waves would, at that distance, blend into areas of brightness rather than shimmer, basically looking smoother. There'll be a mathematical function defining how high frequency waves have less influence on a point's appearance based on distance. Blurring by distance ought to be a suitable fix to remove the worst of the shimmers. Plain old mipmapping would do it, but they'd have to be created on the fly per frame. Or in the creation of the waves, frequency is computed from distance to camera, but that'd require a per-pixel Z test. What they're probably doing is just calculating displacement and normals on the surface and then outputting to render.

 

[warb]

The pics (are they from the latest build?) are sourced from
http://www.watch.impress.co.jp/game/docs/20070608/rfl.htm
and the second pic is 1080p actually with armies and dragons fighting each other.
http://www.watch.impress.co.jp/game/docs/20070608/rfl05.jpg
As long as the water realistically animates when flying by it will be less noticeable depending on angle and height.

Wow, the scale of this game is amazing. That second pic looks to me like a 720p image resized in Photoshop. Have there been any proper 1080p pics released yet? I haven't been following Lair closely, but I am aware that it's rendered at 1080p.

 

[Laa-Yosh]

It's more like 960*1080 with 2xAA, if I understood the tech talk correctly... That's why it looks resized, only half the pixels are shaded.

 

[patsu]

Would be great if one can translate the original articles
http://www.neogaf.com/forum/showpost.php?p=6691922&postcount=108

 

[phat]

Physics based doesn't mean accurate. Unless they are doing a calculation of every single water molecule and taking into account the riverbed's geometry there is definately room for error. I don't know just how they are doing the water, but I have seen a few shots that didn't look quite right...

The explosion ripples are definitely very wrong. Since they are surface ripples, they shouldn't be moving any faster than the other waves in the water. All waves of the same type should travel at the same speed in the same medium.

 

[Crossbar]

The explosion ripples are definitely very wrong. Since they are surface ripples, they shouldn't be moving any faster than the other waves in the water. All waves of the same type should travel at the same speed in the same medium.

I am not sure they are so far off, this is partly a shock wave just moving water sideways and there is the shock wave travelling through the air as well. When they work together the waves created will definitaly be faster than the waves created when you throw a stone in the water.

 

[phat]

I am not sure they are so far off, this is partly a shock wave just moving water sideways and there is the shock wave travelling through the air as well. When they work together the waves created will definitaly be faster than the waves created when you throw a stone in the water.

A pressure shockwave through water would travel at 1500 m/s, or 4.5 times the speed of sound in air. And it would not cause such a large, visible displacement of the transmission medium (the water).

A surface ripple shockwave would not travel any faster than the surface ripples. An example of a surface ripple shockwave is the wake of a fast-moving ship. The wake does not propagate faster than any other surface ripple.

Trust me, the explosion ripples are all wrong.

 

[Capeta]

The wake does not propagate faster than any other surface ripple.

I'm sorry but you are wrong...trust me.

 

[phat]

I'm sorry but you are wrong...trust me.

A wake is a superposition of waves. Each wave in the superposition propagates no faster than any other wave through the medium.

The angle of the wake gets smaller as the boat travels faster, but that's merely its shape, not its propagation speed.

Another way of thinking of wakes is to consider sonic booms, which are wakes in air. The time between a supersonic jet being directly overhead and the time you hear its sonic boom is no shorter than the time for any other sound to travel the same distance (the altitude of the jet). The "shadow" of the sonic boom moves at the supersonic speed of the jet, but the boom itself doesn't.

 

[Capeta]

A tsunami travels hundreds of MPH. Are you saying those 3 foot beach waves or boat wakes travel as fast?

 

[phat]

A tsunami travels hundreds of MPH. Are you saying those 3 foot beach waves or boat wakes travel as fast?

A tsunami is not a surface wave at its origin. It starts at the ocean floor. Think of a cookie sheet filled with water and then you suddenly tilting it.

 

[Capeta]

A tsunami is not a surface wave at its origin. It starts at the ocean floor. Think of a cookie sheet filled with water and then you suddenly tilting it.

Wrong again, a tsunami or a large wave like a tidal wave could be caused by any number of things that does not necessarily originate from the ocean floor, regardless the cause is irrelevent since it still creates a surface wave that travels at very high speed. You are wrong get over it. Maybe it's time for you to actually watch some military footage of destroyers shooting their canons...or atomic explosions at sea...

 

[Crossbar]

a tsunami or a large wave like a tidal wave could be caused by any number of things that does not necessarily originate from the ocean floor,

One example is a land-slide sliding into the water, that should not be that different form the case of a shock wave caused by an explosion.

 

[nico1982]

Maybe it's time for you to actually watch some military footage of destroyers shooting their canons...or atomic explosions at sea...

Just fo reference and courtesy from wikipedia, USS Iowa full broadside