Any visual evidence of PS2 normal mapping?

[_phil_]

They BOTH use specular maps, NOT normal maps.

A spherical alpha-ed envmap should do the trick most of the time.

 

[London Geezer]

A spherical alpha-ed envmap should do the trick most of the time.

There u go.

 

[ihamoitc2005]

That's what i'm saying! The coat on Trinity (T H E B L A C K S H I N Y O N E) is leather. Black leather. The top on Primal's girl is leather. Red leather. The BOTH are trying to achieve the same effect (whether you call this "simulating textures" - which makes no sense cause you don't "simulate" textures - or "simulate geometry" it doesn't matter). They BOTH use specular maps, NOT normal maps..

No they are trying to achieve different effects. Top in primal is simulating leather texture, Trinity coat is simulating not merely texture but actual deformation of surface. This is because Trinity coat is loose fiting and can deform, bend, wrinkle, etc, while primal top is skin-tight and hence no deformation effect needs to be simulated.

Regarding GT3/4 and Primal, it appears we have been saying same thing only in different words. But regarding trinity it seems we disagree so perhaps we should agree to disagree.

 

[London Geezer]

No they are trying to achieve different effects. Top in primal is simulating leather texture, Trinity coat is simulating not merely texture but actual deformation of surface. This is because Trinity coat is loose fiting and can deform, bend, wrinkle, etc, while primal top is skin-tight and hence no deformation effect needs to be simulated.

Uhm no, Primal was also using the effect to create the same effect (creases, deformations, bends...)
If that's the reason why you think Trinity uses normal mapping, you're wrong. But to be honest i really, honestly don't care. I didn't buy Primal and i certainly won't be playing Neo.

Believe what you have to believe.

 

[pixelbox]

Naughty Dog said the Jak engine "supported" bump mapping, but personally i haven't played much of Jak3 to tell. From experience it's the same effects they used in Jak2, which uses the same engine as Jak3. Could be bump map but i'd have to see it in motion.

Since the jak engines didn't have any type of dynamic lights except for the sun, the only place you'll find bump-mapping is in the desert. as the day goes on in the desert, the sun will reveal different creases and hills on the sand.

 

[Phil]

Personally I think there are two different sets of discussions taking place in here: Some are convinced that bump mapping as in the DOT3 is not possible on PS2 - while there are others that believe it is and is used to a degree in few games.

I think the point is, it is possible, it's just not apparent to which degree and in which games/scenarios/quantity it is.

Dunno if the above sets of examples are accurate in terms of their claims - I however think neither arguing has the right to dismiss the others solely on those screenshots which for all it's worth, could be or maybe isn't. We don't have more evidence, so there's really no point in dismissing it and just take in the arguments that would support it or not.



As far as I see it, ihamoitc2005 thinks the Neo game does use normal mapping. Can we at least look at it objectively before dismissing it completely on members credibility / posting history / aditude etc? Because there really aren't any hard facts of either party arguing supporting anything at the moment...

 

[London Geezer]

COUGH...page 1... COUGH

ihamoitc, do you even know what normal mapping is? Just because something "looks bumpy" in a screenshot, it doesn't mean it's normal mapping, or even bump mapping. Good textures can look bumpy on screenshots. That's why it's bloody useless to judge graphics from screenshots. I will NEVER repeat this enough.
.

 

[Phil]

COUGH...page 1... COUGH

True, but you're forgeting how this thread started:

I'd just like to see some shots or footage of the claimed normal mapping in the new Shiny game, Path of Neo.

Anyone have any?

Obviously, the developer is claiming that the game in question (Path of Neo) is using normal mapping, which is why we are looking at screenshots to figure out which textures could be using it or not to validate the probability of the above claim being true or not.

Obviously, no-one has definitve proof at the moment, but I'd say at least through screenshots we can limit us to areas that could be using it and then validating them through footage or actually playing the game. I think given the fact that we're even arguing it's credibility, there could be the good possibility that it in fact normal mapped. At the end of the day, you can't say it isn't without seeing it in motion, right?

 

[ihamoitc2005]

?

Uhm no, Primal was also using the effect to create the same effect (creases, deformations, bends...)
If that's the reason why you think Trinity uses normal mapping, you're wrong. But to be honest i really, honestly don't care. I didn't buy Primal and i certainly won't be playing Neo.

Believe what you have to believe.

Based on our difference experiences we draw different conclusions and there is nothing wrong with that. You see simple texture and specular map, I see actual geometry simulation techniques such as bump-mapping and variants. Since neither of us have access to the code both of us can only speculate so neither of us can say anything with absolute certainty no?

As I said before my friend, you and I will not see eye to eye on this so lets simply agree to disagree on this issue.

 

[Shifty Geezer]

Question : Would the developer outright ball-faced lie? Would they claim normal-mapping if there wasn't any of any sort at all? Devs may stretch the truth somewhat, but such a claim is surely going to need some substance. Looking at the IGN movie it wasn't at all clear, but then the scenes were too short to notice anything. It needs some in-game footage really. Until such evidence surfaces the debate is based purely on faith, making this a religious discussion that belongs in the RPS forum.

 

[therealskywolf]

Question : Would the developer outright ball-faced lie? Would they claim normal-mapping if there wasn't any of any sort at all? Devs may stretch the truth somewhat, but such a claim is surely going to need some substance. Looking at the IGN movie it wasn't at all clear, but then the scenes were too short to notice anything. It needs some in-game footage really. Until such evidence surfaces the debate is based purely on faith, making this a religious discussion that belongs in the RPS forum.

Nobody is saying it isn't using normal mapping. It's just using it in a very crappy way, probably because PS2 just couldn't do more.

Look, Delta Force Black Hawk Down was actually 1 of the 1st games using a technique like normal mapping ON characters. That's right, they used a technique just like PolyBump, where you had this high poly models, and then whatever happens in the middle with lightning and whatnot, for the final game they had characters with like 600 polys.

That looks like What Path of Neo is basically using.

 

[_phil_]

Polybump IS normalmapping.

 

[London Geezer]

Polybump IS normalmapping.

Yep, just the "fancy name" i think Crytek found for their "revolutionary technology" err...

 

[Phil]

Nobody is saying it isn't using normal mapping. It's just using it in a very crappy way, probably because PS2 just couldn't do more.

The definition on how you're using a texturing-method isn't "crappy" - it's either done the way it should be or it isn't, in which case it wouldn't be normal mapping. If it's "looks" crappy is an entirely different argument, one that wasn't debated here.

 

[darkblu]

What happens when one rotates camera from one camera location is intensity of reflected light changes on all real and simulated surfaces if portal.

ok, i did neglect any possible specularity component on those shots, on the basis that they were showin rough stone surfaces. e.g. concrete surfaces never have specularity component to them. they're essentially diffuse therefore do not depend on the camera angle. unless you want to simulate any possible quartz/glass/metal grains in the stone using a notable specularity component on it would not make much sense. of course, that's entirely upon artist's discretion.

As for Neo game, yes, most character models bodies are not very impressive, but the Trinity character model has very nice simulation of real shiny and flexible coat material.

well, that's most likely plain old environmental mapping with reflection vector, it doesn't take normal mapping to achieve the look of trinity's coat seen in those screenshots.

 

[ihamoitc2005]

ok, i did neglect any possible specularity component on those shots, on the basis that they were showin rough stone surfaces. e.g. concrete surfaces never have specularity component to them. they're essentially diffuse therefore do not depend on the camera angle. unless you want to simulate any possible quartz/glass/metal grains in the stone using a notable specularity component on it would not make much sense. of course, that's entirely upon artist's discretion.

Not so much specularity but so I am not sure what you refer to but as for portal simulation of geometrical detail, not specularity is quite apparent.

well, that's most likely plain old environmental mapping with reflection vector, it doesn't take normal mapping to achieve the look of trinity's coat seen in those screenshots.

Environment mapping good for shininess but not good for simulation of surface deformation.

 

[darkblu]

Not so much specularity but so I am not sure what you refer to but as for portal simulation of geometrical detail, not specularity is quite apparent.

i was referring to what you were talking in your previous post - that the 'reflection of things changes depending on the view angle alone' - reflection is what is also known as specularity. a material with low/no specularity (as the one i assumed on those shots) would be predominantly characterized by dispersed (aka diffuse) light, which does not interact with the view angle as the former does not have a reflection direction, IOW it depends only on the interaction of lightsource-vs-receptor, viewer's position being irrelevant.

Environment mapping good for shininess but not good for simulation of surface deformation.

that highly depends on the actual mesh resolution.

 

[darkblu]

We had this discussion months back, but I can't keep away...! CON's bumps are as goos as anything I've ever seen. If you look at modern dungeon games with their normal mapped stone walls and the way light interacts as though it's a 3D surface, that's exactly what CON has. Check out the dark caves when you shoot a fireball down and the light creeps over the surface.

I think it's worth noting that this effect is only on some surfaces though. If an intensive rendering process it'd make sense that it's confined to simpler scenery. It's also worth noting that CON has every other effect including full screen AA, so why not normal mapping as well? Though in the previous discussion LondonBoy suggested it was just lots of geometry, and the developers have never confrimed what technique they used. However, their engine is available for licensing so all we need is someone to try it out and report back to us...

well, i was just commenting on the screenshots under consideration. in a very lame way too : )

 

[Squeak]

What about EMBM with the VUs? The VUs can be used for vertex texturing and has quite a bit of memory to store the hightmaps (even VU0 can store two 64x64x4 textures).
It would be a matter of perturbing the UV coordinates according to the hightmap, don't know if that is possible with the GS in any way.

 

[overclocked]

Retro.....................

What a RETRO thread this is! It reminds of 2002 and forward with various threads about Dot3 BM, CAN the PS2 do it etz? Oh The dreamcast vs PS2 also is a classical.
Continue with the post while i go and get some popcorn...