3dsmax Rendering Support Topic
- GametagAeonFlux
- Posts: 9320
- Joined: Sun Jun 06, 2004 7:27 pm
- Location: Lincoln, NE
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
Halo2Rules' Chief has a 4096x4096 bump map and it looks pretty dang nice in my opinion.Veegie wrote:It's called Z Brush, making those details is as easy as clicking and dragging.. And, try painting a 4000x4000 bump map on say... the MasterChief, by your logic it would appear absolutely beautiful. Wrong.
But anyways, much better Darco, and post production would be pretty sweet as well.

- Veegie[Temp]
- Posts: 2849
- Joined: Thu Jan 29, 2004 11:30 pm
- Location: Logan
- Contact:
![]() |
I'm still waiting for you to make a point.trepdimeflou wrote:Height map != Bump map. They are both greyscale, sure, but used for completely different purposes.Veegie wrote:It's called Z Brush, making those details is as easy as clicking and dragging.. And, try painting a 4000x4000 bump map on say... the MasterChief, by your logic it would appear absolutely beautiful. Wrong.
Bump maps are only there to create the illusion of geometry that isn't polygonally existent. It's a fake, and illusions only work for so long. Do me a favour, back up your claims and show me an extreme-close up a 4096x4096 hand-painted height map on a box. I'm sure it'll turn out material that rivals Da Vinci.
Bump maps are used to add extra detail to a material, to make it appear more realisitic or to increase the depth of the shader through surface effects.
Height maps are used to simulate geometry when it is not possible to implement a model with many subdivisions and a large amount of polys.
Just because they both go in the "Bump Map" slot does not mean they are the same thing.
Oh, wait, didn't your "3d expert" friends know that?
I hope you're not trying to make the point that I starting speaking about normals then to height maps.
- trepdimeflou
- Posts: 727
- Joined: Sat Dec 25, 2004 12:50 pm
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
You stated that bump maps were not meant to be used for close up renders, which is incorrect.Veegie wrote:I'm still waiting for you to make a point.
Your implications then proceeded to state that bump and height maps were used for the identical purposes, which is also incorrect. You were in dire need of education regarding the matter.Veegie wrote:Bump maps are only there to create the illusion of geometry that isn't polygonally existent.
I will repeat what I said, maybe you will understand it better this time.
Bump maps are meant to be used for close renders; that's why they are most often tiled.Height map != Bump map. They are both greyscale, sure, but used for completely different purposes.
Bump maps are used to add extra detail to a material, to make it appear more realisitic or to increase the depth of the shader through surface effects.
Height maps are used to simulate geometry when it is not possible to implement a model with many subdivisions and a large amount of polys.
By saying that bump maps are for renders with a medium - large camera distance from the focal point, you are claiming to have intelligence on the matter far superior to that of the very company whose games you worship.
I guess bungie has no clue what's the "right" way of doing things, since they made a bump map for the masterchief for the sole purpose of added close-up detail.

- Veegie[Temp]
- Posts: 2849
- Joined: Thu Jan 29, 2004 11:30 pm
- Location: Logan
- Contact:
![]() |
As I said before. Height mapping is not meant for close-ups. Though I doubt you're going to accept that, due to the fact that we obviously share differnet interpretations of the words "close up".
Also, since when are staff members told to contribute to being off-topic in an arguement.
Also, I still hold an open invitation for an incredibly high resoultion bump map placed on a box, then rendered at extreme close range.
I find it odd that you link a picture of Halo 3, where the focus is extreme close-up detail to take advantage of the extreme HD support, when the topic of discussion originated from Halo 2.I guess bungie has no clue what's the "right" way of doing things, since they made a bump map for the masterchief for the sole purpose of added close-up detail.
Also, since when are staff members told to contribute to being off-topic in an arguement.
Also, I still hold an open invitation for an incredibly high resoultion bump map placed on a box, then rendered at extreme close range.
- trepdimeflou
- Posts: 727
- Joined: Sat Dec 25, 2004 12:50 pm
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
I'm not offtopic. This is a SUPPORT thread, and you're obviously in need of support. Besides, it's my thread. According to your own philosophy, the maker of the thread is the one who sets the rules for that thread.
Veegie wrote:No.
Bump maps aren't really meant for that close of rendering.
Doesn't look the same to me...Veegie wrote:As I said before. Height mapping is not meant for close-ups.

- Veegie[Temp]
- Posts: 2849
- Joined: Thu Jan 29, 2004 11:30 pm
- Location: Logan
- Contact:
![]() |
- trepdimeflou
- Posts: 727
- Joined: Sat Dec 25, 2004 12:50 pm
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
You can have however many opinions as you like; it's not going to make me "lose any sleep" over anything.
However, you are not allowed to tell a staff member to "fuck off" because they posted information to cure a small portion of your overwhelming ignorance.
I tried to explain where you were wrong, so you could learn and accept the truth, but you're too pompous and stubborn to accept it.
Don't attempt to teach things that are false. It's not benificial to anyone and just makes you look like an idiot.
This discussion has ended. Any further posts regarding this matter will be deleted at my discretion.
However, you are not allowed to tell a staff member to "fuck off" because they posted information to cure a small portion of your overwhelming ignorance.
I tried to explain where you were wrong, so you could learn and accept the truth, but you're too pompous and stubborn to accept it.
Don't attempt to teach things that are false. It's not benificial to anyone and just makes you look like an idiot.
This discussion has ended. Any further posts regarding this matter will be deleted at my discretion.

Bump mapping is a general term referencing a lighting technique used to trick the viewer into seeing something that isn't really there geometrically. Bump maps can be either grayscale (heightmaps) or full color (normal maps); the additional information in a full color bump map provides more information to the renderer so it may dazzle you with extra prettiness - like pink ponies, and such.
Let's take a look at publicly reviewed and accepted definitions:
Bump Mapping:
Additionally, bump maps applied to a large object are not intended for close up viewing as the visual effect of the lighting technique erodes upon closer inspection. To offer high detail upon close inspection, detail maps must be used in combination with bump maps to achieve the desired result (as with the MC in Halo 3). Detail maps allow an artist to greatly enhance the detail of a material to make it seem more real; this is typically achieved with a very low resolution texture meaning it is tiled more often and provides greater detail upon closer inspection.
Nick
Let's take a look at publicly reviewed and accepted definitions:
Bump Mapping:
Normal Mapping:Bump mapping is a computer graphics technique where at each pixel, a perturbation to the surface normal of the object being rendered is looked up in a heightmap and applied before the illumination calculation is done (see, for instance, Phong shading). The result is a richer, more detailed surface representation that more closely resembles the details inherent in the natural world.
The two discussed implementations of bump mapping can both be used for the same purpose - adding more detail to an object. However, the full color bump map will naturally have greater detail and look better. While it is true that both heightmaps and normal maps can be used for terrain generation, that is just an additional use for them. Heightmaps are used for displacement mapping, but that does not imply that is their only use.In 3D computer graphics, normal mapping is an application of the technique known as bump mapping. Normal mapping is sometimes referred to as "Dot3 bump mapping". While bump mapping perturbs the existing normal (the way the surface is facing) of a model, normal mapping replaces the normal entirely. Like bump mapping, it is used to add details to shading without using more polygons.
Additionally, bump maps applied to a large object are not intended for close up viewing as the visual effect of the lighting technique erodes upon closer inspection. To offer high detail upon close inspection, detail maps must be used in combination with bump maps to achieve the desired result (as with the MC in Halo 3). Detail maps allow an artist to greatly enhance the detail of a material to make it seem more real; this is typically achieved with a very low resolution texture meaning it is tiled more often and provides greater detail upon closer inspection.
Nick
Enough. You aren't above any form of law, Nick. Simply because you've been here longer gives you no right at all to command a staff member. The information you provided, while useful, can be found elsewhere as well, so Trep can do whatever he wants to your post. (Even though he wasn't the person who deleted the post in the first place.) Final word, continue assigned topic conversation.
I'll make this simple for you; don't repost when I tell you not too. So stop being an idiot and just follow one simple instruction. Your cooperation is appreciated. -Shout
Last edited by Nick on Wed Aug 30, 2006 2:58 pm, edited 1 time in total.
Well if you want facts here they are plain and simple. I am going to explain just how normal maps work for y'all
Let me start by explaining how basic shading works, because the two go hand in hand.
When a computer renders shading it first finds the normal of the polygon, to find where it is facing. Next it finds where the light is, and makes a vector from the light to the "base" of the normal, this is the Light Vector. The computer then measures the angle between the two vectors (normal and light vectors). Why do this? It is simple, if it is a small angle that means that the polygon is facing towards the light, so it needs to be bright. If the angle is big, the polygon is facing away from the light so it needs to be darker. The basic formula for lighting is brightness = N dot L. With N being the surface normal, and L being the light vector "Dot" is the angle between the two vectors (dot product if you remember geometry class)
Well that's just great if you want each polygon to be shaded by its self with no interpolation! But we don't. That's where this smart dude named Henri Gouraud comes in. In 1971 he came up with Gouraud Shading. This basically means that only the verticies get the N dot L treatment with linear gradients between them on for the shading. This is how the shading works in your viewport. Try making a low polly sphere (like 8 segments) and adding a light... You see what I am talking about, right? The problem is, this looks nasty!
How do we fix this? Pixel Shaders! Nvidia's Gforce2 graphics cards came out with hardware that allowed the N dot L formula to be calculated on every Pixel instead of every vertex, or every polygon.
This is where Normal Maps come in. In order to calculate shading for every little pixel the N dot L formula needs a Normal for each pixel. That is why they are called Normal maps. In a normal map the light blue pixels (127,127,255) are "flat", or, parallel to the polygon. Pink pixels are facing more to the right than the polygon, Green are more up, Purple are more down, and dark blue/green are more left. (With Nvidia, ATI are different, but that's beside the point
) PRESTO! you now know how normal maps work!
Now that you know this and your brains are pulsing with information and swelling in your skulls, you should know that normal maps are no better for rendering in 3ds max than good 'ol grey scale maps.
"WTF? LIES!!!!11" you may say BUT I have proof!

Can you tell my who is sporting the normal map? You can't can you! Buahahahahaha!!!!!!
The real glory of normal maps in in a real time environment. Like in HALO 2. For example, if you look at your MC in the viewport he looks plain and non bumpy. Now if you set up a directX shader with the camo bump you will be able to rotate around and see his beautiful bumps
Whereas the closest thing you could get to that without normal maps is something like active shade, or hitting quick render all the time...
"What about ZBrush?" you may ask "it is fancy and expensive and good looking and it uses normal maps" True, and the reason is simple. You can rotate and paint directly onto the model, all this happens in Real Time and thus Zbrush needs to use Normal maps in it's viewports so you can see what you are doing.
On a final note, because normal maps use pixels for everything, the more the better. If you had pixels that were the size of the polygons there would be no benefit. So that is why you see people like Jean running around with 4096x4096 sized normal maps.
I hope you are all smarter now and can get back to talking about how good my renders are !
PS: Woah! Spartans joined at birth! Creeeeeeeepyyyyyyyyy

Let me start by explaining how basic shading works, because the two go hand in hand.
When a computer renders shading it first finds the normal of the polygon, to find where it is facing. Next it finds where the light is, and makes a vector from the light to the "base" of the normal, this is the Light Vector. The computer then measures the angle between the two vectors (normal and light vectors). Why do this? It is simple, if it is a small angle that means that the polygon is facing towards the light, so it needs to be bright. If the angle is big, the polygon is facing away from the light so it needs to be darker. The basic formula for lighting is brightness = N dot L. With N being the surface normal, and L being the light vector "Dot" is the angle between the two vectors (dot product if you remember geometry class)
Well that's just great if you want each polygon to be shaded by its self with no interpolation! But we don't. That's where this smart dude named Henri Gouraud comes in. In 1971 he came up with Gouraud Shading. This basically means that only the verticies get the N dot L treatment with linear gradients between them on for the shading. This is how the shading works in your viewport. Try making a low polly sphere (like 8 segments) and adding a light... You see what I am talking about, right? The problem is, this looks nasty!
How do we fix this? Pixel Shaders! Nvidia's Gforce2 graphics cards came out with hardware that allowed the N dot L formula to be calculated on every Pixel instead of every vertex, or every polygon.
This is where Normal Maps come in. In order to calculate shading for every little pixel the N dot L formula needs a Normal for each pixel. That is why they are called Normal maps. In a normal map the light blue pixels (127,127,255) are "flat", or, parallel to the polygon. Pink pixels are facing more to the right than the polygon, Green are more up, Purple are more down, and dark blue/green are more left. (With Nvidia, ATI are different, but that's beside the point

Now that you know this and your brains are pulsing with information and swelling in your skulls, you should know that normal maps are no better for rendering in 3ds max than good 'ol grey scale maps.
"WTF? LIES!!!!11" you may say BUT I have proof!

Can you tell my who is sporting the normal map? You can't can you! Buahahahahaha!!!!!!
The real glory of normal maps in in a real time environment. Like in HALO 2. For example, if you look at your MC in the viewport he looks plain and non bumpy. Now if you set up a directX shader with the camo bump you will be able to rotate around and see his beautiful bumps

"What about ZBrush?" you may ask "it is fancy and expensive and good looking and it uses normal maps" True, and the reason is simple. You can rotate and paint directly onto the model, all this happens in Real Time and thus Zbrush needs to use Normal maps in it's viewports so you can see what you are doing.
On a final note, because normal maps use pixels for everything, the more the better. If you had pixels that were the size of the polygons there would be no benefit. So that is why you see people like Jean running around with 4096x4096 sized normal maps.
I hope you are all smarter now and can get back to talking about how good my renders are !
PS: Woah! Spartans joined at birth! Creeeeeeeepyyyyyyyyy
