Stone Textures For Terrain Materials, From Scratch -Part II
by JR Wilde · in Artist Corner · 12/27/2012 (4:57 pm) · 3 replies
Stone Textures For Terrain Materials, From Scratch -Part II
Welcome to the second part of this instructional. Now that we know how to create a stone in a pinch, we're going to finish it off properly and take it all the way with parallax mapping. This turned out to be a bit deeper than I had originally thought. So let's get to it.
I'm going to cover the following in this tutorial:
1. Using nvidia's normal map filter.
2. Altering an image to "prep" it for the normal mapper.
3. Generating a low frequency height map.
4. Saving a DDS normal map with an alpha channel using nVidia's DDS Tools (dxt5_NM.dds)
5. Gamma Correction and image sharpening filters in nVidia's DDS Tools.
First things first, let's talk about image frequency. By knowing what this is, if you don't already, you can make logical and aesthetic descisions for your final materials and the results will be sharper, crisper and impart a more believable illusion of depth than if you did not take the time and use this knowledge.
Image frequency can be translated somewhat to image complexity. Take a plain black 512x512 texture for instance (if one color can be considered a "texture"): it has just about the lowest frequency possible. Now take the same texture and make all 262144 pixels a uniquecolor, and that would be about the highest frequency you could possibly get. To put this in terms of our stone, noise = high frequency, cracks = medium frequency, bumps = low frequency; as you can see, we have a fairly high frequency in our stone.
Consider the stone as a whole and what areas should be pushed-out, or rise above the terrain. This is going to be achieved with a low frequency Height Map to be used in Torque's default parallax shader. This shader is a gross approximation of a parallax effect. This is important to understand. If the parallax effect is set too high, and you have a high frequency in the height map, it will look like _ _ _ _ and destroy all disbelief because everything will be swirling all over the place.
I must admit though, I've had a great deal of fun shamelessly abusing the parallax for "dream sequences" or other surrealistic states: kalidascopic colors swirling in an animated sky box, all of the materials very high in parallax and the whole world goes topsy turvy. I was striving for a "Tak on Acid" world with no terrain save floating stones (with contrary animated textures) to traverse and nasty things to pound on, I digress...
Parallax is a subtle effect, imo, and should be used with care when players are going to be up close and personal with it. I tend to have a sandbox mentality so nearly all of my textures can be seen up close and personal. It is here where the effect is at its most voulnerable to disbelief. Like SSAO, parallax is oftentimes abused. Keep this is mind however: the lower the frequency of your height map, the more parallax effect you can give the material before it breaks down and "swims." Why? We'll need to know what parallax is, ergo..
What is parallax exactly? a rather difficult and bumpy subject, but parallax is the notion that surfaces appear to move relative to one another. Here's one way to visualize it: Imagine you are riding as a passenger in someone's car. Look at their speedometer from the angle you are sitting at. You might notice that your perception of where the needle lies on the guage is probably different than the driver's perception of it. Depending on the country, your distance and viewing angle, you will either see it as "faster" or "slower" than the driver. This is because the Parallax Effect is dependant on the viewing angle or point of view. As one moves their viewing angle to the left or right, through the driver's head (this is a mental exercise so we can do whatever we want), while looking at the same spot, the needle appears to "move" over to and beyond the "actual" or direct-on position that the driver sees. So surfaces seem to move relative to one another. This is a clue for us in our perception of depth. It should be noted again, that every surface that is'nt flat bears this phenomena.
So now we know what parallax is, how do we use it? We need to know one last thing: what a normal map is (and ogres think they have layers).
The normal map normally comes from the normal of a model but what is a normal? Simply put, a normal (or normal vector) is the direction one polygon of the model is facing -the visible side. A polygon in this context is a triangle. So, how does this equate to a flat 2-dimentional surface and the abillity to give it the illusion of depth? Very simply put, as its rather complicated: Imagine that the stone image you've just finished was created with a 3D model (In CS6, it actually was) and therefore has normals. If we encode the directions of the model into a Normal Map and superimpose that map onto the lower polygon terrain we can give subtle clues to the eye that were contained in the higher poly model.
To reiterate, the normal vector (that is encoded in the normal map as described above) tells Torque which direction a particular area (traingle) of the terrain model will reflect the directional/sun and ambient lights. When we describe the terrain surface, by creating a material, we use these "maps" normal vector and reference it with the height map, we can then tell Torque how to render said light striking that particular area, from what "depth" and to which direction(s) it will reflect/scatter, thereby giving us the perceived extra depth on the terrain's low tesselated mesh. The terrain itself doesn't have the geometry that the normal map is "providing." Normals give extra detail because of the detail's (crack, bump, blister what have you) shadow, not its height. A normal map "shadows" things and only gives the illusion of a geometry that is not there. A Normal Map has everything to do witht lighting and the reason why every surface should have one if you want your material to respond properly to light in T3D. A diffuse and normal are the barest minimum.
Now we can make some informed descisions as to what part of the stone that we want to stick out boldly via the [i]height map, and those areas where the extra detail will be filled in through the normal and detail maps. There is an underlying philosophy behind this instructional which is this: The height map will be used for large scale depth, the normal map will take care of the medium details, and the detail map will, er, take care of the details ;)
1. Creating the Height Map.
1. Delete any alpha channels.
2. Switch to layers tab.
3. Right-click on Detail layer and select Duplicate layer. Use the drop-down to select new and give it the O.K. This new document will be discarded later so you don't need to name it.
4. Ctrl+Tab to switch to the new document.
5. Alt+I and select Mode>Greyscale.
6. Alt+T and select Sharpen>Unsharp Mask.
7. Adjust Settings. Adjust the settings as seen. Play with the radius and note what happens. The lower the radius is, slider to the left, the more embossed it looks -not good for our purposes here. We want to remove as much intricate detail as possible. We also want extremes in black and white. Black being "low" and white being "high" in the height calculations.
As an aside, a combination of the high pass filter and the unsharp mask and one has something that looks strikingly like an occlusion map...
8. Alt+T and select Blur>Blur[i/].
9. Ctrl+F a few times -I did seven times or so to get the result in the image below. It should look "softer."
10. Ctrl+J to duplicate the layer.
11. Set duplicated layer's blending mode to [i]overlay.
12. [b]Ctrl+F twice (remember that this repeats the last filter used, which should be blur).
[image]blurTwice.jpg
13. Ctrl+J to duplicate -this should duplicate the layer's blending mode (overlay) as well; if not, make certain that you've duplicated the correct layer. I hope you're starting to see what's happening here.
14. Continue along your merry way duplicating and blurring to your hearts content. see the image below to see where I stopped -do whatever you want. Notice that this one did not require many layers at all; some will require several more depending upon your initial unsharp settings and type of blurring used.
NOTE: you can use gaussian and surface blur[i] or a combination of all three of them as well. If you do so, also dink around with layer opacity and fill opacity sliders. Each algorithm imparts its own unique tone or feeling to the whole.
15. Right-click on any selected layer and select [i]Merge visible.
16. Ctrl+Click on the merged layer's thumbnail. This action selects every pixel on the layer.
For future reference: you can do this to multiple layers to "add to the selection" (Ctrl+Shift[b]+[b]Click[b]) and then move that selection to any layer, by selecting the layer you want the selection on, including a newly created one; very handy indeed. Take it a step further: think of storing selections in alpha channels. Why keep reselecting the same thing over and over the tedious way?
17. [b]Ctrl+C to copy the selected pixels to the clipboard -which can be pasted into any layer/document.
18. Create a new Channel. Switch to the channels tab and create a new channel.
19. Ctrl+V to paste the copied layer into the new channel.
20. Right-click on the channel and select Duplicate channel and send it to the original document from whence this one came and make that document active (the original document).
21. Switch to layers tab and select the detail layer.
22. Ctrl+J twice. You should now have three "detail" layers (detail, detail copy, detail copy 2). We're done with the height map, and have paved the way for normal map creation.
2. Creating the Normal Map.
1. Select diffuse copy 2l layer. We're not going to do anything special with this layer; this normal map will be for comparative purposes only. To me this is the bare minimum way. I will admit that there are a few extremely rare instances where this has worked better for me than the method below. I have noticed this specifically with some vegetation -this is because I also use detail normals for vegetation and if the normal is too "big," it completely overshadows the intricacy of the detail normal. For any other purposes I highly recomend you take the extra thirty seconds or so and do it right. When I say "do it right," I don't necessarily mean "do it my way." I mean "do it right by you: do it from the heart." Which translates into "take the time, but be quick about it."
2. [b]Ctrl+T and select nVidia Tools>normal mapper. Use the following settings (I believe that they're default, but don't quote me on it)
Add height to normal map.
Height Generation:
4 Sample
Wrap
Invert Y
Height Source:
Average RGB
Alpha Field:
Unchanged
NOTE: Although I haven't dinked around with it, I believe it might be possible to set the alpha channel to height and, whithout using an alpha channel as the height source (have an empty alpha channel), let the filter create a height map and place it in the alpha channel. I just thought of this, so I'll check it out and see what happens (I'm writing it here so I can remember to test it out -maybe someone else can as well). It certainly has all the data it requires. Even if it can, it may not be plausable or useful. I have serious doubts. On the other hand, if it works it could be awesome. I love experimenting.
3. Give it the O.K. There's our vanilla flavored normal map. Now let's do something more interesting with the other layer. Go ahead and hide the lazy layer. You can do so by clicking on the eyeball next to the layer's thumbnail.
4. Select the detail copy layer[/i].
5. Alt+T and select Sharpen>Unsharp mask. Yes, we're going to visit our old friend.
6. Adjust settings as shown. This time we don't want big details, so the radius and amount need to be dropped down. It should still be fairly "contrasty" with the edges well defined; the contrast between the light edges and the darker regions is what will give us the shadowing effect of the details. It's possible to take the sharpening too far and the shadows will look fake and painted-on rather than giving the illusion of depth through the use of shadow; in fact, it can go so far that they don't even look like shadows at all, just dots, bands and swishes. When you're happy with your results, give it the O.K.
7. Alt+T and select Blur>Blur. Don't worry, the sharpening effect is still there, you didn't undo it. You can see it clearly near the larger cracks; this is actually a subtle enhancement. In-game its more of a "feeling" than something the eye can distinctly see, although some eyes will be observant enough I'd imagine.
8. Ctrl+J to duplicate layer. and switch the mode to Overlay.
9. Alt+T and select Blur>Surface Blur.
10. Repeat steps 8 and 9 if there's a need for it. Remember that this technique brings out large detail, so for this normal, its probably best to follow the "less is more" philosophy.
11. Select all of the duplicated/blurred layer(s) along with the original. Right-click and select Merge layers.
12. Ctrl+M and enhance the medium and light details that we want. I've torqued the curves to a great extent to illustrate this dialog's versatiliy. Notice the "notches" highlighted in yellow. These are similar to audio's "notch filters." Mixing consoles make heavy use of them in the on board EQ's. For this example, put two points on either side of the frequency(ies) you want to enhance (cut or boost). As you can see in the image I didn't do this on the notch to the right. This is, of course, to illustrate another point: clipping. Notice the curve highlighted in red. The color (signal) can't go beyond that point (It's really bad if you clip the lighter colors). Just to be contrary, I'm going to do the "bad thing" and leave it. For this example, it brought out the mids while leaving some ot the lights intact, and it clipped the low end. ie It also enhanced the darkness of the cracks; probably too much, but we'll see. . .
13. Alt+T and select nVidia Tools>Normal mapper and use the settings in the image as a guide. Keep the alpha setting at Unchanged. When you're happy with it, give it the O.K.
14. Ctrl+Shift+S and select DDS as the Save As... file type. Adjust the settings as you see in the image. Here's something to remember: normal maps need no gamma correction, diffuse and details do. This has to do with linear space and I'll not go into that. Just remember to uncheck the gamma correction for normals, and you'll be fine ;) .
[image www.synchronicity3d.com/wp-content/uploads/2012/12/noGamma.jpg[/image]
3. Altering The Material.
In the World Editor press F3 to enter terrain painting mode. Find the stone material you created in the first part. Open the material dailog by double clicking on it. Add your normal map and set parallax to around .015 to start. Its a big height map.
You're done. Start to finish.
Soon I'll have a PDF of this and the prior instructional on my website. For the moment, I need to get back to my environment kit.
I hope these methods inspire you to try all kinds of new things to further your art. I hope mostly though that I have described the [i]WHY[i] behind the how. Lastly, I'd like to reiterate that I don't necessarily think you should do things my way or want you to think that this is the only way to do things -this is just one of many. You and you alone must blaze your own trail using the compass of your knowledge as your guide.
thanks for reading.
J.R. Wilde
PS. Nearly forgot the screen shots. Not as many, but you should get the point.
[image]www.synchronicity3d.com/wp-content/uploads/2012/12/blend.jpg[image]
[image]www.synchronicity3d.com/wp-content/uploads/2012/12/rockyBeach.jpg[image]
[image]www.synchronicity3d.com/wp-content/uploads/2012/12/distance.jpg[image]
[image]www.synchronicity3d.com/wp-content/uploads/2012/12/hump.jpg[image]
[image]www.synchronicity3d.com/wp-content/uploads/2012/12/underwater.jpg[image]
[image]www.synchronicity3d.com/wp-content/uploads/2012/12/beach.jpg[image]
Welcome to the second part of this instructional. Now that we know how to create a stone in a pinch, we're going to finish it off properly and take it all the way with parallax mapping. This turned out to be a bit deeper than I had originally thought. So let's get to it.
I'm going to cover the following in this tutorial:
1. Using nvidia's normal map filter.
2. Altering an image to "prep" it for the normal mapper.
3. Generating a low frequency height map.
4. Saving a DDS normal map with an alpha channel using nVidia's DDS Tools (dxt5_NM.dds)
5. Gamma Correction and image sharpening filters in nVidia's DDS Tools.
First things first, let's talk about image frequency. By knowing what this is, if you don't already, you can make logical and aesthetic descisions for your final materials and the results will be sharper, crisper and impart a more believable illusion of depth than if you did not take the time and use this knowledge.
Image frequency can be translated somewhat to image complexity. Take a plain black 512x512 texture for instance (if one color can be considered a "texture"): it has just about the lowest frequency possible. Now take the same texture and make all 262144 pixels a uniquecolor, and that would be about the highest frequency you could possibly get. To put this in terms of our stone, noise = high frequency, cracks = medium frequency, bumps = low frequency; as you can see, we have a fairly high frequency in our stone.
Consider the stone as a whole and what areas should be pushed-out, or rise above the terrain. This is going to be achieved with a low frequency Height Map to be used in Torque's default parallax shader. This shader is a gross approximation of a parallax effect. This is important to understand. If the parallax effect is set too high, and you have a high frequency in the height map, it will look like _ _ _ _ and destroy all disbelief because everything will be swirling all over the place.
I must admit though, I've had a great deal of fun shamelessly abusing the parallax for "dream sequences" or other surrealistic states: kalidascopic colors swirling in an animated sky box, all of the materials very high in parallax and the whole world goes topsy turvy. I was striving for a "Tak on Acid" world with no terrain save floating stones (with contrary animated textures) to traverse and nasty things to pound on, I digress...
Parallax is a subtle effect, imo, and should be used with care when players are going to be up close and personal with it. I tend to have a sandbox mentality so nearly all of my textures can be seen up close and personal. It is here where the effect is at its most voulnerable to disbelief. Like SSAO, parallax is oftentimes abused. Keep this is mind however: the lower the frequency of your height map, the more parallax effect you can give the material before it breaks down and "swims." Why? We'll need to know what parallax is, ergo..
What is parallax exactly? a rather difficult and bumpy subject, but parallax is the notion that surfaces appear to move relative to one another. Here's one way to visualize it: Imagine you are riding as a passenger in someone's car. Look at their speedometer from the angle you are sitting at. You might notice that your perception of where the needle lies on the guage is probably different than the driver's perception of it. Depending on the country, your distance and viewing angle, you will either see it as "faster" or "slower" than the driver. This is because the Parallax Effect is dependant on the viewing angle or point of view. As one moves their viewing angle to the left or right, through the driver's head (this is a mental exercise so we can do whatever we want), while looking at the same spot, the needle appears to "move" over to and beyond the "actual" or direct-on position that the driver sees. So surfaces seem to move relative to one another. This is a clue for us in our perception of depth. It should be noted again, that every surface that is'nt flat bears this phenomena.
So now we know what parallax is, how do we use it? We need to know one last thing: what a normal map is (and ogres think they have layers).
The normal map normally comes from the normal of a model but what is a normal? Simply put, a normal (or normal vector) is the direction one polygon of the model is facing -the visible side. A polygon in this context is a triangle. So, how does this equate to a flat 2-dimentional surface and the abillity to give it the illusion of depth? Very simply put, as its rather complicated: Imagine that the stone image you've just finished was created with a 3D model (In CS6, it actually was) and therefore has normals. If we encode the directions of the model into a Normal Map and superimpose that map onto the lower polygon terrain we can give subtle clues to the eye that were contained in the higher poly model.
To reiterate, the normal vector (that is encoded in the normal map as described above) tells Torque which direction a particular area (traingle) of the terrain model will reflect the directional/sun and ambient lights. When we describe the terrain surface, by creating a material, we use these "maps" normal vector and reference it with the height map, we can then tell Torque how to render said light striking that particular area, from what "depth" and to which direction(s) it will reflect/scatter, thereby giving us the perceived extra depth on the terrain's low tesselated mesh. The terrain itself doesn't have the geometry that the normal map is "providing." Normals give extra detail because of the detail's (crack, bump, blister what have you) shadow, not its height. A normal map "shadows" things and only gives the illusion of a geometry that is not there. A Normal Map has everything to do witht lighting and the reason why every surface should have one if you want your material to respond properly to light in T3D. A diffuse and normal are the barest minimum.
Now we can make some informed descisions as to what part of the stone that we want to stick out boldly via the [i]height map, and those areas where the extra detail will be filled in through the normal and detail maps. There is an underlying philosophy behind this instructional which is this: The height map will be used for large scale depth, the normal map will take care of the medium details, and the detail map will, er, take care of the details ;)
1. Creating the Height Map.
1. Delete any alpha channels.
2. Switch to layers tab.
3. Right-click on Detail layer and select Duplicate layer. Use the drop-down to select new and give it the O.K. This new document will be discarded later so you don't need to name it.
4. Ctrl+Tab to switch to the new document.
5. Alt+I and select Mode>Greyscale.
6. Alt+T and select Sharpen>Unsharp Mask.
7. Adjust Settings. Adjust the settings as seen. Play with the radius and note what happens. The lower the radius is, slider to the left, the more embossed it looks -not good for our purposes here. We want to remove as much intricate detail as possible. We also want extremes in black and white. Black being "low" and white being "high" in the height calculations.
As an aside, a combination of the high pass filter and the unsharp mask and one has something that looks strikingly like an occlusion map...
8. Alt+T and select Blur>Blur[i/].
9. Ctrl+F a few times -I did seven times or so to get the result in the image below. It should look "softer."
10. Ctrl+J to duplicate the layer.
11. Set duplicated layer's blending mode to [i]overlay.
12. [b]Ctrl+F twice (remember that this repeats the last filter used, which should be blur).
[image]blurTwice.jpg
13. Ctrl+J to duplicate -this should duplicate the layer's blending mode (overlay) as well; if not, make certain that you've duplicated the correct layer. I hope you're starting to see what's happening here.
14. Continue along your merry way duplicating and blurring to your hearts content. see the image below to see where I stopped -do whatever you want. Notice that this one did not require many layers at all; some will require several more depending upon your initial unsharp settings and type of blurring used.
NOTE: you can use gaussian and surface blur[i] or a combination of all three of them as well. If you do so, also dink around with layer opacity and fill opacity sliders. Each algorithm imparts its own unique tone or feeling to the whole.
15. Right-click on any selected layer and select [i]Merge visible.
16. Ctrl+Click on the merged layer's thumbnail. This action selects every pixel on the layer.
For future reference: you can do this to multiple layers to "add to the selection" (Ctrl+Shift[b]+[b]Click[b]) and then move that selection to any layer, by selecting the layer you want the selection on, including a newly created one; very handy indeed. Take it a step further: think of storing selections in alpha channels. Why keep reselecting the same thing over and over the tedious way?
17. [b]Ctrl+C to copy the selected pixels to the clipboard -which can be pasted into any layer/document.
18. Create a new Channel. Switch to the channels tab and create a new channel.
19. Ctrl+V to paste the copied layer into the new channel.
20. Right-click on the channel and select Duplicate channel and send it to the original document from whence this one came and make that document active (the original document).
21. Switch to layers tab and select the detail layer.
22. Ctrl+J twice. You should now have three "detail" layers (detail, detail copy, detail copy 2). We're done with the height map, and have paved the way for normal map creation.
2. Creating the Normal Map.
1. Select diffuse copy 2l layer. We're not going to do anything special with this layer; this normal map will be for comparative purposes only. To me this is the bare minimum way. I will admit that there are a few extremely rare instances where this has worked better for me than the method below. I have noticed this specifically with some vegetation -this is because I also use detail normals for vegetation and if the normal is too "big," it completely overshadows the intricacy of the detail normal. For any other purposes I highly recomend you take the extra thirty seconds or so and do it right. When I say "do it right," I don't necessarily mean "do it my way." I mean "do it right by you: do it from the heart." Which translates into "take the time, but be quick about it."
2. [b]Ctrl+T and select nVidia Tools>normal mapper. Use the following settings (I believe that they're default, but don't quote me on it)
Add height to normal map.
Height Generation:
4 Sample
Wrap
Invert Y
Height Source:
Average RGB
Alpha Field:
Unchanged
NOTE: Although I haven't dinked around with it, I believe it might be possible to set the alpha channel to height and, whithout using an alpha channel as the height source (have an empty alpha channel), let the filter create a height map and place it in the alpha channel. I just thought of this, so I'll check it out and see what happens (I'm writing it here so I can remember to test it out -maybe someone else can as well). It certainly has all the data it requires. Even if it can, it may not be plausable or useful. I have serious doubts. On the other hand, if it works it could be awesome. I love experimenting.
3. Give it the O.K. There's our vanilla flavored normal map. Now let's do something more interesting with the other layer. Go ahead and hide the lazy layer. You can do so by clicking on the eyeball next to the layer's thumbnail.
4. Select the detail copy layer[/i].
5. Alt+T and select Sharpen>Unsharp mask. Yes, we're going to visit our old friend.
6. Adjust settings as shown. This time we don't want big details, so the radius and amount need to be dropped down. It should still be fairly "contrasty" with the edges well defined; the contrast between the light edges and the darker regions is what will give us the shadowing effect of the details. It's possible to take the sharpening too far and the shadows will look fake and painted-on rather than giving the illusion of depth through the use of shadow; in fact, it can go so far that they don't even look like shadows at all, just dots, bands and swishes. When you're happy with your results, give it the O.K.
7. Alt+T and select Blur>Blur. Don't worry, the sharpening effect is still there, you didn't undo it. You can see it clearly near the larger cracks; this is actually a subtle enhancement. In-game its more of a "feeling" than something the eye can distinctly see, although some eyes will be observant enough I'd imagine.
8. Ctrl+J to duplicate layer. and switch the mode to Overlay.
9. Alt+T and select Blur>Surface Blur.
10. Repeat steps 8 and 9 if there's a need for it. Remember that this technique brings out large detail, so for this normal, its probably best to follow the "less is more" philosophy.
11. Select all of the duplicated/blurred layer(s) along with the original. Right-click and select Merge layers.
12. Ctrl+M and enhance the medium and light details that we want. I've torqued the curves to a great extent to illustrate this dialog's versatiliy. Notice the "notches" highlighted in yellow. These are similar to audio's "notch filters." Mixing consoles make heavy use of them in the on board EQ's. For this example, put two points on either side of the frequency(ies) you want to enhance (cut or boost). As you can see in the image I didn't do this on the notch to the right. This is, of course, to illustrate another point: clipping. Notice the curve highlighted in red. The color (signal) can't go beyond that point (It's really bad if you clip the lighter colors). Just to be contrary, I'm going to do the "bad thing" and leave it. For this example, it brought out the mids while leaving some ot the lights intact, and it clipped the low end. ie It also enhanced the darkness of the cracks; probably too much, but we'll see. . .
13. Alt+T and select nVidia Tools>Normal mapper and use the settings in the image as a guide. Keep the alpha setting at Unchanged. When you're happy with it, give it the O.K.
14. Ctrl+Shift+S and select DDS as the Save As... file type. Adjust the settings as you see in the image. Here's something to remember: normal maps need no gamma correction, diffuse and details do. This has to do with linear space and I'll not go into that. Just remember to uncheck the gamma correction for normals, and you'll be fine ;) .
[image www.synchronicity3d.com/wp-content/uploads/2012/12/noGamma.jpg[/image]
3. Altering The Material.
In the World Editor press F3 to enter terrain painting mode. Find the stone material you created in the first part. Open the material dailog by double clicking on it. Add your normal map and set parallax to around .015 to start. Its a big height map.
You're done. Start to finish.
Soon I'll have a PDF of this and the prior instructional on my website. For the moment, I need to get back to my environment kit.
I hope these methods inspire you to try all kinds of new things to further your art. I hope mostly though that I have described the [i]WHY[i] behind the how. Lastly, I'd like to reiterate that I don't necessarily think you should do things my way or want you to think that this is the only way to do things -this is just one of many. You and you alone must blaze your own trail using the compass of your knowledge as your guide.
thanks for reading.
J.R. Wilde
PS. Nearly forgot the screen shots. Not as many, but you should get the point.
[image]www.synchronicity3d.com/wp-content/uploads/2012/12/blend.jpg[image]
[image]www.synchronicity3d.com/wp-content/uploads/2012/12/rockyBeach.jpg[image]
[image]www.synchronicity3d.com/wp-content/uploads/2012/12/distance.jpg[image]
[image]www.synchronicity3d.com/wp-content/uploads/2012/12/hump.jpg[image]
[image]www.synchronicity3d.com/wp-content/uploads/2012/12/underwater.jpg[image]
[image]www.synchronicity3d.com/wp-content/uploads/2012/12/beach.jpg[image]
About the author
see website: http://www.synchronicity3d.com email me at syncyronicity3d "at" gmail.com if you'd like.
asdf asdf