Modelling and Rendering the TMP Enterprise
- Thread starter Science Officer
- Start date
Think I told you this somewhere at the start of the thread: the original paintjob, and the one you see at trekmodeler, aswell as todays paintjobs on cars, have "two specular colors". First there is the specular color the paint reflects, yellow/bleu/red whatever, reflected by the pigment, then there is a colorless specular color, reflected by the smooth surface, this one would also be the reason lights etc. reflect, or mirror in it.
When you put a bleu light on them, lets say the yellow one, the first specular color will remain yellow, mixed with the (in this case more bleuish) diffuse color you gave to that paint, the second one, the colorless, will remain colorless. If you put on more light, like a bridge spotlight, this colorless will become more white, reflecting more like a metal plate, and the yellow one underneath will not be seen anymore.
I can mimic that in 3dsmax with a multilayer shader, it has two specular colors, first the yellow or bleu or whatever color, second the colorless, in which if desired I can mix some dirt in.
Told you this about the multilayer shader earlier, but I think you said C4D does not have that one, and so my suggestions come to and end
, but you came this far, so don't give up, it is allready looking superb.Hi Wil,
I should of tried a lot harder to see how your terminology for Max would port to C4D.
Thinking back, I did some tests interpreting multiple layers to mean multiple materials on a single object (I found that the specular highlights didn't combine very well). I also quickly assumed that Max had a pearlescent shader and couldn't relate that to C4D's LUMAS/DANEL shaders, hence the statement that C4D had nothing similar.
I had a very crude idea of how this would work around post 47. But then I latched onto something that was giving the right kind of results, not realising I was optimising for a subset of possible lighting conditions! That was a lack of discipline on my part - never take shortcuts.
But on a positive note, the progress to date isn't wasted - a detailed Aztec pattern and a good intensity balance between colour and specular channels.
It looks like the next steps will be:
I apologize Wil if it looked like I'd dismissed your suggestions too easily. You've always encouraged me on this project and your input is always welcome and valued!
Mistakes made, fingers burnt, lessons learnt!
Cheers,
S.O.
I should of tried a lot harder to see how your terminology for Max would port to C4D.
Thinking back, I did some tests interpreting multiple layers to mean multiple materials on a single object (I found that the specular highlights didn't combine very well). I also quickly assumed that Max had a pearlescent shader and couldn't relate that to C4D's LUMAS/DANEL shaders, hence the statement that C4D had nothing similar.
I had a very crude idea of how this would work around post 47. But then I latched onto something that was giving the right kind of results, not realising I was optimising for a subset of possible lighting conditions! That was a lack of discipline on my part - never take shortcuts.
But on a positive note, the progress to date isn't wasted - a detailed Aztec pattern and a good intensity balance between colour and specular channels.
It looks like the next steps will be:
- Use a pearl colour in the colour channel (Y/B/R/G).
- Put the required colour inversion in one specular channel of the LUMAS shader.
- Put a light, colourless specular into another specular channel of the LUMAS shader with some anisotrophy.
- Put a broad, low intensity blue specular into another specular channel or the LUMAS diffuse channel.
I apologize Wil if it looked like I'd dismissed your suggestions too easily. You've always encouraged me on this project and your input is always welcome and valued!
Mistakes made, fingers burnt, lessons learnt!
Cheers,
S.O.
Sorry if I gave you the impression that you did something you need to apologize for, because that was not my intention, oops.
So I read MY post and I think this is what gave you a wrong impression "and so my suggestions come to and end" which I maybe should have said like this "I do not have any ideas how to solve the problems"
I really was thinking C4D did not have the multiple layer shader, but it has. And now it has, I would have suggested your steps 3 and 4, but you figured that out yourself allready
So I read MY post and I think this is what gave you a wrong impression "and so my suggestions come to and end" which I maybe should have said like this "I do not have any ideas how to solve the problems"
I really was thinking C4D did not have the multiple layer shader, but it has. And now it has, I would have suggested your steps 3 and 4, but you figured that out yourself allready
Sorry if I gave you the impression that you did something you need to apologize for, because that was not my intention, oops.
So I read MY post and I think this is what gave you a wrong impression "and so my suggestions come to and end" which I maybe should have said like this "I do not have any ideas how to solve the problems"
I really was thinking C4D did not have the multiple layer shader, but it has. And now it has, I would have suggested your steps 3 and 4, but you figured that out yourself allready
Hi Wil,
That's the trouble with the written word - it has a habit of removing the context in which a comment is made. Too many a post has been taken the wrong way by people and led to all out forum war! It's much better to talk!
But anyway all confusion has been sorted - all the way back to around post 47! Now I know exactly what you were trying to tell me back then, I should be able to take any future comments and make much better use of them.
Cheers,
S.O.
It's back to the drawing board as I now try to get the pearl effect to work with four colours (Red, Green, Yellow, Blue) and a blue coating that is obvious in the spotlight movie snapshot.
In the following animation I have a white and blue light moving around a cylinder. There are 4 pearl colours, each with an additional blue specular to mimic the coating. You can see the colour inversion in action.
http://img839.imageshack.us/img839/2181/testg.mp4
This is very rough and it'll take some time to dial everything in. But it's a start.
Cheers,
S.O.
In the following animation I have a white and blue light moving around a cylinder. There are 4 pearl colours, each with an additional blue specular to mimic the coating. You can see the colour inversion in action.
http://img839.imageshack.us/img839/2181/testg.mp4
This is very rough and it'll take some time to dial everything in. But it's a start.
Cheers,
S.O.
P.S. In case you're wondering what you're looking at, here's an image using a default light:
Hi Dac,
If you've got C4D v8.2, you'll have exactly the same material/shader capability as me. If you have v9 or later, I envy you! I interchange the term material and shader a lot as I'm trying not to be C4D specific - but as you saw with Wil, it can lead to confusion as to what is meant. Also note that with v9 onwards, I think Maxon bought the bhodiNUT shaders and properly incorporated tham into C4D, dropping the name.
So as a fellow C4D user, let me explain - you might want to have a go!
The above scene uses 4 materials - one for each pearl colour (5 if I include the larger scale checkerboard mask). Each of the 4 materials is specific to an area of the cylinder - they're not mixed to produce an effect.
I have some colour in the colour channel (one of pastel red, green, blue or yellow). In the luminosity channel I use a bhodiNUT Lumas shader, which gives you access to:
- 1 diffuse channel
- 3 specular channels
- an anisotrophy channel
- yellow colour - blue specular
- blue colour - yellow specular
- red colour - green specular
- green colour - red specular
All other channels in the material (with the possible exception of the Alpha channel) should be switched off.
So for example. if we take the red pearl material:
Colour Channel
RGB = 255, 240, 240. Brightness = 95%.
Luminance Channel
RGB = 255, 255, 255. Brightness = 100%.
Mix = Normal.
Texture = bhodiNUT Lumas.
In the Lumas Shader
Switch off Diffuse, Specular 1 and Anisotrophy. We might use these later.
In Specular 2...
RGB = 128, 255, 128. Brightness = 80%.
Intensity = 33%. Size = 25%. Contrast = 0%.
Glare = 200%. Falloff = 50%.
In Specular 3...
RGB = 128, 128, 255. Brightness = 100%.
Intensity = 100%. Size = 85%. Contrast = 0%.
Glare = 75%. Falloff = 50%.
As it's very early days, Glare and Falloff should be set to 0%. This will be done when I do this in anger.
The other three materials just use different colours.
The mix of colour channel, speculars and light colour generate a variety of hues and shades. It could lead you to think the materials were quite complex, but actually it's just how the channels interact with each other. The real pain is try to balance the channels to get the correct effect for the Enterprise!
Cheers,
S.O.
Wow, I use 11.5, and I think a fair bit has changed since 8.5 
Having said that though, I've found the lumas function and the multiple specular option that gives you. I think I may have a go with that now to see what I can come up with, although I doubt it will be half as good as yours as texturing really isnt my strong point. I tend go by the "If it looks good enough its done" school of modeling, not the pixel perfect one you've got down to a T. I look forward to seeing how you finish yours though, and thanks for the in depth reply
Having said that though, I've found the lumas function and the multiple specular option that gives you. I think I may have a go with that now to see what I can come up with, although I doubt it will be half as good as yours as texturing really isnt my strong point. I tend go by the "If it looks good enough its done" school of modeling, not the pixel perfect one you've got down to a T. I look forward to seeing how you finish yours though, and thanks for the in depth reply
Nice S.O., a ton better allready
Let me see, try to think with me here, when I see the ship in the movie, and the one that is painted on shipmodeler,
I can see a perfect "grey/white" aztec pattern, but when a Ent spot hits it (or sun, or drydock spot), I can see a broken up aztec pattern.
IMHO, the perfect pattern is from the reflection colorless layer, either it reflects or it doesn't, and the broken up pattern is from the colored specularity.
Have the impression that Olsen first put in the different specularity colors, and a aztec pattern layer of different finish over it.
If this is so, you will need to give the different (broken up) colored polies the right colorless layer to get the overall aztec back.
Does this make sense? maybe you figured out something like that allready yourself?
Anyway, it is going to look great
Let me see, try to think with me here, when I see the ship in the movie, and the one that is painted on shipmodeler,
I can see a perfect "grey/white" aztec pattern, but when a Ent spot hits it (or sun, or drydock spot), I can see a broken up aztec pattern.
IMHO, the perfect pattern is from the reflection colorless layer, either it reflects or it doesn't, and the broken up pattern is from the colored specularity.
Have the impression that Olsen first put in the different specularity colors, and a aztec pattern layer of different finish over it.
If this is so, you will need to give the different (broken up) colored polies the right colorless layer to get the overall aztec back.
Does this make sense? maybe you figured out something like that allready yourself?
Anyway, it is going to look great
Hi Folks,
Dac - I'm not all that good! For example, my saucer took a long time to do as I tried to overcome C4D's interpolation "problems" for rendering adjacent polygons. I look at others peoples work like your own, Wils or those on Foundation 3D and SciFi Meshes and realize I've got a lot to learn if I'm ever to have my own complete Enterprise.
Wil - I think I know what you mean. If we take the Cloudster images and the movie snapshot I use, you can see very bright patterns in the darker parts of the saucer. They don't look like part of the main material colour and don't coincide with an obvious specular highlight. For example the back half of the saucer on the first Cloudster image (top right-hand corner of post 82). You might want to confirm this.
If this is the pattern you're talking abount, I think I know what it is. My guess is that the brighter areas are due to a fine roughness of the surface as they also correspond to the whiter/less specular parts of the hull. Therefore the light is scattered/diffused more. It's the kind of effect you get if you rub sandpaper over something.
As you add more layers of paint, the surface becomes smoother and consequently the scattering/diffusion decreases. So the aztec pattern is part material colour/part scattering. I need to look at TrekModeler's images to see if this is the case.
In C4D we can mimic this in the Illumination channel, where we can specify the Phong Falloff. A high value gives the surface an almost chalky appearence.
I did play with this, but I haven't any renders done to demonstrate the effect. At one point I overdid it and it looked terrible - top value should be 10% for the brightest parts of the Aztec and 0% for the darkest.
Cheers,
S.O.
Dac - I'm not all that good! For example, my saucer took a long time to do as I tried to overcome C4D's interpolation "problems" for rendering adjacent polygons. I look at others peoples work like your own, Wils or those on Foundation 3D and SciFi Meshes and realize I've got a lot to learn if I'm ever to have my own complete Enterprise.
Wil - I think I know what you mean. If we take the Cloudster images and the movie snapshot I use, you can see very bright patterns in the darker parts of the saucer. They don't look like part of the main material colour and don't coincide with an obvious specular highlight. For example the back half of the saucer on the first Cloudster image (top right-hand corner of post 82). You might want to confirm this.
If this is the pattern you're talking abount, I think I know what it is. My guess is that the brighter areas are due to a fine roughness of the surface as they also correspond to the whiter/less specular parts of the hull. Therefore the light is scattered/diffused more. It's the kind of effect you get if you rub sandpaper over something.
As you add more layers of paint, the surface becomes smoother and consequently the scattering/diffusion decreases. So the aztec pattern is part material colour/part scattering. I need to look at TrekModeler's images to see if this is the case.
In C4D we can mimic this in the Illumination channel, where we can specify the Phong Falloff. A high value gives the surface an almost chalky appearence.
I did play with this, but I haven't any renders done to demonstrate the effect. At one point I overdid it and it looked terrible - top value should be 10% for the brightest parts of the Aztec and 0% for the darkest.
Cheers,
S.O.
Yes S.O. that's the pattern I mean, and indeed it looks on the B&W like it is sanded in, but in fact it is the aztec pattern, it somehow pops up on areas where there is no specularity from the lights, that's why I suspect there are two different finishes on the model, one with big reflection(not specularity) and one with less reflection. I think this because I do not believe they sanded the aztecs in.
And there is a better pic of it, see post 54, the little (gold colored)picture, one part of the aztecs are highly glossy and the other part seems to be "black".
BUT, I can see also breakup ):
So, I had a look again at other B&W pics, the area I mean (as you identified in post 82) has also a little breakup on other pictures of the saucer.
Because of this, I checked again what Olsen said about painting the refit.
There is NO finish, all is made with the paints he bought.
The thing we see are the BASIC aztecs he put first on the model, before he got wild and made all of the little panels on top.
From this I can say, the way you are doing this, is the right way, just keep going, I am confident that in the finished saucer the same effect will appear (mostly coming from the two different diffuse channels).
In other words, just forget what I said, and stick to your plan.
Also, if you do not have it allready, DL the pdf file from Hobbytalk in the first post: http://www.hobbytalk.com/bbs1/showthread.php?t=188494
(need to register)
And there is a better pic of it, see post 54, the little (gold colored)picture, one part of the aztecs are highly glossy and the other part seems to be "black".
BUT, I can see also breakup ):
So, I had a look again at other B&W pics, the area I mean (as you identified in post 82) has also a little breakup on other pictures of the saucer.
Because of this, I checked again what Olsen said about painting the refit.
There is NO finish, all is made with the paints he bought.
The thing we see are the BASIC aztecs he put first on the model, before he got wild and made all of the little panels on top.
From this I can say, the way you are doing this, is the right way, just keep going, I am confident that in the finished saucer the same effect will appear (mostly coming from the two different diffuse channels).
In other words, just forget what I said, and stick to your plan.
Also, if you do not have it allready, DL the pdf file from Hobbytalk in the first post: http://www.hobbytalk.com/bbs1/showthread.php?t=188494
(need to register)
Hi Wil,
Looking at Paul Olsen's site it would appear that the plastic of the model had to be sanded. Also they used a white primer that was coarse and grainy, which required sanding too. The pearls would then be sprayed, which could not be sanded.
So my take on it is that some areas may be white primer with either no or very few pearl coats and therefore still rough. Other areas will have more coats, appear darker and be much smoother.
I'd say that bright colour (non/low specular) parts of the Aztec are due to having fewer coats and a slightly rougher surface. After a certain number of coats the surface would be smooth and applying additional coats would make no difference.
Anyway progress report. I'm just getting a feel for the new materials by trying to emulate one of the TrekModeler images. Result below:
Of course it doesn't account for the environment, but I'm just trying to get a feel for what's required.
Cheers,
S.O.
Looking at Paul Olsen's site it would appear that the plastic of the model had to be sanded. Also they used a white primer that was coarse and grainy, which required sanding too. The pearls would then be sprayed, which could not be sanded.
So my take on it is that some areas may be white primer with either no or very few pearl coats and therefore still rough. Other areas will have more coats, appear darker and be much smoother.
I'd say that bright colour (non/low specular) parts of the Aztec are due to having fewer coats and a slightly rougher surface. After a certain number of coats the surface would be smooth and applying additional coats would make no difference.
Anyway progress report. I'm just getting a feel for the new materials by trying to emulate one of the TrekModeler images. Result below:
Of course it doesn't account for the environment, but I'm just trying to get a feel for what's required.
Cheers,
S.O.
Hi Wil,
I just run off a render demonstrating the impact of Phong Diffuse Falloff.
The above image is from an earlier post. The bottom one uses a falloff value that depends on the colour/specular highlight colour. Phong falloff is 0% for the darkest colour/brightest specular and 10% for the lightest colour/darkest specular. The higher the value, the "rougher" the surface.
This makes the biggest impact between 11 and 3 o'clock with virtually no impact between 6 and 11 o'clock. At 2 o'clock near the sensor dome, you can see how the pattern stands out even though there is very little light and no specular.
Cheers,
S.O.
I just run off a render demonstrating the impact of Phong Diffuse Falloff.
The above image is from an earlier post. The bottom one uses a falloff value that depends on the colour/specular highlight colour. Phong falloff is 0% for the darkest colour/brightest specular and 10% for the lightest colour/darkest specular. The higher the value, the "rougher" the surface.
This makes the biggest impact between 11 and 3 o'clock with virtually no impact between 6 and 11 o'clock. At 2 o'clock near the sensor dome, you can see how the pattern stands out even though there is very little light and no specular.
Cheers,
S.O.
Hi Folks,
Had another break from the project after the frustration of trying to get the pearl finish working. So I'm dropping work on it in favour of some more modelling work.
The pearl finish is a bit like baking a cake. You know what a cake looks like, you know what the key ingredients are. But to make it, you're only given the end product and have to deduce the amount of each ingredient, the temperature of the oven, and how long to leave it in! Get these wrong and a decent cake is last thing you'll have (unless you're really lucky).
In order to help the materials along, I need more HQ hull to test against reference images and movie snapshots. I've started work on the top side of the saucer. Image below...
The five dots in centre of the image are for the front spotlight. I've put them in, so I can work out the shape of deck two so that its shadow doesn't fall too far down the saucer.
Cheers, S.O.
Had another break from the project after the frustration of trying to get the pearl finish working. So I'm dropping work on it in favour of some more modelling work.
The pearl finish is a bit like baking a cake.
You know what a cake looks like, you know what the key ingredients are. But to make it, you're only given the end product and have to deduce the amount of each ingredient, the temperature of the oven, and how long to leave it in! Get these wrong and a decent cake is last thing you'll have (unless you're really lucky).In order to help the materials along, I need more HQ hull to test against reference images and movie snapshots. I've started work on the top side of the saucer. Image below...
The five dots in centre of the image are for the front spotlight. I've put them in, so I can work out the shape of deck two so that its shadow doesn't fall too far down the saucer.
Cheers, S.O.
Last edited:
YARN
Fleet Captain
Hi Folks,
Had another break from the project after the frustration of trying to get the pearl finish working. So I'm dropping further work on it in favour of some more modelling work.
The pearl finish is a bit like baking a cake.
In order to help the materials along, I need a lot more HQ hull so I can test against reference images and movie snapshots. I've started work on the top side of the saucer. Image below...
The five dots in centre of the image are for the front spotlight. I've put them in, so I can work out the shape of deck two so that its shadow does fall too far down the saucer.
Cheers, S.O.
I think you are making magnificent progress.
Best of luck with the pearl finish.
Hi Folks,
I was hoping to have a completed impulse engine this week - but instead it's only half done! It's been an interesting exercise, especially the supposedly simple outer ring.
For ages I lined up my simple shapes, but couldn't get the correct alignment. Then having dug around for images of the impulse engine, I realized the ring is saddle shaped - the front and back dip slightly.
Happy with the result, not so happy with the rate of progress.
Cheers,
S.O.
I was hoping to have a completed impulse engine this week - but instead it's only half done! It's been an interesting exercise, especially the supposedly simple outer ring.
For ages I lined up my simple shapes, but couldn't get the correct alignment. Then having dug around for images of the impulse engine, I realized the ring is saddle shaped - the front and back dip slightly.
Happy with the result, not so happy with the rate of progress.
Cheers,
S.O.
Wow! Really... Wow!
