Oh, I like videos! I'm one of the idiots...lol.

Laurie

Videos are cool... I just have such a hard time remembering everything... :m_think:

I agree that if somebody doesn't know the really low-level mechanics of nodes, then they can't even begin to follow. Basics like you say - that whatever is plugged into a node parameter is multiplied with the value entered in the node parameter. That color math is the same as regular math, but done on each component. That plugging a color into a number parameter averages the three components (or conceptually converts it to gray scale). That plugging a number into a color parameter duplicates the number in all three components.

Beautiful! That's what I'm refering to, Bill! Thank you!

So, what, a chapter in your tome? A wiki somewhere central? What do you reckon?

Quote - these are all great ideas!

i second the notion for a wiki defining nodes.  i've been the the CastlePoser tutorials several times, and i still have no clue why Bias (x, y > 0.5) makes x = black into a grey, but Bias(x, y < 0.5) leaves x = black and x = white untransformed.  what equation does Bias actually apply mathematically?  and Gain? 

What gave you that idea? Bias never transforms 0 and 1. Bias(0, any) = 0. Bias(1, any) = 1.

Bias is actually an isomorph of Pow. (See how I like to use the shorthand possible when everybody learns all the words?)

Bias(a, b) = Pow(a, Log(b) / Log(.5))

Why is this particular isomorph useful? If you think about the graph of the Pow function, the exponent bends a diagonal line up or down. Pow(x, 1) is a straight diagonal line. When the exponent is greater than 1, the line bends down. When the exponent is less than 1, the line bends up. This means that the second argument is a contra, which is hard to think about. (By contra, I mean to make something bigger you must supply something smaller and vice versa. Contra parameters are always head scratchers. Put three or four in a row and your head explodes.) (Note: I hate frequency for this reason. Frequency is a contra. To make a repeating element bigger, you must decrease the frequency. I hate that. But sometimes frequency is not a contra, like when I say that a parameter in the Loom shader script defines how many threads there will be. Then a bigger number is more threads (not contra), and though it is contra to thread size, that's conceptually clear.)

So - one thing we don't like about Pow for the purpose of bending a function up or down is it is contra. The other is we have no idea how much it bends.

But the isomorph defined by Bias is really nice. It is no longer contra. Increase b, and the line bends up. Decrease b, and the line bends down. How much? Well, when "a" is .5 (exactly half way across) the output is b. Whaaaat? Sure it is.

To show why, we need to remember an important identity from algebra:

Log(Pow(a, b)) = b * Log(a)

y = Bias(.5, b)
{ Using definition of Bias }
y = Pow(.5, Log(b) / Log(.5))
{ Take the Log of both sides }
Log(y) = Log(Pow(.5, Log(b) / Log(.5))
{ using identity Log(Pow(a, b) == b* Log(a) }
Log(y) = (Log(b) / Log(.5)) * Log(.5)
{ on the right hand side, re-arrange }
Log(y) = Log(b) * (Log(.5) / Log(.5))
{ P/P = 1, where P is Log(.5) }
Log(y) = Log(b)
{ take anti-log of both sides }
y = b

QED

The Bias function was invented/discovered/popularized by Ken Perlin. Perlin also invented the Perlin Noise function and variants, AKA the Poser Fractal_Sum node, the fBm node, and the Turbulence node. He won an Academy Award in special effects for inventing the noise functions.

http://mrl.nyu.edu/~perlin/

He also invented the Gain function. It is a bit more complicated, but defined in terms of Bias.

gain(g,t)
{
    return if(t<0.5,(.5bias(1.0-(g),2.0(t))),(1.0-.5bias(1.0-(g),2.0-2.0(t))));
}

And I thought algebra was tricky...

;o)

Laurie

so, what BB has just told us is....

www.youtube.com/watch

(corrected for proper point in video)

Well if you look in my shaders, you'll almost always find a Bias or 2 in there.

Whenever I have any sort of gradient, and I don't like the gradient because it doesn't rise fast enough or it rises too fast, I can fix it with Bias.

Bias can take all the gray values and shift them up or down, while keeping black and white unchanged.

For example, suppose I have a grayscale lip mask for lipstick. Suppose the edge of this mask is soft, and I want it to be a little harder. I use Bias(mask, .8) and where the mask is .5 it becomes .8. Darker values are increased more, lighter values are increased less so they don't go over 1.

On the other hand, if I want a gradient to rise less quickly, I'll say something like Bias(x, .2). So then where x = .5 it becomes .2. Dark values get darker but not by a lot. Brighter values get much darker. But white is still white.

The third one is more than just Bias. I will be explaining this in my book. (I was already working on a shader book before Coop started this process. I plan to let Deecey handle the basics for free on the SM site, and you'll have to buy the book to get the intermediate and advanced stuff.)

Robyn and CobaltDream have been reviewing the book, so they already know some of the stuff that are necessary to understand what I'm doing here in the third one. I'm using the "Complement" function.

C(x) = 1 - x

The last one is a function I came up with called IBias:

IBias(a, b) = C(Bias(C(x), 1-b))

It's very handy.

Some thoughts:

Saving poses - What each option in select poses does and when to use them. 

Mat room - explain the beginner nodes (easiest to use and most commonly used ones) how they work and show what a change to a value will do.

Lighting - explain and show how all the basic lights work and show how to produce some simple effects.

Lighting - what HDRI and IBLs are, how, when, and why to use them.

Tips - general basic tips for beginners that everyone should be using.  Then do intermediate and advanced tips.

Setting up external runtimes, when, and why I should use them.
How to set up my poser preferences and what each one does.
Rendering options - show how to set up render options for what you are doing, and include tips to speed up the renders or to make them look better.

Mat room - tell me if a node or an option is beginner, intermediate, advanced, or mostly useless.  Show me how it works and what the changes do. 

NVIATWAS - what this means.

Tips for realism
Tips for commonly used artistic effects (ie using a filter to produce a monocrome image).

And finally - how to make a bloody decent toon render with appropriate outlines.

http://www.castleposer.co.uk/articles/maths_bias.html

as far as i can tell from the tutorial's picture, the result of Bias(U, 0.8) has a transformed black isn't touching the bottom any more. even if i thought i couldn't trust the material room preview, which has a darkest color of 46, 46, 46, i have to believe this image.  unless i'm missing something?

going below 0.5, white doesn't seem to be transformed.  i totally don't get it.

Quote - Detailed but intelligible instructions regarding the Material Room.

Detailed but intelligible instructions regarding the Hair Room.

I will 2nd that suggestion.

Quote - Good instructions that cover bringing a face pic onto a Poser model, something that doesn't look incrediably lame.

Yet another Great suggestion.

Using point lights

Quote - IAre you talking about using the Face Room, or using some other third party application like Photoshop or one of the 3D Painting applications?

Start to Finish:
Take Photos and apply to Poser Figure to Make plausable resemblance.

I would steer clear of PhotoShop and other High Priced Software.
Use something like GIMP or PSP to get the body to match the Face Room Output.

This wil be great, so far everything I want is listed so waiting im...patiently.    8^)

 NVIATWAS - what this means. <<<<<<

ROFL  that is one of the best requests in here, though it won't take a tutorial to answer it.

It SHOULD be added to an FAQ though. 8-)

Information about all the "hidden" features would be good.

• how to start python scripts in the scripts menu via keyboard shortcuts
• how to change existing keyboard shortcuts
• how to change the port library and poser communicate (or just explain all the Poser.ini parameters)

Great thread Denise!

Yup, that it is, getting some great feedback.

I suspect I'll be busy for a while. 8-)

Poser users are visual, so tutorials should focus on effects and how to evolve them.

For example, a tutorial on creating a stone shader needs to show the following types of steps:

• start with an object and add a stone type color
• add bump and displacement and now it looks a little more like a rock
• add a granite or other node to give it more realism
• tweak the granite settings to show how it adds or detracts from that realism
• maybe add a second color to vary the material
• tweak those settings

Artists aren't interested in the details involved in the math. They're interested in how changes affect their image. It's more work and makes it harder to cover the larger ground that's needed, but it's what artists need.

An example of what not to show in a tutorial would be the BB link
www.castleposer.co.uk/articles/maths_chart.html
This is excellent information for somebody that's getting deep into node math, but the average artist only wants to know how it can be used to enhance an existing material node effect being created.

A tutorial on the render settings, and which should be changed for an optimal image, would be useful. I remember having trouble with atmospherics in an image and it wasn't very straightforward how to achieve the best results.

Agree with the visual approach, and that's how I usually do it

actually, that link is only example of what not to show if people never hope to actually understand why any of those steps to making a stone shader  make sense.  i could show you several examples of tutorials for specific shaders, and Acadia's actually catalogued most of them.  over the years, i've seen at least 4 different step-by-step skin shader threads.  but people keep asking for the same sets of tutorials because they don't understand how any of those nodes work or how it made sense to put them together.

no, don't start with math.  but a visual representation of what you're doing when you, for instance, use sine to make something repeat, and how other nodes work when you use them to control the frequency and amplitude of repetition, is really important.  it's important to understand exactly how each one of the math nodes works and responds to variable input when you use it in a shader.  as it is, most people don't even understand how to reasonably tweak any complicated shader they inherit, because they don't understand what the nodes do, what's valid input and what isn't.  as a result they might be able to duplicate a shader in a tutorial, but rarely can they do more with it than change colors.  i think leaving everything a black box for everyone and just doing the recipe tutorials is exactly why there's so many requests for the same materials over and over and over again.  and even from the same people.

since people are in this thread asking for more than what already exists, and asking how to use the material room in general, i think it's much better to present a stone tutorial with graphs like the ones at CastlePoser every time you involve a math node than to pretend you don't need to understand the nodes to use them well. 

consider more than the simple shader.

let's say your stone shader has a  whole set of nodes creating a "texture" controlled by  the input of a "roughness" parameter and outputting into bump, displacement and specular, whole set of nodes creating color variation controlled by a "variation" parameter, and another whole set of nodes controlling veins taking a set of parameters determining vein width, vein color and vein frequency.  how do these vary according to their parameters?  it's much better to show with a graph how those sets behave according to their parameters than just show each piece of the shader, and show some examples of different settings.  doing all three is best.

Just so you understand, a visual approach is only ok to show somebody how to adjust a shader that has been spoon fed to them.

When I was learning Poser in my first couple years, I asked many people for help on a good leather shader. Leather is something everybody is intimately familiar with. We all know what it looks like. But no Poser "artist" was able to point me to any leather shaders except those that were part of products, that were drawn.  Nobody was able to make a convincing procedural leather.

Now, to me, seeing examples of how to draw leather is useless. I can't/won't draw, because I suck at drawing, and because after I'm done I have a drawing, basically one shader, not 1000 variations on it.

Slowly I developed an understanding of the math used to manipulate Perlin Noise, and was able to connect the dots and make a convincing procedural leather shader that I can instantly apply to all sorts of things, in infinite variations of tone, color variation and patterns, roughness, shine, etc. And I mean instantly.

Now if you want to know how this leather shader works, to the point that you could make one or something similar (for example, a real granite shader, which has a lot in common with leather), then you have to learn the math. I don't mean calculus. I mean basic algebra. Like how Bias works. If you can't look at a bump map and understand how to manipulate it with a couple nodes, then you're going to have to draw the bitmap over again, which is not shader work, but texture work.

You want a book on shaders, there will be math. You want a book on textures, there will be drawing.

I don't think anyone here is saying "don't give us math." But the point is: when you give us math, show us concretely what it's doing. Give math in a way that assumes we don't "get" math in the abstract.

Quote - Give math in a way that assumes we don't "get" math in the abstract.

Bingo.

There has to be a balance between both.  And it has to be presented in laymen's terms that can be understandable to people of varying experience levels.

I'll get it, doncha worry. 8-)

but to "get' math in the concrete you absolutely have to understand how it a node or group of nodes responds to various inputs.  giving only a few samples (ex: here's stone with a roughness of 0.25, a variation of 0.5, and a vein of width = 0.1, color = white, density = 0.5) gives those who don't understand math less of an understanding of how things work, not more.  someone who's great at it will look at the nodes and know pretty much what's happening.  but those who are weakest at math need to see the graphs to get the whole continuum.  they need to see where it peaks, becomes 0, or

and gagnonrich specifically said, "don't present the graphs." 

You always have folks who can't understand something as it's explained by the current curriculum. For those cases, another way must be found to explain it in a way they do understand ;o). It's something teachers have been up against for years.

BB, I love that the pathways of your mind get right to the superhighways. People like me though, spend my time meandering along backroads, getting lost, getting carjacked... Math is like that for me. I can add, subtract, multiply and divide. It gets me through life. But long algebra is not one of my strong suits...lol. Even if I DO want to understand why something does what it does, it doesn't help me to understand the math because math just doesn't make sense to me. If you could put it another way, maybe give a real world example, that would help a lot more ;o). It's not that we don't want to understand, it's that a lot of us just can't. That doesn't make us stupid or unwilling to learn, but it does make us harder to reach...lol.

I've give you an example: I make chainmaille in my spare time (like I actually have any) and there are things you must know when you cut your own rings like I do - such as having to figure out aspect ratio. Your rings must be a certain size for the amount of rings that will be going thru them at a certain gauge. I did manage to get that one. But as for the different weaves, there was one that was incredibly tricky for me. I read pattern after pattern and just could not get it. I did see a video of it being done and BAM! - I discovered how simple it really was. So, some can get away with just being told how, but some have to see how ;o). I'm one of the "seers", and I'm sure I'm not alone...lol.

Anyway, just didn't want you to think that your efforts weren't appreciated or were wasted. They're not. They're just not well understood by some of us, even those of us that do want to know "WHY does it do that?" ;o).

Laurie

If I can make a suggestion: perhaps like what you did here with Bias, Bill:
market.renderosity.com/mod/forumpro/showthread.php

I keep refering back to those sorts of things. We could have a section on Bias... or we could have a section on how to shift colour values.

For example, I've been wanting to shift colour values for a skyDome image: I want to be able to incrementally darken the image within the shader without loss of image quality. It plugs into the Ambient_color channel on PoserSurface. Dusk is approaching, so the stormy sky needs to be darkened.

I get the impression from my renders that simply lowering the Ambient_Value number degrades the image quality, which i don't want.

I have no idea whether my conclusions are even correct... this is just my impression.

But anyway, I see an application of the Bias node, here, potentially.

Quote - You always have folks who can't understand something as it's explained by the current curriculum. For those cases, another way must be found to explain it in a way they do understand ;o). It's something teachers have been up against for years.

Bingo!

I have been told by MANY publishers that the target educational level for computer books is generally geared for a high-school aged audience. This conversation is actually a GOOD example of why the materials haven't been documented extensively up to this point. With all of the various node types, all of the various settings and examples that have to be done for them, and yes, even getting into what math nodes do (and don't even get into the dPdu ones yet! LOL), you start to see the enormity of the task. And given that publishers usually don't like to make a lot of books that are more than 500-700 pages (due to costs involved in production and printing), this explains why it gets only basic coverage.

I am watching, listening, and evaluating the best approach as I read the comments, so keep them coming. I know whatever is decided might not initially make everyone happy (that will be an impossible task), but it can only get better over time.  

8-)

Quote - but to "get' math in the concrete you absolutely have to understand how it a node or group of nodes responds to various inputs.  giving only a few samples (ex: here's stone with a roughness of 0.25, a variation of 0.5, and a vein of width = 0.1, color = white, density = 0.5) gives those who don't understand math less of an understanding of how things work, not more.  someone who's great at it will look at the nodes and know pretty much what's happening.  but those who are weakest at math need to see the graphs to get the whole continuum.  they need to see where it peaks, becomes 0, or

and gagnonrich specifically said, "don't present the graphs." 

And gagnonrich was right. You don't present a graph in a tutorial. It isn't a matter of no principial information vs a graph. That's why I keep saying: teach us to fish, don't just give us one. It's a matter of always presenting the principial information in a concrete context. A graph is fine for an appendix or an endnote as further info; that's great.

no graph means you don't know how to fish.  it's the fishing line and tackle.  unless you're enough of a math wiz to have a graphing calculator or see how a node works in your head.  if you're as bad at math as everyone says they are, you cannot understand what's going on in any shader with most math nodes or anything variable without a visual of what it's doing.  no matter how it's presented.  there simply is no english that better describes a power curve than actually seeing it.

"You always have folks who can't understand something as it's explained by the current curriculum." look at the forums.  the current curriculum, other than some of bagginsbill's posts, is ignore the math and just plug stuff in randomly and see what gives the right result in certain lighting.  we already have what people are asking for here.  go thorugh Acadia's thread.

It's a bit like wanting to write but not wanting anything to do with grammar.

The whole purpose of this exercise is to provide information from Those-That-Know to those artists who have been searching in vain for information about nodes and what they do. Are we going to selectively exclude information because we can't readily assimilate it ourselves, and use that inability as a barometer of how useful or relevant this information is going to be to artists in general?

This strong differences in philosophies to shader development is what has prompted me to suggest the wiki for the more technical (maths and such) side to nodes. As yet, no one (besides CobaltDream) has made any reference to this suggestion. Is it a rejected notion? And if so, why?

Quote - This strong differences in philosophies to shader development is what has prompted me to suggest the wiki for the more technical (maths and such) side to nodes. As yet, no one (besides CobaltDream) has made any reference to this suggestion. Is it a rejected notion? And if so, why?

Not ignoring that suggestion, I am just not in a position to decide whether or not it can be done, as setting up a wiki on SM's site is beyond my authority.

Oh ... I also want to mention, that this discussion is about tutorials in general that are needed for Poser, at all skill levels.  Though I want to see continued discussion on how to handle materials (and yes, I AM reading with an open mind as I'm trying to formulate my approach on them), I would also like some discussion on other tutorials that will be needed as well. 

Keep it coming. 8-)

A compendium of valid shaders (which is what my perception is of the direction of this project) has validity in the same sense as DS presets. People who just want to have a material shader for turbulent water can get that shader for that sort of water.
Those who want to understand why it works the way it does - yes, there are a few clock-builders among us artists - should have access to that information.

The reason I suggest a wiki for that set of information is that it is constantly update-able by an endorsed set of experts in the field, whereas your web-pages are in the hands of a few, and once published, are pretty-much set in stone.
Setting up a wiki is no big deal: I've set one up myself. Shouldn't be a big challenge to SM: they can use the Daz model as an example, perhaps.

Quote - no graph means you don't know how to fish.  it's the fishing line and tackle.  unless you're enough of a math wiz to have a graphing calculator or see how a node works in your head.

I don't buy this. A graph is not a fishing lesson. It isn't fishing line and tackle either. It's more like a reference chart to tell you what line supports what weight. That's handy, but it doesn't do you a bit of good unless you know a lot more basic stuff (what fish are in your location, how big they are etc). A graph is a reference (that's why I said it's great to have in e.g. an endnote). And it will be helpful to do refining after general uses are understood. Unless those general uses are understood, nobody but nobody is gonna care about the graph. It's the more basic stuff that will suck people into the graphs as they need them.

IMO.

Deecey, I think the big thing is to work with stuff that isn't very well-supported already. (Although some things that are "supported" by threads etc are not supported in way that the average user will ever have that support at his or her fingertips.) That's why I begged regarding the hair room. The tutorials I've seen are few in number and not all that in depth at all.

Beginners :
A tutorial on how to use the Poser manual and included tutorials.

harsh , but ...........

Explain and how to use:

Normal Mapping
Tone Mapping

Too much read.

How about make the tutorials in a form of videos!!!

or better yet both!

Hmmm... I'll mention it again.... RENDER NETWORKING...

A good tutorial showing the end user how to set up a render farm and how to use the render queue manager. Yes, I've seen the video on SM's site for a VERY basic look at the render queue.
What would be nice is a complete guide on how to connect the machines together, how to set up the IP protocols, what kind of hardware works the best and what the minimum requirements are for hardware that will benefit the user on a small budget.

I'm actually surprised that I haven't seen more interest in render networking around here, considering that we already have one version of poser that is capable of it.

I am about 40% through the MS Poser 7 book.   I am retired with a 100% disabled wife to care for. I am just now finding time to study Poser.  I would love examples of how you should do things.  A walk through.  I am having problem with what to make the Parent of things.  I guess that I am just too much of a newbie at this point.   If I buy a package and it comes with a pair of shoes, how do I separate them so I can make each foot the parent on each shoe?  I could probably post to the Poser Newsgroup and somebody would feel sorry for me and offer to help. But, I am trying to do this on my own. I'm not that dumb, I hope, so I am keeping the Newsgroup in reserve.  I exchanged an email with an old Poser user and he suggest that the best way to learn is just play with everything.  That is fun and I am learning but the book is so much better if it just had more details.

If this is not clear, I want somebody to hold my hand and walk be from start to end with some samples in each chapter.  I am just now starting to understand some of the very basic elements. 

In sumnary - more details. especially in the beginning.

Hal

janhal - welcome and don't hesitate to ask questions in this forum. There are thousands of people lurking and for each one who comes forward like you and speaks his question, 999 benefit. We love to answer questions here. Don't worry if your question is a repeat, we'll just point you to the thread where it was already answered. We understand that the search engine is of limited use, and those of us hanging out here a lot know where the good threads are.

Quote - ...those of us hanging out here a lot know where the good threads are.

Yep: 9 times out of 10, it's one of your threads...lmao.

Laurie

I can do it now - but just getting a character to sit in the chair was hard for me to learn.  I sincerely wish that somebody would have told me/taught me how to move the joints so the character actually looked like they are sitting in the chair.  I am still having problems with some of it - more on that later if I don't get it figured out today or tomorrow.

I have set aside two hours a day to work with Poser 7.  I have just about got to the point that I am not going to buy anything else until I have a fairly good working knowledge.  And there are a lot of things that I see for sale both here and on Daz, DNA etc., 

I guess there is just no easy way of doing things and I just need to take more time and learn.  There is no 'sit in chair' pose, at least I don't have one. 

I made a mistake with my first post - so let me correct it  - I am working my way through Poser 7 reveled by e frontier, not by Micro Smith.  I am doing it one page at a time and not jumping ahead but at this point, I have no idea how to use magnets or morphs.  Have not got to the material room yet.  As I said, it took me almost ten days to get used to using the dials to get my character to sit on a bench and look natural.  Who would have thought that you had to move the hips back in the z zone to get the butt on the bench.  Trust me, I tried just using the x and y for two of three days.      

Maybe this is where I should have started.  Been following the newsgroup for years.  I am getting better each day but being able to ask dumb newbie questions and get an answer would have saved a lot of time.  Somebody needs to write the book - "Poser for Dummies" Just simple poses are not that simple to somebody who has never used Poser.   Maybe the kids in school today have no problem with x,y and z but at age 69 it took  a while to understand it. It is still a problem with some poses.  Can not wait until I get into reflections - I think they are so neat. 

Bottom line, I think I am wanting to do way more then I should be doing at this time.  I continue to marvel at some of the pictures that are posted to the newsgroup - I keep wanting to say, "How did you do that".  Rina seems willing to help anybody, but unless I really get stuck or finish the book - and the second one that is waiting  - which is using Poser and Photoshop, I will keep my questions here.   Trust me that exchange on CG 2.2 zoomed over my head but I did save some of it to a text file for later use.  Todays project is turning my character 180 degrees without moving the camera.  Again, can not find the 'turn around" icon to click on so, it is back to the dials - because I know it can be done.  Do I just move the feet and hope the body follows?  At least when I get back down stairs, that is where I am going to start. 

TIA - Hal

janhal,

We obviously take a lot for granted. Never occured to me that simply rotating the figure (not the camera) could be unclear.

Start a new thread. Call it Total Beginner Poser Room Info.

By the way, the answer is select "Body" and use yRotate.

 Again, can not find the 'turn around" icon to click on so, it is back to the dials - because I know it can be done. <<

First, turn off inverse kinematics on both of the feet. Otherwise, they will "stick" to the floor. To do this, choose Figure > Use Inverse Kinematics, and uncheck both feet.

Secondly, select the BODY actor of the figure (not the clothing).

Then go into the Parameters palette, and set the yRotate dial to 180.

Quote - janhal,

We obviously take a lot for granted. Never occured to me that simply rotating the figure (not the camera) could be unclear.

Start a new thread. Call it Total Beginner Poser Room Info.

By the way, the answer is select "Body" and use yRotate.

Hello BB
I know for me, after almost 40 years At Sea, it generally was a faster process to assume nothing.
I found it most often took far less time to explain something If I started from Ground Zero.

I frequently encounter this with my grandson and grandaughter.

Cheers
DR