Hey Guys,
I've just written an intro tutorial on writing an xray shader using VOP's, its up on the techimage website:
http://www.techimage.co.uk/products/houdini/houdini_tutorials.html [techimage.co.uk]
Check it out and let me know what you think.
Cheers
Andre
Full Version: New Tutorial
andre - I went to download the pdf but when acrobat tries to display it, it cant find the font (on windows xp), just rows of blobs!
mark
mark
thanks for the tut andre…
Mark - I just opened it up in Acrobat reader 6 in WinXP and all was good….
Mark - I just opened it up in Acrobat reader 6 in WinXP and all was good….
yup, that'll be the solution!
Good tute. One question. Why did you normalize the N,I vectors out of the global attrib node? Those ops were put down without explanation.
Shannon,
well spotted. I do apologize for that.
Basically N and I are vectors of some length and in order to use them with predictable results we need to normalize them, meaning converting them to unit length. If you use normals of non-unit length you'll get unpredictable results.
I hope this explains it sufficient.
Andre
well spotted. I do apologize for that.
Basically N and I are vectors of some length and in order to use them with predictable results we need to normalize them, meaning converting them to unit length. If you use normals of non-unit length you'll get unpredictable results.
I hope this explains it sufficient.
Andre
As Mario(we call him “Domain-0 to 1-Master” here at Axyz 
) is fond of pointing out, on a sometimes hourly basis 
, the only way you can get predictable results in shaders is to tightly control the domain that your calls work within. It's always best to have as much as possible fall between 0 to 1(i.e. normalized), so you minimize/track down any bizarre results that are blowing out white or coming up solid black. It's just one less thing to struggle over. For instance, you *could* have a noise VOP generate values that go from -23 to +1223, but it's not a smart thing to do - if you wanted to actually have that sort of range, you should get a clean 0-1 range from the generator and then fiddle with it afterwards to get what you want. Much easier to track down incorrect shader design.
When he was porting his rman noise over to VEX/VOPs, after much begging he finally put in a tab that let us optionally skew the output range. The original name for that tab was “Evil”.
Cheers,
J.C.
) is fond of pointing out, on a sometimes hourly basis , the only way you can get predictable results in shaders is to tightly control the domain that your calls work within. It's always best to have as much as possible fall between 0 to 1(i.e. normalized), so you minimize/track down any bizarre results that are blowing out white or coming up solid black. It's just one less thing to struggle over. For instance, you *could* have a noise VOP generate values that go from -23 to +1223, but it's not a smart thing to do - if you wanted to actually have that sort of range, you should get a clean 0-1 range from the generator and then fiddle with it afterwards to get what you want. Much easier to track down incorrect shader design.When he was porting his rman noise over to VEX/VOPs, after much begging he finally put in a tab that let us optionally skew the output range. The original name for that tab was “Evil”.
Cheers,
J.C.
I understand fully now. I'd love to see more vex tutorials. Also, thanks for the math tute link. That cleared up alot for me. :wink:
Thanks for that.
Nice one John.
I have already started writing another one and I hope (time permitting) to write a few VOP tutorials which will increase in difficulty and explain the math to some degree.
I found when I was starting out that there wasn't enough material for beginners or non-programmers/technical people, which explains more in detail why you would use certain functions and what they actually do.
Nice one John.
I have already started writing another one and I hope (time permitting) to write a few VOP tutorials which will increase in difficulty and explain the math to some degree.
I found when I was starting out that there wasn't enough material for beginners or non-programmers/technical people, which explains more in detail why you would use certain functions and what they actually do.
I found there to be a lack of tutorials that address those who are beyond the beginner level, but need explaination for some of the higher level material. Practical examples using those math ops. Your vex tute is just what the doctor ordered.
There are many people out there who have years of 3d experience, but only rudimentary “3d math” and little experience writing shaders. Those who come from the 3d Studio max or lightwave world are used to prefabs where you tweak params and throw on maps to make shaders. Some can make quite sophisticated shaders this way and never learn about the underlying operations. Moving from that world to houdini, vex is quite intimidating.
There are many people out there who have years of 3d experience, but only rudimentary “3d math” and little experience writing shaders. Those who come from the 3d Studio max or lightwave world are used to prefabs where you tweak params and throw on maps to make shaders. Some can make quite sophisticated shaders this way and never learn about the underlying operations. Moving from that world to houdini, vex is quite intimidating.