Beta testers: Simple test of new functions

[chkman]

Hey config writers,

I was doing a little test of the new functions in the 3.1 beta 6 release. There's a little bug with < and >, they're reversed. Andy will fix next beta.
For now here's what I got for a colormap:
TestGT.txt

C0="max(cos(t*.5),cos(t*.3))"
R="((float)(i < .5))*i",
G="i*( (float)(c0 < .4 && c0 > .79) )"
B="( (float) ( (i < .25) && (i > .85) ) )* (i + ((float) (sin(t) < .29) && (sin(t) > .5)) )"

Simple but it works. The new functions are documented at
http://www.soundspectrum.com/g-force/Do ... mming.html

Have fun!

Edited the face out of the math....

[JayPro]

I had a feeling this was why a lot of the waveshapes and flowfields in my collection weren't working in the beta.

Now I'm even more (edit) intrigued than before. I'd like the ones that I know aren't working to get back to doing so; but how?

I suppose I'll hafta run everything thru the beta and see on a case by case basis???

I'm terrible at math as y'all know already, and I'm just afraid that the pithy explanations rendered in the new funcs aren't enough.

PS: ***wyvill??????***

PPS: Please don't get me wrong. These new addons are interesting and will (it'd seem to me) certainly expand GF's versatility to the power of it-boggles-the-mind-how-much; but again, someone of my math-challenged abilitles may hafta see some documentation/commentary on how, if you will, function converts to form, i.e., what I see.

[chkman]

Yeah, the new beta is flagging errors in the configs. I had some with unmatched parens, an operator hanging out there, a ",pi" in the wrong spot, misspelled functions, and the new G is announcing the errors.

I had one flow that looked like the face of E.T. but now that I "corrected" it, it's got extra curves in the eyes and ruins the effect. I don't know math that well either so I don't know if I can redo it right.

Having the error messages makes it easier to experiment.

[chkman]

Not sure about the paren it worked two different ways right after the 1 and at the end:

Aspc=0
D0 = "(float)(((int)1)>>(int)(R))"
D1 = "(float)(((int)1>>(int)(theta)&(int)(127)))"

srcR="r-D0"
srcT="T-d1"
Vers=310

Now, if it could do automatic data type conversions like a high level language like Basic.

BTW, wyvill is documented but not available till next beta.

[JayPro]

I'm still at a loss (but I'll try figuring it out later); so from the looks of it the capital T is now the alternate designation for theta? as opposed to the lowercase used for WaveShapes?

[chkman]

Whoops that's an error!

The compiler didn't catch it. I put theta in and it looks way different. Must have been compiled as 0 - D1.

Thanks, pretty cool now.