Patch 6 to Fractint version 20.4
Patch 6 to Fractint version 20.4 is now available on the Fractint Developer's web sites: www.fractint.org or www.fractint.net This is unrelated to Rich's port. What's new: Added Per Image Settings to the formula parser. When initially selected from the fractal type formula screen, the symmetry and any per image settings are enforced. The per image settings will be saved to PARs and GIFs, and can be changed from the appropriate screen. The possible values for per image settings are the same as the parameters that appear in PAR files, with the exception of 'params='. Params can't be set in the per image settings section, because they get set to zero when the formula is parsed. The formula syntax looks like this: mandel(symmetry)[per image settings] { per pixel settings: iteration bailout } For example: IslandOfChaos(XAXIS_NOPARM)[float=y function=sqr/sin/cosxx] {; Jonathan Osuch ; Generalized by Tobey J. E. Reed [76437,375] ; Try p1=0, p2=4, fn1=sqr, fn2=sin, fn3=cosxx ; Note: use floating point z = p1, x = 1: IF (x < 10) z=fn1(z) + pixel ELSE z=fn2(z) / fn3(z) + pixel ENDIF x = x+1, |z| <= p2 } Jonathan
Jonathan, Thanks adding the Per Image parameter Settings to the formula parser! Also, thanks for allowing parameters to be set in the formula file. A question: I have never known what the axis symmetry named: (XAXIS_NOPARM) [in your example formula] actually does. It is listed in the 20.0 docs but no explanation is given. I also don't understand right off what: XAXIS_NOREAL and XAXIS_NOIMAG do. I found the following description of ORIGIN symmetry at: http://www.math.hmc.edu/calculus/tutorials/transformations/symmetry_origin.p df f(?x) = ?f(x) for all x in the domain. Here is an intuitive way to view this symmetry: If you start at a point on the curve, draw a line segment through that point and the origin, and extend it an equal distance past the origin, you arrive at another point on the curve. I can see how that could be extended to: pixel(-x,y) = pixel(x,-y) Thanks for any help on XAXIS_NOPARM, XAXIS_NOREAL and XAXIS_NOIMAG. (Are these described in any later unreleased developer docs?) [I know I should create a parm file with an asymmetrical fractal in it and apply all the symmetries to see what they look like, but I was hoping that there was an explanation of these already typed in somewhere. I am also in the process of moving from a failing Win98 machine to an XP machine and am temporarily up to my ears in alligators. ;o] Perhaps there should be descriptions in the docs of what the more obscure symmetries do, since some of them are non-intuitive -- at least to me... - Hal Lane ######################### # hallane@earthlink.net <mailto:hallane@earthlink.net> # #########################
Hal Asked:
A question: I have never known what the axis symmetry named: (XAXIS_NOPARM) [in your example formula] actually does. It is listed in the 20.0 docs but no explanation is given.
Jonathan is in Colorado. He'll be back soon. I am actually going to see him soon because for the first time in years I am driving not flying to Minnesota from Texas. I will see Jonathan in Iowa either coming or going on my way to or from my mother's memorial service. Suppose you have a fractal that is symmetrical with respect to the x axis if the paramaters are zero but not otherwise. Then XAXIS_NPPARM would give you what you want. If you enter a parameter, exploitation of symmetry is turned off.
I also don't understand right off what: XAXIS_NOREAL and XAXIS_NOIMAG do.
Same idea - I presume the NOREAL means the real part of the first parameter, I'd have to look to see but I'm lazy :-)
Perhaps there should be descriptions in the docs of what the more obscure symmetries do, since some of them are non-intuitive -- at least to me...
Once you figure it out write it up and we'll put it in the docs :-) Tim
On Mon, 2007-06-04 at 15:23 -0400, Hal Lane wrote:
A question: I have never known what the axis symmetry named: (XAXIS_NOPARM) [in your example formula] actually does. It is listed in the 20.0 docs but no explanation is given.
As Tim indicated, (XAXIS_NOPARAM) means there is x-axis symmetry as long as the parameters are zero (p1=(0,0) and p2=(0,0)). In this case it no longer makes much sense since the generalization implies there will always be a p2 value other than (0,0). You need to have a bailout other than zero to produce an image. My original formula used "|z| <= 4" as the bailout criteria. Also, when the functions are changed, symmetry goes out the window. I believe my original formula had the three functions hard coded. For the _NOPARAM variants, p1 through p5 are checked to make sure they are all equal to (0,0). If they are all zero, the symmetry is applied.
I also don't understand right off what: XAXIS_NOREAL and XAXIS_NOIMAG do.
For the _NOREAL variant, p1 through p5 are checked to make sure that the real part is equal to zero (0,?). If the real part of p1 through p5 is zero, the symmetry is applied. For the _NOIMAG variant, p1 through p5 are checked to make sure that the imaginary part is equal to zero (?,0). If the imaginary part of p1 through p5 is zero, the symmetry is applied.
(Are these described in any later unreleased developer docs?)
You have the latest developer's version.
Perhaps there should be descriptions in the docs of what the more obscure symmetries do, since some of them are non-intuitive -- at least to me...
Other than the above, I'm not sure what I can add. Jonathan
Jonathan, Thanks for your explanations about the axis symmetry parameters. As Tim suggested I will try to put your and his explanations into a form suitable for inclusion in the documentation. - Hal Lane ######################### # hallane@earthlink.net <mailto:hallane@earthlink.net> # #########################
-----Original Message----- From: fractdev-bounces+hallane=earthlink.net@mailman.xmission.com [mailto:fractdev-bounces+hallane=earthlink.net@mailman.xmission.com]On Behalf Of Jonathan Osuch Sent: Saturday, June 09, 2007 7:42 PM To: Fractint developer's list Subject: Re: [Fractdev] <Jonathan>What do the axis symmetries XAXIS_NOPARM& XAXIS_NOREAL do?
On Mon, 2007-06-04 at 15:23 -0400, Hal Lane wrote:
A question: I have never known what the axis symmetry named: (XAXIS_NOPARM) [in your example formula] actually does. It is listed in the 20.0 docs but no explanation is given.
As Tim indicated, (XAXIS_NOPARAM) means there is x-axis symmetry as long as the parameters are zero (p1=(0,0) and p2=(0,0)).
In this case it no longer makes much sense since the generalization implies there will always be a p2 value other than (0,0). You need to have a bailout other than zero to produce an image. My original formula used "|z| <= 4" as the bailout criteria.
Also, when the functions are changed, symmetry goes out the window. I believe my original formula had the three functions hard coded.
For the _NOPARAM variants, p1 through p5 are checked to make sure they are all equal to (0,0). If they are all zero, the symmetry is applied.
I also don't understand right off what: XAXIS_NOREAL and XAXIS_NOIMAG do.
For the _NOREAL variant, p1 through p5 are checked to make sure that the real part is equal to zero (0,?). If the real part of p1 through p5 is zero, the symmetry is applied.
For the _NOIMAG variant, p1 through p5 are checked to make sure that the imaginary part is equal to zero (?,0). If the imaginary part of p1 through p5 is zero, the symmetry is applied.
(Are these described in any later unreleased developer docs?)
You have the latest developer's version.
Perhaps there should be descriptions in the docs of what the more obscure symmetries do, since some of them are non-intuitive -- at least to me...
Other than the above, I'm not sure what I can add.
Jonathan
_______________________________________________ Fractdev mailing list Fractdev@mailman.xmission.com http://mailman.xmission.com/cgi-bin/mailman/listinfo/fractdev
participants (3)
-
Hal Lane -
Jonathan Osuch -
Tim Wegner