home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
ftp.xmission.com
/
2014.06.ftp.xmission.com.tar
/
ftp.xmission.com
/
pub
/
lists
/
fractint
/
archive
/
v01.n311
< prev
next >
Wrap
Internet Message Format
|
1998-10-08
|
42KB
From: owner-fractint-digest@lists.xmission.com (fractint-digest)
To: fractint-digest@lists.xmission.com
Subject: fractint-digest V1 #311
Reply-To: fractint-digest
Sender: owner-fractint-digest@lists.xmission.com
Errors-To: owner-fractint-digest@lists.xmission.com
Precedence: bulk
fractint-digest Friday, October 9 1998 Volume 01 : Number 311
----------------------------------------------------------------------
Date: Fri, 9 Oct 1998 10:46:01 -0700 (PDT)
From: Ken Childress <kchildre@uccs.jpl.nasa.gov>
Subject: Re: (fractint) Re: [fractal-art] cost of images
Kerry,
> Ken,
>
> I agree with you in principle, but I think that you can still make good
> enlargements from monitor shots, depending on your equipment, image, and
> standards. Current 17" monitors deliver 1600 x 1200 resolutions, and a
> good video card and shoot that in true color, elminating any jpeg
> artifacts.
I don't mean to say that this isn't a method worth consideration. It
just isn't worth the effort for me because I can get better results for
less effort. If you like the results, by all means use it.
> Also, actual dpi numbers lose relevance when you're talking
> about putting a picture on the wall where people generally don't get their
> noses into it. Of course, you're right about making big and/or detailed
> enlargements--you just can't do that from a monitor shot.
This is true. Which is why religious arguments about Epson vs. HP vs.
Canon printers are a waste of time. Bigger enlargements can suffer less
DPI since you are viewing them from farther away. However, you can't
take a 1600x1200 pixel image and get a very good 30" by 40" print from
it, IMO.
> To print from transparencies like this, you should find a lab that does
> Ilfochrome processing (used to be called, "Cibachrome"). It's a process
> that prints directly from the trans, without an internegative. There is a
> bunch of other reasons why it's good (like color saturation and archival
> life), but the main one is the "wow" factor. Expensive, but definitely
> worth it.
I agree completely here. If anyone hasn't seen an Ilfochrome print, you
should make an effort to see the results. They can be absolutely
stunning. With the brilliant colors in so many of the fractal images,
an Ilfochrome print would be dazzling. Definitely worth it, plus the
image will last for a long, long time.
Ken...
- --------------------------------------------------------------
Thanks for using Fractint, The Fractals and Fractint Discussion List
Post Message: fractint@lists.xmission.com
Get Commands: majordomo@lists.xmission.com "help"
Administrator: twegner@phoenix.net
Unsubscribe: majordomo@lists.xmission.com "unsubscribe fractint"
------------------------------
Date: Fri, 9 Oct 1998 18:49:42 +0100
From: "Les St Clair" <les_stclair@crosstrees.prestel.co.uk>
Subject: (fractint) September par collection
Hi everyone,
A little later than usual, but the September par collection is now available.
As usual there are two formats:
1. Just the pars - http://ourworld.compuserve.com/homepages/Les_StClair/
2. The pars with the text of the original messages -
http://www.homeusers.prestel.co.uk/crosstrees/fractasi.htm
Now for a shameless plug...
For some really nice fractal animations (Java applets) check out my new page at
Prestel
http://www.homeusers.prestel.co.uk/crosstrees/cube.htm
(I first saw these on Sharon Webb's site and fell in love with them - thanks
Sharon!).
cheers, Les
- --------------------------------------------------------------
Thanks for using Fractint, The Fractals and Fractint Discussion List
Post Message: fractint@lists.xmission.com
Get Commands: majordomo@lists.xmission.com "help"
Administrator: twegner@phoenix.net
Unsubscribe: majordomo@lists.xmission.com "unsubscribe fractint"
------------------------------
Date: Fri, 9 Oct 1998 11:52:11 -0700 (PDT)
From: Ken Childress <kchildre@uccs.jpl.nasa.gov>
Subject: Re: (fractint) September par collection
>
> Hi everyone,
>
> A little later than usual, but the September par collection is now available.
>
> As usual there are two formats:
> 1. Just the pars - http://ourworld.compuserve.com/homepages/Les_StClair/
WOW!!! I have arrived. :-)
I made one of the thumbnail images on Les's page.
Thanks Les, I'm truly honored.
Ken...
(I'm not worthy. :-) )
- --------------------------------------------------------------
Thanks for using Fractint, The Fractals and Fractint Discussion List
Post Message: fractint@lists.xmission.com
Get Commands: majordomo@lists.xmission.com "help"
Administrator: twegner@phoenix.net
Unsubscribe: majordomo@lists.xmission.com "unsubscribe fractint"
------------------------------
Date: Fri, 9 Oct 1998 21:00:52 +0100
From: "Kim Bach Petersen" <kimb@post8.tele.dk>
Subject: Sv: (fractint) Re: [fractal-art] cost of images
Hi Steve - and all of you other great people outthere,
>I would LOVE....I say again LOVE...to hear a detailes description of the
whole
>process....starting with "This is the size I render to print a ??X??
photo."
I've been doing some 15"x10" digital prints (A3 paper). Many a photo shop
provides such prints, usually made on a digital color photocopying machine,
that can print as well. The machine my local printshop uses, prints at 400
dpi allowing lots of detail. The colors are vivid and clear, I'm very
satisfied with the results. Still, intense colors seems to do best, while
more pale or dusty colors have a lesser "wow-factor".
A print cost some 5 $ (30 danish kroner), but if you want multible prints of
the same picture, the price can be reduced greatly, even halfed. The method
is easy, all you need is fractint, a disk and maybe a photoeditor, and you
can make impressive posters.
In need of a portable harddisk or the like, I usually stick to 100 dpi
prints, since it allows me to have the file on a standard floppy disk. I
render a picture on 4500x3000 or 6000x4000 and then reduces it to 1500x1000
to anti-alias it. Then I use the photoeditor to specify the wanted size
(15"x10") and resolution (100 dpi), and the saves the ready-to-print file as
jpeg (with lowest compression, unlike gif jpeg allows the specification of
parameters such as size and resolution). Some photoeditors allows all this
to be done in one process. Notice that I use a resolution that is dividable
with the printers resolution (100 = 400 / 4). If I did not, it would
generate aliasing while printing - and it's nicer to avoid.
Kind regards, Kim.
- --------------------------------------------------------------
Thanks for using Fractint, The Fractals and Fractint Discussion List
Post Message: fractint@lists.xmission.com
Get Commands: majordomo@lists.xmission.com "help"
Administrator: twegner@phoenix.net
Unsubscribe: majordomo@lists.xmission.com "unsubscribe fractint"
------------------------------
Date: Fri, 9 Oct 1998 13:19:48 -0700 (MST)
From: Kerry Mitchell <lkmitch@primenet.com>
Subject: (fractint) 5 point star coloring methods
Fresh from the, "Just because you can doesn't mean that you should" files,
:-) here are some formulas built around 5 point stars
- -------------------------------------------------------------------------------
Kerry Mitchell
lkmitch@primenet.com www.primenet.com/~lkmitch/
- -------------------------------------------------------------------------------
comment { ; narrative copyright Kerry Mitchell 09oct98
Seeing Stars
Most fractal images involve circles in some respect: either stopping
the iteration when the orbit moves outside of a given circle, or
coloring by how close the orbit comes to a certain circle, or some
variation. These formulas use a 5 point star instead of a circle.
The "starnear" formulas color by how close the orbit comes to a star.
The "starbail" formulas are set up to stop iterating either when the
orbit comes into a star or when it leaves a star.
The star is represented by the 5 outer points. These are equally spaced
on a circle. The center and size of the circle are user-specified, as
is the rotation angle of the star.
How z at any iteration relates to the star (inside, outside, how close)
is determined by looking at each of the 10 sides, one at a time. Each
side can be represented by a line A*x + B*y + C = 0, where A, B, and C
come from the coordinates of the 2 outer points that are joined to make
the side. Given the numbers A, B, and C, the quantity
q = A*real(z) + B*imag(z) + C
is computed. If q is positive, then z is on one side of the line, and
if q is negative, then z is on the other side of the line. If |q| is
very small, that means that z is very close to that side of the star.
Taking the signs of q for all 10 sides will determine if z is inside or
outside of the star. Using the smallest value of |q| gives the distance
from z to the star.
To see this in action, use the formula "star". It will draw the outline
of a star, with the position, orientation and size that you choose.
This formula, and the "starnear" formulas that are built on it, use a
logarithmic transfer function from distance to angle (for decomposition
coloring in Fractint). This helps highlight the star without needing to
find out if the point is exactly on the star or not. The "star-inout"
formula will draw a star in 2 colors. Everything inside the star is one
color, and everything outside of the star is the other color. These 2
formulas are simply to give an idea of how the method works.
The fractal formulas need a bit of extra explanation. In the
"starnear_jul" formula, you have the option of whether or not to check
the pixel (initial value of z) for its distance from the star. If the
star is centered at the origin (p2=0/0), this can make quite a difference.
For example, use the "starnear2" parameter set. As written, the real
part of p3 is -0.28. The magnitude of real(p3) is the radius of the circle
containing the 5 outer points. If real(p3) is negative, that tells the
formula to check the pixel for its distance before iterating. In the
image, this shows up as a star in the center. Now, make real(p3) +0.28.
You'll see that the outer portion of the image is unchanged, but the star
in the middle has gone away. In the "starbail" formulas, the flag to
check the pixel before iterating (pixelflag) has been set in the formula,
and is not interactively changeable (too many parameters).
The "starnear" formulas use a standard bailout condition--stop when the
iterate gets too big, or leaves a very large circle. Conversely, in the
"starbail" formulas, the bailout condition is built around a star. If
the star is centered at the origin and is big enough, then you can
essentially use the star instead of the traditional bailout circle. This
is what has been done in parameter set "starbail1". The magnification
was decreased to show the bailout star, and the fractal can clearly be
seen in the middle. It is also possible to bailout when the orbit enters
a star. This is similar to the orbit trap methods that others have
written, and works particularly well for Julia sets that are dusts. An
example of this is in parameter set "starbail2". Some care must be taken
if the star is not centered on the origin. The "starbail" formulas set
z to a default (1,0) if the orbit was not caught by a star. If the star
is too far away or too small to catch many orbits, the majority of the
screen will be one color (color 0 when using decomposition coloring). If
you use stars off of the origin, you may want to increase the size and
set "pixelflag=0" in the formula, to prevent the image from being taken
over by one huge rogue star.
One final note: This method uses angles a lot to figure out which side
belongs to whom. Consequently, the ability to rotate the star is a bit
limited. If the rotations are kept to the range of 0 to 36 degrees, there
shouldn't be any problems. Rotations outside this range (try it with the
"star" formula) will result in branch cuts as the formula gets its angles
confused. Fortunately, since the star is symmetric, there shouldn't be
a need to rotate it beyond those bounds.
}
starnear1 { ; copyright Kerry Mitchell 09oct98
;
; sample of "near to 5 point star" coloring
;
reset=1960 type=formula formulafile=fractint.frm
formulaname=starnear_jul center-mag=-0.358/0.651/8
params=0/1/0/0/0.25/18 float=y maxiter=256 inside=0
decomp=256 periodicity=0 colors=zrM<40>zzzzzzyyz<60>QQzPPzPP\
zPPy<61>000000110<60>yn0zo0zo0zo1<19>zrL cyclerange=0/255
}
starnear2 { ; copyright Kerry Mitchell 09oct98
;
; sample of "near to 5 point star" coloring
;
reset=1960 type=formula formulafile=fractint.frm
formulaname=starnear_jul center-mag=0/0/1.1/1/-90
params=0.28/0/0/0/-0.28/36 float=y maxiter=256 inside=0
decomp=256 periodicity=0 colors=vlv<23>i1ih0hf0f<28>303\
101100<29>x99zAAzBB<30>zzz<30>i2ih0hg0g<29>202000000<30>\
y99zAAzCC<30>zzz<5>wnw cyclerange=0/255
}
starnear3 { ; copyright Kerry Mitchell 09oct98
;
; Mandelbrot sample of coloring by nearest approach to 5 point star
;
reset=1960 type=formula formulafile=fractint.frm
formulaname=starnear_man center-mag=-1/0/0.56/1/-90
params=-2/0/0.25/36 float=y maxiter=100 inside=0 decomp=256
periodicity=0 colors=000<35>x00z00z10<34>zx0zz0zz1<36>zzz<51>\
zz2zz0zy0<51>y20x00w00<30>200000000000000 cyclerange=0/255
}
starbail1 { ; copyright Kerry Mitchell 09oct98
;
; bails out when orbit leaves a 5 point star
;
reset=1960 type=formula formulafile=fractint.frm
formulaname=starbail_jul center-mag=0/0/0.25 cyclerange=0/255
params=0.3795135923319061/0.3349323012073617/0/0/5/18 float=y
maxiter=256 inside=0 decomp=256 periodicity=0 colors=000<40>x00z\
00z00<40>zy0zz0zz1<39>zzxzzzzzz<40>1zz0zz0yz<39>02z00z00z<41>000
}
starbail2 { ; copyright Kerry Mitchell 09oct98
;
; bails out when orbit enters a 5 point star
;
reset=1960 type=formula formulafile=fractint.frm
formulaname=starbail_jul passes=1 center-mag=0/0/1.1/1/-90
params=0.3/0/0/0/-0.5/36 float=y maxiter=256 inside=0
decomp=256 periodicity=0 colors=000<15>zo0<14>zyvzzzxxz<13>\
SSzPPzOOw<14>000<15>zo0<14>zywzzzxxz<13>RRzPPzOOv<11>55B337\
223110<13>tj0xm0zo1<13>zxtzyxyyz<12>VVzTTzRRzPPyNNu<12>3361\
12220<14>yn0zo2<14>zyyyyz<13>SSzQQzOOx<14>111 cyclerange=0/255
}
starbail3 { ; copyright Kerry Mitchell 09oct98
;
; Mandelbrot sample of bailout out when moving inside a 5 point star
;
reset=1960 type=formula formulafile=fractint.frm
formulaname=starbail_man center-mag=-0.407/0/0.813
params=1/0/-1/36 float=y maxiter=100 inside=0 decomp=256
periodicity=0 colors=000<35>x00z00z10<34>zx0zz0zz1<36>zzz<51>\
zz2zz0zy0<51>y20x00w00<30>200000000000000 cyclerange=0/255
}
frm:starbail_man { ; Kerry Mitchell 09oct98
;
; Mandelbrot, bails out when orbit enters/leaves 5 point star
; p1 = center of star
; cabs(real(p2)) = star size of star (try 1)
; sign(real(p2)) = in/out flag:
; + = bailout when orbit leaves star
; - = bailout when orbit enters star
; imag(p2) = star rotation angle, degrees
; only use angles from 0 to 36 degrees
;
; initialize iteration parameters
;
zc=0, c=pixel, done=0, iter=1, bailout=1e12
pixelflag=0 ; set by hand
;
; star parameters
;
center=p1, xcen=real(center), ycen=imag(center)
r=cabs(real(p2)), inout=1
if(real(p2)<0)
inout=0
endif
phi=imag(p2)/180*pi, twopi=2*pi, temp=twopi/10
t0=phi, t1=t0+temp
t2=t1+temp, t3=t2+temp, t4=t3+temp, t5=t4+temp
t6=t5+temp, t7=t6+temp, t8=t7+temp, t9=t8+temp
;
; set up control points
;
x0=r*cos(t0)+xcen, y0=r*sin(t0)+ycen
x1=r*cos(t2)+xcen, y1=r*sin(t2)+ycen
x2=r*cos(t4)+xcen, y2=r*sin(t4)+ycen
x3=r*cos(t6)+xcen, y3=r*sin(t6)+ycen
x4=r*cos(t8)+xcen, y4=r*sin(t8)+ycen
;
; check pixel to see if it bailed
;
if(pixelflag!=0)
x=real(zc), y=imag(zc), t=imag(log(zc-center))
if(t<0)
t=t+twopi
endif
flag=0
if((t>t0)&&(t<=t1))
f=x*(y0-y2)+y*(x2-x0)-x2*y0+x0*y2
if(f<0)
flag=1
endif
elseif((t>t1)&&(t<=t2))
f=x*(y1-y4)+y*(x4-x1)-x4*y1+x1*y4
if(f>0)
flag=1
endif
elseif((t>t2)&&(t<=t3))
f=x*(y1-y3)+y*(x3-x1)-x3*y1+x1*y3
if(f<0)
flag=1
endif
elseif((t>t3)&&(t<=t4))
f=x*(y2-y0)+y*(x0-x2)-x0*y2+x2*y0
if(f>0)
flag=1
endif
elseif((t>t4)&&(t<=t5))
f=x*(y2-y4)+y*(x4-x2)-x4*y2+x2*y4
if(f<0)
flag=1
endif
elseif((t>t5)&&(t<=t6))
f=x*(y3-y1)+y*(x1-x3)-x1*y3+x3*y1
if(f>0)
flag=1
endif
elseif((t>t6)&&(t<=t7))
f=x*(y3-y0)+y*(x0-x3)-x0*y3+x3*y0
if(f<0)
flag=1
endif
elseif((t>t7)&&(t<=t8))
f=x*(y4-y2)+y*(x2-x4)-x2*y4+x4*y2
if(f>0)
flag=1
endif
elseif((t>t8)&&(t<=t9))
f=x*(y4-y1)+y*(x1-x4)-x1*y4+x4*y1
if(f<0)
flag=1
endif
else
f=x*(y0-y3)+y*(x3-x0)-x3*y0+x0*y3
if(f>0)
flag=1
endif
endif
if(flag==inout)
done=1
z=zc-center
endif
endif
:
; standard iteration if pixel didn't bail
;
if(done==0)
iter=iter+1, zc=sqr(zc)+c
x=real(zc), y=imag(zc), t=imag(log(zc-center))
if(t<0)
t=t+twopi
endif
;
; see if orbit is inside or outside of star
; by checking each side one at a time
;
flag=0
if((t>t0)&&(t<=t1))
f=x*(y0-y2)+y*(x2-x0)-x2*y0+x0*y2
if(f<0)
flag=1
endif
elseif((t>t1)&&(t<=t2))
f=x*(y1-y4)+y*(x4-x1)-x4*y1+x1*y4
if(f>0)
flag=1
endif
elseif((t>t2)&&(t<=t3))
f=x*(y1-y3)+y*(x3-x1)-x3*y1+x1*y3
if(f<0)
flag=1
endif
elseif((t>t3)&&(t<=t4))
f=x*(y2-y0)+y*(x0-x2)-x0*y2+x2*y0
if(f>0)
flag=1
endif
elseif((t>t4)&&(t<=t5))
f=x*(y2-y4)+y*(x4-x2)-x4*y2+x2*y4
if(f<0)
flag=1
endif
elseif((t>t5)&&(t<=t6))
f=x*(y3-y1)+y*(x1-x3)-x1*y3+x3*y1
if(f>0)
flag=1
endif
elseif((t>t6)&&(t<=t7))
f=x*(y3-y0)+y*(x0-x3)-x0*y3+x3*y0
if(f<0)
flag=1
endif
elseif((t>t7)&&(t<=t8))
f=x*(y4-y2)+y*(x2-x4)-x2*y4+x4*y2
if(f>0)
flag=1
endif
elseif((t>t8)&&(t<=t9))
f=x*(y4-y1)+y*(x1-x4)-x1*y4+x4*y1
if(f<0)
flag=1
endif
else
f=x*(y0-y3)+y*(x3-x0)-x3*y0+x0*y3
if(f>0)
flag=1
endif
endif
;
; if the orbit was on the appropriate side,
; set "done" flag
; set z to iteration variable for coloring
;
if(flag==inout)
done=1
z=zc-center
;
; safety valve in case star doesn't catch orbit
; if orbit goes to infinity or max iterations reached:
; set "done" flag
; set z to 1 to differentiate it from star trap
;
elseif((|zc|>bailout)||(iter==maxit))
done=1
z=1
endif
endif
done==0
}
frm:starbail_jul { ; Kerry Mitchell 09oct98
;
; Julia set, bails out when orbit enters/leaves 5 point star
; p1 = Julia parameter
; p2 = center of star
; cabs(real(p3)) = star size of star (try 1)
; sign(real(p3)) = in/out flag:
; + = bailout when orbit leaves star
; - = bailout when orbit enters star
; imag(p3) = star rotation angle, degrees
; only use angles from 0 to 36 degrees
;
; initialize iteration parameters
;
c=p1, zc=pixel, done=0, iter=1, bailout=1e12
pixelflag=1 ; set by hand
;
; star parameters
;
center=p2, xcen=real(center), ycen=imag(center)
r=cabs(real(p3)), inout=1
if(real(p3)<0)
inout=0
endif
phi=imag(p3)/180*pi, twopi=2*pi, temp=twopi/10
t0=phi, t1=t0+temp
t2=t1+temp, t3=t2+temp, t4=t3+temp, t5=t4+temp
t6=t5+temp, t7=t6+temp, t8=t7+temp, t9=t8+temp
;
; set up control points
;
x0=r*cos(t0)+xcen, y0=r*sin(t0)+ycen
x1=r*cos(t2)+xcen, y1=r*sin(t2)+ycen
x2=r*cos(t4)+xcen, y2=r*sin(t4)+ycen
x3=r*cos(t6)+xcen, y3=r*sin(t6)+ycen
x4=r*cos(t8)+xcen, y4=r*sin(t8)+ycen
;
; check pixel to see if it bailed
;
if(pixelflag!=0)
x=real(zc), y=imag(zc), t=imag(log(zc-center))
if(t<0)
t=t+twopi
endif
flag=0
if((t>t0)&&(t<=t1))
f=x*(y0-y2)+y*(x2-x0)-x2*y0+x0*y2
if(f<0)
flag=1
endif
elseif((t>t1)&&(t<=t2))
f=x*(y1-y4)+y*(x4-x1)-x4*y1+x1*y4
if(f>0)
flag=1
endif
elseif((t>t2)&&(t<=t3))
f=x*(y1-y3)+y*(x3-x1)-x3*y1+x1*y3
if(f<0)
flag=1
endif
elseif((t>t3)&&(t<=t4))
f=x*(y2-y0)+y*(x0-x2)-x0*y2+x2*y0
if(f>0)
flag=1
endif
elseif((t>t4)&&(t<=t5))
f=x*(y2-y4)+y*(x4-x2)-x4*y2+x2*y4
if(f<0)
flag=1
endif
elseif((t>t5)&&(t<=t6))
f=x*(y3-y1)+y*(x1-x3)-x1*y3+x3*y1
if(f>0)
flag=1
endif
elseif((t>t6)&&(t<=t7))
f=x*(y3-y0)+y*(x0-x3)-x0*y3+x3*y0
if(f<0)
flag=1
endif
elseif((t>t7)&&(t<=t8))
f=x*(y4-y2)+y*(x2-x4)-x2*y4+x4*y2
if(f>0)
flag=1
endif
elseif((t>t8)&&(t<=t9))
f=x*(y4-y1)+y*(x1-x4)-x1*y4+x4*y1
if(f<0)
flag=1
endif
else
f=x*(y0-y3)+y*(x3-x0)-x3*y0+x0*y3
if(f>0)
flag=1
endif
endif
if(flag==inout)
done=1
z=zc-center
endif
endif
:
; standard iteration if pixel didn't bail
;
if(done==0)
iter=iter+1, zc=sqr(zc)+c
x=real(zc), y=imag(zc), t=imag(log(zc-center))
if(t<0)
t=t+twopi
endif
;
; see if orbit is inside or outside of star
; by checking each side one at a time
;
flag=0
if((t>t0)&&(t<=t1))
f=x*(y0-y2)+y*(x2-x0)-x2*y0+x0*y2
if(f<0)
flag=1
endif
elseif((t>t1)&&(t<=t2))
f=x*(y1-y4)+y*(x4-x1)-x4*y1+x1*y4
if(f>0)
flag=1
endif
elseif((t>t2)&&(t<=t3))
f=x*(y1-y3)+y*(x3-x1)-x3*y1+x1*y3
if(f<0)
flag=1
endif
elseif((t>t3)&&(t<=t4))
f=x*(y2-y0)+y*(x0-x2)-x0*y2+x2*y0
if(f>0)
flag=1
endif
elseif((t>t4)&&(t<=t5))
f=x*(y2-y4)+y*(x4-x2)-x4*y2+x2*y4
if(f<0)
flag=1
endif
elseif((t>t5)&&(t<=t6))
f=x*(y3-y1)+y*(x1-x3)-x1*y3+x3*y1
if(f>0)
flag=1
endif
elseif((t>t6)&&(t<=t7))
f=x*(y3-y0)+y*(x0-x3)-x0*y3+x3*y0
if(f<0)
flag=1
endif
elseif((t>t7)&&(t<=t8))
f=x*(y4-y2)+y*(x2-x4)-x2*y4+x4*y2
if(f>0)
flag=1
endif
elseif((t>t8)&&(t<=t9))
f=x*(y4-y1)+y*(x1-x4)-x1*y4+x4*y1
if(f<0)
flag=1
endif
else
f=x*(y0-y3)+y*(x3-x0)-x3*y0+x0*y3
if(f>0)
flag=1
endif
endif
;
; if the orbit was on the appropriate side,
; set "done" flag
; set z to iteration variable for coloring
;
if(flag==inout)
done=1
z=zc-center
;
; safety valve in case star doesn't catch orbit
; if orbit goes to infinity or max iterations reached:
; set "done" flag
; set z to 1 to differentiate it from star trap
;
elseif((|zc|>bailout)||(iter==maxit))
done=1
z=1
endif
endif
done==0
}
frm:starnear_man { ; Kerry Mitchell 09oct98
;
; Mandelbrot, colors by nearest approach to 5 point star
; p1 = center of star
; cabs(real(p2)) = star size of star (try 1)
; sign(real(p2)) = pixel flag:
; + = don't consider pixel--start after first iteration
; - = consider pixel
; imag(p2) = star rotation angle, degrees
; only use angles from 0 to 36 degrees
; use decomp=256, bailout hardcoded to 10^12
;
; initialize iteration parameters
;
zc=0, c=pixel, done=0, iter=1
bailout=1e12, fmin=bailout
;
; star parameters
;
center=p1, xcen=real(center), ycen=imag(center)
r=cabs(real(p2)), pixelflag=0
if(real(p2)<0)
pixelflag=1
endif
phi=imag(p2)/180*pi, twopi=2*pi, temp=twopi/10
t0=phi, t1=t0+temp
t2=t1+temp, t3=t2+temp, t4=t3+temp, t5=t4+temp
t6=t5+temp, t7=t6+temp, t8=t7+temp, t9=t8+temp
;
; set up control points
;
x0=r*cos(t0)+xcen, y0=r*sin(t0)+ycen
x1=r*cos(t2)+xcen, y1=r*sin(t2)+ycen
x2=r*cos(t4)+xcen, y2=r*sin(t4)+ycen
x3=r*cos(t6)+xcen, y3=r*sin(t6)+ycen
x4=r*cos(t8)+xcen, y4=r*sin(t8)+ycen
;
; if pixel flag =/= 0, then check pixel for how close
; it is to star
;
if(pixelflag!=0)
x=real(zc), y=imag(zc), t=imag(log(zc-center))
if(t<0)
t=t+twopi
endif
if((t>t0)&&(t<=t1))
f=x*(y0-y2)+y*(x2-x0)-x2*y0+x0*y2
elseif((t>t1)&&(t<=t2))
f=x*(y1-y4)+y*(x4-x1)-x4*y1+x1*y4
elseif((t>t2)&&(t<=t3))
f=x*(y1-y3)+y*(x3-x1)-x3*y1+x1*y3
elseif((t>t3)&&(t<=t4))
f=x*(y2-y0)+y*(x0-x2)-x0*y2+x2*y0
elseif((t>t4)&&(t<=t5))
f=x*(y2-y4)+y*(x4-x2)-x4*y2+x2*y4
elseif((t>t5)&&(t<=t6))
f=x*(y3-y1)+y*(x1-x3)-x1*y3+x3*y1
elseif((t>t6)&&(t<=t7))
f=x*(y3-y0)+y*(x0-x3)-x0*y3+x3*y0
elseif((t>t7)&&(t<=t8))
f=x*(y4-y2)+y*(x2-x4)-x2*y4+x4*y2
elseif((t>t8)&&(t<=t9))
f=x*(y4-y1)+y*(x1-x4)-x1*y4+x4*y1
else
f=x*(y3-y0)+y*(x0-x3)-x0*y3+x3*y0
endif
fmin=cabs(f)
endif
:
; standard iteration, find polar angle of iterate
;
iter=iter+1, zc=sqr(zc)+c
x=real(zc), y=imag(zc), t=imag(log(zc-center))
if(t<0)
t=t+twopi
endif
;
; compute how close iterate is to each side of star
;
if((t>t0)&&(t<=t1))
f=x*(y0-y2)+y*(x2-x0)-x2*y0+x0*y2
elseif((t>t1)&&(t<=t2))
f=x*(y1-y4)+y*(x4-x1)-x4*y1+x1*y4
elseif((t>t2)&&(t<=t3))
f=x*(y1-y3)+y*(x3-x1)-x3*y1+x1*y3
elseif((t>t3)&&(t<=t4))
f=x*(y2-y0)+y*(x0-x2)-x0*y2+x2*y0
elseif((t>t4)&&(t<=t5))
f=x*(y2-y4)+y*(x4-x2)-x4*y2+x2*y4
elseif((t>t5)&&(t<=t6))
f=x*(y3-y1)+y*(x1-x3)-x1*y3+x3*y1
elseif((t>t6)&&(t<=t7))
f=x*(y3-y0)+y*(x0-x3)-x0*y3+x3*y0
elseif((t>t7)&&(t<=t8))
f=x*(y4-y2)+y*(x2-x4)-x2*y4+x4*y2
elseif((t>t8)&&(t<=t9))
f=x*(y4-y1)+y*(x1-x4)-x1*y4+x4*y1
else
f=x*(y3-y0)+y*(x0-x3)-x0*y3+x3*y0
endif
f=cabs(f)
;
; update minimum distance
;
if(f<fmin)
fmin=f
zmin=zc
endif
;
; upon escape or maximum iterations:
; set "done" flag
; use minimum distance from star as polar angle of z
; for use with decomp coloring
;
if((|zc|>bailout)||(iter==maxit))
done=1
t=log(fmin)
z=cos(t)+flip(sin(t))
endif
done==0
}
frm:starnear_jul { ; Kerry Mitchell 09oct98
;
; Julia set, colors by nearest approach to 5 point star
; p1 = Julia parameter
; p2 = center of star
; cabs(real(p3)) = star size of star (try 1)
; sign(real(p3)) = pixel flag:
; + = don't consider pixel--start after first iteration
; - = consider pixel
; imag(p3) = star rotation angle, degrees
; only use angles from 0 to 36 degrees
; use decomp=256, bailout hardcoded to 10^12
;
; initialize iteration parameters
;
c=p1, zc=pixel, done=0, iter=1
bailout=1e12, fmin=bailout
;
; star parameters
;
center=p2, xcen=real(center), ycen=imag(center)
r=cabs(real(p3)), pixelflag=0
if(real(p3)<0)
pixelflag=1
endif
phi=imag(p3)/180*pi, twopi=2*pi, temp=twopi/10
t0=phi, t1=t0+temp
t2=t1+temp, t3=t2+temp, t4=t3+temp, t5=t4+temp
t6=t5+temp, t7=t6+temp, t8=t7+temp, t9=t8+temp
;
; set up control points
;
x0=r*cos(t0)+xcen, y0=r*sin(t0)+ycen
x1=r*cos(t2)+xcen, y1=r*sin(t2)+ycen
x2=r*cos(t4)+xcen, y2=r*sin(t4)+ycen
x3=r*cos(t6)+xcen, y3=r*sin(t6)+ycen
x4=r*cos(t8)+xcen, y4=r*sin(t8)+ycen
;
; if pixel flag =/= 0, then check pixel for how close
; it is to star
;
if(pixelflag!=0)
x=real(zc), y=imag(zc), t=imag(log(zc-center))
if(t<0)
t=t+twopi
endif
if((t>t0)&&(t<=t1))
f=x*(y0-y2)+y*(x2-x0)-x2*y0+x0*y2
elseif((t>t1)&&(t<=t2))
f=x*(y1-y4)+y*(x4-x1)-x4*y1+x1*y4
elseif((t>t2)&&(t<=t3))
f=x*(y1-y3)+y*(x3-x1)-x3*y1+x1*y3
elseif((t>t3)&&(t<=t4))
f=x*(y2-y0)+y*(x0-x2)-x0*y2+x2*y0
elseif((t>t4)&&(t<=t5))
f=x*(y2-y4)+y*(x4-x2)-x4*y2+x2*y4
elseif((t>t5)&&(t<=t6))
f=x*(y3-y1)+y*(x1-x3)-x1*y3+x3*y1
elseif((t>t6)&&(t<=t7))
f=x*(y3-y0)+y*(x0-x3)-x0*y3+x3*y0
elseif((t>t7)&&(t<=t8))
f=x*(y4-y2)+y*(x2-x4)-x2*y4+x4*y2
elseif((t>t8)&&(t<=t9))
f=x*(y4-y1)+y*(x1-x4)-x1*y4+x4*y1
else
f=x*(y3-y0)+y*(x0-x3)-x0*y3+x3*y0
endif
fmin=cabs(f)
endif
:
; standard iteration, find polar angle of iterate
;
iter=iter+1, zc=sqr(zc)+c
x=real(zc), y=imag(zc), t=imag(log(zc-center))
if(t<0)
t=t+twopi
endif
;
; compute how close iterate is to each side of star
;
if((t>t0)&&(t<=t1))
f=x*(y0-y2)+y*(x2-x0)-x2*y0+x0*y2
elseif((t>t1)&&(t<=t2))
f=x*(y1-y4)+y*(x4-x1)-x4*y1+x1*y4
elseif((t>t2)&&(t<=t3))
f=x*(y1-y3)+y*(x3-x1)-x3*y1+x1*y3
elseif((t>t3)&&(t<=t4))
f=x*(y2-y0)+y*(x0-x2)-x0*y2+x2*y0
elseif((t>t4)&&(t<=t5))
f=x*(y2-y4)+y*(x4-x2)-x4*y2+x2*y4
elseif((t>t5)&&(t<=t6))
f=x*(y3-y1)+y*(x1-x3)-x1*y3+x3*y1
elseif((t>t6)&&(t<=t7))
f=x*(y3-y0)+y*(x0-x3)-x0*y3+x3*y0
elseif((t>t7)&&(t<=t8))
f=x*(y4-y2)+y*(x2-x4)-x2*y4+x4*y2
elseif((t>t8)&&(t<=t9))
f=x*(y4-y1)+y*(x1-x4)-x1*y4+x4*y1
else
f=x*(y3-y0)+y*(x0-x3)-x0*y3+x3*y0
endif
f=cabs(f)
;
; update minimum distance
;
if(f<fmin)
fmin=f
zmin=zc
endif
;
; upon escape or maximum iterations:
; set "done" flag
; use minimum distance from star as polar angle of z
; for use with decomp coloring
;
if((|zc|>bailout)||(iter==maxit))
done=1
t=log(fmin)
z=cos(t)+flip(sin(t))
endif
done==0
}
frm:star-inout { ; Kerry Mitchell 09oct98
;
; draws 5 point star--1 color inside, 1 color outside
; p1 = center of star
; real(p2) = size of star (try 1)
; imag(p2) = rotation angle, degrees (0 for point at top)
; use decomp=256
;
zc=pixel, done=1
center=p1, xcen=real(center), ycen=imag(center), r=real(p2)
phi=(imag(p2)+18)/180*pi, twopi=2*pi, temp=twopi/10
if((phi<0)||(phi>=temp))
phi=0
endif
t0=phi, t1=temp+phi, t2=2*temp+phi, t3=3*temp+phi
t4=4*temp+phi, t5=5*temp+phi, t6=6*temp+phi
t7=7*temp+phi, t8=8*temp+phi, t9=9*temp+phi
x0=r*cos(t0)+xcen, y0=r*sin(t0)+ycen
x1=r*cos(t2)+xcen, y1=r*sin(t2)+ycen
x2=r*cos(t4)+xcen, y2=r*sin(t4)+ycen
x3=r*cos(t6)+xcen, y3=r*sin(t6)+ycen
x4=r*cos(t8)+xcen, y4=r*sin(t8)+ycen
:
x=real(zc), y=imag(zc), t=imag(log(zc-center))
if(t<0)
t=t+twopi
endif
flag=0
if((t>t0)&&(t<=t1))
f=x*(y0-y2)+y*(x2-x0)-x2*y0+x0*y2
if(f<0)
flag=1
endif
elseif((t>t1)&&(t<=t2))
f=x*(y1-y4)+y*(x4-x1)-x4*y1+x1*y4
if(f>0)
flag=1
endif
elseif((t>t2)&&(t<=t3))
f=x*(y1-y3)+y*(x3-x1)-x3*y1+x1*y3
if(f<0)
flag=1
endif
elseif((t>t3)&&(t<=t4))
f=x*(y2-y0)+y*(x0-x2)-x0*y2+x2*y0
if(f>0)
flag=1
endif
elseif((t>t4)&&(t<=t5))
f=x*(y2-y4)+y*(x4-x2)-x4*y2+x2*y4
if(f<0)
flag=1
endif
elseif((t>t5)&&(t<=t6))
f=x*(y3-y1)+y*(x1-x3)-x1*y3+x3*y1
if(f>0)
flag=1
endif
elseif((t>t6)&&(t<=t7))
f=x*(y3-y0)+y*(x0-x3)-x0*y3+x3*y0
if(f<0)
flag=1
endif
elseif((t>t7)&&(t<=t8))
f=x*(y4-y2)+y*(x2-x4)-x2*y4+x4*y2
if(f>0)
flag=1
endif
elseif((t>t8)&&(t<=t9))
f=x*(y4-y1)+y*(x1-x4)-x1*y4+x4*y1
if(f<0)
flag=1
endif
else
f=x*(y0-y3)+y*(x3-x0)-x3*y0+x0*y3
if(f>0)
flag=1
endif
endif
z=0
if(flag==1)
z=(0.0,1.0)
endif
done==0
}
frm:star { ; Kerry Mitchell 09oct98
;
; draws 5-point star, not a fractal
;
; p1 = center of star
; real(p2) = size of star (try 1)
; imag(p2) = rotation angle, degrees (0 for point at top)
; use decomp=256
;
zc=pixel, done=1
center=p1, xcen=real(center), ycen=imag(center), r=real(p2)
phi=imag(p2)/180*pi, twopi=2*pi, temp=twopi/10
t0=phi, t1=temp+phi, t2=2*temp+phi, t3=3*temp+phi
t4=4*temp+phi, t5=5*temp+phi, t6=6*temp+phi
t7=7*temp+phi, t8=8*temp+phi, t9=9*temp+phi
x0=r*cos(t0)+xcen, y0=r*sin(t0)+ycen
x1=r*cos(t2)+xcen, y1=r*sin(t2)+ycen
x2=r*cos(t4)+xcen, y2=r*sin(t4)+ycen
x3=r*cos(t6)+xcen, y3=r*sin(t6)+ycen
x4=r*cos(t8)+xcen, y4=r*sin(t8)+ycen
:
x=real(zc), y=imag(zc), t=imag(log(zc-center))
if(t<0)
t=t+twopi
endif
if((t>t0)&&(t<=t1))
f=x*(y0-y2)+y*(x2-x0)-x2*y0+x0*y2
elseif((t>t1)&&(t<=t2))
f=x*(y1-y4)+y*(x4-x1)-x4*y1+x1*y4
elseif((t>t2)&&(t<=t3))
f=x*(y1-y3)+y*(x3-x1)-x3*y1+x1*y3
elseif((t>t3)&&(t<=t4))
f=x*(y2-y0)+y*(x0-x2)-x0*y2+x2*y0
elseif((t>t4)&&(t<=t5))
f=x*(y2-y4)+y*(x4-x2)-x4*y2+x2*y4
elseif((t>t5)&&(t<=t6))
f=x*(y3-y1)+y*(x1-x3)-x1*y3+x3*y1
elseif((t>t6)&&(t<=t7))
f=x*(y3-y0)+y*(x0-x3)-x0*y3+x3*y0
elseif((t>t7)&&(t<=t8))
f=x*(y4-y2)+y*(x2-x4)-x2*y4+x4*y2
elseif((t>t8)&&(t<=t9))
f=x*(y4-y1)+y*(x1-x4)-x1*y4+x4*y1
else
f=x*(y3-y0)+y*(x0-x3)-x0*y3+x3*y0
endif
t=log(cabs(f))
z=cos(t)+flip(sin(t))
done==0
}
- --------------------------------------------------------------
Thanks for using Fractint, The Fractals and Fractint Discussion List
Post Message: fractint@lists.xmission.com
Get Commands: majordomo@lists.xmission.com "help"
Administrator: twegner@phoenix.net
Unsubscribe: majordomo@lists.xmission.com "unsubscribe fractint"
------------------------------
Date: Fri, 9 Oct 1998 17:41:38 EDT
From: SKarl52884@aol.com
Subject: Re: (fractint) Re: [fractal-art] cost of images
In a message dated 10/9/98 1:44:48 PM Eastern Daylight Time,
kchildre@uccs.jpl.nasa.gov writes:
<< The best way I know to do
that is to use slides for photographs for underwater photographs, and
go from digital to a print for digital images. >>
Hi there...
Which brings to mind the digital camera.
My brother is bringing it over tonight.
It produces jpg and he claims that to get prints on paper all one needs to do
is deliver them to the right service agency.
Appearently the files aren't huge as he sent me a few shots of the of the
mountains last week while on vacation. They seemed to be at
1280x768...downloaded fast
and looked very very good!
Steve
- --------------------------------------------------------------
Thanks for using Fractint, The Fractals and Fractint Discussion List
Post Message: fractint@lists.xmission.com
Get Commands: majordomo@lists.xmission.com "help"
Administrator: twegner@phoenix.net
Unsubscribe: majordomo@lists.xmission.com "unsubscribe fractint"
------------------------------
Date: Fri, 9 Oct 1998 17:51:19 EDT
From: SKarl52884@aol.com
Subject: Re: Sv: (fractint) Re: [fractal-art] cost of images/videomode?
In a message dated 10/9/98 3:06:47 PM Eastern Daylight Time,
kimb@post8.tele.dk writes:
<< I
render a picture on 4500x3000 or 6000x4000 and then reduces it to 1500x1000
to anti-alias it. >>
Hi Kim
Have i missed a fractint up-date??? The highest videomode I have
[in 19.6.... disk ram ] is 2048x2048 or 2048x1536.
How are you getting 4500x3000 and 6000x4000?
Steve
- --------------------------------------------------------------
Thanks for using Fractint, The Fractals and Fractint Discussion List
Post Message: fractint@lists.xmission.com
Get Commands: majordomo@lists.xmission.com "help"
Administrator: twegner@phoenix.net
Unsubscribe: majordomo@lists.xmission.com "unsubscribe fractint"
------------------------------
Date: Fri, 09 Oct 1998 17:02:28 -0500
From: "Paul N. Lee" <Paul.N.Lee@Worldnet.att.net>
Subject: Re: Sv: (fractint) Re: [fractal-art] cost of images/videomode?
SKarl52884@aol.com wrote:
>
> Have i missed a fractint up-date??? The highest videomode
> I have [in 19.6.... disk ram ] is 2048x2048 or 2048x1536.
> How are you getting 4500x3000 and 6000x4000?
Fractint still generates images at 2048x2048, but if you want higher
resolution images for printing quality prints, then use Fractint's
"pieces" divide-and-conquer feature. You can create multiple PAR
entries that break an image up into pieces so that you can generate the
image pieces one by one. There are two reasons for doing this. The
first is in case the fractal is very slow, and you want to generate
parts of the image at the same time on several computers. The second is
that you might want to make an image greater than 2048x2048. The
parameters for this feature are:
X Multiples - How many divisions of final image in the x direction
Y Multiples - How many divisions of final image in the y direction
Video mode - Fractint video mode for each piece (e.g. "F3")
The last item defaults to the current video mode. If either X Multiples
or Y Multiples are greater than 1, then multiple numbered PAR entries
for the pieces are added to the PAR file, and a MAKEMIG.BAT file is
created that builds all of the component pieces and then stitches them
together into a "multi-image" GIF. The current limitations of the
"divide and conquer" algorithm are 36 or fewer X and Y multiples (so you
are limited to "only" 36x36=1296 component images), and a final
resolution limit in both the X and Y directions of 65,535 (a limitation
of "only" four billion pixels or so).
The final image generated by MAKEMIG is a "multi-image" GIF file called
FRACTMIG.GIF. In case you have other software that can't handle
multi-image GIF files, MAKEMIG includes a final (but commented out) call
to SIMPLGIF, a companion program that reads a GIF file that may contain
little tricks like multiple images and creates a simple GIF from it.
Fair warning: SIMPLGIF needs room to build a composite image while it
works, and it does that using a temporary disk file equal to the size of
the final image - and a 64Kx64K GIF image requires a 4GB temporary disk
file!
P.N.L.
- -------------------------------------------------
http://www.fractalus.com/cgi-bin/theway?ring=fractals&id=43&go
- --------------------------------------------------------------
Thanks for using Fractint, The Fractals and Fractint Discussion List
Post Message: fractint@lists.xmission.com
Get Commands: majordomo@lists.xmission.com "help"
Administrator: twegner@phoenix.net
Unsubscribe: majordomo@lists.xmission.com "unsubscribe fractint"
------------------------------
End of fractint-digest V1 #311
******************************