From: fractint-owner@xmission.com (fractint Digest) To: fractint-digest@xmission.com Subject: fractint Digest V1 #6 Reply-To: fractint@xmission.com Sender: fractint-owner@xmission.com Errors-To: fractint-owner@xmission.com Precedence: fractint Digest Friday, August 22 1997 Volume 01 : Number 006 In this issue: Re: (fractint) Hello (fractint) Periodicity checking (fractint) video-driver (fractint) Fractal Creations 2nd edition Re: (fractint) video-driver (fractint) 2nd Question: New Hole (fractint) Holey Fractals, Batman! Re: (fractint) 2nd Question: New Hole Re: (fractint) 2nd Question: New Hole (fractint) 2nd Question: New Hole Re: (fractint) 2nd Question: New Hole (fractint) Re: Fractint formulas Re: (fractint) Tru.c workings Re: (fractint) Tru.c workings (fractint) New Coloring Methods (fractint) question & introduction (fractint) PARs and FRMs See the end of the digest for information on subscribing to the fractint or fractint-digest mailing lists and on how to retrieve back issues. ---------------------------------------------------------------------- Date: Wed, 20 Aug 1997 09:59:44 -0700 From: "Jay Hill" Subject: Re: (fractint) Hello Russel wrote: >Here is my request, could you (being all who post formulas or parameters) >post them as e-mail attachments? A few have done this and it makes >it so much easier. If we must use attachments please make it in addition to the email, when I export my email to an ascii file, attachments are lost. Par files without the description are soon orphins. I have found the Fractal of the Day email works great if I export the whole thing and name it with a .par extension. Fractint 19.6 reads it fine. Jay - ------------------------------------------------------------ Thanks for using Fractint, The Fractals and Fractint Discussion List Post Message: fractint@xmission.com Get Commands: majordomo@xmission.com "help" Administrator: twegner@phoenix.net Unsubscribe: majordomo@xmission.com "unsubscribe fractint" ------------------------------ Date: Wed, 20 Aug 1997 13:07:49 -0400 From: "Damien M. Jones" Subject: (fractint) Periodicity checking - --=====================_872111269==_ Content-Type: text/plain; charset="us-ascii" Alice, - And the periodicity setting....Les St. Clair has tried to explain that - to me, and it's like a big brick wall. OK, I'll take a stab at it. Let me know if I obfuscate... ;-) Your average garden-variety fractal takes each screen point and maps it to a complex number, running that number through the same equation over and over again to see what happens. Points "outside" the set head off towards infinity (they get bigger than the "bailout" value) and points "inside" the set don't--they just kind of hang around. Eventually you've run the point through so many iterations you "give up"--this is the maximum iterations value. The problem (as far as speed is concerned) is that all of your inside points are run through your equation the *full* number of times. If there was a way to detect early on that a point isn't going to head to infinity, you could quit early, safe in the knowledge that the point is "inside". Well, for some points, you *can* find out early. Some points, as they are pushed through the equation over and over, produce a repeating cycle of numbers. For example, in the Mandelbrot set, the point at (-2,0), as it is run through the iterated equation z = z*z + c, produces the sequence (0,0), (-2,0). One number produces the other. Obviously that's not heading towards infinity, so if that can be detected, the program can quit early and not calculate 10,000 iterations of that point. To see what's going on, start FractInt and let the M-set fully draw. Press "O" (to turn on the Orbits window) and "L" (to connect all the points with Lines). Now move the pointer over the M-set. What FractInt is showing you is how the point you start at moves around as it is run through the iterative equation. Move the pointer into the main bud at the left side, and you'll see the point ping-pongs between two areas. Now move the pointer into the big bud at the top. You'll see a rough triangle, as the point moves through three small areas that make up the triangle corners. (Press L again to see the point clusters.) For many fractal types, inside points show this kind of periodic (cyclic) behavior. But here's the catch: some points may *look* like they're doing this, but in fact the numbers are just a little bit different each time; and if you iterate long enough, they suddenly become unstable and shoot off towards infinity. With FractInt's periodicity checking, sometimes these points are mistaken early on as periodic points, when in fact they just need to be iterated more to show that they shoot off to infinity. To see an example of this, I've attached a PAR file of a Julia set. If you generate this as is, you'll see the middle of the looping spiral is missing, even though the maximum iterations is set to 1000 and it should appear. (And it uses two-pass rendering, so it's not being dropped out by guessing.) If you press "G" and type "periodicity=no" to turn off periodicity checking, the fractal will redraw and you'll see the inner loops of the spiral. The effect is much more pronounced at lower resolutions (like 640x480) than my normal working res of 1024x768. This type of drop-out can be baffling if you forget to try turning off periodicity checking. - --=====================_872111269==_ Content-Type: text/plain; charset="us-ascii" Content-Disposition: attachment; filename="Period.par" periodicityexample { reset=1960 type=julia passes=2 center-mag=0/0/0.6666667 params=-0.7482893185793418/-0.025423566101695 float=y maxiter=1000 bailout=128 inside=0 colors=@atomic.map } - --=====================_872111269==_ Content-Type: text/plain; charset="us-ascii" Damien M. Jones / temporary sanity designs / http://www.emi.net/~dmj/ dmj@emi.net / my gallery: http://www.geocities.com/SoHo/Lofts/2605/ - --=====================_872111269==_-- - ------------------------------------------------------------ Thanks for using Fractint, The Fractals and Fractint Discussion List Post Message: fractint@xmission.com Get Commands: majordomo@xmission.com "help" Administrator: twegner@phoenix.net Unsubscribe: majordomo@xmission.com "unsubscribe fractint" ------------------------------ Date: Wed, 20 Aug 97 19:29:05 +0200 From: "Guenther Pfannhauser" Subject: (fractint) video-driver hello! I have a question about graphic-adapters with fractint. Specially about the ATI MACH-64 chipset. I read in the DOC that anyone (how knows everything about a graphics-adapter) can write his own video-driver. So, does anybody know where I can find these driver-files on the internet? Thanks Guenther Pfannhauser - ------------------------------------------------------------ Thanks for using Fractint, The Fractals and Fractint Discussion List Post Message: fractint@xmission.com Get Commands: majordomo@xmission.com "help" Administrator: twegner@phoenix.net Unsubscribe: majordomo@xmission.com "unsubscribe fractint" ------------------------------ Date: Wed, 20 Aug 1997 11:09:04 -0700 From: "Jay Hill" Subject: (fractint) Fractal Creations 2nd edition Fractal Creations 2nd edition: What is the interest in this out of print book? Send my email if you are looking for a copy. JAY.R.HILL@cpmx.saic.com Jay - ------------------------------------------------------------ Thanks for using Fractint, The Fractals and Fractint Discussion List Post Message: fractint@xmission.com Get Commands: majordomo@xmission.com "help" Administrator: twegner@phoenix.net Unsubscribe: majordomo@xmission.com "unsubscribe fractint" ------------------------------ Date: Wed, 20 Aug 1997 11:43:41 -0600 From: Rich Thomson Subject: Re: (fractint) video-driver In article <199708201738.RAA126038@out1.ibm.net> , "Guenther Pfannhauser" writes: > I have a question about graphic-adapters with fractint. > Specially about the ATI MACH-64 chipset. > I read in the DOC that anyone (how knows everything about a > graphics-adapter) can write his own video-driver. So, does anybody know > where I can find these driver-files on the internet? You'll want to start with the source code for fractint. It is available in all the same places that executable version is. Here's a URL for the one at spanky - -- ``Between stimulus and response is the will to choose.'' -- Steven Covey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= 3D Paint: The Power to Create in 3D; Rich Thomson email me for more info rthomson@ptc.com - ------------------------------------------------------------ Thanks for using Fractint, The Fractals and Fractint Discussion List Post Message: fractint@xmission.com Get Commands: majordomo@xmission.com "help" Administrator: twegner@phoenix.net Unsubscribe: majordomo@xmission.com "unsubscribe fractint" ------------------------------ Date: Wed, 20 Aug 1997 17:52:22 -0400 (EDT) From: A M Kelley Subject: (fractint) 2nd Question: New Hole - --1920402471-551766607-872114546:#19925 Content-Type: TEXT/PLAIN; charset=US-ASCII Well after that exciting success with my first holed fractal, which is now dressed up in a nice colormap and is ready for my gallery, I went running for this fractal, already on my gallery and one of my favorites, that also has a hole in its spiral vortex; I was eager to apply what I've learned to fill in its hole. I'll bet since you see there's an attached par file you can guess what happened. Since this one's formula does not allow for a bailout value, I wrote one in the par file. No good. Playing with the maxiter and logmap values makes no difference. This fractal has a distest of 1000, and I tried varying that. Raising it to 50000000 or so fills in the hole with some busy looking stuff, but still, it's a hole. I am almost certain this is my last holed fractal, by the way. The par has a zoom into the hole included. Thank you all in advance.--Alice - --1920402471-551766607-872114546:#19925 Content-Type: APPLICATION/octet-stream; name="quest2.par" Content-ID: Content-Description: Fractalj.gif { ; Pulse ; by amkelley@freenet.columbus.oh.us reset=1950 type=formula formulafile=fractint.frm formulaname=CGNewtonSinExp passes=2 center-mag=-0.992426/0.931621/4.63121 params=1.4/1 float=y maxiter=5000 inside=bof60 logmap=5 distest=1000/71/80/60 colors=834pJE<43>A45945834723502301000<13>ehlhkoilokmp<9>WTL834<173>834V\ 0BW0B } Pulse_zoom { ; amkelley@freenet.columbus.oh.us reset=1950 type=formula formulafile=fractint.frm formulaname=CGNewtonSinExp passes=2 center-mag=-0.950575/0.9244/46.3121 params=1.4/1 float=y maxiter=5000 inside=bof60 logmap=5 distest=1000/71/400/300 colors=834pJE<43>A45945834723502301000<13>ehlhkoilokmp<9>WTL834<173>834V\ 0BW0B } - --1920402471-551766607-872114546:#19925-- - ------------------------------------------------------------ Thanks for using Fractint, The Fractals and Fractint Discussion List Post Message: fractint@xmission.com Get Commands: majordomo@xmission.com "help" Administrator: twegner@phoenix.net Unsubscribe: majordomo@xmission.com "unsubscribe fractint" ------------------------------ Date: Wed, 20 Aug 1997 18:11:04 -0400 (EDT) From: A M Kelley Subject: (fractint) Holey Fractals, Batman! I forgot to mention that I did try various periodicity values with this latest fractal, and there was no effect.....--Alice - ------------------------------------------------------------ Thanks for using Fractint, The Fractals and Fractint Discussion List Post Message: fractint@xmission.com Get Commands: majordomo@xmission.com "help" Administrator: twegner@phoenix.net Unsubscribe: majordomo@xmission.com "unsubscribe fractint" ------------------------------ Date: Wed, 20 Aug 1997 15:43:04 -0700 From: "Jay Hill" Subject: Re: (fractint) 2nd Question: New Hole Now these are strange par files. Look what happened when I zoom out. There is a sudden jump in the image. Why? JumpPoint { ; zoom out just a bit ; see suddenly a different picture reset=1950 type=formula formulafile=fractint.frm formulaname=cgnewtonsinexp passes=2 center-mag=-0.992426/0.931621/0.006024541 params=1.4/1 float=y maxiter=5000 inside=bof60 logmap=5 distest=1000/71/80/60 viewwindows=2/0.75/yes/0/0 colors=834pJE<43>A45945834723502301000<13>ehlhkoilokmp<9>WTL834<173>834V\ 0BW0B } - ------------------------------------------------------------ Thanks for using Fractint, The Fractals and Fractint Discussion List Post Message: fractint@xmission.com Get Commands: majordomo@xmission.com "help" Administrator: twegner@phoenix.net Unsubscribe: majordomo@xmission.com "unsubscribe fractint" ------------------------------ Date: Wed, 20 Aug 1997 19:33:46 -0400 From: "Damien M. Jones" Subject: Re: (fractint) 2nd Question: New Hole Alice, This one took me a while, but it is a bailout problem. The fractal type you're using in this image is a Newton-type, which uses a different bailout test than some fractals. With a Newton-type, you test to see if your iterated value homes in on a "solution", rather than if it heads to infinity. So typically you check to see if a value gets very small, instead of gets very big. In this case, the bailout is at the end of the function, and is .0001--change it to .0000001 and your hold will close right up. BTW, this was one of my favorite images of yours. Very interesting. Here's the corrected FRM (just paste this in your own FRM file): newCGNewtonSinExp (XAXIS) { z=pixel: z1=exp(z) z2=sin(z)+z1-z z=z-p1*z2/(cos(z)+z1) .0000001 < |z2| } Better yet, replace .0000001 with p3, and you can adjust the bailout to taste. Just remember that in Newton types, a *smaller* bailout means points iterate longer; with other types (where points shoot off to infinity), *larger* bailouts means points iterate longer. Ain't fractals fun? :) Damien M. Jones / temporary sanity designs / http://www.emi.net/~dmj/ dmj@emi.net / my gallery: http://www.geocities.com/SoHo/Lofts/2605/ - ------------------------------------------------------------ Thanks for using Fractint, The Fractals and Fractint Discussion List Post Message: fractint@xmission.com Get Commands: majordomo@xmission.com "help" Administrator: twegner@phoenix.net Unsubscribe: majordomo@xmission.com "unsubscribe fractint" ------------------------------ Date: Wed, 20 Aug 1997 19:49:44 -0400 From: Sylvie Gallet Subject: (fractint) 2nd Question: New Hole Hi Alice, >> I'll bet since you see there's an attached par file you can guess what= >> happened. I added a second parameter (p2) to the formula and the pars. Now, you can use p2 to reduce the size of the hole but it will never disappear. - Sylvie FRM:CGNewtonSinExp-M (XAXIS) { z=3Dpixel: z1=3Dexp(z) z2=3Dsin(z)+z1-z z=3Dz-p1*z2/(cos(z)+z1) p2 < |z2| } Fractalj-m { ; Pulse ; by amkelley@freenet.columbus.oh.us reset=3D1950 type=3Dformula formulafile=3Dfractint.frm formulaname=3DCGNewtonSinExp-m passes=3D2 center-mag=3D-0.992426/0.931621/4.63121 params=3D1.4/1/.0001/0 float=3D= y maxiter=3D5000 inside=3Dbof60 logmap=3D5 distest=3D1000/71/80/60 colors=3D834pJE<43>A45945834723502301000<13>ehlhkoilokmp<9>WTL834<173>8= 34V\ 0BW0B } Pulse_zoom-m { ; amkelley@freenet.columbus.oh.us reset=3D1950 type=3Dformula formulafile=3Dfractint.frm formulaname=3DCGNewtonSinExp-M passes=3D2 center-mag=3D-0.950575/0.9244/46.3121 params=3D1.4/1/.0001/0 float=3Dy maxiter=3D5000 inside=3Dbof60 logmap=3D5 distest=3D1000/71/400/300 colors=3D834pJE<43>A45945834723502301000<13>ehlhkoilokmp<9>WTL834<173>8= 34V\ 0BW0B } - ------------------------------------------------------------ Thanks for using Fractint, The Fractals and Fractint Discussion List Post Message: fractint@xmission.com Get Commands: majordomo@xmission.com "help" Administrator: twegner@phoenix.net Unsubscribe: majordomo@xmission.com "unsubscribe fractint" ------------------------------ Date: Wed, 20 Aug 1997 17:21:57 -0700 From: "Jay Hill" Subject: Re: (fractint) 2nd Question: New Hole Well, so here is what I tried, but the spot does get very small if you make p2=1.e-12. What are the limits for these parameters. Is extended arithmetic available so we can zoom in? Looks like if it is, set p2 = (desired_spot_size)^^2. Here I go in with magnification 10^^11 and p2 = 10^^-24 and there is the spot. frm:newCGNewtonSinExp (XAXIS) { z=pixel: z1=exp(z) z2=sin(z)+z1-z z=z-p1*z2/(cos(z)+z1) p2 < |z2| ; p2 small like .0000001 or smaller, not zero } NewFract_Deepzoom { ; Pulse zoomed in ; by amkelley@freenet.columbus.oh.us reset=1950 type=formula formulafile=quest2.par formulaname=newcgnewtonsinexp passes=2 center-mag=-0.95145378834490950/+0.92132932950487430/1.703328e+011 params=1.4/1/0/0/9.999999999999999e-025/0 float=y maxiter=5000 inside=bof60 logmap=5 distest=1000/71/80/60 colors=834pJE<43>A45945834723502301000<13>ehlhkoilokmp<9>WTL834<173>834V\ 0BW0B } - ------------------------------------------------------------ Thanks for using Fractint, The Fractals and Fractint Discussion List Post Message: fractint@xmission.com Get Commands: majordomo@xmission.com "help" Administrator: twegner@phoenix.net Unsubscribe: majordomo@xmission.com "unsubscribe fractint" ------------------------------ Date: Thu, 21 Aug 1997 16:32:56 -0700 From: "Jay Hill" Subject: (fractint) Re: Fractint formulas Hi all, Since I would also like to know how to do these things with Fractint, I am pushing this out to the fractint list. For example, I want to color a point inside an MSet component according to its period or some other function of its position, c. And if it is outside, color it according to its iteration count and some other function of c and z[n]. Jay Forewarded message.... Subject: Fractint formulas Date: 21 Aug 1997 05:08:53 GMT From: "Phong" Organization:Coteric Continuum Newsgroups: sci.fractals I'm trying to write a formula that will bail out the normal way for some pixels, but will specify a specific color for ones that meet a certain criteria at some point during their iteration. Basically what I am trying to do is the exact same thing as the parameter inside=epsiloncross, except I want to specify how close the orbits must come to the axes (or some arbitrary points or lines in the complex plane) to bail out (and get colored a specific color rather than the iteration count). Right now, I'm keeping track of what iteration I'm currently on, and if I want to bail out to a certain color, I set a variable then bail out when the color wraps around to the right one. That is a total kludge, slows things down and is awkward. Is there a better way? If not, is it something that is planed for a future version of Fractint (i.e. a variable you can set to specify that a pixel should be a certain color rather than the number of iterations). -- Tom Schumm ---- http://www.ismi.net/~phong/ ----__/\__---- Phong -- -- v3.1 GCS/M d- s: a20 C$++>++++ W++(--) P++++ M-- \ / w---(++) -- -- UBLS$+++(++++) L+ t+* X+ b+ DI++++ G+ !r y? __/\__/ \__/\__ e -- - -- the Coteric Continuum -- O- -- phong@ismi.net / tgschumm@mtu.edu -- - ------------------------------------------------------------ Thanks for using Fractint, The Fractals and Fractint Discussion List Post Message: fractint@xmission.com Get Commands: majordomo@xmission.com "help" Administrator: twegner@phoenix.net Unsubscribe: majordomo@xmission.com "unsubscribe fractint" ------------------------------ Date: Thu, 21 Aug 1997 18:57:04 -0600 From: "Tim Wegner" Subject: Re: (fractint) Tru.c workings > Date: Mon, 18 Aug 1997 08:20:48 -0600 (MDT) > From: Jason Hine > To: fractint@xmission.com > Subject: (fractint) Tru.c workings > Reply-to: fractint@mail.xmission.com > Howdy all, > After a weekend of playing around with the little piece of C code that > comes with Fractint called TRU.C, I've determined the following (maybe!): > > 1) TRU.C is designed to read in the file ITERATES.TGA produced when > Fractint is given the "truecolor=yes" command; TRU.C first reads in the file > header (see next note!), then reads in the actual iterations data. There is a > subroutine, RGBMAP, which produces a new .TGA file containing color data instead > of iterations data. > > 2) There seems to be a problem reading in the header which looks like > this: > Who-knows-what... 12 bytes > xdots............ 2 bytes > ydots............ 2 bytes > Who-knows-what... 2 bytes > Max_iteration.... 4 bytes > > All these values are being read correctly except for ydots, which always > ends up being a large number, like 6147489 or such... > > 3) If you set a viewwindow size of 50x38 and create iterates.tga, then > not only is the ydots in the header corrupt, but the actual iteration data is > also incorrect. > > Can someone provide me with the supposed binary file format for the > iterates.tga file? Other suggestions welcome... I have another question, but > I'll post it in a separate email. Thanks, all! > > Jason (Iteration) Hine > > ------------------------------------------------------------ > Thanks for using Fractint, The Fractals and Fractint Discussion List > Post Message: fractint@xmission.com > Get Commands: majordomo@xmission.com "help" > Administrator: twegner@phoenix.net > Unsubscribe: majordomo@xmission.com "unsubscribe fractint" > > - ------------------------------------------------------------ Thanks for using Fractint, The Fractals and Fractint Discussion List Post Message: fractint@xmission.com Get Commands: majordomo@xmission.com "help" Administrator: twegner@phoenix.net Unsubscribe: majordomo@xmission.com "unsubscribe fractint" ------------------------------ Date: Thu, 21 Aug 1997 18:57:04 -0600 From: "Tim Wegner" Subject: Re: (fractint) Tru.c workings Hi everybody, I'm back from a quick dash to Baltimore to tack my son to the maryland Institute College of Art. It was a good trip to a part of the country that's new to me. I see the fractint list has been lively while I'm gone. I don't see many messages that that a response from me; thanks to various list members for answering questions. Jason Hine asked: > Can someone provide me with the supposed binary file format for the > iterates.tga file? I remind myself about this over the weekend. I just glanced at the code, but to much time has gone by for me to give you a quick answer :-) Tim - ------------------------------------------------------------ Thanks for using Fractint, The Fractals and Fractint Discussion List Post Message: fractint@xmission.com Get Commands: majordomo@xmission.com "help" Administrator: twegner@phoenix.net Unsubscribe: majordomo@xmission.com "unsubscribe fractint" ------------------------------ Date: Fri, 22 Aug 1997 00:32:43 -0400 From: "Damien M. Jones" Subject: (fractint) New Coloring Methods Hello all, In my first post to this list, I asked if anyone knew how to add new coloring options to FractInt, preferably using the formula parser (instead of recompiling the C code). Now I see some others are asking the same thing. I've done some playing around, and I found out how to do it. Let me give a little background, first. I've often wondered what the bof60 inside coloring option would look like applied to outside points. I got my first taste by fudging the distance estimator method (DEM) pixel percentage value. If you turn DEM on, you can set a percentage distance (relative to a pixel's size) such that if the point is within that distance to the set's inside, it is colored as an inside point, even though the point is outside. Well, I set this to a ridiculous number (32767) and suddenly *all* my points are colored like inside points. This produces some pretty neat pictures, and I've been moderately happy with the results. Except for three things. First, most of the inside coloring options aren't very tweakable. Second, because I'm relying on points being within 327 pixels of the set, sometimes I get points that don't quite reach, and get colored using the outside coloring method--and that really screws up poster-size prints. And third, it still only gives me a few more coloring options, only a couple of which (bof60 and bof61) produced really good results. So I thought of a way to use the formula parser to add new coloring techniques. Now, I'm sure some of you old hands and formula-writing already knew this trick, but it was quite a discovery for me. The basic trick is that just before you bail out, you figure out what your color should be, and make it your z value's real part. Then you can use FractInt's "real" coloring option to extract the color value. Of course, since the "real" coloring option adds in the iteration count too, you track that and subtract it out of the z value's real part; this cancels it out. What you're left with is a whole new avenue of coloring fractals. So, my original goal was implemented like this: Color01 { ; New coloring technique #1 (bof60 outside) ; outside = real: closest approach to p1 ; p2: color scaling factor ; p3: bailout z = 0, c = pixel, closest = 100000, point = 0, done = 2: z = fn1(z) + c done = done + 1 IF (|z-p1| < closest) point = z closest = |z-p1| ENDIF IF (|z| > p3) point = point - p1 z = |point| * p2 - done done = -1 ENDIF done >= 0 } The formula remembers the orbit point closest to p1 (so it doesn't have to be 0,0) and allows you to specify a color scaling factor and a bailout. (Set fn1 to the sqr function to get the M-set.) The color scaling factor is important; it lets you tweak the colors as you zoom in. And because the images don't rely on the distance estimator method to fake out the coloring routine, they're totally resolution-independent, and can be generated poster-size. The above formula only colors outside pixels. I wanted to be able to color inside pixels at the same time (as an option). Doing that is pretty easy--you just bail out one iteration early, so FractInt never thinks you've quite made it to an inside point: Color01i { ; New coloring technique #1i - same as #1, does inside too ; might want to turn off periodicity checking ; outside = real: closest approach to p1 ; p2: color scaling factor ; p3: bailout z = 0, c = pixel, closest = 100000, point = 0, done = 2: z = fn1(z) + c done = done + 1 IF (|z-p1| < closest) point = z closest = |z-p1| ENDIF IF (|z| > p3 || done >= maxit) z = |point-p1| * p2 - done done = -1 ENDIF done >= 0 } It's important to remember to turn off periodicity checking--since this formula basically fools FractInt into thinking all pixels are "outside", if you leave periodicity on, it might end up thinking some points are "inside" and not color them the way you want. bof60 colors based on the distance of closest approach to the origin. bof61 colors based on the iteration of closest approach to the origin. In one picture I generated, I wanted to plot the *angle* of closest approach to the origin. This was several months ago, before I had figured out this new trick, so I wrote a program to do the job. But this week I wrote a FRM to do the trick: Color05 { ; New coloring technique #5 (angle of closest approach) ; might want to turn off periodicity checking ; outside = decomp: angle of closest approach to p1 ; p2: bailout z = 0, c = pixel, closest = 100000, point = 0, done = 2: z = fn1(z) + c done = done + 1 IF (|z-p1| < closest) point = z closest = |z-p1| ENDIF IF (|z| > p2) z = point-p1 done = -1 ENDIF done >= 0 } Since writing angle-decoding formula stuff is obnoxious, I thought to take advantage of FractInt's decomposition option--which takes the final z value and colors based on its angle. So the above formula remembers the actual point of closest approach, and just before bailing out, resets z to this value--which FractInt then considers the "last value" and colors based on its angle. I have an inside variant of this, too: Color05i { ; New coloring technique #5 (angle of closest approach) ; might want to turn off periodicity checking ; outside = decomp: angle of closest approach to p1 ; p2: bailout z = 0, c = pixel, closest = 100000, point = 0, done = 2: z = fn1(z) + c done = done + 1 IF (|z-p1| < closest) point = z closest = |z-p1| ENDIF IF (|z| > p2 || done >= maxit) z = point-p1 done = -1 ENDIF done >= 0 } And here's a PAR that produces an effect similar to what I had to write a program for earlier: mandelc { reset=1960 type=formula formulafile=dmj.frm formulaname=Color05i function=sqr center-mag=-0.938416/0/0.6666667 params=0/0/4/0 float=y maxiter=256 inside=0 invert=0.5/-0.5/0 decomp=256 periodicity=0 colors=@bluegrey.map } I am still playing with this technique, and so far I've used it to produce some moderately spectacular images. I'll post some in my gallery soon. Some of the coloring techniques I'm using require one-pass rendering to work properly, so it's faster to download them than to generate them. Damien M. Jones / temporary sanity designs / http://www.emi.net/~dmj/ dmj@emi.net / my gallery: http://www.geocities.com/SoHo/Lofts/2605/ - ------------------------------------------------------------ Thanks for using Fractint, The Fractals and Fractint Discussion List Post Message: fractint@xmission.com Get Commands: majordomo@xmission.com "help" Administrator: twegner@phoenix.net Unsubscribe: majordomo@xmission.com "unsubscribe fractint" ------------------------------ Date: Fri, 22 Aug 1997 17:31:01 +0200 From: Jacco Burger Subject: (fractint) question & introduction Hello everyone, I recently joined the Fractint-mailing list and I am very surprised to discover that there are other people in this world that use Fractint and are dealing with the same problems as I am. I really learned a lot from it already. So I figured that if I could download all previous messages, I would probably have a lot of interesting stuff to learn from. Another reason to do this is that I am considering sending some questions to the list, and I hate to ask questions which already have been asked and answered. So I thought: how does one do this? I have send a message to majordomo=40xmission.com with =A1index fractint=A2 in the body. Very soon = I found a reply in my mailbox, saying: (begin) > -- >=20 > >>>> index fractint > .: > total 4 > drwxrwxr-x 2 domo domo 512 Aug 15 17:52 archive > drwxrwxr-x 2 domo domo 512 Aug 18 14:20 latest >=20 > ./archive: > total 826 > -rw-r--r-- 1 domo domo 208578 Aug 18 14:20 fractint.9708 > -rw-r--r-- 1 domo domo 41682 Aug 12 20:59 v01.n001 > -rw-r--r-- 1 domo domo 41916 Aug 13 17:50 v01.n002 > -rw-r--r-- 1 domo domo 52474 Aug 14 20:57 v01.n003 > -rw-r--r-- 1 domo domo 63586 Aug 15 17:52 v01.n004 >=20 > ./latest: > total 68 > -rw-r--r-- 1 domo domo 1726 Aug 15 17:55 001 > -rw-r--r-- 1 domo domo 2947 Aug 15 17:55 002 > -rw-r--r-- 1 domo domo 1281 Aug 15 17:55 003 > -rw-r--r-- 1 domo domo 1468 Aug 15 17:55 004 > -rw-r--r-- 1 domo domo 2133 Aug 16 11:19 005 > -rw-r--r-- 1 domo domo 4910 Aug 16 18:08 006 > -rw-r--r-- 1 domo domo 2457 Aug 17 18:53 007 > -rw-r--r-- 1 domo domo 1980 Aug 17 20:31 008 > -rw-r--r-- 1 domo domo 1334 Aug 18 04:20 009 > -rw-r--r-- 1 domo domo 2167 Aug 18 08:22 010 > -rw-r--r-- 1 domo domo 2499 Aug 18 08:35 011 > -rw-r--r-- 1 domo domo 1085 Aug 18 09:37 012 > -rw-r--r-- 1 domo domo 1988 Aug 18 14:20 013 > >>>>=20 >=20 >=20 (end) Well, ehemmmmm..... I guess this is some sort of directory list, but I still haven=A2t got a clue which files are available for downloading , = what they contain and what their names are. Can anyone help me out on this???? I also noticed it is a sort of custom to introduce yourself. Well, = allright then - if you read on don=A2t blame me if you get bored. Also please take = in consideration that english is not my native language and I am to lazy to get the dictionary. At the age of 12 I was drawing my first fractal, a Pythagoras-tree (some = of you might remember that in the old days drawing was a proces done by = people on flat white stuff called paper with a little stick called a pencil or pen). This tree had 1023 branches so you can understand it was a hell of a job. I used ink so one tiny mistake would spoil the whole image, and that is why I never managed to finish it. In those days the only computer I had ever seen was in a comic-book (no, I was born AFTER the war), but I already thought how wonderful it would be = if I had a machine that would be able to do the drawing. Seven years later I had my first computer, a Sinclair ZX81 with a RAM-memory of - hold your breath - 16 Kbyte (born after WHICH war?). The graphic screen was = something like 80 x 46 x 2, I am not sure, but it wasn=A2t much more. I could even = save my BASIC-programs on a tape=21 But I never came to make this machine draw = a pythagoras-tree for me. I had read an article about Mandelbrot fractals in some magazine at the library, and I thought: this is very exiting and I want to learn all about it=21 When I had access to a proper computer for the first time, and = learned some TurboPascal, I started making my own fractal-drawing program. The program was very slow, and I had only a four color CGA-screen available, but I managed to zoom in on the Mandelbrot-set. I also started thinking about how to build features like Fractint has now, something like solid-guessing and periodicity-checking. But I am not exactly a programming engineer and I realised this was going to be very complicated.= =20 And then one day... a friend of mine gave me a present for my birthday. It was floppy-disc with a copy of Fractint 17.2 for DOS. Yes, it was a cheap present because the program is freeware and even the disc was used, but until this day I am very grateful to him. In the mean time I had bought a 80286-computer and the machine must have calculated thousands of fractals by now. I wouldn=A2t have been surprised if one day the poor thing would = have put a message on its screen like =A1Are you nuts??? at the moment when I feeded him another 200 line batch-file. I have also spend many hours editing color maps to get good looking images, and I must say I managed to get some real nice ones. A few weeks ago I bought a new computer with a Pentium 166 MHz processor. It was a terrible shock to see how fast Fractint runs on this. I had a fractal that took more than 11 hours to compute on my old 80286, on the Pentium it only took 10 minutes......and there are still people complaining= that Fractint is SLOW???=21=21? My next plan is to buy a colour printer and put some of my images on that old fashioned flat white stuff. I am also planning to get an Internet-account at home and maybe put a gallery on a homepage. I=A2ll = keep you posted. Bye=21 Jacco Burger from Delft in The Netherlands you can e-mail me at Jacco.Burger=40BU.TUDelft.nl - ------------------------------------------------------------ Thanks for using Fractint, The Fractals and Fractint Discussion List Post Message: fractint@xmission.com Get Commands: majordomo@xmission.com "help" Administrator: twegner@phoenix.net Unsubscribe: majordomo@xmission.com "unsubscribe fractint" ------------------------------ Date: Fri, 22 Aug 1997 16:48:40 -0400 From: "Damien M. Jones" Subject: (fractint) PARs and FRMs My apologies to all who experimented with the FRMs and PAR that I posted last night. I should know better to write technical messages just before going to bed; I leave things out that are important. :-) For those who tried the FRMs and found they generated a blank screen, switch to floating-point math. Also don't forget to set a bailout; the default is zero, and the FRM doesn't substitute a different value if you forget to put one in. 4 is good to start, but depending on the point you've selected for p1, you might want to use something higher to prevent discontinuities. And if you don't set the color scaling factor (for the color01 formula) to something besides 0, you won't see the new coloring method. As for the PAR... well, boneheaded me didn't realize until after I posted it that it referred to a map file rather than include the colors directly. Color me stupid. I've attached a new PAR file below. mandelc { reset=1960 type=formula formulafile=dmj.frm formulaname=color05i function=sqr center-mag=-0.938416/0/0.6666667 params=0/0/4/0 float=y maxiter=256 inside=0 invert=0.5/-0.5/0 decomp=256 periodicity=0 colors=000<63>oozooznny<61>AACAACBBD<61>zzzzzzyyy<60>333 } And, for easier reference, here's the FRM file with the four formulae I posted last night (save as dmj.frm): Color01 { ; New coloring technique #1 (bof60 outside) ; might want to turn off periodicity checking ; outside = real: closest approach to p1 ; p2: color scaling factor ; p3: bailout z = 0, c = pixel, closest = 100000, point = 0, done = 2: z = fn1(z) + c done = done + 1 IF (|z-p1| < closest) point = z closest = |z-p1| ENDIF IF (|z| > p3) point = point - p1 z = |point| * p2 - done done = -1 ENDIF done >= 0 } Color01i { ; New coloring technique #1i - same as #1, does inside too ; might want to turn off periodicity checking ; outside = real: closest approach to p1 ; p2: color scaling factor ; p3: bailout z = 0, c = pixel, closest = 100000, point = 0, done = 2: z = fn1(z) + c done = done + 1 IF (|z-p1| < closest) point = z closest = |z-p1| ENDIF IF (|z| > p3 || done >= maxit) z = |point-p1| * p2 - done done = -1 ENDIF done >= 0 } Color05 { ; New coloring technique #5 (angle of closest approach) ; might want to turn off periodicity checking ; outside = atan: angle of closest approach to p1 ; p2: bailout z = 0, c = pixel, closest = 100000, point = 0, done = 2: z = fn1(z) + c done = done + 1 IF (|z-p1| < closest) point = z closest = |z-p1| ENDIF IF (|z| > p2) z = point-p1 done = -1 ENDIF done >= 0 } Color05i { ; New coloring technique #5 (angle of closest approach) ; might want to turn off periodicity checking ; outside = atan: angle of closest approach to p1 ; p2: bailout z = 0, c = pixel, closest = 100000, point = 0, done = 2: z = fn1(z) + c done = done + 1 IF (|z-p1| < closest) point = z closest = |z-p1| ENDIF IF (|z| > p2 || done >= maxit) z = point-p1 done = -1 ENDIF done >= 0 } If there are still problems with these, let me know. I've produced lots of pictures with these over the past week, so I know they work. But my posting of them (as history proves) might not. :) Damien M. Jones / temporary sanity designs / http://www.emi.net/~dmj/ dmj@emi.net / my gallery: http://www.geocities.com/SoHo/Lofts/2605/ - ------------------------------------------------------------ Thanks for using Fractint, The Fractals and Fractint Discussion List Post Message: fractint@xmission.com Get Commands: majordomo@xmission.com "help" Administrator: twegner@phoenix.net Unsubscribe: majordomo@xmission.com "unsubscribe fractint" ------------------------------ End of fractint Digest V1 #6 **************************** To subscribe to fractint Digest, send the command: subscribe fractint-digest in the body of a message to "majordomo@xmission.com". If you want to subscribe something other than the account the mail is coming from, such as a local redistribution list, then append that address to the "subscribe" command; for example, to subscribe "local-fractint": subscribe fractint-digest local-fractint@your.domain.net A non-digest (direct mail) version of this list is also available; to subscribe to that instead, replace all instances of "fractint-digest" in the commands above with "fractint". Back issues are available for anonymous FTP from ftp.xmission.com, in pub/lists/fractint/archive. These are organized by date.