From: owner-fractint-digest@lists.xmission.com (fractint-digest) To: fractint-digest@lists.xmission.com Subject: fractint-digest V1 #66 Reply-To: fractint-digest Sender: owner-fractint-digest@lists.xmission.com Errors-To: owner-fractint-digest@lists.xmission.com Precedence: bulk fractint-digest Thursday, January 8 1998 Volume 01 : Number 066 ---------------------------------------------------------------------- Date: Wed, 7 Jan 1998 22:00:25 -0800 From: "Jay Hill" Subject: (fractint) Piers Anthony and Cliff Pickover chat with people on the Internet Forwarded from sci.fractals: Piers Anthony and Cliff Pickover (author of many fractal books) will have a live chat with people on the Internet at http://www.barnesandnoble.com Thursday, January 8th at 7pm ET. Go to the Barnes and Noble site today to learn more and see photos. Piers and Cliff will be discussing their new highly-acclaimed science-fiction novel Spider Legs. - - - ------------------------------------------------------------ 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, 8 Jan 1998 12:42:06 -0000 From: Andrew Jenner Subject: (fractint) Ideas for future versions of fractint Here are some ideas I've had for the next versions of fractint. Some or all of these ideas might be impractical or impossible, so don't shoot me down with flames if I'm spouting rubbish. Previous versions of fractint have generated fractals by using several layers of program. These are the layers for escape time fractals: 1) Drawing mode algorithm (solid guess, boundary trace, 1-pass, etc.) - this gives a screen pixel 2) Zooming and panning (also rotation, stretching and skewing) - this transforms the screen pixel into a point in the complex plane. 3) Non-linear transformations (i.e. inversion) - this transforms that complex number into another one, which is used as the starting point for... 4) The iterated function - the outputs of which are: the iteration count what caused the calculation to stop (maxiter reached, bailout reached, or periodicity check kicked in) the last value of z These are used in... 5) The colouring algorithm, i.e. the inside=, outside=, decomp=, biomorph=... options. This gives a number 0...255, which is used to dereference the... 6) Colour map - which is used to colour the pixel specified by the drawing algorithm (this is done by the hardware). The next versions of fractint should follow the same plan, but the addition of truecolour modes means that colouring needs to be done a little more inventively. Rather than have (5) output a single integer, it should output a real number. This can be used to look up the colour in a colour map which, rather than consisting of discrete list of colours, consists of a short program which accepts a single real number and outputs three real numbers (red, green and blue). For example, if all three outputs were set to be the same as the input, you would get a smooth transition between black and white. Of course, old fashioned maps would still be supported, but there should be an option to make the transition between adjacent colours in the map smooth, so it can be used for continuous potential images. This way, assuming that the screen can be redrawn fast enough truecolour images can be colour cycled. Better still, some colouring algorithms might output several real numbers, each of which can be assigned a different colour map. These could be colour cycled seperately, with different speeds and directions. Then you also need an algorithm to specify how to combine the outputs from the colour maps. For more information on this technique, have a look at the amazing program "Chromazone for windows" (written entirely in assember). It isn't a fractal program (at least the version I have isn't) but it does have a fantastic colouring facility. Similar layering can be worked out for orbit fractals, .IFS and Lsystem - the main difference is that (2) and (3) are done backwards, and in reverse order (an ordered pair of real numbers are used to work out a screen pixel). For some fractals, e.g. bifurcations, you need to do (2) and (3) both forwards and backwards, because you scan along the screen but plot vertically. Diffusion fractals and any other odd ones do their own thing. Now, this is a Windows 95 program, which means a) We can have images at any size - full screen mode or window mode. For window mode we need to specify the size of the window. For both modes we need an option to specify how many pixels are calculated for each screen pixel. No need for disk-video modes (the image must be stored in memory anyway, in case the window is hidden) - for images larger than the screen, you can view the image at half size, say, as a "preview" for what the "disk-video mode" is doing (of course, windows handles whether memory or swap file is used for this). I hope this makes sense. b) Printing is much neater using the windows printing engine. A "print preview" should be implemented which supports scaling and centering (see the equivalent feature on Paint Shop Pro). (The print option in the DOS version of fractint, if it is to continue (and it should) might as well be taken out, and maybe put into another program. I'm sure there are much better dedicated programs for printing .GIF files.) c) Background fractals: fractint can take over all the idle cycles, thus making background fractals without slowing down your computer (much). d) Screensaver. If background fractals are enabled, presumably this would simply show the fractal currently being drawn. If no fractals are being drawn, it could just display a (colour cycled) slideshow of saved images. e) Output file format: you can have as many as you like using output format filters. Ditto input format filters. f) Mouse controlled 3D rotation to set parameters for setting 3D parameters. For background fractals, it would be useful to be able to have a "batch list" which you can add your pars to so it can calculate all the images, one after another. You should be able to promote or demote images within this list, and maybe keep multiple lists. Fractint should have a built in editor for .PAR, .FRM, .L, .IFS and .MAP files (especially with the new, complicated colour maps), plus any other new similar filetypes. Also features to help organize these files, such as having a hierarchy of "subdirectories" inside .PAR files, which can contain other images and subdirectories, etc. Email-ready output: output a .PAR file which contains all the information needed to generate the image - colour maps, formulae, everything. That suggestion will be popular with this list. When reading such an email-ready file back into fractint, the necessary bits and pieces can be seperated out and put in the right places in the hierarchy, thus eliminating wasted disk space if the reciever already has the right bits. Drawing mode algorithms. Instead of having 1-pass, 2-pass, solid guess seperately, you could have a single "n-pass" mode where you specify "n" and an option to "calculate all pixels" or not. A new algorithm could be to calculate the pixel which is furthest away from all the other pixels already calculated. This isn't as hard as it sounds, and would give a quick preview of the image. It could also be adapted to be an improvement on the solid guessing algorithm. Maybe there could be an option to recalculate pixels where it looks like the guessing algorithm has guessed wrong, like a paint program (just wave your mouse over the pixels you want to recalculate). More options and user defined options for stage (3) above. Currently fractint only has inversion. Maybe this could be extended to twisting, z+1/z transformation... How about the equivalent of a "zoom box" for these (of course, for inversion, it would be what is outside the box that is actually drawn, rather than inside). Put built in types on the same level as formulas. It seems that most fractals on this mailing list are of type=formula. You should be able to write type=. Lots more colouring algorithms, including field lines (which I think has been cracked now - the thread on field lines, decomposition and texture mapped fractals can be read at "http://homepages.enterprise.net/berrypark/andrew/field.txt". I gather these lines are at right angles to the isolines of the continuous potential colouring (which is not be the same as the electrical potential if the inside of the set contains uniformly distributed electric charge, something else altogether). If so, using these and cont-pot you can do all sorts of exciting colourings, including texture maps. User defined colouring algorithms, to make .FRM files simpler (at the moment we seem to have .FRMs for iteration formulas, .FRMs for colouring algorithms and .FRMs for each possible combination. User defined bailout conditions. Now, if (the most significant bits of) the output from (5) is stored rather than the colour, you can change colour maps without recalculating. Better still, if you store the output from (4), you can change the colouring algorithm without recalculation, although this could be impractical because of the amount of data that needs to be stored, especially if you start storing lots of values of "z" rather than just the last one. Store a flag for each pixel saying whether it has been calculated or not, and you can switch drawing modes without starting all over again. Symmetry could be smarter - currently switching on decomp switches off symmetry. Is this really necessary? Higher level functions - it would be nice if some of these could be built into fractint, but if not it would also be nice if user defined functions could be written and used in "fn" types as well as formulae (and even new colouring, colour map, bailout algorithms). Algorithms for working out the optimum values of bailout and maxiter? Wow, this file worked out to be a lot longer than I thought it would be, and fractint is already a fantastic program. Sorry I've been lurking so long - I wish I had more time for messing about with fractals. Sadly that may be one problem that cannot be solved with a new version of fractint. Andrew Jenner 1st year undergraduate, Mathematics with Physics, Queens' College, Cambridge - - - ------------------------------------------------------------ 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, 08 Jan 1998 06:11:37 -0800 From: Wizzle Subject: Re: (fractint) Ideas for future versions of fractint Andrew..... Clearly you are headed toward a PhD in Fractint........good suggestions!!! In the mean time can we just have a Winfract that does everything the current dos version does??? Angela At 12:42 PM 1/8/98 -0000, you wrote: >Here are some ideas I've had for the next versions of fractint. Some >or all of these ideas might be impractical or impossible, so don't >shoot me down with flames if I'm spouting rubbish. > > >Wow, this file worked out to be a lot longer than I thought it would >be, and fractint is already a fantastic program. Sorry I've been >lurking so long - I wish I had more time for messing about with >fractals. Sadly that may be one problem that cannot be solved >with a new version of fractint. ><> >Andrew Jenner >1st year undergraduate, Mathematics with Physics, >Queens' College, Cambridge > - - - ------------------------------------------------------------ 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, 8 Jan 1998 11:43:35 -0500 (EST) From: Ian Kaplan Subject: Re: (fractint) RE: fractint-digest V1 #63 OK, I'll have one last go at this (rather off-) topic: Godel's theorem, which Penose presumably cites, states that in no system (meaning set of rules and axioms) complex enough to describe itself (ordinary formal logic could be an example of such a system) can it be shown that the system both complete (meaning that all true statements about the system can be proven in the system) and consistent (meaning that no paradoxes can be derived in the system.) This means, for example, that it cannot be proven that any of our systems of mathematics is 'perfect'. This proof is not particularly recent; it occured in the first half of the century, though I find I can't remember when... As for the paper mentioned stating that Turing machines can compute uncomputable problems, I'd be fascinated to hear this... Ed Avis made a comment which I believe to be erroneous regarding the ability of Turing machines to model physical law. Nondeterministic sets of rules are quite within the understanding of automata theory; they are described as "nondeterministic finite automata" and are certainly modelable as Turing machines. The universe might not be so describable were it not finite (possessing a finite number of rules and a finite space) but it is generally felt, though certainly not proven, that it is. I suppose that when I refer to a finite number of laws or rules, I am implying that the universe is "algorithmic" in the sense that all of its behaviors are describable by some set of rules. It is certainly concievable that this is not so, but I would have to say that such a claim falls outside of anything that can be considered scientific yet... - - - ------------------------------------------------------------ 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, 8 Jan 1998 10:00:15 -0700 From: Tim Riley Subject: Re: (fractint) 3d curses From: Andrew Fraser Subject: (fractint) 3d curses Date: Wed, 07 Jan 1998 00:13:25 +0100 > I've been enjoying generating lots of the wonderful fractals posted on > the list, especially the fantastic 3Dballs series with variations. This > all came to an abrupt end when I downloaded somebody's files infected > with the dreaded "=3D" disease. Full of confidence, I found and replaced > all of these, and now none of the ball fractals work either! Guess I > must have stripped out too many 3D's. - ---------------End of Original Message----------------- Instead of removing all "3D"s, replace all "=3D"s with "="s, since there are valid uses of the string "3D" in some pars and frms. ----------------------------------------- | Tim Riley | | Institute for Telecommunication Science | | National Telecommunications and | | Information Administration | | US Dept. of Commerce | | Boulder, Colorado | | E-mail:triley@its.bldrdoc.gov | | Per favore spenga le Sue scarpe | | davanti abbandonare il nave spaziale. | ----------------------------------------- - - - ------------------------------------------------------------ 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, 8 Jan 1998 12:01:20 -0500 From: Sylvie Gallet Subject: (fractint) mndbulbs Hi All, Here is a new bunch of =3D3D's... - Sylvie - ------------------------------------------------------- Here's a formula you may find useful. The PAR file will produce a black and white image of a "hollow" Mandelbrot set. fastmand { ; Copyright (c) Paul W. Carlson, 1997 ; This formula performs NO iterations if the point is ; in the large cardoid (the "lake") or the large ; period-2 bulb. ; float=3Dy must be used. ; z =3D 0 c =3D pixel pi3 =3D pi / 3 bailout =3D 0: ; ;***************************************************** ; Check if point can be in either circle or cardoid ;***************************************************** aic =3D abs(imag(c)) IF (aic < 0.6495190528) rc =3D real(c) ;************************************************* ; Check if point is in the circle ;************************************************* IF (aic < 0.25) delta =3D sqrt(0.0625 - aic * aic) IF ((rc > (-1 - delta)) && (rc < (-1 + delta))) bailout =3D 1 ENDIF ENDIF ;************************************************* ; Check if point is in the cardoid ;************************************************* IF (rc > -0.75 && rc < 0.375) IF (rc > 0.25 && aic < 0.2165063508) t =3D abs(acos(0.5 * (1 + sqrt(3 - 8 * rc)))) ELSE t =3D abs(acos(0.5 * (1 - sqrt(3 - 8 * rc)))) ENDIF card_i =3D 0.5 * sin(t) * (1.0 - cos(t)) IF ((t > pi3 && aic < card_i) || (t <=3D pi3 && aic > card_i)= ) bailout =3D 1 ENDIF ENDIF ENDIF ;***************************************************** ; If point is in neither we have to iterate ;***************************************************** IF (bailout =3D=3D 0) z =3D z * z + c ENDIF (bailout =3D=3D 0) && |z| < 4 } mndbulbs { ; Copyright (c) Paul W. Carlson, 1997 reset=3D1960 type=3Dformula formulafile=3Dfm.frm formulaname=3Dfastmand passes=3D1 center-mag=3D3.665e-006/-8.65e-007/0.6666663 float=3Dy maxiter=3D2000 inside=3D0 colors=3D000zzz<253>zzz } - - - ------------------------------------------------------------ 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, 8 Jan 1998 12:01:17 -0500 From: Sylvie Gallet Subject: Re: (fractint) =3D disease Hi Les, >> >>CompuServe mail, using New Mail, is MIME compliant, which performs >> the encoding and decoding of your mail. >> >>As long as the recipient is using MIME compliant mail, then this >> should not be a problem. So, we just need a MIME compliant list server.... - Sylvie - - - ------------------------------------------------------------ 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, 8 Jan 1998 12:01:29 -0500 From: Sylvie Gallet Subject: Re: (fractint) Yet another unreadable formula from Paul Carlson Peter wrote: >> Looks lovely on my machine. Do you have a collection of pars and frms >> somewhere? Peter Check out my web page: http://ourworld.compuserve.com/homepages/Sylvie_Gallet/homepage.htm - Sylvie - - - ------------------------------------------------------------ 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, 08 Jan 1998 10:28:03 -0700 From: Rich Thomson Subject: Re: (fractint) =3D disease In article <199801081202_MC2-2E90-41C9@compuserve.com> , Sylvie Gallet writes: > So, we just need a MIME compliant list server.... Err... no, the list server doesn't have anything to do with this really. It isn't the source of the =3D's nor is it interfering with their removal. The problem lies in having everyone use MIME compatible software. The compuserve software generates Content-Transfer-Encoding: printed-quotable when it isn't needed. If that software were doing something more reasonable (like assuming US ascii, or at least letting you set it as a preference), then there wouldn't be a problem. I've looked back over all my fractint messages that are plagued with =3D disease and they all come from compuserve. - -- ``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: Thu, 8 Jan 1998 18:26:23 -0000 From: "Bagpuss" Subject: Re: (fractint) Client-server Fractint >If I can process the fractal as a background application while reading my >email or whatever, I don't really care that my processing time goes from 10 >minutes to 5. And by next year I'll cut that time anyway by going from my >100mhz pent to 233 or something anyway. hmm, I DO care if the render time goes up... Win95 has a horiffic processor overhead, and while this is fine on a fast machine (most programs dont use very much processing power) for processor intensive work (like fractal and 3d rendering) DOS is better by far - up to 2.5 times faster on my machine. Until microsoft come up with ways to multitask DOS apps I think I'll stick with the good old reliable DOS mode, even if they came up with an easy way to allocate system resources Id be happier. I dont see any reason why I should have a fast machine yet only use a fraction of its power just because thats what Uncle Bill has decreed. There is another issue as well...ie graphics modes. Windows is very strict about screen resolutions, and would you really want to restart your machine every time you wished to change resolution? - - - ------------------------------------------------------------ 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, 8 Jan 1998 18:29:24 -0000 From: Edward Avis Subject: (fractint) Client-server Fractint I've been reading some responses to my suggestion that Fractint v20 should run on a client-server model. The consensus seems to be that it would be nice, but might be difficult to implement (especially in DOS) and that other features should come first. In some ways I think that client/server Fractint could actually be easier to develop than the present form of the program. The reason for this is that the server part would be the same across all platforms (DOS, Windows, Unix, whatever). The platform-specific code could be isolated into the client part. This would mean that Winfract and Xfractint would stop lagging behind the DOS version in terms of features. If a new formula or new calculation engine is introduced, it gets added to the server, which is the same on all versions of Fractint. (Of course, some would also require enhancements to the client, but most would not). There is a TCP/IP library for DJGPP (and threads packages), so if the DOS version will be compiled with DJGPP (as seems likely), portability to DOS should not be a problem. Similarly there are good Winsock implementations for Windows. On the other hand, I'm not one of the Fractint developers, so their opinions on what would be "difficult", and what features are higher priority than others, carry more weight than mine. - -- Ed Avis epa@datcon.co.uk http://members.tripod.com/~mave/index.html - - - ------------------------------------------------------------ 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, 8 Jan 1998 13:43:05 -0500 (EST) From: Ian Kaplan Subject: Re: (fractint) Client-server Fractint The client-server model is clearly elegant, but definitely introduces overhead. Would anyone with a really deep knowledge of communications protocols, processor cycles, and fractint's current design care to comment on how much overhead we're talking about? I mean, if calculation is 80% of fractint's drawing time on a typical Pentium, and drawing is 20%, would some sort of multithreaded client-server design add on another 20% of time, or 600%, or what? I suppose 'multithreaded' is something of an assumption; to some extent you could separate out the two sets of routines (calculating and drawing) without really slowing down the code... hmm. Don't know enough about fractint's design, do I. - - - ------------------------------------------------------------ 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, 08 Jan 1998 11:53:10 -0700 From: Rich Thomson Subject: Re: (fractint) Client-server Fractint In article , Edward Avis writes: > In some ways I think that client/server Fractint could actually be easier > to develop than the present form of the program. The reason for this is > that the server part would be the same across all platforms (DOS, Windows, > Unix, whatever). The platform-specific code could be isolated into the > client part. The server part also has platform-specific code. For instance, the Pentium has 80-bit floats, but IEEE doubles are 64-bit floats. The new 64-bit SGI machines might have a "double" that is 128 bits. Since fractint switches from floating-point arithmetic to arbitrary precision arithmetic when the floating point code runs out of precision, it needs to know the sizes of the various floats and doubles in order to know when to switch. > This would mean that Winfract and Xfractint would stop lagging behind the > DOS version in terms of features. Xfractint doesn't lag the DOS version at all. They use the same code base. However, winfract lags because it is a completely different kind of port to windows that doesn't share the code the way the xfractint port does. Fractint should be made more portable so that contributions to the engine part of fractint can be shared on all platforms, modulo the compatability issues discussed above. > There is a TCP/IP library for DJGPP (and threads packages), so if the DOS > version will be compiled with DJGPP (as seems likely), portability to DOS > should not be a problem. Similarly there are good Winsock implementations > for Windows. If any networking is introduced into fractint, it should be TCP/IP at this point. There was a time when IPX seemed important in the PC networking world, but except for corporate LANs IPX seems to be giving way to TCP/IP in most markets, certainly it looks that way in all new networking stuff that's created. > On the other hand, I'm not one of the Fractint developers, so their > opinions on what would be "difficult", and what features are higher > priority than others, carry more weight than mine. I think people generally work on what they think is neat :). Fractint is currently going through some teething pains as it hits puberty and outgrows its 16th bit. Developer discussions I've seen say that the roadmap for the future looks like: 32-bit flat DOS application, more integrated ports between X11-unix/Windows/DOS, 24-bit rendering. I think you'll eventually get what you're asking for out of fractint, but you're going to have to wait a while if you want other people to do the coding :) - -- ``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: Thu, 08 Jan 1998 12:01:25 -0700 From: Rich Thomson Subject: Re: (fractint) Client-server Fractint In article <199801081843.NAA08040@force.stwing.upenn.edu> , Ian Kaplan writes: > The client-server model is clearly elegant, but definitely introduces > overhead. Would anyone with a really deep knowledge of communications > protocols, processor cycles, and fractint's current design care to > comment on how much overhead we're talking about? I mean, if calculation > is 80% of fractint's drawing time on a typical Pentium, and drawing is > 20%, would some sort of multithreaded client-server design add on another > 20% of time, or 600%, or what? How the duty cycle splits between display and computation depends very much on all the inputs. For deep zooming, distributed processing becomes a huge win because computing each pixel takes so long. For the default M-set screen, the computation can probably be completed locally before you can even assemble all the distributed pieces into a picture over a modem. Typical factors affecting the performance would include communication speed, processor speed, and scatter/gather overhead. I'm not aware of any way of optimally allocating the resources such that you can always guarantee that you produced the image in the shortest time possible. However, its probably reasonable to build a heuristic that does the right thing 90% of the time. I like to use fractint with showdot=b/20, but I have noticed that when the processing time per pixel is short, that this actually slows down the display. I think it ends up spending lots of its time bitblting the little triangle bitmap in and out of the video memory compared to the time it spends computing the pixel. However, for very slow (i.e. deep zoom or complicated formula) renderings the showdot=b/20 overhead is minimal compared to the total rendering time. Also, with slower renderings, I want to know where the computation is more than with a faster rendering. - -- ``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: Thu, 8 Jan 1998 14:09:23 -0500 From: Sylvie Gallet Subject: Re: (fractint) =3D disease Rich Thomson wrote: >> The compuserve software generates Content-Transfer-Encoding: >> printed-quotable when it isn't needed. But if the message doesn't have any accented letter, what makes it "think" printed-quotable is needed? >> I've looked back over all my fractint messages that are plagued with >> =3D3D disease and they all come from compuserve. Not all!!!! There are also messages from Justin A. Kolodziej, Linda Allison, Cesare Gianuzzi, Gedeon Peteri and Jay Hill. Cheers, - Sylvie - - - ------------------------------------------------------------ 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, 08 Jan 1998 14:06:56 -0600 From: "Paul N. Lee" Subject: Re: (fractint) Client-server Fractint Bagpuss wrote: > > There is another issue as well...ie graphics modes. > Windows is very strict about screen resolutions, and > would you really want to restart your machine every > time you wished to change resolution? > Microsoft has some useful items for downloading from their site. They are usually referred to as "Power Toys". One of these is a small executable to be put into your StartUp path, called "QuickRes". It will stay in your System Tray and allow you to instantly alter your monitors resolution without rebooting. It has always worked fine for me without any problems. P.N.L. - - - ------------------------------------------------------------ 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, 8 Jan 1998 13:52:57 -0700 From: Tim Riley Subject: Re: (fractint) Sylvie Gallet exhibit at MOCA: Museum of Computer Art From: davides Subject: Re: (fractint) Sylvie Gallet exhibit at MOCA: Museum of Computer Art Date: Wed, 31 Dec 1997 20:56:18 -0500 To: fractint@lists.xmission.com >A new exhibit at MOCA: Museum of Computer Art features some >30 fractals by Sylvie Gallet at http://www.dorsai.org/~moca. "URL not found on this server"...; using Netscape. Is this viewable only w/IE or is there another problem? - ---------------End of Original Message----------------- This problem happens when you are using an e-mail program that is part of a web viewer or linked to a web viewer, and the URL (as above) appears as a link. When the web viewer croaks, make sure that any superfluous characters (usually punctuation, such as the concluding period in the example above) aren't included in the URL. If they are, edit them out of the URL in the web viewer's location window. ----------------------------------------- | Tim Riley | | Institute for Telecommunication Science | | National Telecommunications and | | Information Administration | | US Dept. of Commerce | | Boulder, Colorado | | E-mail:triley@its.bldrdoc.gov | | Per favore spenga le Sue scarpe | | davanti abbandonare il nave spaziale. | ----------------------------------------- - - - ------------------------------------------------------------ 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, 08 Jan 1998 14:22:44 -0700 From: Rich Thomson Subject: Re: (fractint) Client-server Fractint In article <34B531DF.3492@Worldnet.att.net> , "Paul N. Lee" writes: > It will > stay in your System Tray and allow you to instantly alter your monitors > resolution without rebooting. This is only true if you don't change the depth of the pixels. So if you switch from 1280x1024 256 colors to 1024x768 16-bit color, you will still have to reboot. - -- ``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: Thu, 08 Jan 1998 14:24:12 -0700 From: Rich Thomson Subject: Re: (fractint) Sylvie Gallet exhibit at MOCA: Museum of Computer Art In article , Tim Riley writes: > This problem happens when you are using an e-mail program that is part > of a web viewer or linked to a web viewer, and the URL (as above) > appears as a link. When the web viewer croaks, make sure that any > superfluous characters (usually punctuation, such as the concluding > period in the example above) aren't included in the URL. If they are, > edit them out of the URL in the web viewer's location window. You can solve this potential problem by enclosing the URL in , as in . The enclosure has been proposed as a way to embed URLs safely within other text files and have them be easily recognized and extracted by programs. - -- ``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: Thu, 8 Jan 1998 13:24:18 -0800 From: "Jay Hill" Subject: Re: (fractint) =3D disease Sylvie wrote >> I've looked back over all my fractint messages that are plagued with >> =3D disease and they all come from compuserve. > Not all!!!! There are also messages from Justin A. Kolodziej, Linda > Allison, Cesare Gianuzzi, Gedeon Peteri and Jay Hill. Me?!?! I checked v01.n034 through v01.n055. All my posts from ehill1 are fine! We'll see about this one from nosc.mil. Use Carlson's formula, see http://home.san.rr.com/jayrhill/CARLSON.PAR 4058_36_1 { ; Copyright (c) Jay Hill, 1998 reset=1960 type=formula formulafile=rast.frm formulaname=Astroid_Mset passes=t center-mag=0.3396842133902784/0.0646134485573786999/298482.0 params=0.016/0.3/8/30 float=y maxiter=3000 inside=253 outside=summ colors=000fOz<28>I0Kz0f<28>O08z88<28>O00zW0<28>c40zz0<28>aG00z\ R<28>0C40zz<28>0CCGGz<28>00O000<10>000z88000000 savename=4058_36_1 } - - - ------------------------------------------------------------ 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 #66 *****************************