home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!uwm.edu!rutgers!ub!dsinc!satalink!bert.tyler
- From: bert.tyler@satalink.com (Bert Tyler)
- Newsgroups: sci.fractals
- Subject: Re: Fractint 17.2: Intere
- Message-ID: <1398.1004.uupcb@satalink.com>
- Date: 2 Jan 93 15:45:00 GMT
- Reply-To: bert.tyler@satalink.com (Bert Tyler)
- Organization: Datamax/Satalink Connection * Ivyland, PA (215) 443-9434
- Lines: 28
-
- (discussing the zoom limit of Fractint...)
-
- PH>Is there any way to increase this limit? ...or do I have to start from
- PH>scratch and write my own version of fractint?
- PH>Just how many bits of fraction does fractint use?
-
- If you zoom beyond the limits of Fractint's integer math algorithms, Fractint
- quietly switches over to floating-point math. Its precision limits
- are essentially those of the Intel architecture at that point, and are
- based on its ability to store the coordinates of successive pixels (in
- 'double' format) using different values. (I'm on vacation and checking my
- mail on a laptop that doesn't have a "C" compiler, so I can't dig
- up the exact limit at the moment.) To get around that limit,
- you would have to rewrite Fractint (including its asembler routines)
- using an alternate format - or use an Xfract port on a machine which has
- more significant digits in its floating-point format. The latter is probably
- *far* easier.
-
- PH>I also find a limit of 1000 on iterations to be rather small.
-
- I'm confused - Fractint's current iteration limit is 32,768 (it would have
- been 65536 had I not been stupid enough to use signed integers for a field
- that never goes negative - maybe we'll fix that someday).
-
- Bert Tyler (bert.tyler@satalink.com)
- ---
- . DeLuxe./386 1.25 #343sa . Did you expect mere proof to sway my opinion?
-
-