Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!ucsd!dog.ee.lbl.gov!george.lbl.gov!antony From: antony@george.lbl.gov (Antony A. Courtney) Newsgroups: comp.graphics Subject: Re: Marching Cubes Message-ID: <6794@dog.ee.lbl.gov> Date: 5 Sep 90 21:56:52 GMT References: <5300038@ux1.cso.uiuc.edu> Sender: usenet@dog.ee.lbl.gov Organization: Lawrence Berkeley Laboratories Lines: 59 X-Local-Date: Wed, 5 Sep 90 14:56:52 PDT In article hallettJ@gemed writes: >Uh, sorry. Since Marching Cubes is a proprietary algorithm to GE >Medical Systems, it is doubtful you would find the full description or >source in the public domain. I see. And keeping marching cubes a "trade secret" really helps make it a standard, well accepted approach to doing volume visualization. As someone else pointed out, all people care about is an implementation of the PUBLISHED algorithm. Whatever hacks GE makes to this are of little interest to anyone else unless they publish what they are and why they are good. Furthermore, here's a thought you can take back to your marketing trolls who tell you that 'Our PROPRIETARY version is undoubtadly better': The person who posted the request for an implementation of the algorithm did so from an academic instition, meaning he is either a student or researcher. And students or researchers tend to go out into the working world eventually. Now, had you posted an article with a complete description of the algorithm, a description of what enchancements GE has made, and source to this version(although with no support for this product), this person would have probably been left very pleased by GE's kind offering, and would, when consulted for some purchasing decision in future, be able to recommend that they purchase their software from GE(so as to have it be supported software rather than just a 'freebie'), with full confidence in the quality of the product, because he has actually used it for some time. After your rather arrogant article, with unsupported conjecture about it being better than other implementations, he is more likely left with a bad taste in his mouth about GE's attitude, which is(as your article implies) "Ours is a better implementation, but we're not going to tell you how or why, you just have to take our word for it and pay us cash money to use it." Since he is clearly not going to pay for the software(and why should he?! there are several perfectly adequate implementations out there for FREE), your article only LOST you a future sale. Not to mention all of the rest of the people who read this newsgroup who will hopefully feel the same way. >More probably, you would find a facimile >algorithm - same results, but more slowly. "More slowly?" Do you have any factual basis for this claim?!! If so, I would be very interested to see a complete analysis of every marching cubes implementation available compared with your own and some FIGURES indicating the substantial speed difference. >-- > Jeffrey A. Hallett, PET Software Engineering > GE Medical Systems, W641, PO Box 414, Milwaukee, WI 53201 ~antony -- ******************************************************************************* Antony A. Courtney antony@george.lbl.gov Advanced Development Group ucbvax!csam.lbl.gov!antony Lawrence Berkeley Laboratory (415) 486-6692