home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!portal!cup.portal.com!blackbelt
- From: blackbelt@cup.portal.com (Ben - Williams)
- Newsgroups: comp.sys.amiga.graphics
- Subject: Re: Trying to find the perfect morpher
- Message-ID: <69774@cup.portal.com>
- Date: Wed, 18 Nov 92 12:46:55 PST
- Organization: The Portal System (TM)
- Distribution: world
- References: <1992Nov14.204633.20305@newshub.ariel.yorku.ca>
- <1992Nov15.034518.26357@mnemosyne.cs.du.edu> <69612@cup.portal.com>
- <1992Nov18.070137.21238@mnemosyne.cs.du.edu>
- Lines: 71
-
- Ben Scott writes:
-
- "a 244K Addendum file? That's way beyond reasonable. I will look forward to
- the new manual provided my client decides to pay for any upgrades."
-
- Reasonable is one of those slippery, individually defined terms. You are
- welcome to your opinion, of course. We think it IS reasonable,
- considering the price of the product. That is not to say that we like the
- situation. If Imagemaster was priced where it should be for the features
- it provides, it would be financially feasable to do a new manual on a
- regular basis - but we believe that the Amiga market absolutely will not
- support that kind of pricing.
-
- "Yes, I realize that [bug was easy to get around] as well. But with bugs
- like that, I can't help wondering how they got through the testing phase?
- With new features coming out every few weeks, IS there a testing phase? (OK,
- loaded question, I know)"
-
- Sure it's loaded. :^) Yes, there is a testing phase; we have several
- versions ``cooking'' at one time. Some things get tested more than
- others, and furthermore, some things are added VERY late in the game and
- so are tested minimally. The HAM defaults got by us simply because no one
- here uses HAM as a default - we like the register modes better. Note that
- the bug ONLY exists based upon a particular default configuration, of
- which there are many thousands - it's not a fully testable situation, in
- short.
-
- "I do not expect a file called "addendum" to contain the only mention of
- probably half the major features in a software package. I have this funny
- idea that it should be possible to at least find the word "morphing"
- somewhere in the manual of a software package that advertises morphing."
-
- The addendum is part of the manual. Your failure to treat it that way is
- not in accordance with the organization of the package we sold you - your
- choice, again, but your implied assertion that the morphing is not
- documented is unsupportable. It *is* documented, there ARE mentions of it.
-
- [lotta Morph+ stuff deleted - see Ben's msg if curious]
- "Gee, this is turning into the mini-review I said I wouldn't do..."
-
- Yeah... looks that way, doesn't it. :^)
-
- "I'd recommend you add that [vector preview] to Imagemaster if you haven't
- already (I couldn't find mention of it anywhere)."
-
- It's possible we'll do that - it's not a very high priority, though.
- Imagemaster can already generate an image-based morph preview in just a
- few minutes on a 3000-class machine. There are much higher priority items
- in the queue first (you DO want us to test this stuff don't you?)
- <ducking and running>
-
- "Also, I had a lot of trouble keeping the (non-morphing) background stable
- while the foreground faces were warped - even placing dozens of non-moving
- points around the background, and outlining the faces with them as well, I'd
- still get distortion even inches away. (it was a curtain with lots of
- vertical folds)"
-
- Hmm. You should be able to get a high degree of isolation by adding edges
- to the points surrounding the areas to be controlled, and then
- surrounding the morphing region with an enclosed, edged, IShape that does
- not move. Edge features are v21+, I'm not sure what version you have, btw.
-
- "Perhaps I missed something... that would allow you to draw a freehand SOLID
- line separating the morphing area from a background, and assuring (better, at
- least) the confinement of the warping effect?"
-
- Nope - we don't have a freehand tool, but you can place points and then
- connect them with edges, which should provide the same capability in the
- end result.
-
- Ben (Williams... we could get confusing here... :^)
-