#Morphed Letters
20 messages in this thread
Is there any simple way to turn a letter into a grid suitable for morphing
with Ripple?
Mitch
Sure is… use the Create/Object/Tessellate function to increase the
density of the mesh. The Face-Center option might be more appropriate than
the Edge option, but you can try both and see which one works best for the
model.
– G
Do I have to loft the letter, then boolean it with the grid, or can I just
tesselate the lofted letters?
Just loft the letter and then Tessellate it. Boolean's should only be used
if you need to sculpt an object or drill a piece out of it.
– G
Gary I am still confused (and I did read the manual <g>).
Specifically what I want to do is make a letter like the one in the
M_WAVE1.3DS example. It is the letter M made up of a grid of cubes. I have
spent hours trying to do this and nothing seems to work. There is no
example for how to do it in the manual either. You can't just tesselate a
letter, if you do that you get a bizarre surface to it. And as the next
message points out, its not clear how to use booleans to get the
intersection of a set of grids and a letter. HELP.
Mitch
I think I've finally figured out what you're talking about. The file
you're referring to is MELT_1.3DS… not M_WAVE1.3DS… isn't it? If you
want to make text out of a fine "grid", like the letter "M" in MELT_1.3DS,
all you've got to do is turn off the Optimize button in the Lofter's Make
Object dialog. As long as you've got as many interpolated path steps along
your path of extrusion as you want grid interpolations in your text, you'll
be fine (the default of 5 ought to work perfectly for you). Plus, when you
loft the object, make sure the Tween button is on, so the interpolations
actually happen.
It sounds like you need to read the section on object "optimization" on
page 6-101 of the Reference manual as well as the discussion about "The
Graphic Components of Mesh Objects" on pages 6-12 through 6-15. Plus,
Tutorial #4 goes into quite a bit of detail about how to loft objects with
or without interpolated steps, and the use of the optimization button (page
4-1 through 4-16, with highlights on 4-14 and 4-15 relating to
optimization).
– G
>
I think I've finally figured out what you're talking about. The file
you're referring to is MELT_1.3DS… not M_WAVE1.3DS… isn't it?
>
Opps. Ding. I must have snapped a synapse. Yes I meant MELT_1.3DS.
>
If you want to make text out of a fine "grid", like the letter "M" in
MELT_1.3DS, all you've got to do is turn off the Optimize button in the
Lofter's Make Object dialog.
>
Now I understand what is going on. You must also have Tween on for this to
work. What still puzzles me is the cryptic sentence on page 6-11..When you
convert a model into a mesh object using the Objects/Make command, the
rezulting objec t does not consist of splines, but rather of
interconnecting triangular faces of three vertices and three sides each. No
matter how many times I look at Figure on 6-12 I can't see where the little
triangle is your refering to.
BTW I had a major breakthrough in understanding when I was trying to
duplicate the MELT_1.3DS example. What I never got was that the morphed
version of the letters are more like "targets". I thought from using MWAVE
and RIPPLE that you had to use every morphed object to get it right. But in
your melt example you use just two of the morphed letters (2 and 4 I think)
and just ignore 0, 1, 3, 5). Once I saw that it dawned on me that 3DS was
filing in the frames between these targets and I could just put the morphed
object 2 on frame 12 and object 4 on frame 24 and have the whole thing. In
you example you have another morphed object at frame 56, but you only
render to 24, is that last one needed?
if you ever go over the manual you might want to have the reader try this
exercise. After setting up the morphed letter, try adjusting the itmes in
the dialog box and watching what happens, as well as moving the morph
locations to different places then 12 aznd 24. Also undestanding that all
the morphs have a parent object without the numbering was important to
understand.
Thanks Gary, this is very educational for me.
Regarding the description on page 6-11, turn on Display/Geometry/All Lines
in the 3D Editor to see the triangular faces of which all objects are made.
Regarding the morphing, I'd have to ask again: did you _really_ read
through the tutorials? From beginning to end? The morphing tutorial that
begins on page 14-12 clearly only uses a few keyframes to morph a column.
In addition, on page 14-20, is the statement: "You're going to morph the
first column into the other columns. Remember, it will look like the first
column has _become_ the other columns, but it always remains the first
column with only its vertices keyed to other positions." Then, on the
following page, I repeat: "We've said it several times, but we'll say it
again: The bent column is still the first column. Its vertices have been
moved to be the same as column01. If, for example, you were rendering this
and each column had a different material assigned to it, the bending column
would appear with the same material because, although its vertices are
moved, its material is the same."
– J
Yes I read the tutorials. While I think the product is great I don't think
the manauls are as good as they could be…they are too short. They lack
reinforcement to make sure you understand the vital elements. They don't
go into enough depth. They lack examples. They lack good explanatory
illustrations. I don't think its frair to just say "read the manuals" and
it will all be clear. There is a crying need for a book here (that is my
business ya know) that goes into more depth. BTW I am sure if you took a
survey you find a great deal of agreement with my points.
Actually, in the last couple of years, you're the first person who has ever
said a negative thing about the manuals. I'm not kidding. And the truth is
that we've gotten dozens of comments about how great the 3DS docs really
are. They're CERTAINLY not perfect. It would be great if they had lots
more illustrations, 4-color printing, and were 2 or 3 times longer.
(although at over 300 pages, they're certainly not short)
And you're right! That's where book publishers are supposed to come in and
say "this looks like an opportunity for more information."
Regarding taking a survey, I'm sure it would be very informative. I'd be
happy to see any other folks on the forum give us their opinion about the
quality of the 3DS documentation.
– G
Personally, I though the 3ds2 Docs were great. When I finished them, I felt
I had a good solid understanding of how the program worked. In my opinion,
they were the clearest and most understandable computer manuals I've ever
come across, and I've been through my share. The only reason I would buy a
full fledged book on 3ds2, is if it were on tips and tricks from the people
who use it, work-arounds n' stuff, things I would have never though of or
taking hours of my time to reinvent the wheel, so to speak. I don't suppose
there is a newsletter floating around somewhere? I didn't have any problems
with the manuals.
Gary,
I think the the 3DS docs are probably the best manuals I've ever worked
with — far superior to most others in layout and very well written (and a
great binding).
I believe that all manuals have inherent limitations and because 3DS is
such an enormously complex program those inherent limitations become more
apparent with 3DS than with other programs. One problem arises because 3DS
is really four different programs each with different menus, concepts, and
procedures. The total amount of information one needs to keep active in his
or her brain ram while working in 3DS is substantial. I often forget how to
do something I've done a few days or weeks ago. And sometimes when I sit
and read through the manual while rendering, I find all sorts of
interesting and valuable information which I'm sure I've read before but
have either totally forgotten or they are items which had no meaning to me
when I first read them because I was still then struggling with some
simpler concept or procedure and wasn't yet ready for that advanced
information.
I really like the short suggestions you have throughout the manual and I
think the little "how to's" you provide (like the "This is how to clone a
selection set of objects" on page 7-98) are very, very valuable. I would
like to see more of such features. One section which badly needs such a
feature is in the chapter on batch rendering. In addition to the long list
of parameters, it would be so helpful to have had several examples of
typical command lines using many of the parameter switches. A simple line
stating, "If you wanted to render a file named "horse.3ds" at 640 x 480
with low anti-aliasing and using a custom palette (etc) your command line
should look like this . . . " would have saved me hours of frustration.
[More]
[Continued]
But the central problem (to me) with the manuals is not the manuals. I
think you and Jack forget that sometimes people just don't get what they
need from the doc even if the information is in there. You guys are much
too defensive about any criticism or implied criticism about your docs and
you just plain don't like stupid questions. I understand that, believe me
— I'm that way too. But I think that in the software support business,
you've got to be able to deal with stupid questions — it's part of the
job. When I call Aldus to ask SQ's about Pagemaker, I'm not told to RDFM;
they just politely and patiently me how to do what I need to do. It's very
nice. What you guys need is some genteel person to handle the SQ's -someone
who doesn't have those (very justifiable) protective, maternal feelings
about your docs.
I do agree with Mitchell Waite that it is not fair to expect everyone to
understand everything just because a description is provided. There are
some sections which would benefit from more reinforcement and depth. It may
be that you guys know too much to be able to understand how baffling many
of the concepts are for us common folk — especially at first. But to be
honest, with the six pounds (?) of docs you provide, I don't know if its
fair to expect any more. There is certainly room for an aftermarket "tips
and techniques" book.
Alec Jason
Mitch,
Everyone please excuse me for butting in, but I've been following this
thread for some time now and feel something has to be said.
While I totally agree that a third-party book which focuses on "tricks &
tips", work-arounds, specific how-to's and the like is a good idea, I must
take issue with your characterizations of the 3DSr2 documentation.
As a user since it's it's original release (I was saving for TOPAS at the
time), and an active member of the San Diego 3DS user's group (I personally
field anywhere up to 25 phone calls a week for technical support from
fellow users who are not on CIS yet), I feel I am in a position to comment
responsibly on the content of the docs.
It is a widely held opinion that the 3DS documentation is of the HIGHEST
caliber, quite possibly the _best_ ever written for software on the
personal computer platform. Of the many, many users I have come in contact
with over the last few years, there have been virtually NO COMPLAINTS about
the documentation at all. No, it doesn't tell you _everything_ you can do
with the software, but it supplies a comprehensive basic understanding of
_every_ operation 3DS is capable of.
Yes, it is complicated, yes sometimes you have to follow it _very_ closely.
Yes it can become a little tedious at times. But the close follower will be
rewarded with the knowledge to do incredible things!…
continued…
..continued
Mitch, please accept this message with the good will with which it is
intended. I honestly believe that if you had _followed_ (not just read) the
tutorials step-by-step, and _understood_ the reference manual, nearly all
your recent questions would have been answered. Hey, "it's in there" —
really!
With best regards,
Alan Iglesias
Along this same line suppose I want my first letter to remain whole until
say the 12th frame and then start its morph. I thought I could accomplish
that by just copying all the keys in the frame 0 to frame 12. But if I do
that I notice my original letter actually grows a bit rather then staying
still, as if it is still trying to morph. I found that if I copied just the
morph key to keys 1 thru 11, it accomplished the goal. But I don't
understand why.
Mitch
It's bad practice to lock the morph keys by copying the first key to all
the other keys. You're not only reducing the number of interpolated frames
that way, but you're also blowing any future opportunity to rescale the
overall time of the animation without having to screw around with all those
keys.
The reason why the morph started before frame 12 is because the default
splined motion (for all keys) causes a little bit of undershoot and
overshoot to happen unless you compensate for it. That applies to Position,
Rotation, Scaling, Lighting, AND Morphing keys. This is explained in the
reference manual on pages 8-33 through 8-36, and there's an entire tutorial
(#12) about dealing with the same issue (as it applies to Position and
Scaling). Morphing is no different.
The correct way to start a splined motion precisely on a key is described
in the Note about Key Bias on page 8-36 of the reference manual. And as it
says in the second paragraph of page 8-34, "If you want completely linear
motion (such as for a machine), set the continuity to 0…" Of course, if
you want this to affect ALL keys (not just the one at frame 12, for
example), you also turn on the G button beside it to make the continuity
adjustment global.
Simple! You only have to be a mutant to do it!
– G
Mitchell (famous) Waite,
If you want 2d letters: I've turned letters into grids by means of a
boolean intersection with a 2d GRID (100/20 x 100/20 x 0/0 with a letter 90
units high.) I had to loft the letter first, then intersect with the grid,
then delete all vertices except the desired plane. This gave me a 2d
letter made up of a rectangular grid. If you don't need rectangles, you
can tessellate an planar (unlofted) letter extracted directly from the
shaper (CREATE/OBJECT/GET SHAPE, I think…)
If I didn't loft the letter to give it some thickness before the boolean
(i.e. planar letter with planar grid), 3DS would crash every time with a
reboot.
JKJ (forever in your debt, you C)
Thanks Jordon for the reply. I tried lofting a letter and applying the
manta ray IPAS function but it didn't work and I don't know why. I also
tried using the Boolean technique you described but did not try deleting
the grid. Jack says I don't need to use booleans.
Thanks
Hope you don't mind me butting in, but I follow all of these threads in
my never-ending search for knowledge…
The boolean method I described will leave you with a grid on the top
face of a 3d letter only. Retaining the top face only will give you a 2d
grid.
As per your message (28197) about the letter M, I don't seem to have
M_WAVE1.3ds but if you mean MELT_1.3ds, you can duplicate the big M as
follows:
In Shaper, create the desired letter outline
In Lofter, set PATHS/STEPS to 10 and SHAPE/STEPS to 10,
choose OBJECT/MAKE, and turn Optimization OFF and Tween ON
If you Optimize or leave Tween off, you won't get all those little
intermediate steps that look like cubes on the sides of the letter.
Note thate this still will not give you the FACE of the letter made up
of a rectangular grid. I've found no other way to do this than to resort
to the booleans. As you've seen, both tessellation methods give lots of
*interesting* triangles on the surface!
JKJ (never met a challenge I didn't like)