#3D-Express soft wear
7 messages in this thread
I use 3D-Express.
I am under sub-contract with a company that Synthonics(the developer of 3D
Express) has contracted to produce a CD-ROM of meshes. I, along with other
sub-contractors like myself, have been hired to produce 3D wireframe meshes,
for a 300+ wireframe CD-ROM that's supposed to be realeased sometime around
christmas with 3D Express (which will then be called WireFrame Express).
FIRST OFF: Don't buy 3D Express. Not YET. I consider the program still in
it's infancy. Truely, 3D E can make some models that 3D Studio could only
dream about. It isn't anything at all like 3D Studio. It's strictly a Windows
modeling program. And it can only model things that you can take a picture of.
For instance (and this is a slightly abreviated example) take 2 pictures of a
friends face, scan those pictures into your computer (from prints, or have them
developed to Photo-CD which I do), load them into 3D express, plot points all
around the contours of one-half of the face, connect 3 points at a time to make
"faces", mirror that one half of the face(your friends face) to make a whole
face, export DXF, load into 3D Studio and wa-la!
The program is extremely simple to use. The hardest part to learn is how to
take the pictures correctly, which the documentation fails to explain how to
do. As far as I know the documentation is being re-written. It HAS to be.
It's really useless. The sub-contractors were all offered a 2 session
training class to learn 3D Express.
I don't know if I answered your question. I didn't mean to go on and on. If
you want some more details of how it works, and the model building process
involved, I'd be glad to tell you what I know about it.
Anyway, I like it because it's really easy to use, and we get paid top dollar
for our meshes. 🙂
But don't buy it yet, because you wouldn' t know how to use it, and you'd get
really mad and then throw it away, and you'd punch yourself because you paid
395$ for it.
-Paul
>> The hardest part to learn is how to take the pictures correctly, which the
documentation fails to explain how to do. <<
I have this program. Could you elaborate on just how the pictures should be
taken?
– ej
Ernie,
Have you tried taking any pictures yet?
Maybe if you can tell me what you know I can give you some usefull hints that
I've learned in class and by trial and error.
Here's a few things to remember, though.
Almost any 35mm(except disposable)camera with an adjustable lens or
interchangable lenz will do.
Don't use auto-focus (if your camera has it).
The longer the lens length the better (100mm to 200mm is best) (I use 90, which
works fine. 50mm should be a minimum).
Be sure your calibration unit is visible in both (or all 4 photographs if the
object isn't symmetrical).
-Paul.
>> Have you tried taking any pictures yet?
Maybe if you can tell me what you know I can give you some usefull hints that
I've learned in class and by trial and error. <<
[Red faced] No. Not yet. It's on my "to do" list. Been on it for awhile. I
just couldn't pass up the chance to ask a pro while the thread was hot.
>> Be sure your calibration unit is visible in both (or all 4 photographs if
the object isn't symmetrical). <<
4 photos? In my version you can only display 2 at a time. Do you have a newer
version?
– ej
Ernie,
Depending on how much of an object you want to model you need between 2 and 8
pictures.
In the example of modeling a face you only need 2 photos. (And yes, you only
work from 2 photos at a time). Since a human face is symmetrical (for the
most part) you only need to model one half, and afterwards mirror that
wireframe when you go to export it DXF. Draw a line of reference points down
the center of the face, and all around the contours of that half (so you can
easily find the exact same points in both Left and Right photographs).
Let's say you want to model the whole head. In that case you would have to
take 2 photos for the face (of one half the face, from the center of the nose
to the ear) and 2 photos for the back of the head (of the center to the back of
the ear). You can model these 2 objects seperately, and then you can use the
Merge function, as you output DXF, to connect the 2 pieces together, creating
an entire head. This method works exactly the same for modeling a car, or any
other symmetrical object.
If you want to model an entire NONsymmetrical you need 8 pictures. 2 for the
front left half, 2 for the front right, 2 for the back left, and 2 for the back
right.
There. Was that a long-winded enough way to answer your picture question?
-PAul
>> Was that a long-winded enough way to answer your picture question? <<
Very much so. Thanks a way-lot.
On another topic: Do you know if Synthonics ever completed their IPAS version
of 3D Maker for 3DS?
– ej
>> On another topic: Do you know if Synthonics ever completed their IPAS
version of 3D Maker for 3DS? <<
I don't know.
Synthonics has a BBS in LA, where they run all of their 3D Express
subcontractor business from, and they also have 3D Maker stuff on there.
(818)865-0496
-Paul