#Architectural Animations
31 messages in this thread
Our office has been creating shaded presentations for architectural projects.
We are now trying to do rendered images, with average success so-far. What is
good for creating walk-thrus of buildings? Autovision vs. Accurender or
anything else? What Cd's are available for additional materials to attach to
our building models? Is there anyway to add titles to the presentations?
I saw a project that someone had created and it had ceiling fans turning in an
animation presentation format – how'd they do that? Last but not least where
can I get more info. about 3-d presentations? any good books?
Thanx
Jeff
>> Autovision vs. Accurender or anything else? What Cd's are available…
The answer you are most likely to get from members of this forum is 3D Studio.
As a complete 3D modeling and animation package, it will fit your needs
extremely well. As far as I know, Autovision and Accurender produce decent
stills and walkthrus, but for "real" animation (turning fans, opening doors,
setting sun, lens effects) 3DS is the only way to go.
CD's? Check out Autodesk's Texture Universe, Ketiv's Blocks and Materials for
starters.
Books? If you opt to go with 3DS there are TONS of books available at the
major chains. Look for the Tips & Tricks series, 3DS Applied, and Inside 3DS.
– Dave
Jeffrey:
That's a lot of questions!! There are plenty of others on here who are more
knowledgeable and experienced who will jump in, but I'll give ya my two
pennies' worth as well:
>>What is good for creating walk-thrus of buildings? Autovision vs. Accurender
or anything else?
I prefer 3D Studio (surprise!) for my architectural images and animations, but
either of the two products you mention can do _simple_ walk-thrus, but 3D
Studio is the most flexible. AutoVision and Accurender run directly inside
AutoCAD, which may be simpler for you, but at some sacrifice of speed. 3D
Studio is a separate product.
>>What Cd's are available for additional materials to attach to our building
models?
I am not sure about AutoVision and AccuRender, but 3D Studio includes a CD with
many additional texture maps. There are also a few third party CDs available
with texture maps and meshes.
>>Is there anyway to add titles to the presentations?
We use Corel Draw to add titling, but you can also use desktop publishing
packages or Animator Pro (now upgraded to Animator Studio) – it depends on what
type and quality of output you need!
>>I saw a project that someone had created and it had ceiling fans turning in
an animation presentation format – how'd they do that?
They turned on the switch (ha-ha – just kidding). This can be done with 3D
Studio, but I don't think it is possible with AccuRender or AutoVision.
Good luck on your quest!
Bob Keefer, AIA
If you are going to have anything, other than your camera and target point,
moving in your animation, then you definitely need a product like 3dStudio.
AcuRender and AutoVision only support camera and target point movement. The
other members' suggestions for resources are all very good. Ketiv's Blocks and
Materials CD has numerous ready-to-use models and materials.
You didn't indicate whether you had purchased AccuRender or AutoVision as yet.
If you are going to be using 3dStudio for your animations, then I would
recommend AutoVision over AccuRender. AutoVision uses the same material
libraries, lights, and camera definitions as 3dStudio (surprise! They're both
AutoDesk products), and will export your entire environment as a ready-to-use
.3ds file.
If you have any questions, either post them here or call me at my office. I
work for an AutoDesk dealer in San Diego named Imperatives, Inc. We deal in
CAD, animation, video editing and production, and multimedia / presentation
hardware and software.
Good luck! It's a fun and lucrative industry to be in at this time!
I've done a couple of building walkthroughs with 3D Studio. I can't think of
any other program that I've seen on an Intel-based system that I would rather
use for this kind of work. Most of what I do is animated illustrations of
equipment – I did a couple of animations to help explain how a recent
microgravity experiment worked. Some of the techniques are a little different
in a walkthrough, particularly the lighting but I would recommend 3D Studio for
any kind of 3D animation. It is perfect for dropping in little details like the
ceiling fans, opening doors, etc.
Chris
– Chris Johnston, Written 11-May-1995 @ 23:09:32
Using: Windows NavCIS PRO 1.62
Chris,
May I quote you for one of our Promotional Ad's?
Murph'
3DS Dealer
Sure.
What did I say? <g>
Chris
– Chris Johnston, Written 15-May-1995 @ 20:55:56
Using: Windows NavCIS PRO 1.62
I'm Looking……I'm Looking.
I've got a hard copy of your post #169710 but I can't find it on my computer.
You mentioned that you can't think of any other porgram that you've seen on an
Intel-based system that you would rather use for this kind of work.
Murph'
Oh, yeah – I remember that now and I see no reason why you can't use it as a
personal opinion. (The people at work are touchy about anything that sounds
like an endorsement that has their name attached to it)
Chris
– Chris Johnston, Written 17-May-1995 @ 21:01:18
Using: Windows NavCIS PRO 1.62
I hadn't planned on using any AEC Firms Names just Chris Johnston
Architect if that's OK? You see the Architects in Wisconsin are
still debating on whether or not it's a good idea to get off the
boards and use this new fangled computer thing! <BG> Murph'
Chris Johnston is OK, but I'm not an architect. I'm a chemist, actually, but
I'm also my group's animation and visualization specialist.
Chris
– Chris Johnston, Written 19-May-1995 @ 20:44:09
Using: Windows NavCIS PRO 1.62
Maybe "animation & Visualization specialist"…Sound Good?
Murph'
Sure, that sounds good!
Chris
– Chris Johnston, Written 22-May-1995 @ 20:29:28
Using: Windows NavCIS PRO 1.62
>>in a walkthrough, particularly the lighting but I would recommend 3D Studio
for any kind of 3D animation. It is perfect for dropping in little details like
the ceiling fans, opening doors, etc.
Chris
Ok, you touched upon a particular question I have. It may be basic, but I
learn by "leaping around" a lot instead of sequentially. What I was wondering
is if there are any IPAS's to keep regular motions active trhoughout the
animation. E.g. keeping a ceiling fan turning in a circle at a specified rate
throughout your animation, no matter how short or lon it may be, and no matter
how you might scale the various segments. So far, the way I've been doing such
things has been to manually turn them – but if I were going to do a roomful of
fans for an extended period, it would be nice to specify their repetetive
motion and let the KXP have at it. Any ideas?
A quick question – how do you usually do your modeling? DO you use Acad and
import it in?
I also would be interested in hearing more about your microgravity work, it
sounds fascinating, provided you're free to discuss it. You can send me mail
if you don't want to tie up the forum.
Thanx!
>> it would be nice to specify their repetetive motion <<
Use Object/tracks/loop, I believe. It will do nasty things to your
keyframes, like make them invisible in the track info box, but it will give you
repetitive movement/rotation.
So, if you want 360 degrees in 10 frames, put in that motion, then loop
the tracks for that object.
Paul S.
>> Use Object/tracks/loop, I believe. It will do nasty things to your
keyframes, like make them invisible in the track info box, but it will give you
repetitive movement/rotation.
Ahh, much oblige. I hadly really read about/played around with that one
particular feature, I will have to look it up and give it a try.
There is a way to set up a loop in the keyframer which looks like it should
work if you set it up right.
I usually do my modeling in 3D Studio, but for complex models I use Acad.
In general I would rather model in 3DS to avoid the problems with importing a
geometry, but Acad can do some things a lot better than 3DS. The booleans in
Acad are better and I like the snaps and relative coordinates.
Chris
– Chris Johnston, Written 15-May-1995 @ 21:09:07
Using: Windows NavCIS PRO 1.62
I thought you might have done your modeling in Acad 😉 It seems to be the
general concensus. Frankly, it would be hard to dream of doing complex
architectural elements without relative coordinates, geometrical entities,
object snaps, or even being able to key in a given coordinate if you like. As
for importing geometries, that can be a problem, but the booleans in 3DS take
so much fussing to get around their inherent problems it would appear that
often importing geometries cause less work than making the booleans work in
3DS.
And… I can't imagine many of my architectural peiece being created without
booleans. They could be done, but you'd have MANY associated pieces for each
simple element. Complex fenestration would be almost impossibly complex with
ye olde boolean subtraction. SImple fenestration isn't so bad, but the stuff
*I* come up with…. heh. -JK
Jerome
Excuse me for jumping in here, but I've found that for complex architectural
elements 3DS is a much more powerful modeler.
The deform tools in the Lofter, in conjunction with multiple shapes on the
path, let you create things like fancy chair legs and backs, picture
frames/moulding, plumbing fixtures, etc., quickly and efficiently with no face
flipping required.
Granted, if you need accurate arc radii you can do it better in acad, but will
1/100 inch snap resolution, lines can be drawing very accurately.
Try experimenting with the deform tools and let me know what you think.
I have an extensive acad/autoarch background, BTW, so I know how easy it is to
resort to a comfortable workplace<G>
Have fun.
TedB
-Ted Boardman
!^NavFont01F0007MGHHI582899
>> Excuse me for jumping in here, but I've found that for complex architectural
elements 3DS is a much more powerful modeler.
Oh, ok… I might have misrepresented myself in earlier comments. I believe I
recognize the true capacity for 3DS' lofter to create complex ELEMENTS within
architecture. Indeed, I have often sung the praises of 3DS' lofting
capabilities, despite the limitations of accuracy in the shaper. INdeed, for
archiotectural ELEMENTS, the snap/grid is usually sufficient. The limitations
I encounter in modeling in 3DS are usually in the development of the building's
superstructure and envelope. Dealing with the complex angles, precise angular
measurements, precise curves, and sheer number of precise walls gets to be
problematic in a shaper thats only accuracy is a snap feature. Though the
lofter is powerful (and even the shaper iss powerful for more ORGANIC and
freehand forms), I find even a basic, simple sshareware package to be
necessary. The other problem I encounter is once I loft such shapes done in
CAD, the trouble with booleans makes it exceedingly difficult to add
fenestration, chamfers, and envelope character which couldn't be easily pieced
together by lofting. Hence the desire to model in a more CAD accurate
envirnment with functional booleans.
>> The deform tools in the Lofter, in conjunction with multiple shapes on the
path, let you create things like fancy chair legs and backs, picture
frames/moulding, plumbing fixtures, etc., quickly and efficiently with no face
flipping required.
Agreed there. I wouldn't want to do WITHOUT 3DS for such mouldings, legs,
woodwork, trim, columns, furniture and fixtures. That's why I wish that ACAD
could perform booleans on imported 3DS meshes that 3DS itself can't handle.
What's more, is then the 3DS lofted elements could then be combined with the
ACAD superstructure of the building, etc… I don't know, perhaps ACAD 13 can
do this, I just know that 12 cannot do much with a polyface mesh in the way of
solid modeling.
Heh – more reason for someone to generate a program to build CSG or BREP solid
databases out of 3DS meshes! 🙂
>> Granted, if you need accurate arc radii you can do it better in acad, but
will 1/100 inch snap resolution, lines can be drawing very accurately.
True. Actual ACCURACY isn't a problem, as long as you're dealing with objects
of such aspect ratios as to fill the screen at one time. When you're dealing
with walls 100' long and 6" thick, then have a few 10-15 degree walls @ 50' or
so, then a half dozen arcs of 25' radius that vary from 6" to 2' in thickness,
you wind up having to zoom in and out so much it's very hard to keep track of
absolute coordinates, unlike being able to draw a line at a certain angle and
have it STAY that angle while you zoom in and calulate where you want it to
intersect.
Yess, it indeed can be done, but the more complex and irregular your building,
the more work this can generate. 3DS' shaper is GREAT for certain things, not
as great for others.
>> Try experimenting with the deform tools and let me know what you think.
I have an extensive acad/autoarch background, BTW, so I know how easy it is to
resort to a comfortable workplace<G>
—
I agree, as well, that my difficulties are compunded by not being as accustomed
to working with snap only and relying on deform tools, etc. A *LOT* can be said
for proficiency. Not that I believe greater proficiency will cure are woes,
but it can only help. You have very valid points about the elements, as I said
earlier. I appreciate your input, and would welcome any other wisdom about
modeling in 3DS you might have time to share. We might have a lot to talk
about.
Thanx! -JK
Thank you, Jerome!
I think a good discussion of the pluses and minus' of various modeling options
is good to get newer forum members thinking.
One important thing I didn't put into the formular is, if you are developing
an architectural model for presentation purposes only it is easier to do more
in 3DS, as opposed to building the model in conjunction with working plans. I
have tried to keep the two seperate, but that depends on office procedure and
deadlines<G>.
Thanks again for taking the time for a well thought out response.
TedB
-Ted Boardman
!^NavFont01F0007MGHHHBE215D
Yes, I agree that good modeling discussions are quite desireable here!
As for my particular uses, I am developing a practice where I can do
ssome presentation work for architectural clients (myself included
once I get my license), but more often I seek to create the work
directly from concept sketches to be used as sets/backdrops, some VR,
and even the occasional video game (i.e. prototyping the scenes in
3DS and exporting the finisshed models/images). I also intend to
explore future trends in architecture taking into account social,
technological and economic trends, combining them with a little
philosophy and imagination. There's less of a market for the latter,
but they can begin as my personal art, and perhaps find a market
either as a form of art, released videos, or as part of various shows
on futurism and science.
But – before I elaborate too much – my primary uses will not be
leaning towards developing working drawings alongside the model, but
rather using the model as an exploratory tool either towards the end
of a finished image, or towards design development progressing
towards future stages. Take this as you will for how you might see
me doing the majority of my modeling. I just might add – what I dream
isn't conventional or all that orthagonal!
🙂
Thanx again for the input! -JK
>> not be leaning towards developing working drawings <<
In that case, I'd think it would be easier and more productive to model
everything in 3DS to avoid face normal and welding problems.
I'd like to see a model or jpg sometime if you have something available.
Thanks for discussion.
>> In that case, I'd think it would be easier and more productive to
model everything in 3DS to avoid face normal and welding problems.
Unless there's something MAJOR I've missed in doing architectural
models in 3DS, they're NEVER *EASIER* than other packages, at least
for the models I create. Yes, the welding and face normal problems
can be headaches, but the SHAPING, at least, needs to be from another
source. Perhaps the answer iss a happy comprimise in using CAD
shapes to plug into the lofter for some decent finished meshes.
However, the problem with booleaning in some fenestration,
grooves/channels, fillets, chamfers, other voids and articulations,
and etc., remains a problem. Not an eternally insurmountable one,
but one nonetheless.
>> I'd like to see a model or jpg sometime if you have something available.
I will have to shoot a couple off to you sometime. there are a
couple things in progress, but I'd like to tweek them a little before
showing them (they just need a little sweat to make them look
great). To make part of a demo reel, and as a learning experience, I
am taking various projects from architecture school and creating
models of them, and eventually animating them. There are two
projects which are posing some interesting problems that should be a
good learning experience to solve. One is a very kinaesthetic
approach to a museum outside the Acropolis. A complex and marvelous
project in my opinion, but modeling *IT* isn't the main concern. It's
the acropolis and the surrounding topography/cityscape this musuem is
immeshed in that's proving the be the challenge. If this were just
one building displaying one angle of the acropolis, a well placed
still would suffice. The surrounding buildings might do as abstract
massing should I get enough data, but it's STILL a plethera of work
for a project that's just an architectural demo I'm not getting paid
for. The Acropolis itself is a great challenge. Just in getting its
topography and massing data without forking over major $$$. And it's
the A. and the topography of the scene that is the celebration of
this facility. How would you go about a project that requires one to
move through a city of many buildings, and show how the movement thru
this facility dramatizes a world famous landmark?
The other project has equally as demanding a landscape insofar as it
s importance to the scene goes. However, it's not based upon real
buildings and massing. There's some latitude for creativity. Without
going too far into its nature, it's an observatory in the shouthwest,
that requires desert scenes complete with buttes.
Anyways, I ramble on. I will send some images to you via mail before
long (and will be posting some, too, as promotion once I'm satisfied
with them).
I had a lot of trouble with the walls in my last walk through. They did not
translate properly from ACAD 12 to 3DS my any method that I tried so I ended up
remaking them in the 2D Shaper and Lofter. The walls imported from ACAD had
black streaks that looked like a missing face, but I couldn't find anything
wrong.
If 3DS added relative coordinates, snaps, and improved the Booleans I would be
less likely to use ACAD. Of course, if ACAD 13s 3DS conversions work, I might
use it more!
Chris
P.S. Did you ask me for more details on my microgravity animations? I only
have one that I can't talk about (It was made with data that is proprietary to
the company that did the work). Most of them have no problems on that subject
and I would be happy to tell you whatever you wanted to know about them.
– Chris Johnston, Written 17-May-1995 @ 20:58:58
Using: Windows NavCIS PRO 1.62
Hmm…. that wall sounds like it might have had some flipped normals,
or even a missing or misplaced vertex. I noticed that when I boolean
in 3DS, I very often get to complete the command, yet vertecies are
strung all over hell's half acre and make for a rather unappealing
picture.
What you said about possibly using ACAD less if they improved shape
building in 3DS is exactly why I'm afraid they haven't spent more
effort in doing so. Since Adesk's primary product is ACAD as opposed
to 3DS, they probably figure "if you want CAD accuracy, just buy
ACAD", and it makes them MORE money, theororetically. I'm betting if
3DS were ADesk's only product, you would have seen much better
shaping tools back in r2 or at least r3. As for booleans… it
could be something similar. Most who need linear precision will go
with ACAD or UStation, while the lack of booleans but a VERY powerful
lofter was great for more organic modelers who rarely needed numerous
booleans…
As for you microgravity animations, YES, I am indeed interested. It's
a subject I know little about but have heard just enough to get me
interested. I'd like details, but if you give me an overview of what
the general concept of microgravity visualization is about, I can get
more specific. I very much appreciate your willingness to discuss
it.
I thought the same thing – a flipped normal or a missing face, but I
couldn't find anything wrong with the faces in the area of the dark
streaks. I assume that there **was** something wrong but I just
couldn't find it.
We've got a group at work that develops and flys experiments that
take advantage of the low gravity environment that is available on
the space shuttle or, if they ever build it, the space station. The
acceleration levels are in the range of 0.001 to 0.000001 g depending
on the amount of residual air drag and the activity level on the
orbiter.
Density driven convection is what makes warm air rise and cold air
sink. If you take away gravity, convection stops. If you take away
gravity there is no pressure difference between the top and bottom of
a column of liquid. There are a lot of processes that are hard to
study on the ground because convection or the hydrostatic head
pressure gets in the way. These are the kind of experiments that you
can get answers to by taking them into space.
The hardware that you use for these experiments is often complex, so
I sometimes end up getting the job of putting together an animation
that illustrates how it all works. I did a series of animations to
illustrate the concepts behind a new-technology furnace that we were
designing that showed how it was assembled. In the process they
showed why the design was so good.
I've also done some work on a couple of other experiments, doing
essentially the same thing – showing how the hardware worked.
Fascinating…. I have always appreciated such work, as atronomy was
actually a first career choice as a child before architecture and
computer graphics, until I saw the eminent problems in such a career
until you gain prestige. Actually, I guess you could say cosmology
was my forte', as I loved to theorize about how the whole worked.
That'ss one reason why your experiments/ visualizations fascinated
me.
I would definitely be interested in seeeing a sample of your
animations, provided you're not under any non-disclosure clauses.
Indeed, I would also be interested in learning how I could get
involved in a similar vein, if at all possible.
Please let me know if I could see a small sample of your work.
Thanx again! -Jerome Kalisz
With only one animation as an exception, I am not under any
non-disclosure restrictions with my work, so I can probably put
together some samples. I'm not sure what the best format would be –
maybe I'll just have you send me a blank VHS tape and I'll copy some
examples from my 3/4" masters.
– Chris Johnston, Written 22-May-1995 @ 20:28:24
Using: Windows NavCIS PRO 1.62
Format wars?
The biggest problem in selecting formats is trying to decide how much
degradation you can stand in the duplication process. Feeding a 3/4
Umatic machine with a Par is fairly good quality but it goes soft
when using this 3/4 tape to produce VHS copies. Since the PAR can
output real time it is always much better to output the animation
directly to the VHS tape. The exception to this original recording
to VHS from the PAR is if your PAR output is recorded to a component
recording process such as betacam or 1". Now you can edit and make
several generations of dubbing in the component format without
noticeable loss. In fact, your VHS tapes will look about as good
from the betacam tape as it would if the VHS was recorded live from
the PAR.
Hey, that would be GREAT. I can do so however you wish, should you
wish me to get your address thru mail or whatever. Unusual work of
many types interests me, and I grew up as a science fanatic long
before computers became readily available to the home… Let me
know, I will send one off. Thanx!