Friesen@HIS-PHOENIX-MULTICS.ARPA (04/06/88)
Someone asked, "Who out there is a Cyberpunk?"
I guess I am one. I own CAD 3-D 2.0 and Cyberpaint.
I actually am not as pleased as I could be with CAD 3-D however.
The reasons are as follows: 1. The drawing tools are not that good,
for example, for a presentation I wanted to construct and manipulate a
model of DNA (which resembles a twisted ladder) and found this to be
impossable with the tools supplied.
2. The slowness of the program (especially when joinning)bothers me. I
know its the hardware, and it does not bother me that much, but it gets
a little tiresome.
3. The shading is sometimes splotchy, meaning it is to unrealistic, for
example, there might be a sphere, withe light coming from the top right,
almost all of the parts facing the light are the same shade, then the
next row has a sudden decrease, 3 or 4 shades in color, it needs to be
more gradual, also the fact that it shades even a sphere with squares, I
know it is because everything is drawn with triangles, but then that is
obviously not to good of a way to draw a sphere!
4. I also do not like the fact that objects you create do not cast
shadows.
5. The most annoying thing is the fact some of the files (Animator,
cubit, and more) seem to be missing, and no one I have contacted has
them either.
6. It also takes a whole double sided disk to save an animation file,
which is a pain!
I like Cyberpaint, but I also have some complaints about it.
1. It needs fill patterns.
2. It needs features so you can select dotted lines, ect.
3. It needs user-definable fill paterns.
4. It needs better text input abilities.
Cyberpaint is fairly good, but CAD 3-D definitly has room for
improvement. (especially for the price)
Aric Friesen
"Exterminate! Exterminate!"
/~\-* --<
/*** /*****
Want to argue my complaints, E-mail me.greg@bilbo (Greg Wageman) (04/09/88)
In article <880405205724.800048@HIS-PHOENIX-MULTICS.ARPA> Friesen%PCO@BCO-MULTICS.ARPA writes: > >Someone asked, "Who out there is a Cyberpunk?" > >I guess I am one. I own CAD 3-D 2.0 and Cyberpaint. > >I actually am not as pleased as I could be with CAD 3-D however. > >The reasons are as follows: 1. The drawing tools are not that good, >for example, for a presentation I wanted to construct and manipulate a >model of DNA (which resembles a twisted ladder) and found this to be >impossable with the tools supplied. Depending upon exactly how you are trying to model such an object, I think it possible. Difficult, but possible. You can certainly do it with the new CyberControl language since it allows you to write equations which can be used to generate objects, using splines. Sounds like a perfect application, to me. (CyberControl drives CAD-3D using the applications messaging event system provided by the Atari's AES.) >2. The slowness of the program (especially when joinning)bothers me. I >know its the hardware, and it does not bother me that much, but it gets >a little tiresome. I have to agree, joining anything reasonably complex is coffee-break time. Not having seen his source code, I don't know how optimal his algorithms are (Tom Hudson, that is, author of CAD-3D). >3. The shading is sometimes splotchy, meaning it is to unrealistic, for >example, there might be a sphere, withe light coming from the top right, >almost all of the parts facing the light are the same shade, then the >next row has a sudden decrease, 3 or 4 shades in color, it needs to be >more gradual, also the fact that it shades even a sphere with squares, I >know it is because everything is drawn with triangles, but then that is >obviously not to good of a way to draw a sphere! Are you using the default shading, or have you defined custom color sets? I much prefer the results from the latter (say 14 shades). However, it's a trade-off between the number of object colors vs. the number (gradiation) of shades. That is a hardware limitation, at least until someone combines the Spectrum 48 colors/scan line code with CAD-3D (are you listening, Antic??? :-) ) >4. I also do not like the fact that objects you create do not cast >shadows. I agree. However, there is a technique where you can fake them. Make a copy of the object which should cast a shadow. Scale the object down along one axis until it is nearly two-dimensional, then re-color it to black. Position it as appropriate. Not automatic, not elegant, but it works. >5. The most annoying thing is the fact some of the files (Animator, >cubit, and more) seem to be missing, and no one I have contacted has >them either. My disk came with the "animate" program, if that's what you mean. The "cubit" objects were there too, but you have to walk through the tutorial to produce the animation (it's worth the investment of time). If you got the program legitimately, (and I'm not casting aspersions), Antic's support is really excellent. Make a backup, send them the original disk, explain the problem, and they'll fix it. >6. It also takes a whole double sided disk to save an animation file, >which is a pain! That depends, of course, on the length of the animation and the complexity. You DID install the "Cybersmash" desk accesory, didn't you? Said accessory performs a delta-compression on each frame as it is generated, so that it doesn't consume nearly the 32,000 bytes it would otherwise require. Without it, you can fill the disk up quite quickly. Even so, you can chain an animation across multiple disks. The real limitation is memory. With a Meg, you are still limited to about 2 minutes of animation. >I like Cyberpaint, but I also have some complaints about it. > [ ...complaints deleted...] I own it, but confess I haven't had a chance to play with it, so I can't comment intelligently. >Cyberpaint is fairly good, but CAD 3-D definitly has room for >improvement. (especially for the price) WHAT???? CAD-3D is about the most incredible software product I have ever seen at the price (what, $90?). We're talking SOLID MODELLING here, folks, no simple pixel editor. Animation capabilities. Color. I can do things with my Atari that, maybe 5 years ago, people would have KILLED to be able to do for under $10,000 (hardware + software). >Aric Friesen >Want to argue my complaints, E-mail me. No, you can't flame someone like this in public without a public rebuttal. Besides, you don't give a plausible e-mail address. Note that I have no connection with Antic Publishing except as a satisified customer. Greg Wageman Schlumberger Technologies UUCP: ...!decwrl!spar!snjsn1!blfca1!greg 1601 Technology Drive ARPA: greg@blfca1.com@spar.slb.com San Jose, CA 95110 BIX: gwage (408) 437-5198 CIS: 74016,352 "Nest deeply, and carry a big stack." ------------------ The opinions expressed herein are solely the responsibility of the author.
Steve_Godun@rubbs1.UUCP (Steve Godun) (04/13/88)
CAD 3-D was designed to be a modeling tool, primarily for things sich (oops - such) as buildings and such. Yes, you can do molecular modeling (as in that DNA example), and contrary to your comments, the results are very agreeable. However, you can get better results if you use Antic's Molecular Design disk (or something like that -- it's in The Catalog). Processing speed is fairly slow, but I have played with it for a while and I find it bearable. With the addition of a Blitter chip (yes, I have played with CAD 3-D on a Mega ST-4 w/Blitter) and expanded memory, processing time is *quite* rapid. Please be more objective in your comments. -- Mail coming to you from the Fidonet Gateway - Node 1:107/330 To respond, please send mail to ...rutgers!rubbs1!net!node!user_name