|
|
|
Newbie
Group: Forum Members
Last Login: Monday, April 19, 2010 3:00 AM
Posts: 15,
Visits: 193
|
|
I am working on a large (15+ miles) visualization project of a highly complex road. My original intent was to loft the highway using a similar technique as in the Rules of the Road tutorial, but because of the constantly changing right of way, lane changes and on and off ramps lofting just did not seem like it was going to work for the models complexity and our very short time frame to produce the model.
Attached is just a sample of the variables i need to deal with and this is constant for miles and miles.
Could/should of I used a lofting technique to create these highways?
To model the road thus far I have created 3d polyline center lines made in Autocad by extending the vertical profile information along the highway, which are very accurate. Created a surface in between two of these splines in which an extruded form of the roadway pavement intersects the surface and is then cut out with procutter.
While this technique has worked I cant help but feel like there was an easy way especially as I get into some of the limitations this has for texturing and lofting details such as barriers.
Any insight will be greatly appreciated. Thank you.
|
|
|
|
|
Veteran
Group: Administrators
Last Login: Friday, January 27, 2012 4:50 PM
Posts: 280,
Visits: 1,667
|
|
Whatever gets you the result that you want is the right way 
From the looks of it doing a loft for that road would suck. I'll often create a loft for the basic shape of the road then use editable poly techniques to modify the surface to fit the CAD. I think both ways are as labor intensive.
THOMAS SHANNON
SENIOR DESIGN VISUALIZATION SPECIALIST
PB Project Visualization
http://www.pbprojectviz.com/
|
|
|
|
|
"old dog"
Group: Administrators
Last Login: Thursday, February 02, 2012 8:34 PM
Posts: 646,
Visits: 4,051
|
|
|
|
|
|
Newbie
Group: Forum Members
Last Login: Monday, April 19, 2010 3:00 AM
Posts: 15,
Visits: 193
|
|
Yea, I had a freak out moment about the mapping of the textures and oil. But it looks like I can use the same 3d center lines to help with those.
Thanks for the reassurance, I was looking for that more than anything.
Also great forum and resource you guys have here, this is my companies first foray into this type of work and I can see this being a great resource, and I hope that I can one day I can contribute. 
|
|
|
|
|
Newbie
Group: Forum Members
Last Login: Monday, April 19, 2010 3:00 AM
Posts: 15,
Visits: 193
|
|
The project is "finished" we did a great job given the circumstances, but its really been a disaster the whole way. I will try and get some stills up soon, I will need to check on the disclosure agreement. But I would really like some feedback on both the visualization and the business aspect of it.
It was a learning experience for sure, for better or for worse though? Im not sure yet...
|
|
|
|
|
Veteran
Group: Administrators
Last Login: Friday, January 27, 2012 4:50 PM
Posts: 280,
Visits: 1,667
|
|
AEC viz is something that's seriously underestimated. There are so many techniques and technologies to master to ensure your work is both aesthetically pleasing and engineering accurate.
That's why this site is here, because what we do is hard, and people need help. I can't wait to see the final and your future work. You're off to a great start!
THOMAS SHANNON
SENIOR DESIGN VISUALIZATION SPECIALIST
PB Project Visualization
http://www.pbprojectviz.com/
|
|
|
|
|
Newbie
Group: Forum Members
Last Login: Monday, April 19, 2010 3:00 AM
Posts: 15,
Visits: 193
|
|
Without giving too much away because its not finished yet but I really need some advice on managing client expectations. I don't know if its best to reply to me via PM or if here is fine. I'm assuming if anyone knew the budget for this project you would probably laugh, or just feel real sick at your stomach, heh.
I know its hard to tell from a single screen shot, but I tried to pick out something that showed the general level of detail that was proposed for this visualization.
Once again over generalizing, but this condition continues for approximately 11 miles. What would be the modeling time commitment for say a person working full time (me) and a person working part time (intern) on a project of this scale.
Also what type of review procedure do some of you go though, there are so many different variables and ways to make mistakes needless to say ive made a ton and there are still more to fix.
Essentially I have been flying blind and need something to ground me back to reality.
And thanks again for helping me along the way, without help from here I don't know where I would be.
|
|
|
|
|
"old dog"
Group: Administrators
Last Login: Thursday, February 02, 2012 8:34 PM
Posts: 646,
Visits: 4,051
|
|
| Wow! You've certainly got a monster project on your hands there. I think you've got the right approach to this already. With highways, clients almost always request some sort of driving (either the who thing or a portion). There are pros and cons to this for sure. By getting the camera down to a driver's viewpoint you may not have to build out so much of the environment but will need to show more details on the roadway. From your rendering, I would guess to complete the rest of the alignment (without seeing the data) you would be looking at 5-7 weeks. One suggestion would be to pick key locations along the alignment to highlight in more detail rather than try to build the entire thing out. Just managing the cars alone will drive you to drink (heavily). Suggest using proxies for cars and any vegitation and even X-refing them in. More importantly, can your hardware support a model of that size? This may be a good excuse to get 64X if you aren't running it already. Also, for an animation fo that length you would be looking at thousands of frames. Do you have a render farm or can get access to workstations overnight? Hopefully this isn't scaring you. I've built out 55 miles of a proposed highway, but it was rural and no cars, trees or interchanges. It was more for a general fly-to model where we would highlight areas of interest... Keep the questions coming. We can help you get through this! 
Glen Loyd
Lead Design Visualization Specialist | Parsons Brinckerhoff
www.pbprojectviz.com
 
|
|
|
|
|
Newbie
Group: Forum Members
Last Login: Monday, April 19, 2010 3:00 AM
Posts: 15,
Visits: 193
|
|
Thanks for the reply Glen, the animation is actually already done, I just cant seem to make it "perfect" (in the technical sense, they were more concerned in the engineering accuracy than the visuals) for the client.
I should have "kept more questions coming" during the modeling process, but the time frame was so quick for me (about 8 weeks) I was working on it so much I felt like I had little time to look for instruction. I was talked into this project by my boss, and by my ego (This was my first 3d modeling project), and now we are both thinking that projects of this scale are probably not in our best interest.
Now im stuck in a infinite loop of revisions to fix things that probably had I had experience I could of built the model better but now im having to go back and re-edit.
8100 frames, 4000 trees, and about 15,000 cars, as vray proxies didn't xref them in though perhaps I should try that.
In the original proposal we agreed to do 50% of the animation at the drivers level, which then turned into maybe 5%. I did a section down at at the drivers level, which actually turned out to be the best looking scene and the easiest to control all the variables. But the rest was essentially a constant elevation flyover of the conditions of the highway.
|
|
|
|
|
"old dog"
Group: Administrators
Last Login: Thursday, February 02, 2012 8:34 PM
Posts: 646,
Visits: 4,051
|
|
|
|
|
