PabloEscobar

Hello everyone,

Im in a jam---Im a consultant in NJ, and i work with T-mobile presenting photosimulations of there cell phone antennas on roof top's of buildings.
I also design monopoles, and cell towers for them.

HERES the PROBLEM---

my boss takes the work to court and testifies to it. Well sometimes its not that easy, this being the case, a location proposed in Long Island is protesting this monopole antenna going up in a shopping area. Now the lawyers are trying to discredit 3D studio max, and or Viz, which are the programs i work in. I need to know what is the margin of error in max and viz?? I work of of AutoCad files preped for my by the engineers, after importing them, i build the model according to their specs. I aline the camera according to the pictures taken by my boss of the location, I render, and then superimpose using Photoshop.

I need to give an explonation of how accurate is the Max, Viz program, why Im i saying it is that way, and where did i get my information from. Can anyone give me any information? or direct me where to go to ask for info like this. I included a sample of what exactly my work intells.

Thanx,
Pete

Attached Images

Test pic.jpg (62.3 KB, 91 views)

AJLynn

This is a tough one - basically, Max is as accurate as the information you give it. You can model with very high accuracy and Max will not make errors in raytracing what object is seen from where, what gets in the way of what, etc. But photomatching is hard. I think your best bet is probably to have the photo taken professionally by a camera with well-documented specs, and use surveying techniques to pinpoint the camera location and direction as accurately as you can, so that you can say with preponderance-of-evidence certainty that your photomontage is correct. The more real-world-understandable information you can provide, the more you can convince city officials who are not computer experts but are used to dealing with people like surveyors.

What exactly is in question here? Whether you can see the pole? Exactly how large it is? The color scheme?

imgumbydammit

I went through something nearly identical with a 172' tall "Militaristic" ( neighborhood groups language..not mine ) 911 communication tower in the city I work for....residents pitched a hissy fit about the antenna...after it had been approved and erected...they wanted cladding to make it look like a giant redwood tree for one option and a monoplole for another.barring complete removal of course..which was their first demand I made a few renderings showing a monopole in it's place..and had to explain how I got to the elevation I did...
In my case it was easy because I had an in place antenna of the same height to use .Still I had some touch up to do and building areas to replace where the monopole exposed what was essentially covered by the 3 legged monster in place .I imagine the lawyers concerns are over correct scale of the antenna..they want to know that your rendering is going to exhibit EXACTLY what will exist in real world..you might have to explain how you determined the scale in the image....
Good Luck

IGD

PabloEscobar

I totally agree with the simplistic language to be used as most people dont even know that this technology exists. You basically explained to me what i thought all along here. That was the question, so basically its my photomontage that comes to question here, and my skills in Photoshop correct?

PabloEscobar

O man, u hit the nail on the head huh, u know exactly what i mean. And its is the scale that comes in question here. Well my boss uses a crane company to come and park in the location of the proposed monopole and then it rises to a desired height of 80 to 140 feet, then i clean the crane up in photoshop mark the height and put in the rendered pole. Mean time i set the pole up to the sizes given to me by the engineer, for instance 30 inch bottom and 24 inch top, so the pole slims as it gets higher.

In this case I guess both of u r right, and i guess like "AJLynn" said my photomontage accuracy comes into question here.
Hmm any idead of how else i can defend my work here?
I already mentioned that there might be a less then 2% margin of error in Max and Viz cuz the lawyers couldnt understand how a program could be 100% accurate.

Sawyer

I imagine the margin of error is greater in the field with the as built pole. "less thn 2%" sounds arbitrary. What kind of margine of error? And what you are really dealing with is human perception not an absolute height. A camera might perceive and might distort or could be used to intentionally distort an image. A rendering is the same. Perspectives are tricky to judge scale.

__________________
And you may ask yourself
What is that beautiful house?

PabloEscobar

I ment the margin of error in 3D, as in modeling software. I need to defend my work here, at least the 3D part, because Im not the one that went out there and took the field work pictures, so i wont vouch for that, but my 3D needs to have some supportive facts behind it, so Im in the clear.
http://www.cgarchitect.com/vb/19277-...ps-tricks.html
By the way this is a good thread so far ...Im downloading the book right now to c how else this book can help me with the facts.

Koper

one also have to remember that max's unit setup can change automatically. I work in meters, but if I reset max, not close and open again, but 'file -> reset', then the unit setup changes to generic, which is inches, but you still can setup your inches system to be different to the generic one. But if I close max and open it up the system is correct as I have set it up. Thus when i reset i have to change again. Then you also get your system unit setup, this stays the same no matter what.

One has to be sure at all times what max's unit setups are and what the system setup is, otherwise you'll get a big margin of error.

You didn't by any change reset max so that it worked on another unit setup than what you are used to.

PabloEscobar

No --
I imported the Autocad file and set my unit sys to US.
I got cought up with that problem before. Mid way in my project nothing was matching....i was like what the hell.....eventually i found out y!!

Never made that misteak again, but thanx.

I v e mailed ur friend Jeff Mott---Im hoping he can give me some facts here
something i can take to court with me .

BrianKitts

I always work in american standard units, and have never had any problems with accuracy based on units changing in the scene. If you never mess with them you shouldn't have much to worry about..... i realize thats not always true for those who work with the metric system.

As for your whole questioning accuracy. If you used a crane in the field to "mark" the correct height, that mere fact debunks any idea they have about max's accuracy and your photoshop accuracy. The only margin of error I could see is not based on the 3d software but as mentioned is based on the field measurements. Where did the autocad plans come from.... what is the margin of error comming out of the civil dept that gave you the drawings. If you measured in the field.... what is your personal margin of error.

It's like scientific measurement, if you are measuring stuff to the nearest one-hundredth of an inch, than your margin of error is a decimal shift away. (one-tenth of an inch). Since you can set your units in max down to thirty-seconds, than the if you key all the dimensions in correctly, your margin of error will probably be smaller than the contractor who's actually going to errect your structure anyways.

so what the heck am I trying to say.... if you need to defend photoshop, your argument is the crane operator, the only way photoshop could look wrong is if the crane operator was wrong about what height the boom was raised to. If it's max you have to defend, site that max is accurate down to one thirty-second of an inch (or whatever the smallest you set your units to) then as long as you place everything in the scene at the right height and points, your only margin of error would point at the surveyor or civil engineer that gave you the site data.

__________________
Brian Kitts
ANSHEN + ALLEN