Using real time interactive 3D

[phillip@esperient]

hi, my name is Phillip Trotter - I am the Director of Product Managent at Esperient Corporation - developers of Esperient Creator.

 

We have a number of architects and arch viz teams using our software and I would like to ask the CGArchitect community - what challenges you see in adopting real time 3D into your workflows?

 

What barriers for adoption have you come across - cost? deployment?, technology, bad sofware workflows? data challenges?

 

What have you found that works for you and what doesn't? How can your current practises improve. If you don't use real time - why not?

 

What are the business drivers for and against using interactive real time tools or which types of tool best fit your workflows.

 

Obviously I work for a tools vendor but I'm really not looking to start an our software is better than x hype discussion (or hype our own software). I'd much rather discuss industry challenges and look at discussions on workflow, practises, business challenges and what you would like to see happen to make it easier to adopt real time viz.

 

I'm hoping I can add info based on some of things we are discovering and hopefully build a discussion thread which is beneficial for everyone

 

I look forward to reading your thoughts, comments and feedback.

 

cheers

 

Phillip

[Devin Johnston]

- what challenges you see in adopting real time 3D into your workflows?

Most of the good VR engines seem to come from the gamming community, unfortunately it's very difficult to convert the model you create into a format the these engines can run.

 

What barriers for adoption have you come across - cost? deployment?, technology, bad sofware workflows? data challenges?

Bad software workflows are the primary hindrance in my opinion, but cost is also a factor as well as time. What ever the solution it needs to function as a plug-in within the 3D package and allow for easy translation into the VR engine. If not the VR package needs to be able to import the model along with textures and lighting from a large variety of different formats.

 

If you don't use real time - why not?

We don't use it because we haven't found anything commercially available that offers the quality we want. Like I've said before there difficult to use and don't deliver the wow factor that an animation does.

 

VR is an untapped resource that the arch viz community really wants to use, the problem is that no one has come up with a good solution. There is also a big quality difference between animations and current VR that needs to be addressed.

[cleblanc]

Phillip,

 

This is the first time I have been asked about the obstacles for realtime by someone in the realtime industry. I applaud your approach to the issue. Most of the time I find the industry just assumes that real time has an inevitable place in arch presentation.

 

The first reason that I do not use real time is that clients don't ask for it. Still images are an expectation that most clients already have.

 

I do not promote real time for a couple reasons...

 

Interactivity does not seem appropriate for most final presentations. There may be an argument for it in design development, but 95% of my work is for final presentation.

 

The compromise in quality is not something I am willing to accept as an artist. When I have done real time in the past, I have had to accept its limitations and swallow my artistic pride. The quality is not such an issue for schematic/dd, but sketchup is probably an acceptable solution for those uses.

 

I find real time work harder to market for future projects. Screen grabs of real time presentations simply can't compare to pre rendered images, and making the actual presentation available online is cumbersome for the artist and user.

 

Chris

[phillip@esperient]

Hi Devin, Chris

 

Firstly thank you for the thoughtful responses. You both mention the visual quality and not willing to compromise what you offer to the client and Chris - you said that interactivity is not relevant to 95% of your final presentation work, outside of schematic//dd. Do either of you see areas where it is beneficial? or could be if Devin's workflow challenges were largely solved?

 

To anyone else following the thread - I am kinding of hoping that we can get some of the community who do use interactive 3D to chime into this thread to see if there is a counterpoint perspective -or rather to see what areas of arch viz VR/ interactive real time 3D, can benefit.

 

Chris, as you mentioned, I know 13 years back when I was doing research in VR - one of the expectations from the visualization industry was that VR presentations would one day replace stills and animations - and i thought then (and still do now) that it misses the point. Each media has a way of conveying a story - and people are often most confortable with the media that is familiar - so still shots and rendered animation walkthroughs work for most of todays clients; its a familiar presentation experience for the client and the medium affords the artist (with a lot of work) an established toolset for conveying mood, space and focus within the presentation.

 

Interactive environments afford a different experience - with visual quality trade offs compared to current generation non-real time animations - but is that line changing and demand increasing? or is real time interaction for arch viz still very new, niche and with out the workflow and nuance for the viz artists to want to use or experiment with it at its current costs of entry?

 

cheers,

 

Phillip

Edited November 11, 2008 by phillip@esperient

[Devin Johnston]

Hi Devin, Chris

 

Firstly thank you for the thoughtful responses. You both mention the visual quality and not willing to compromise what you offer to the client and Chris - you said that interactivity is not relevant to 95% of your final presentation work, outside of schematic//dd. Do either of you see areas where it is beneficial? or could be if Devin's workflow challenges were largely solved?

 

VR would be most beneficial on large projects with lots of different spaces both exterior and interior where a client really want's to get into the details. It doesn’t need to be as high quality as a rendering but if you've ever played Call of Duty or Halo then you will understand the detail and quality that I'm thinking of. Workflow needs to be as simple as possible because most VR projects will need to work in parallel with what ever renderings and animations you’re producing for your client and it would be a big pain to have to redo everything twice. That's the biggest disadvantage that I can see because unless it works that way you'd almost have to have a separate department creating VR environments and that isn't going to be cheep.

[ronll]

I have been doing (what is now called, "Arch Viz") full-time since 1989, having had to develop my own software back then to do so. I have long been interested in real-time for architecture and used to try to adapt game engines for that purpose with little success. In 2002, I took most of a year off to develop my own real-time engine from scratch using DirectX. It was very successful as a high quality real-time rendering engine but at the time the graphics cards in laptops weren't up to speed to handle it. So in order to market it as a service to my clients I had to build a "lugable" computer in a suitcase that had a fast nVidia card.

 

I sold the real-time service to one client for a pre-sales tool for a large condo project. In spite of all of my hands on with the project, I could not get their sales team to effectively use the product. To my mind, the navigation was very simple with just a mouse or even the keyboard arrow keys if they weren't comfortable with the mouse. The users seem to be afraid of it. It had the ability to walk thru the space at eye-level or rise up and fly above a (ceilingless) unit by using the scroll wheel. I found that they only would use it by flying above the 3D plan. They would not attempt to walk thru the space at eye-level for fear of bumping into walls or something. Whatever the reason, they just would not use its full potential.

 

Trust me, it was a very good implementation with good textures and lighting and very easy navigation. And of course laptops soon were able to run it full screen. But all my marketing efforts failed to get anyone interested in using it. Frankly, I personally think that it is a case of having a lot of gee-whiz factor but much less actual usability beyond that in the real world. To a certain extent I think the same is true of architectural animations: more techno-whiz than real world usefulness.

 

Real world builders, architects, and their clients are more interested in seeing their buildings than in the gadgets that show it. Still images do this VERY well and are much more portable. Stills can be hung on a wall in a sales office. They can be printed in magazines, newspapers and brochures and easily shown on the web. Stills are the heart of my business and far, FAR outstrip any demand or requests for animations or real-time. Maybe this will change over time. But as yet, I see it as a solution without a problem. Of course, YMMV.