Gaming Engine as a Presentation Tool

[LukeC]

I was having a little think different presentation methods and was thinking about using a gaming engine as an interactive method for clients.

 

Now i know this isnt as simple as it may sound with the costs of the actual engine (or maybe use a suitable open source engine?).

 

A few downfalls that quickly came to mind were the portability of it all; do you have a machine dedicated that you cart around for clients to use or do they have to install the whole thing themselves.

 

Usability; how hard it would be to use it. As in there is a bit of a learning curve involved (not necessarily too steep) in learning how to play a fps and since the controls would be pretty much the same, would a older non game player be able to use this?

 

Copywrite laws; in distributing a copy of your work are you breaking the copywrite on the produce but distributing the engine. This however would be eliminated by using an open source engine.

 

Cost; would it be worth the money and effort for a company to set this up?

 

 

Now these are all just ideas off the top of my head i thought i would just throw in for everyones thoughts. I know this has been somthing that has been thought of before by many people but there has to be a reason why there isnt as such a viable "interactive" solution.

 

Another thing to keep in mind is the low poly counts within games..

 

One last thing, sorry if this is been spoken about before (as im sure it possibly has), I had a quick scan and didnt see anything really relating to this.

 

 

Cheers

Luke

[Richard McCarthy]

...

 

 

 

This thread should be in Realtime/3D/VR/VRML forum.

 

 

 

...

[Richard McCarthy]

 

Now i know this isnt as simple as it may sound with the costs of the actual engine (or maybe use a suitable open source engine?).

 

A few downfalls that quickly came to mind were the portability of it all; do you have a machine dedicated that you cart around for clients to use or do they have to install the whole thing themselves.

 

Usability; how hard it would be to use it. As in there is a bit of a learning curve involved (not necessarily too steep) in learning how to play a fps and since the controls would be pretty much the same, would a older non game player be able to use this?

 

Copywrite laws; in distributing a copy of your work are you breaking the copywrite on the produce but distributing the engine. This however would be eliminated by using an open source engine.

 

Cost; would it be worth the money and effort for a company to set this up?

 

 

Now these are all just ideas off the top of my head i thought i would just throw in for everyones thoughts. I know this has been somthing that has been thought of before by many people but there has to be a reason why there isnt as such a viable "interactive" solution.

 

Another thing to keep in mind is the low poly counts within games..

 

 

 

 

There are currently many engine avaliable that can do such thing, but you have to forget about game engine as such. Quake2/3/DOOM3 cost upward of few $250,000 USD for the license use of engine (commercial) same goes with any other commercial 3D game engine.

 

Currently the best solution I can see is Quest3D (http://www.quest3d.com)

and rTre (http://www.cubicspace.com) that let you author and distribute unlimited copies. So copyright issue is not really an issue if you use these engines.

 

As for the presentation, I think it's not really a problem as most computer nowadays have built in 3D card/chip (ATI/Nvidia mobility chip), even if it's a problem you can always bring your notebook with built in 3D chip to your client to showcase your design.

 

 

Lastly, is it really worth the money to set this up? I think eventually, clients would all demand this solution... but for now, it's really all anyone's guess.. it's NICE to have realtime walkthrough, it's NICE and impresses a lot, but it takes A LOT MORE TIME right now to produce such thing (more process, texturebaking, UV, decimating poly, adjusting and maximising LOD) so is it really justify to bill your client for such unless he specify?

[LukeC]

Ah cheers sorry bout that and posting in the wrong spot. I notice a thread already on that topic.. so ill leave this one

 

 

Just quickly in response.. "so is it really justify to bill your client for such unless he specify?" Probably not when you put it that way. But what if you put in your own initiative, researched and completed a demo project under your own steam to proove it could be done then you can use it as a "do you want this" sort of thing. just thoughts anyway

[Pablo Zurita]

Lastly, is it really worth the money to set this up? I think eventually, clients would all demand this solution... but for now, it's really all anyone's guess.. it's NICE to have realtime walkthrough, it's NICE and impresses a lot, but it takes A LOT MORE TIME right now to produce such thing (more process, texturebaking, UV, decimating poly, adjusting and maximising LOD) so is it really justify to bill your client for such unless he specify?

I think having a real-time presentation with proper tools is going to be really important and will impress the client a lot. Why it isn't used now? Because no one has done it properly, or because some artists and architects are too concerned with some aspects that they see when they compare a single frame from a real-time presentation against a rendered image. I'm currently working at Irradiance Software on a software to do real-time presentations, and I think once we show a good demo showing how much better and dynamic a real-time presentation can be, people will start to realize that even though a real-time presentation may not look as good as a render, its really good in its own.

[Manuel Rivera]

This thread is interesting. I have a number of similar thoughts. I also have tried out macromedia drector's w3d, and vrml import tools. There are a couple of hoops to jump through but you can get high quality models right out of MAX. The learning curve for scripting on director's end for collioion dection is steep. But once you have it you can apply it to almost all situations. The only thing required on the clients end is shockwave player, which comes standard with most browsers. ( Firefox needs it to be manually installed, but still works fine)

[Pablo Zurita]

This thread is interesting. I have a number of similar thoughts. I also have tried out macromedia drector's w3d, and vrml import tools. There are a couple of hoops to jump through but you can get high quality models right out of MAX. The learning curve for scripting on director's end for collioion dection is steep. But once you have it you can apply it to almost all situations. The only thing required on the clients end is shockwave player, which comes standard with most browsers. ( Firefox needs it to be manually installed, but still works fine)

I think one of the problems is that everybody expects to just do so export/import and everything has to look right. I think that models have to be tuned for interactive demos; I don't think there is such thing as just using all the assets from a normal render. Even with tools such as rtre you need to customize to make it work right. I believe if people actually evaluated real time architectural demos on its own instead of comparing those demos to still frames generated by a renderer that took 3 hours or more to generate a single frame, then people would realize that this kind of demos are great, useful, and a huge plus for a presentation.

[rued]

I must agree with Pablo - that the dataset needs to be tuned. Even though it's quite usable to just use native formats like VRML, 3DS, etc. There is really alot of performance to get in optimizing your dataset in different ways before using it for real-time purposes!

 

About using game engines directly, there are issues pointed out in earlier posts. In addition we actually tried using the original Quake 3, when it was released. We had our own render, but tried to use the editor for architectural visualizations. Even though we got alot of stuff for free: Visibility calculations, lightmaps, collision volumes - we found the content creation process quite bad. First we had to work with "brushes" and carve our way through solid volumes. But even worse, we where bound by too strict units, making it very difficult to make exact models based on blueprints or similar.

 

Even though this works great for games, it was simply too limited for our goals. Our artists (Max users) had to use a lot of time getting used to the carving paradigm. In addition the lighting was simply not good enough, seen from a modern GI point of view. And since the lightmaps where saved directly to the binary bsp file it was very difficult to finetune problematic areas in PhotoShop.

 

So all in all, we concluded that the content creation process was simply too time consuming compared to what we could do in Max.

 

Regards

Thomas Rued

Digital Arts

[Pablo Zurita]

About using game engines directly, there are issues pointed out in earlier posts. In addition we actually tried using the original Quake 3, when it was released. We had our own render, but tried to use the editor for architectural visualizations. Even though we got alot of stuff for free: Visibility calculations, lightmaps, collision volumes - we found the content creation process quite bad. First we had to work with "brushes" and carve our way through solid volumes. But even worse, we where bound by too strict units, making it very difficult to make exact models based on blueprints or similar.

 

Even though this works great for games, it was simply too limited for our goals. Our artists (Max users) had to use a lot of time getting used to the carving paradigm. In addition the lighting was simply not good enough, seen from a modern GI point of view. And since the lightmaps where saved directly to the binary bsp file it was very difficult to finetune problematic areas in PhotoShop.

 

So all in all, we concluded that the content creation process was simply too time consuming compared to what we could do in Max.

 

Regards

Thomas Rued

Digital Arts

I agree, I believe you can't use a game engine with its tools directly. The requirements of an architects are very different of the ones of a game designer. Everything was to be tuned for architects.

[Manuel Rivera]

I think also that there my be a dynamic shift in the way in which architectural work is represented. Not only for clientele purposes, but for academics. I believe if we push these tools to their absolute maximum poetential, it will then generate an industry. I have already heard of people mentenioning companies like cebas, or turbosquid. But what about going hardcore like scripting something in c++, or the other end of the spectrum like gmax, I guess I'm making both a statement, and asking a question?