I am always aware of the "language" of a software solution to a problem (and specially a field of study)… you will perfectly build a great ICE based population tool and you will be probably super-happy with the solution… in a few years though things will potentially look weird and that is because as we learn and learn we will see there is not a knowledge of the design applied because the system does not have the knowledge and does not provide the right language to define it.
To put an example, if you are about to build a town you should have quite a lot of insider knowledge on architecture and urbanism, the software tool should know about the elements and rules, the mechanics and the various aspects that govern city design to produce something that makes sense. A bunch of boxes won't cut it any more and the same happens with vegetation. Another good example is crowd simulation, it is great you can move particles and avoid each other, have sort of control and sort of a IA approach but only a correct model will produce the results that in 5 years time will still look great. There is the option of doing it yourself, and you know what? may be look even better but the fact is that the very difficult 7 years studying architecture I dealt with gave me the tools that I suspect will take you at least 7 years to learn. (if you are a genius less but this won't be a matter of months.) Therefore, crowds, vegetation, ecosystems, architecture, engineering are fields I tread very carefully and first of all look at which tools deal with the necessary language to model the system, then I see how can I bring the data to my favourite tool (XSI for example) and this approach forces me to really get inside that particular field. So my approach with vegetation is to use Vue and World maker and then export the data to the tool of choice for rendering. The process is always very painful and flexibility is the price in many many occasions, but if you find the right workflow the speed increases tenfold and the results are unbeatable. Please remember this is my opinion and certainly there are many other valid ways around. Hope it helps... Jordi Bares [email protected] On 1 Feb 2014, at 05:49, Enrique Caballero <[email protected]> wrote: > Hey everyone, > I was hoping to pick your brains. > > > I am currently doing some research on how to simplify, and make cheaper our > set dressing process over at this small shop that I work at. > > In the old days, when we did small simple shows, set dressing wasn't very > difficult for us as there were limited amounts of props and environment > pieces. We would just import referenced models and place them by hand/ > > But now that we do full forests and large amounts of rocks and plant life, we > need to upgrade our techniques. As our original techniques are no longer > sufficient. > > > > I am looking at how ICE can help us with the initial set dressing. Maybe by > painting weight maps we can spawn different types of mushrooms and trees etc. > > > I am currently experimenting with Dart Throw, and it looks very promising. > Thanks Julian! > > Although I need to figure out how to get it to switch between different geo > types, IE. Trees, mushrooms, etc. Currently I can only get it to spawn one > geo type at a time. > > I was wondering if anyone had any ideas about how we could simplify this > stuff, as I said before, currently I am looking at Dart throw and ABScatter > to see if they can help us out, but I am open minded. Has anyone tried Vue > Studio or anything similar that they can recommend? > > best, > Enrique
