Lumion General Connection Failure

0 views

Skip to first unread message

William Dupere

unread,

Aug 3, 2024, 2:13:09 PM8/3/24

to wenregitan

As twinmotion is coming up with realtime raytracing next year - live link to archicad and connection to bimobjects.com, the same with Chaos group with Lavina, (unfortunately no connection to bimobjects there?) and real time raytracing generally as the new standard; what will be left for people today having archviz as their profession? Will they be obsolete? If not, why, and how would they need to transform to survive?

Do archviz people need to adapt to using twinmotion or should they strive for that hi end quality, and will that hi end quality eventually be overtaken by one click solutions from real time raytracers?

With ai architecture algorithms coming up, already proving better results with working out architecture than the actual architects themselves, will the architects need to transform and taking more of archviz to get their days filled, just so that archviz people are even more screwed?

So what will be in it for architecture visualizers in the coming time, and what should they do to survive and convincing their bosses and collaborators/customers to not push the archviz job into the hands of non-nerds and non-artists?

Always endeavor to improve technical and creative skills along with emotional intelligence and client empathy. In then end, value your time appropriately and all will be fine. If one stays hungry to get better and provide a level of service that clients desire and appreciate, there will be work.

Yes, the sky is falling sentiment comes around every 5 years or so. First, it was Gen1 rendering software, then it was ray tracing, then GI, then globalism, realtime engines, VR and now the cycle starts over again with realtime-raytraced tech. I thought GI would kill the business back in the early 2000's because it became so easy to render scenes without having to perform elaborate tricks and rigs to get realistic lighting. How wrong I was!

With all of the advances made throughout the years, I still see animations with cameras crashing through trees, roller coaster paths and Ferraris parked in front of elementary schools. You are more valuable than you realize and needed more now than before.

Thank you for your thoughts. I am well aware of the old doomsday threads. But we never saw live links from the drafting/drawing programs straight to engines that actually are visualization specific programs, which directs straight to architects. Programs that gives you the exact correct objects with direct links to manufacture verified bim products in bimobjects.com.

Why should we spend time on importing and detailing objects when this is done in a few seconds with one click from archicad? Why should we spend time on making rigs with nice lighting and skies, when this is automatically done in the architect targeted program, but also with animated skies with light matches with coordinates and datetime as well?

Why should we spend time on updating model on changes, when this happens automatically by one click directly from the drafting program, by the drafter himself? Why should we spend time on making projects in anima when people animation comes easily with twinmotion? Comes with animated vegetation as well. Why should we spend time and money on time and licenses with growfx, forest pack, Phoenix FD or other packages?

This is specifically the case when it comes to archicad to twinmotion. Live raytracing in twin coming first quarter 2020. So my guess is that the reason why everything is not falling spart for you guys, would be that your customers dont know of archicad and the live link to twinmotion yet.

Ours are about to, and it seems like the senior directors in our company seems like it is ok to leave our customers to decide whether twinmotion quality is enough. It also seems like the customers actually are well okay with going a bit down on quality to get more images with less quality. I'm not sure about you, but my guess is that your bread and butter arent all in that high profiled international projects neighter. You know it, I know it. Visualization is one of the things they look at as an unnecessary expenditure. They also dont know how much job it is to go from idea to image, so why not let the architect herself do it correct the first time? They are after all very well trained, where the new ones claims they have almost 50 percent of their education as visualization.

So seems like we need to go down on quality to not loose our jobs. But if we go down on quality, then the company could as well just use the architects to make the visuals. So we loose our jobs anyway. What is left for us to do?

While reading and understanding what the contributors here have written, I still have some issues with coming to peace with all this. 3ds max is falling behind. We need animated objects right out of the box. 3ds max (or any other modeling program) needs live link to bimobjects.com. Things like that. Or else we are spending so much time on double work that we are going out just by that. Should we stick to what gives great quality or should we sell our souls and start with twinmotion? It hurts me to write it.

I'm an in-house visualization person as well, so here was my experience from 2019. When I had met with my team for our last 2019 team meeting, I had really stressed one singular point. That was the way we are rendering in Dec 2019 will not be the same as we will be rendering in Dec 2020. For us, 2020 is going to represent a another paradigm shift in the rendering world. It is both scary and exciting, but will ultimately lead to great things for our industry.

We saw our workload drop by almost 50% when Enscape was rolled out. This was both good and bad. Good in that the 500 iteration "what does this look like" rendering process was off our plate. Bad when those renderings were considered "good enough" during client review stages. Where we saw our workload rebound at the end of the year was when these images were made public and tried to sell the product to the public, they were roundly rejected. This was even more so if there was a vote on the project in either city or county votes for bonds to fund it. The public is a lot smarter about CG imagery that we think they are, so they know bad CG when they see it. Even if they don't understand why. I mean, look at the cat-a-strophic results of the terrible CG in Cats. The real benefit of Enscape was realized in those projects that had been through Enscape produced better quality Revit models for us to import, decisions about materials were already made, and we were able to focus back on how well we are visualizing the design intent versus do we like the design at all?

Same as Enscape. Architects at the controls of this software is good for internal review, but bad for public review. Architects tend to want to show the functional see everything view and twirl-n-hurl animations. They want to break the living hell out of the images with 2-sun set ups, perfectly clear glass on a brilliant sunlit day, and dusk shots where every surface is the same 50% illumination value. These images tend to fail hard when presented to the public, but again, now that the project as at least started in Lumion, we can focus more on art than the importing process. As far as Twinmotion goes, as long as Epic fixes the multiple UI issues then this software is going to be a very difficult competitor for Lumion. Twinmotion has generally always looked better than Lumion, but was beyond frustrating to use so no one ever really adopted it.

This is where we are starting to recover our workload. Rather than fight the trend of Enscape, we are now shifting to support our designers. It is no real secret that many architects and interior designers are dreadful material creators and many don't have the extra time to learn how to create good materials. So we can now help them create better base materials using Substance, which in turns gives us as visualizers a better start when it comes time for us to get involved.

These are great and all, but are not really that much of a threat. The reason being is that when working with designers and clients, you can't ever tell them that you can't represent the material or furniture they want because one of those sites doesn't carry it. These sites are great for projects when you just need stuff, but how much work does that really represent? If an architect or interior design has picked out a brick sample, have you ever been able to show them a render that has the brick that has a 50% "close enough" match?

Again, great technology but still has a long way to go to really get adopted. Right now, even with Datasmith, it takes far too long to get into Unreal than it does for Vray rendering or Enscape/Lumion/Twinmotion. If you don't want to spend time in optimization, then somehow you need to convince your studio and possibly your end client that you will need $5,000 or more worth of graphics cards just to view your project. There was an excellent webinar from HOK about using Unreal to visualize the new Texas Rangers baseball stadium, but they were happy with 5-10 FPS when walking around in Unreal with the client. That frame rate was with a Quadro RTX 5000, so no slouch of a GPU. Again, that worked for client reviews but would fail hardcore if that was released to the public.