Karin Skaug

I appreciate the feedback, I am in the process of writing down (what might possibly be no more than) a "wishlist" and I want to ask you what things makes the most trouble when cleaning Revit files for visualization. My points so far is: - Avoid overlapping geometry. Two different elements can't occupy the same space in real life, it should be the same in the digital model. - Avoid duplicates for the same reason as mentioned above. - Be cautious when using models off the internet, check that the detail level is appropiate and the file size is acceptable. - Whenever possible; use square profiles instead of circular ones for railings and other small details. - Do a visual check to ensure Revit has snapped the walls together the way you intended. What would you add?

Thank you for the feedback. One reason I don't stay in Revit is that most often the building is only a part of the illustration (albeit an important part) and there is much landscaping and vegetation happening that I think is better handled by SketchUp and Skatter (plugin). Another reason is that most part of what I do is photomontages and the Revit camera does - to my knowledge - not have the physical parameters I need to correctly match a view to a photograph. Enscape is promising to making the entourage bit more manageable from within Revit, so maybe in the future I'll work more Inside Revit, but for now I need/want my workflow to be via SketchUp. Walking around in Enscape seems to catch many errors in my firm as well, but it seems I am in no position to really force their behaviour, I can only "encourage" them to do so.

I work as an inhouse visualizer for an architectural firm. The usual workflow is to import Revit files via the Simlab plugin to SketchUp, fix them, add context and entourage and then render images with Enscape. I have been struggling with a file for a rather large building, it comes into SketchUp as 20000 objects and crashes when I try to render in Enscape. The (exported) error report from Revit is nearly 2000 lines, which seems to imply that there are issues with almost 10 % of the geometry. Usually I am able to fix things within SketchUp, but it is really a stupid way to work, as I have to repeat it in the next revision. It also uses my allotted time to fix other peoples mistakes, making the visualization process more expensive than it should be, thus making me less competetive. The question I have is how many errors or bad modeling in the Revit file do you forum readers accept, and can you really demand that the architects "clean up their own mess" before sending it off to visualization (or other services)? Do you all get clean, perfect models from your architects, or do you have some sort of quality description/minimum standard for the files you work with? If that is the case, does anyone have the opportunity to share such a description with me? Earlier, they only messed things up in 2 dimensions, but there is an awful lot more that can go wrong i 3D...

I have been working as an inhouse visualizer for an architecure firm for the last 14 years. In my area clients generally don't want to pay for more than two days work for an illustration, so that has put a brutal limit to all my ambitions. I never have the time to mess around. I now use SketchUp with Enscape because they are quick to work with, that's how I can free up most time to be vaguely creative in my imagemaking. (But they also have a lot of limits, so there is still much I can't do.)

I switched from 3ds Max to SketchUp (almost) three years ago. It was hell at first, but now I find Max the hellish option - when I have to go back to Max to do something - it seems like everything need far more clicks to do the same ting as in SketchUp. So you will be able to work faster in SketchUp. I have a lot of extensions. It is hell when they upgrade and you have to load 75 new versions one by one. But some of them make all the difference. You can get instant VR experiences with Enscape (without exporting or prepping anything!), ForestPack-like scattering with Skatter and organic sculping with Artisan. Most of the extensions are free, some cost more than SketchUp. I use the simlab exporter in Revit to get a neatly organized model- with textures- into SketchUp. (It even respects instances, so you get multiple linked copies instead of hundreds of "stupid" unique objects as you got from Revit to Max.) The Revit models often needs more detailing, so you still get revision hell if that is the case. "My" architects usually have a building file, and a topo file with the buildings linked in. I usually import the topo file since I then get the north direction correct and I don't have to adjust the position of the model with each revision import. SketchUp is sadly really weak at handling lots of geometry, but Enscape is amazingly strong. Enscape gives you the possibility of using proxies, and that makes you able to easily work with large projects/lots of detail. But you have to link/proxy almost everything. I'm happy now with how the program works, but nonetheless I still feel a bit ashamed when I tell people which program I use. My bad, I guess.

This is my experience as an in-house visualizer. I have been working as a visualizer in Norway for 16 years and things have been changing a lot. The first years, I was doing "magic" and everybody was amazed to see any kind of 3d- images of their project. Then Revit came along and people realised it actually wasn't magic and started making their own illustrations, leaving the complicated stuff to me. I switched my workflow from FormZ to 3dsmax/V-Ray to keep ahead of the rest. Then Enscape came along and it seemed just plain stupid not to use it. I switched my workflow to Sketchup/Enscape and so far everyone is happy with things as they evolve. Now everyone here is surrounded by magic every day. The single fact that you actually see what you do is enough for me to never go back to any offline render. SketchUp does not have... a lot of things, but most of the time I never used the advanced stuff anyway. Enscape does also not have a lot of things, but in the beginning, when I compared, I was not able to get better renderings out of V-ray that I did out of Enscape. My work isn't at the high end of high end, it's just as good as it can get within the constraints we have. We normally can't charge more than two days work per image in my region. Competition work is different, but we don't do a lot of them anymore. People won't work for free anymore, so it gets too expensive. Covid hasn't really impacted any of our projects, so it's mostly business as usual, ...now in realtime. (The exception beeing VR - I haven't touched the VR goggles since the Covid outbreak) Well, that was my 2p... :-)

I think one of the most important - and most difficult thing - is making it just good enough. Not spending unnecessary resources, but not making it too simple and get a disappointed client. Earlier our company had a slogun saying we would bring "that little extra", and I still try do do that. Surprise them a little in a positive way, like finding a view that includes a local landmark out of the window, putting images of the historical use of the site on the living room wall or shooting background photos which includes local arrangements or other features. Things that bring more feeling/sense of the location, I guess. My clients have always expressed appreciation for this.

Thank you for your reply. 5 days seems to be fair. When I not-very-scientificly compared living expenses using the the Big mac index and average salary statistics for different countries, it gave more or less the same result. If the clients spend their "normal" amount of money, they can get two days work with my company's rates or approximately a week from one of the companies based in Eastern Europe - or more like two weeks for a Vietnamese company. Please have in mind this is a very rough and unscientific conclusion. I don't know what examples of my work you have seen - I haven't published anything I remember, and my companys website features work from other artists as well. (It hasn't been updated with any of my work after I got back from maternity leave (which is generous in this country and of course also adds to the high rates)). I do what I think is regular small city stuff - planning, schools, office- and appartement buildings in any phase, exteriors and interiors, often photomontaged with drone footage capured by me. A competition now and then. I haven't published anything at cgarchitect as most projects are confidential until they are... well, outdated, but I guess I'll start to get permission to publish to get better score from now on.? Thankfully, we have strict law regulations about working hours.

I've been a member here since 2012, but haven't contributed that much. I have been working full time with visualizations for 15 years now, but I still experience the imposter syndrome and get overwhelmed by all the talent people show. The background for my question is this: I started out at roughly the same time as MIR. I used to think I was able to make the same sort of images they could, if I only got the right client. The years went by, and I got the right client. I made some twilight renders with fog and everything and found out I was far from anything remotely MIR-ish. But as for most of us (I guess), the fancy competition renders are not what makes the bulk of the work, so I have continued working full time as a visualization artist/technician, only with lesser self esteem. I guess I can call my work mid-range quality. Not fancy, but always as correct as possible with regards to context, views and lighting conditions. In my early years, competition wasn't really an issue, but now it is. I work in-house at an architectural firm, but the illustration work is most of the time contracted as a separate job. I happen to live in a very expensive and cold country, so globalization does not favor me. To exaggegrate that, I live/work in a part of the country where housing is (relatively) cheap, so the developers have very tight margins. All this result in me having to deliver stills in two days. That's what the clients in this area are willing to pay for. My question to you is, how long time do you use to produce a mid-range quality still image? (Not fancy, but always as correct as possible with regards to context, views and lighting conditions.) The question of "how much should I charge" comes up now and then, and is almost impossible to answer due to globalization and all the different types of assignments and the level of expertise. But the time spent per picture could be easier to define. I once read that the guys at MIR were expected to deliver one image per week. How long time do you get?

Well, I ended up with a mix of simple massing, photos and photoshop actions to make it "painterly". We didt win the competition. The client asked for a simple massing model of a large mixed use area, but the other competitors delivered "the full monty" with detailed plans, material schemes and -lots of detailed renderings... I really don't like these competitions.

I work as an in house visualizer at an architectural firm, but my illustrations are mainly not included in the main architects fee, but extracted as a fixed price job. But every time the project changes, and often the agreed deadlines are missed and we lose money doing changes without getting paid for the work. I struggle with finding a way to manage the changes. Not starting the work until the project planning is "finished" isn't an option, as our clients want to start marketing early. A solution is to have a fixed price but to charge the client extra for changes initiated by the clients, and to put the cost of the architects changes into the architects fee, but experience shows its a mess of who orders which changes at different times in the production. How do you manage changes when you have to have a fixed price agreement? Any sharing of experience with setting up contracts/agreements to cover this would be helpful. Thanks.

Can anyone point me in the direction of good 3d illustrations of planning areas - with buildings as simple mass/volumes without any details? I have tried both Google and Pinterest, but I only find detailed renderings even for so-called "concept work." I have to make something nice -without facades- for a competition and need a leg-up for visual style...

Thank you for helpful answers. I also got a very helpful reply from Des in Ireland on the formZ forum on this topic: http://forums.formz.com/index.php?/topic/3955-documenting-building-render-accuracy/?p=14350

As a client of ours is deploying quality control of renders, another company has been caught cheating in their 3d-illustrations. We are now asked to describe how we arrive at the representation of the new buildings and document their accuracy compared to the existing surroundings. (This is regarding photomontages with a rendered project composited against a photo background.) Have any of the users here been asked to document the accuracy of you renders? How do you - in that case- do it? (It is a timely, yet strange idea to quality check cg-illustrations, but as they often are delivered as a representation of a "future truth", is is a reasonable question. I just wonder how they do the actual checking, as there is so many software packages and different workflows involved in producing this kind of imagery)

Thank you. (All of you).