We interview Juan Manuel Bautista Hoepfner, architectural visualization designer, professional 3d artist and instructor about how he has used Shade.
It all really started back when I was in the university. I was hit from behind in a car crash, and had to wear a cervical collar for more than four months. As I was a studying Architecture at the time, I had to find a way to draw without straining my neck, and computer graphics was the answer. I became aware that I could draw better, faster, and with much more precision than using traditional drawing methods. This led me to be something of a cg pioneer in my university, as I was the first student to successfully create a complete architectural project using the computer alone.
Fresh out of the university, I opened what was probably the first archviz studio in my hometown, Puebla (Mexico). It was quite a success, but I partnered with the wrong type of people in the endeavour (non-creative and lazy partners), so I closed shop. Fast forward some years later, I was fed up with my current job. I was the head of the waterjet division in an architectural stone workshop. Which meant almost I did everything remotely creative, not just waterjet-related. However, I yearned for those good times doing creative computer graphics. So I began researching for a 3d suite to invest in, but the high end packages were prohibitively expensive, and reasonably priced ones were just not good enough. I got an email from Curious Labs on a pre-sale offer on Poser 6 which included Shade 7 LE for free, and it got me curious. I found out that Shade was a 3d program developed in Japan, with almost no exposure in the west, and more importantly, it was good for archviz, and it was free with Poser 6.
So I took my chances, bought the bundle, and I have never looked back since.
Because I like Shade! Seriously, I do. The way Shade works using a parent-child child relationship for everything just makes sense to me. That, and because it has the features I need, at a price I can afford. That, and that Shade forces me to think in an orderly fashion. It might be a bit confusing at first to adhere to a workflow, but in the professional world, a good pipeline workflow is 90% warranty to a job well done.
I don't use Shade alone. Generally, I work with a combination of Shade, Silo, Autocad, Revit, , Poser, 3dcoat, and zbrush. Shade fits nicely in this multi-program pipeline. I also have Lightwave and Vue, but I seldom use either this days. Not because they are not capable (they are, big time!), but because Shade has grown quite a lot, and I have no huge compelling reason to switch from Shade to Lightwave (complex render setups) or Vue (ecosystems) this days.
I think I partially answered this in the first question. But more to the point, I started learning 3D by myself (as nobody else knew anything about it where I lived). It helped me survive a semester in the university, and after that, it empowered me to work faster, with much more attention to detail. This is still the case, and there are even times when I give feedback to the client on mistakes on his projects derived from accurately depicting in 3d. So it has not only helped me solve problems, but indirectly solves my client's problems.
Instancing! Instancing is a blast. You can instance trees, buildings, hairs, cars, all this without impacting memory consumption significantly. And instanced items cast/receive shadows, reflections, refractions, etc. There is no way to distinguish what is instanced and what is not in the final render. Then there is this option to choose an image's channel (r, g, b, or alpha) as a greyscale mask. Modern game engines use this all the time to drive shader attributes, and Shade supports that natively.
I try to divide the composition into smaller chunks. Then divide again into smaller chunks. Divide and conquer. It can make a seemingly impossible task much easier to handle. For example, before "populating" a scene, I create parts (groups) that will contain things. The main item, the secondary items, the background items. In those, I again create parts that dwell into more detail. This is especially important for projects that might have revisions in the future, where it is easy to forget just what you were thinking when you were working on the composition.
I have an AMD 1090T Phenom II x6 with 16 gigabytes of ram, with two 1terabyte hard disk drives, working with an nVidia 560gtx video card. All in a Cooler Master HAF 922 case, and a 700W Cooler Master power supply. With all the fan slots in the case occupied with fans, and an oversize heatsink for the CPU. Using a 22" and a 24" monitors. I hope to upgrade to a SSD drive and 16gb more ram soon!
In other words, I believe that investing on a solid piece of hardware is worthwhile. In this, I mean that when your computer runs cool, your computer almost never crashes. When a render can take several hours, it is not a good idea to have an overheating problem, or have a system crash because your power supply simply cannot give enough power.
And also something very important: clients don't just look at your portfolio. At least here, they also look at how cool your equipment is. The more your rig looks like the batcave, the better the chances you'll land an assignment from a local client.
Experiment! Shade now has a very robust viewport preview renderer, so unless you're working on something terribly complicated (and then you're not just starting in 3D!), the feedback is near realtime, even with global illumination turned on.
Investigate! The Shade community is very helpful, and there people like Larry Bolch are so knowledgeable on applying real life photography experience into computer graphics, and all this without having to pay for all the advice and help!
And, read the manuals! Sometimes the answer is several mouse clicks away.
For archviz, which is most of my cg work, I model the buildings myself, and rely on Poser figures for my human scales (which import great in Shade via PoserFusion), and heavily use pre-made content for foliage, furniture, and things like cars and general scene decorations. When there is something a bit out of the ordinary, like a special brand of a lamp, I model it myself. I used to do everything from scratch, and I sometimes do to brush up on my modelling skills, but sometimes, I'd rather spend time detailing the mood of the scene than modelling the perfect car or the perfect human figure, which being archviz, are only supporting elements in the scene.
This is a tricky question! There are many things I like in Shade. But certainly one of my favorite features is MultiPass rendering support. The possibility of being able to quickly tweak reflections, refractions, depth blur, etc etc etc without having to re-render, is simply wonderful. That, and the viewport preview rendering. Sometimes the preview is so good, I save the preview image and send that to the client, with no need to do a "real" render.
Apart from what Jürgen already mentioned, I'd ask for a faster viewport redraw speed. I don't know how difficult this might be, but I just like to fill my scenes with stuff, and now that we have excellent instancing support, seeing thousands of trees and cars and people and furniture in a viewport can sometimes not be as fast as I see them in my dreams. Oh, that, and a Shade logo as an image file that I can print and iron on to a T-shirt!