This could be interesting.

I think a good render is one where can expect - with a degree of comfort and experience - that the depiction will make it from virtuality to reality with minimal loss. The render should be about the could, not the might.

Is the above a could or a might?

Render

Actual

(xten's sapphire gallery)

so what do you think is successful/unsuccessful about the rendering?

Well, BIG's render is incredibly evocative, yet it is also on the 'could' side of the ledger - architecturally, this could happen.

xten's obviously didn't happen, but even so the render is not good. The junction between the existing roof and new structure is poor. The lighting is not great. It didn't translate.

My argument is that it is not about a accuracy, nor is it about marketing. It lies somewhere in the middle. 

what is the right image format post images in the thread?

I use chrome - right click over suitable image - copy image url - then hit the image button in the add comment box and paste - does the trick.

just so i have skin in the game...

generally, i go for simple, quiet, neat, and clean.  so here are some images from a competition board.  this is just sketchup and photoshop.  i put a lot of emphasis on tuning the range of lineweights and colors.  generally, I'll go with 3, 5, or 9 lineweights.  By doing so, there is a subtle consistency and visual theme that is established.  The same with colors.  I like to limit the palette, generally, just because when I have more colors, it seems to be a more substantial challenge to integrate them well, and i never have the time to do so.  The same with text.  Only 1-3 fonts in 2-5 sizes and 1-3 colors.  I stick with a particular justification, again for consistency. 

I think my technique is serviceable but leaves a lot of room for improvement. 

A good rendering, to me (and my clients), is one that leans towards what a good photograph is.  It should capture reality, but present it in an attractive and captivating way.  Something that engages the viewer, what that really means depends on the audience (a competition is a much different audience than an image created for the general public).

Like photography, there is good and bad.  A good rendering is truly a piece of art.  Most, again like all things artistic, are mediocre at best.

does that mean photo-realism, trace, and if not, what qualities in a non-photo-realistic rendering would qualify it as having similar qualities to a good photograph?

same thing, some of my favorite photos are ambiguous.  I think anything 'art' oriented should engage the viewer, in some manner (that includes architecture, photography, etc.).  It could be many things, but should stimulate some sort of response (for example, the trend with dark skies/harsh highlights and soft glows with competition images would not go over so well for a retired couple looking for an idea of what their place will look like after it is built).

Both of those, imho, could be paintings (or something else), but they are photographs.  I don't think a great rendering has to be photoreal, but it should engage the viewer.  A basic photoreal rendering can look quite bland, too (like some generic shot some tourist would take with their phone).

This is the part many miss.  It isn't [only] about technological proficiency, but about having an 'eye' to what makes something 'good', engaging and captivating.

Back to renderings...some of my favorites blur the line between 3D, photography and paintings.  It is all 'virtual', so the limits are purely up to the artist.

thanks!  beautiful photographs

so it seems you're talking about composition --- about the rendering/graphic/image standing on its own as a composition, independent of its functional purpose --- but also being appropriate for the audience --- does that sound right?  i guess that's what I'm taking from your comment at the moment

Yeah, that sounds about right.  Think about it like this:  

Fine art photos = competition entry renderings

Commercial/advertising photos = presentation renderings (marketing, approvals, etc.)

Lotsa variables of course (demographics, project types, etc.).

I think you have to look at things similar to how old hand drawings were made - you need to convey architectural information and accuracy, but you are also (or can be) creating something that could be framed and put on someone's wall.

(BUT not every single project can be amazing, its just an approach to continually explore and make work you (and your audience) is happy with)

In my opinion, renderings should try and convey an experience, which photorealism does not always convey. Experience is more a collage of senses (real, perceived, thought) and can combine past, present and future as well.

Even when creating photorealistic renderings, we are collaging to a certain extent - why not make it more obvious and more of a diagram to engage the viewer better rather than provide something photorealistic that we have a smaller hope of actually achieving?

Well, again, that depends on your audience.  99% of the people out there won't relate to "diagrams".  I'd love to just make more artistic images and pretty pictures all day, but that wouldn't serve our clients well.

For competitions, you are presenting to a very select, highly educated group of people that will respond well to the more diagrammatical presentations.

As for "hoping to achieve", I would hope that you are presenting images, photoreal or not, that are close enough to what 'can' be done that it won't really matter.  Personally, I've never worked on a project where the rendering was so far from reality that it was misleading.  That kinda defeats the purpose.

I'd go so far as to say that this can be a problem with architects - that they/us get seduced by something (whether it be a pure basswood model or abstract rendering) that they/us fail to really test what the final building will look like.  

I hesitate to call renderings art, though they can be artfully rendered. A rendering has a specific descriptive purpose; art does not. Renderings should only be as evocative as the proposed architecture is in and of itself. Anything more I tend to call "weepy watercolors." Including birds and wheat and pebbles doesn't do anything to describe the architecture. To me it is little better than a Thomas Kinkade painting, even if more skillfully wrought.

This is not to advocate for photorealistic renderings. I actually think most photorealistic renderings are not at all descriptive; they are too bogged done in creating an illusion of reality than they are in describing the architecture. A good rendering brings out the salient features of the architecture, stripping away the extraneous details. It is a communicative tool above all.

So a photography can't be art, either?  I do not prescribe to some idea that art is only "art" if it does not have a purpose beyond its own existence.  Just like great music, the fact that it was created for a movie doesn't mean it has less depth than something with no purpose.

A great rendering is as much art as great architecture or great photography is art (imho).

Your opinion is clearly from the architect's side, where the architecture is all that is important. From a marketing side (I am on both sides, btw) it is about selling an idea.  That means that presenting something in the best way possible.  

Renderings, like architecture, music and photography, can be basic and shallow, only serving some immediate use, or it can be something exceptional that reaches beyond the immediate use or application (think how a great piece of architecture transcends its original program, or a commercial photograph ends up in galleries, or a movie soundtrack outlasts its movie to become a classic).

3D modeling and subsequent rendering process is pretty much an entry level position in architecture. For any young pups reading this, try to distance yourself from doing these professionally as soon as you can. Yes, I know they are fun to do, but from perspective of running an office, renderings are pure expense. 

Good or absolutely horrible? It doesn't matter. It's a marketing tool. "Hey Joe, make it look less like a cardboard box surrounded by other cardboard boxes and more like a fortress that YOU deserve that's totally in the forest and shit." While this is extreme, all renderings try to do the exact same thing: fudge the reality ever so slightly. Note the types of people we put in renderings. No fatties pls.

That's true from an architect's perspective, but all you youngons also keep in mind that you have a better chance at a more creative and lucrative career in graphics/rendering/3D than you do in architecture.  I had a similar attitude when I was on the other (the architect's) side of the fence, but being more in the marketing side now I have a much broader view of everything.

Keep in mind that it is, generally speaking, marketing (and money) is what makes projects happen.  So to everyone outside of the internally focused architect's circle, it is a crucial piece of the puzzle (and pie).

And yeah, it does matter if it is "good" or not.  How well anything marketing related is received translates directly into dollars for those financing/running/investing in the project.  Unlike architecture, quality is directly related to profit and is valued as such.

There is a reason that every major firm, from Gensler to Morphosis to Nouvel, hires out for much of their presentation work.

"There is a reason that every major firm, from Gensler to Morphosis to Nouvel, hires out for much of their presentation work."

This is a great point.  I never presume that I am producing arch viz or working toward doing so.  rather, i think of what i do as quick and dirty working method to explore and communicate ideas.  even so, the 3D modeling and rendering and animation tools are becoming so user-friendly that what can be 'whipped up' is often surprising (in a good way).  having said that, what i do is not in the same league as what real arch viz people do.  many of them are artists.  sometimes i lurk their forums.  it is astonishing how perceptive they are to slight variations in color, texture, and shadow.  they do work at a whole other level.

jmanganelli, when you say whole other level, are you referring to the stuff you'll find on cgarchitect? Those are definitely a step or two up from most of the images I've seen here. And I'd like to note that not all of them are photorealistic. The software is getting easier to learn, but learning to 'see' artistically is still as difficult as ever. To 'see' for rendering is a different, but not unrelated, skill set and I don't think it should be dismissed so easily by architects. Renderings area great way to engage the market, yes as marketing tools, but it's better than no engagement at all.

This could really turn into another post topic, but during the downturn I also found that a firm's visualization personnel are one of the first to go. Back in '08 I was hired on the strength of digital modeling skills picked up at a Chinese firm - Rhino, 3dsMax and Vray - only to be the first to go at my corporate American firm once we lost a much needed competition.  I had struggled to get out of the viz department - BIM, AutoCAD - anything to get back into the actual creation of a building, but the compartmentalization is tough to shake. 

Ironically, I already knew from my experience in China that there are scores of eager, highly skilled Chinese renderers without an architectural education, who would put my skills to shame - and for a fraction of the cost.

Back on topic, I find photo-realist renderings sterile and sometimes even disingenuous.  I appreciate an impressionistic approach; one wants to evoke a quality that describes the experience of the actual architecture, once it is finished.  No small task to be sure, but an ideal rendering in this vein will conjure ephemeral qualities such as light in its changing qualities, movement, and the general sense of occupying a space.  Something that conveys the designer's intent. 

Peter Zumthor famously refuses to have his buildings photographed, as this can't compare to the actual experience, and IM Pei refuses to show his working drawings for similar reasons.  Nonetheless, the reality for most architects is that most clients have little patience for anything other than a literal translation of how their money will be spent.
Board members, historic review boards shareholders, the general public - everyone needs to know what it will look like before a project gets the ok, and I struggle with whether this hinders the final product, not to mention the profession as a whole. 

The website for the Ken Roberts Memorial Delineation Competition has a good selection of different rendering philosophies. 

I think this approaches my personal ideal:

By Kevin Scott/Olson Kundig Achitects

The rendering you posted reminds me of hugh ferriss's renderings

That's a nice example.  

As to whether it hinders the architecture...only if the architecture sucks!  ;-)  Which is most architecture, really.

Generally speaking, the better the architecture the better the chances of making a beautiful image.  You can only put so much lipstick on the pig...

I'm breaking off this comment by rusty:

"3D modeling and subsequent rendering process is pretty much an entry level position in architecture. For any young pups reading this, try to distance yourself from doing these professionally as soon as you can."

and starting it as a tangential thread here.

OFFICE DA

@ fokt, though I love their finished work, and the early hand drawings and physical models are golden, I'm pretty underwhelmed by most of Office da's renderings.  Wanna throw up an example?

yes, example please

I've always been a fan of Lewis-Tsurumaki-Lewis and their renderings. Simple, yet showing a level of complexity. Refined, yet looks like they took a screenshot and cleaned it up with photoshop and then added some linework by hand.

I like LTL's distinctive renderings as well.  The oscillation between color/mass and edge/outline effectively communicates the design while asking the viewer to imagine lighting and materiality. I attended a lecture by Paul Lewis where he was asked about their drawings and stated  that the benefit comes through the overlap between the analogue and digital.  My guess is they quickly create massing in a digital program like sketchup, print and add linework on trace, then color the scanned hybrid with photoshop.  He also stated that they are influenced by humor and how it's produced spatially, citing Buster Keaton :)

On the topic of could vs. would, I think I'd be just over the side of could because you can impart some of the experiential aspect of architecture that a sterile realistic render cannot. LTL's renders, especially their perspective sections like Brian Henry posted, are really good at depicting mass and the experience of exploring the spaces between them. I used their renders as an inspiration in studio and since then perspectival sections are a must. :)

@E.T. Baldosser

I like how the renderings from Office da illustrate an economy in content and production. If you look through their projects, the content/views they choose to render speak for themselves and don't need visual fireworks. The way they are sequenced show the most important aspects and allow for clarity. The content/views are most important - bang for your  buck. They also have great elevations and sections.

They have flash and there aren't lots of renderings posted on other websites. But look through their projects and you'll understand what I mean.

I also love the images of Paisajes Emergentes. They're romantic, nostalgic, ephemeral, and it's all Photoshop.

Finally, not computer generated renderings, but up-close photos of models are always great and there are lots of examples... Holl, BIG, JDS, STAR, etc.