Impression of Autodesk 2009 and the danger of CG Arch industry disappearing

[Iain Denby]

Ah, Ian, you've rumbled me! Did you come to this conclusion after visiting our office?!

 

Ha...certainly not Simon!

[jkletzien]

I am still surprised that people think geometry, and accurate physics are the key to selling speculative real estate.

 

The only thing that sells purely on it's own merits as far as I can tell is heroin, everything else could really use a story as a vehicle.

[Richard McCarthy]

i have trialed max design 2009 with the mental ray feature...

and i am afraid to say that i agree with the begging of the end.

i work as a design draftsman in australia using revit architecture. i have not yet used revit 2009 however just be using revit architecture 2008 and the mental ray features of max design, it is apparent to me that quite good results are rediculously easy to come by and unfortunately to chaos group and the like , quite fast. im sure many people in the industry will be asking the question do i change back to the old slow mental ray or keep on the path of vray etc.

the features of revit 2009 are surely to turn heads and it cannot be dismissed that cg art is becomming more and more easily produced and time efficient. and the question is that will cg high end art only be used for marketing at the very end stages.

thus with the implementation of revit systems within the industry models will already be basically completed with small scale modelling to be composed within 3ds max. taking away more work from the artist and in turn more money.

i surely think studios will be shutting down because i know from my work that alot of jobs are not even put to the client for high end art it is just the inhouse artist that gets to play...

 

 

Excellent point, you hit it right on the head.

That is exactly how I feel, our office use Revit as well, and I think that's exactly what's going to happen, with more and more modeling down inside BIM software (I know the arguement that Revit is still LARGELY limited in modeling complex organic geometries, and highly inefficient in doing that, but nothing is going to say that Autodesk might now include it in the next release to make things easier)

 

Your point of "time efficiency" is exactly what I am thinking. If BIM (Revit) type software are taking away the hours of 3D artist, it is proportionally taking away the value of 3D artist as whole, any rational arch design firms / developer (with the current and oncoming housing slump, and the culture of profit driven/squeezing every last cent out of the poor souls they call it architects) would find it less and less apealing to send job to CG viz firms. It not only takes enormous time amount of time to externalize and liasoning with external contractor (the viz firm, from architect's point of view), and it slows down the whole design process if you are waiting for a preliminary CG rendering to be done by external contractor with uncertain time frame of getting it back. That's exactly the reason many firms now adopt to 3D BIM software, to be able to see it quickly in a fluid design process.

 

And it is also increasingly difficult to convince architect/developer to discard their inhouse modeling and sent it to external contractor for rendering. If majority of the modeling work is already been "DONE" why should we sent all our hard work out to someone to press the magic render pretty button?? And honestly in today's world, in particular in this industry, everyone are all very well inform, and majority of the "interior" work for arch viz (furniture/fittings/entourage) that are not done from BIM can largely be bought for pennies from online libraries, so the "REAL HARD WORK" is just fine tuning camera angle and lighting solution / render style. Is that really all that hard ?

[vizwhiz]

i Think This Thread will outlast The CG Industry as we now know it

[Justin Hunt]

This is a mindset that architects need to get over if they truely want to adopt BIM workflow. The basic idea of BIM is One model, Many sources of input. Architects need to get over the idea that they are the owners of the model (this only applies to BIM). The ability and need to share the model between different consultants is essential. HAVC, Engineering and any other relevant consultant must be able to plug in their information into the model. This is exactly what was said three or four years ago when BIM first started to make inroads to mainstream adoption. Architects traditonally have been very slow to adopt new technology. Thankfully things are getting better, and will be getting even more so as younger, tech savey architect raise up in the ranks into positions of aurthority.

 

This should extend to the 3D visualiser. The model has been done (hopefully properly modeled), why souldn't it be passed onto the visualiser to get those "pretty images"? 70% of the visualisation time is waisted modeling. Just imagine what could be done if that was dropped down to 30% (there is always the need for additional modeling).

 

You should know by now that visualisation isn't just about building a model and hitting render. More offten than not the texturing, lighting, animation and other tricks we employ are sidelined because modeling takes up too much time. How many "bad" animations have you seen or even done because they just wasn't enough time to produce a more polished product?

 

I for one hate modeling.

 

jhv

[Derek Entesano]

This is a mindset that architects need to get over if they truely want to adopt BIM workflow. The basic idea of BIM is One model, Many sources of input. Architects need to get over the idea that they are the owners of the model (this only applies to BIM). The ability and need to share the model between different consultants is essential. HAVC, Engineering and any other relevant consultant must be able to plug in their information into the model. This is exactly what was said three or four years ago when BIM first started to make inroads to mainstream adoption. Architects traditonally have been very slow to adopt new technology. Thankfully things are getting better, and will be getting even more so as younger, tech savey architect raise up in the ranks into positions of aurthority.

 

This should extend to the 3D visualiser. The model has been done (hopefully properly modeled), why souldn't it be passed onto the visualiser to get those "pretty images"? 70% of the visualisation time is waisted modeling. Just imagine what could be done if that was dropped down to 30% (there is always the need for additional modeling).

 

You should know by now that visualisation isn't just about building a model and hitting render. More offten than not the texturing, lighting, animation and other tricks we employ are sidelined because modeling takes up too much time. How many "bad" animations have you seen or even done because they just wasn't enough time to produce a more polished product?

 

I for one hate modeling.

 

jhv

 

But would you reduce your fee by 70% Engineers made a killing a while back when they traded in their drawing boards and could just add a few notes the the architects DWGs! Dont know of many models that arent nearly solely constructed by the architects. Until the ideal world BIM happens then the architect should get a fee for the model, or be happy with a basic shell as there is a lot of IP invested in some of these models in the way of objects and standards.

Personal opinion of course!

 

And I concur with many previously in relying on your artistic talent to set you apart.

[aaron-cds]

Excellent point, you hit it right on the head.

That is exactly how I feel, our office use Revit as well, and I think that's exactly what's going to happen, with more and more modeling down inside BIM software (I know the arguement that Revit is still LARGELY limited in modeling complex organic geometries, and highly inefficient in doing that, but nothing is going to say that Autodesk might now include it in the next release to make things easier)

 

Your point of "time efficiency" is exactly what I am thinking. If BIM (Revit) type software are taking away the hours of 3D artist, it is proportionally taking away the value of 3D artist as whole, any rational arch design firms / developer (with the current and oncoming housing slump, and the culture of profit driven/squeezing every last cent out of the poor souls they call it architects) would find it less and less apealing to send job to CG viz firms. It not only takes enormous time amount of time to externalize and liasoning with external contractor (the viz firm, from architect's point of view), and it slows down the whole design process if you are waiting for a preliminary CG rendering to be done by external contractor with uncertain time frame of getting it back. That's exactly the reason many firms now adopt to 3D BIM software, to be able to see it quickly in a fluid design process.

 

And it is also increasingly difficult to convince architect/developer to discard their inhouse modeling and sent it to external contractor for rendering. If majority of the modeling work is already been "DONE" why should we sent all our hard work out to someone to press the magic render pretty button?? And honestly in today's world, in particular in this industry, everyone are all very well inform, and majority of the "interior" work for arch viz (furniture/fittings/entourage) that are not done from BIM can largely be bought for pennies from online libraries, so the "REAL HARD WORK" is just fine tuning camera angle and lighting solution / render style. Is that really all that hard ?

 

This is exactly the attitude that is behind architects thinking they can produce marketing images that are often times mediocre. Can your BIM software model a specific chair that was selected by the interior designer? Can it place all of the plants species shown on the landscape plan for an exterior rendering? How about chamfering corners or displacement maps? Or perhaps a better question, will the architects who find the "magic render button" have a clue how to do the things that will set a good rendering apart from a bad one? Keep on producing those renderings for client design meetings and leave the marketing renderings to us. As long as the developers need to sell their buildings, we'll always have a place. The architects should be using in-house renderings for design. I've seen a lot of terrible renderings used for marketing because the developers were too cheap or just not savy enough to know we exist. The arch-viz industry is still very young. The magic render BIM button is only going to weed out the arch-viz companies who still do scanline renderings or who simply don't stand out.

[Charles Gaushell]

Amen!

 

Let's close in prayer.

[Justin Hunt]

Quote "But would you reduce your fee by 70% Engineers made a killing a while back when they traded in their drawing boards and could just add a few notes the the architects DWGs! Dont know of many models that arent nearly solely constructed by the architects. Until the ideal world BIM happens then the architect should get a fee for the model, or be happy with a basic shell as there is a lot of IP invested in some of these models in the way of objects and standards.

Personal opinion of course! "

 

With BIM the model is the final product, not the set of printed drawings. The prints become a by-product.

 

In the same light, the image/animation we produce is the final product, unless we agree to hand over the model and charge as such.

 

In both situations agreements need to be in place limiting liability for the mis-use and third party licensed content, which should include any additional contect created by you.

 

jhv

[Charles Gaushell]

 

With BIM the model is the final product, not the set of printed drawings. The prints become a by-product.

 

In the same light, the image/animation we produce is the final product, unless we agree to hand over the model and charge as such.

 

In both situations agreements need to be in place limiting liability for the mis-use and third party licensed content, which should include any additional contect created by you.

 

jhv

 

I hear what you are saying, but the BIM model cannot stand on its own without necessary notes and specs. The final product is the construction/bid documents which includes drawings, specifications, contracts, general conditions, and in many cases now and moving forward the digital models.

 

Many architects already require that a waiver be signed to limit the usage of drawings and other files that they provide to us. So that really shouldn't be a problem. Only issue I see is if they want to charge - quite frankly we usually rebuild since things are a mess - too many polys, poorly done or the like.

[Crazy Homeless Guy]

I hear what you are saying, but the BIM model cannot stand on its own without necessary notes and specs. The final product is the construction/bid documents which includes drawings, specifications, contracts, general conditions, and in many cases now and moving forward the digital models.

 

Many architects already require that a waiver be signed to limit the usage of drawings and other files that they provide to us. So that really shouldn't be a problem. Only issue I see is if they want to charge - quite frankly we usually rebuild since things are a mess - too many polys, poorly done or the like.

 

Yes, I have been involved in a handful of healthcare projects where they are starting to list the BIM model as part of the deliverables. AutoDesk has been marketing to the facilities people at these institutions, showing them how they can further benefit by having a BIM model for their building. It gives them better information on their building for maintenance, repairs. It provides feedback on what finishes are where, and what they are going to need to do to maintain them. It also opens the door for tracking equipment and more....

 

But while we are on the subject of licensing.. I am trying to remember the guys name from the first DMVC conference in Boston. He was a fill in for someone who I believe was from ScreamPoint, but had to cancel. He worked for a housing developer. He hit on a couple of things, the one that stuck with me is that the future is about licensing information and not about creating information. That is where the money is. He implied that there were 3 or 4 architecture firms in the US that were toying around with this. They would license there information to be used by various people involved in projects. I don't know exactly how that works, but it certainly brings up some interesting concepts.

 

This concept always makes me think of how Apple, Verizon, Amazon, etc... all want to license music to you, rather than sell it to you, and let you use it how you want. There is more money in the licensing. Verizon is the one that really honks me off the most.

[Justin Hunt]

Oh I agree that more often than not the Revit models are a mess, going further they are even a mess within Revit because sometimes its easier to bastardise a part from an unrelated family just to get the geometry for something else, instead of creating a new family/part etc. As such the information gathered from the model is inaccurate. Prime example is one modeller striped the legs off a table to use the top for a sunscreen (dont ask why). Then when ever a schedule would be pulled off the model there would be a list of a few 100 tables

 

This doesn't stop with the BIM model, it amaises me still at how badly organised some max models are. Not named, wrong scale, no layer or if layers are used wrong objects on wrong layer, I could go on.

 

As to the licensing, thats exactly what all the software developers are doing and what I was trying to allude to in my previouse post. How to implement it will be a nightmare, just look at the warez industry.

 

jhv

[mat8iou]

Working as an architect, I have deliberately had to work to get involved in the architectural side of things, rather than being allocated only visualisation / presentation tasks (because at that stagej I was the fastest in the office at producing them). It is the same story in many other practises. Architects studied architecture because they wanted to be architects - if they had wanted to only be producing visuals, then they would have done a different (& no doubt shorter course). So, while I can produce visuals for my own projects, when needed, there are many people who I work with who can't produce 3d visuals & I don't wan't my job to become one where I'm spending my time visualising other people's work (I don't have a problem with that type of work as such - its just not where my core expertise is).

 

As technology has improved, so have client expectations. We are contracting out more visual work than ever before & even opened a sub office largejly to deal with this type of stuff. So I don't see the need for people to produce good quality visuals to dissapear any time soon.

[Scott Erstad]

The only thing that sells purely on it's own merits as far as I can tell is heroin, everything else could really use a story as a vehicle.

 

... and sex. Sex is a big seller. Huge.

Scott

[Justin Hunt]

Well we have started to roll out Revit 2009 across the office and I can tell you now It isn't the Magic One Button Solution.

 

The funny thing with the rendering pre-sets, even on low the settings are way too high. One of the projects is very complex, two 30+ story apartment towers sitting on a retail/commercial podium. Its a very detailed model.

 

On Draft at the 75dpi setting FG still had not finished after 22 hours of rendering. This is on a quad core with 2 gig of Ram.

Needless to say we stopped the rendering. I had a look and tweaked a couple of settings. Nothing major and BAM 20 minutes later a render spat out. Not great but at least something.

 

As with anything to do with rendering it is vital that one understands what the settings mean and the impact of changing them. I think this is a prime example of how dangerous "Magic Button" rendering can be.

 

jhv

[Christopher Nichols]

I only juts saw this thread and started reading part of it... but it got really long, and I felt a need to reply as an ex-architect, ex arch-viz, and present visual effects person. So forgive me if my points are repeated.

 

Many of us in the VFX world are pissed at Autodesk for different and similar reasons. Autodesk is killing 3dsmax to make the archviz people happy. Why? Because there are 10 times more archviz customers compared to VFX customers. So we are forced to switch to Maya... which is progressing at an even smaller pace than 3dsmax. Archviz is king in the 3D world actually because there is SO SO SO much of it. While VFX gets a lot of press, there are only so many licenses they can sell. So they gear their licenses towards their audience. When they realized that archviz outsells vfx by a factor of ten, max started dying in the vfx world. For those of us that want things like Vray, it is starting to sting. The move to Maya, Linux, and Vray will be a hard but worthwhile battle, but it is the same enemy. Autodesk is pushing Mentalray like their ugly daughter's dowry.

 

Anyway... to answer your push button great render issue. That only gets you so far, and it gets old real fast. Just take a look at those sun and sky renders from Maxwell what were so popular a few years back... everyone's renders looked like them, and they got old really fast. Eventually, people come back to the artist, not the tool.

[Spooner04]

This is nuts. Five years ago it was gloom and doom because our work was all going to be outsourced. Now all the architects are going to be doing it? Whatever, i'm not buying it. Let them have their revit models and mental ray textures.

 

 

I'll tell you the conversation we'll be having with the architects for the next five years......Send it to me when you're done. No, I'm not going to give you a discount for having it modeled already because it will take me 3 days to clean up all the overlapping faces and horrible geometry. Whatever time is left over will be used rendering out the animation prepasses so that you can have shiny, high poly cars and axyz mocap people that move. We'll be taking the time savings from the front end and putting it on the back end.

 

I will say that I'm worried about chaos group. But, their rate of improvement consistently out paces mentalray. As long as that is the case, the people that know what they're doing will be using vray.

Edited May 7, 2008 by Spooner04

[M V]

I work in at firm that uses Revit and SketchUp. I do not think anyone is in any danger of losing their jobs because of these programs. You need to know the technical and the artistic aspects to get great illustrations. We are artists and most of us are talented technical wizards at what we do.

It takes so much more than clicking a render button to create a decent presentable illustration. In my work I use 3-5 different programs and a multitude of both computer and art techniques that come with years of experience.

The firms that are satisfied with a ‘quick and dirty’ Revit rendering (using Mental Ray or not) are not going to be hiring professional renderers anyway. I guess my point is, that we sometimes take for granted how much back-end knowledge we have when it comes to making the software sing. I have looked at Revit 09 and tried to render a few models already. It's much slower than Max or any other rendering engine I have ever used, the results are good (better than Accurender that’s for sure), but you still have to have a good understanding of the software to get something decent out of it.

Most Revit users only use about 60% of the programs modeling abilities and a rendering is only as good as the model and most firms have crappy family libraries and stick to drafting 2d lines on elevation views.

I agree that we need to evolve. Our profession is constantly evolving to keep up with new technology, faster computers and more intuitive software. What we must do to set secure our livelihood is to take active roles as artists. Talent will never be something Autodesk can through in a box.

[landrvr1]

Yes, I have been involved in a handful of healthcare projects where they are starting to list the BIM model as part of the deliverables. AutoDesk has been marketing to the facilities people at these institutions, showing them how they can further benefit by having a BIM model for their building. It gives them better information on their building for maintenance, repairs. It provides feedback on what finishes are where, and what they are going to need to do to maintain them. It also opens the door for tracking equipment and more....

 

But while we are on the subject of licensing.. I am trying to remember the guys name from the first DMVC conference in Boston. He was a fill in for someone who I believe was from ScreamPoint, but had to cancel. He worked for a housing developer. He hit on a couple of things, the one that stuck with me is that the future is about licensing information and not about creating information. That is where the money is. He implied that there were 3 or 4 architecture firms in the US that were toying around with this. They would license there information to be used by various people involved in projects. I don't know exactly how that works, but it certainly brings up some interesting concepts.

 

This concept always makes me think of how Apple, Verizon, Amazon, etc... all want to license music to you, rather than sell it to you, and let you use it how you want. There is more money in the licensing. Verizon is the one that really honks me off the most.

 

 

Yes, there's been a lot of talk about licensing and other "Let's Take Back The Streets!" money-making schemes involving BIM. The architects and promoters of BIM are all "Hey, we'll get back our respect and increase our earnings by selling the BIM information/data/files instead of giving it away!"

 

So, so, so naive.

 

Let's me honest here, shall we? After all, it's architects we are talking about. The worst business people on the planet. Zero spine. Zero backbone. But I digress...

 

In the not-too-distant future an architecture firm is actually going to sell and/or license BIM data for significant profit

 

Exactly 2.5 seconds later, the architecture firm down the street is going to announce that they will be giving all of their BIM data away for free as part of their 'service'.

 

 

Edited May 7, 2008 by landrvr1

[landrvr1]

Well we have started to roll out Revit 2009 across the office and I can tell you now It isn't the Magic One Button Solution.

 

The funny thing with the rendering pre-sets, even on low the settings are way too high. One of the projects is very complex, two 30+ story apartment towers sitting on a retail/commercial podium. Its a very detailed model.

 

On Draft at the 75dpi setting FG still had not finished after 22 hours of rendering. This is on a quad core with 2 gig of Ram.

Needless to say we stopped the rendering. I had a look and tweaked a couple of settings. Nothing major and BAM 20 minutes later a render spat out. Not great but at least something.

 

As with anything to do with rendering it is vital that one understands what the settings mean and the impact of changing them. I think this is a prime example of how dangerous "Magic Button" rendering can be.

 

jhv

 

Great post, Justin. Thanks. Good to hear from someone who's tested this out.

[aaron-cds]

Let's me honest here, shall we? After all, it's architects we are talking about. The worst business people on the planet. Zero spine. Zero backbone. But I digress...

 

In the not-too-distant future an architecture firm is actually going to sell and/or license BIM data for significant profit

 

Exactly 2.5 seconds later, the architecture firm down the street is going to announce that they will be giving all of their BIM data away for free as part of their 'service'.

 

 

 

lol, so true

[Justin Hunt]

Chris has a point, arch viz is a major driving force in 3D, but it hasn't always been so. For a long time we were considered the poor cousins to VFX folk. It only after many years of pleading and pushing the envelope in terms of render engins that we are in the position we are in now.

 

Also consider the droping of Viz in the push for Max to be an architectural tool. On the other hand Maya has always been touted as the VFX tool of choice. Its only very recently that Maya has made any inroads in arch viz.

Correct me if I am wrong but hasn't Vray been ported to Maya already? if not the some time soon?

 

jhv

[senrikyu]

I had an interesting experience talking to a local renderer who has done some gorgeous images with just Revit and Photoshop. I asked him what the polygon count was on one of his models, and he replied, "What's a polygon?" Wow!

 

It made me wonder how long it will be until anyone can create super realistic images and animations by hitting a few buttons. Knowing about things like polygons, mapping coordinates, textures/materials, etc. will be redundant one day. Luckily, we're not there yet- at least for a few more months!

[Adrian Bold]

On the other hand Maya has always been touted as the VFX tool of choice. Its only very recently that Maya has made any inroads in arch viz.

Correct me if I am wrong but hasn't Vray been ported to Maya already? if not the some time soon?

jhv

 

Justin, our studio has being using Maya and MR since 2002/3. Chaos announced a plugin for Maya back in 05 but the latest we've got from being on their beta team is a release last year that doesn't work very well. But I think we're alone on this matter as my travels and discussions with most firms overseas is using maya for arch-viz is overkill and so the demand for chaos isn;t huge... anyone else use Maya for arch-viz?

 

Comparing the interface between vray and MR, i'd have to say the latest MR engine has come along way in simplicity and physically-accurate lighting. I've been exposed to alot of our clients' architects using revit but the case is always the same. "cant be bothered with anything more than a sketchup or conceptual render, would rather outsource to the pros"

 

Like Nils at neoscape our clients are the developers, not the architects; and like Nils, we are forever picking up work that already has had some warm-fuzzy 3D work performed, and now the client want the pros to add both the static and cinematic touch to communicate a technical and washy concept to stakeholders and the general market; (just like an ad agency would manage a brand.)

 

Often this work that has previously performed 3D is either poorly or surprisingly reasonably well done, from inhouse architects; other freelancers or off-shore sweatshops that screw it up. We generally dont touch any 3D model data files from elsewhere because they are usually a hacked mess.

 

My 2c to add here is that we aren't just Artists. My staff range from engineers to brand managers; and having the intent of solely making a pretty picture with the warm fuzzies I think is the wrong approach. I often cringe when i see an image that i can tell the architect has had a hand at creatively controlling.

 

In our business, we work with clients at a concept stage, DD, DA through to marketing and advertising; so having a fusion of skills that understand technical drawings, council policies, buyer behaviour, market trends and client intent are critical to achieving a result that works, sells, and won't bite the client on the bum with a legal law suit over false advertising once the site is built. Each project is different, and the design of it alone is not the end-all to visualisation.

 

A 'make render' button that can tailor our work for each project to this degree does not exist, and as long as a market exists where clients need to outsell their competition down the road (all hail capitalism) professionals like us with knowledge and infrastructure to deliver are still needed.

 

Someone else's comment about the Crytek engine still holds too. This could change the industry significantly and reduce the need for freelancers photoshopping single images, (rather just screengrab) probably for the better. As per above, Someone, more like a team, will still need to put a fully-comprehensive and believable scene together for the mass-market to take notice of it. Medium and portability for this is another story.

 

cheers.

[Justin Hunt]

I do both work in-house and freelance.

 

Over many years we have built up the in-house 3D department to a state where we are being asked more and more to do marketing images and animations.

 

What I find quite funny. On the occasions where we have gone head to head with a viz studio in pitching for a job. The Viz studio often wins the job, only later the client then askes us to either hold the Viz studios hand in getting the job done or having us redo the job entrily. In fact I am doing one right now. The viz studio is successful )in that its been around for a while) but the images they produced were shocking. We did the arcitectural design and after alot (far too much) of backing and forthing the client has asked us to redo the 3D's.

 

Just because its in-house does not mean that its crap.

 

On the flip side I am doing more and more freelance work as alot of the architects that the developers use are not capable of doing 3D as well. There are more small (less than 100 people) architect firms than there are large one. 3D is still considered a luxury by many architects firms. As such there will be plenty of work for viz studios for a long time yet.

 

jhv