why has GPU rendering not completely eliminated CPU options

[Tommy L]

My only argument against that is until computers that "Think" are mainstream, they will not take over the world.

 

One thing that all those professions have in common is listening to people talk and being able to not only understand but interpret what they are and are not saying. Same goes for our jobs, I for one welcome one button rendering, then it will truly become like photography, leaving me to get on with the "artsy" side of the job. As it is far too much of our time is taken up fiddling with settings.

 

jhv

 

Half of me agrees with you, but the other half knows that the fiddly-ness and hardware intensive nature of the job keeps my income safe. I dont consider myself a talented artist, but extremely technically proficient with a streamlined workflow gleaned from experience. This efficiency gives me an edge, one I would be sad to lose.

[Devin Johnston]

Tom's right about the technical nature of our jobs keeping our positions safe, that advantage goes away when photo real images can be created as easily as someone can print a document. I think that animations, VR, and augmented reality are the future of arch viz, but even animations will become much easier to create and render very soon.

[Tommy L]

Funny how far this thread has strayed from the original question. But the discussion frames the original question in a different context.

Even if GPU does make work quicker, it just broadens the toolset and enables a higher bar, which is good for everyone. It will also pervert the aesthetic somewhat, maybe standard imagery will become a bit 'gamey'?

The median aesthetic follows the path of least resistance. You can see the ingrained workflow of the majority in their work (myself included) and thats because we are geared toward efficiency, not laborious creationism (aka production artist vs 'fine' artist). Strides in tech tow this evolution along and those that remain at the fringes are the ones who really excell, even if commercial recognition does not always accompany their artistic acheivements.

Remember Domelight plugin? Generated a rig of spotlights to emulate GI. I bet if I started using that again on all my exteriors it would eventually become my 'style'. And style never goes out of fashion.

[panagiotiszompolas]

Funny how far this thread has strayed from the original question. But the discussion frames the original question in a different context.

Even if GPU does make work quicker, it just broadens the toolset and enables a higher bar, which is good for everyone. It will also pervert the aesthetic somewhat, maybe standard imagery will become a bit 'gamey'?

The median aesthetic follows the path of least resistance. You can see the ingrained workflow of the majority in their work (myself included) and thats because we are geared toward efficiency, not laborious creationism (aka production artist vs 'fine' artist). Strides in tech tow this evolution along and those that remain at the fringes are the ones who really excell, even if commercial recognition does not always accompany their artistic acheivements.

Remember Domelight plugin? Generated a rig of spotlights to emulate GI. I bet if I started using that again on all my exteriors it would eventually become my 'style'. And style never goes out of fashion.

 

Hi,

 

I agree with pretty much everything you said except for the statement of GPU rendering introducing a more 'gamey' look. Unless you're talking about true realtime GPU renderers (like Unreal, Unity, etc), the majority of GPU ray tracers out there execute very similar rendering algorithms to their offline CPU counterparts, so there shouldn't be any differences in the look.

 

Some of them do have faster rendering modes which are taking shortcuts (like that Octane video above, hence all this blue light leak) but these modes typically exist along with more accurate rendering options.

 

-Panos

[Michael J. Brown]

One additional point to consider is how firms like Gensler are in essence diluting the value of architectural services with the open broadcasting of their use of GPU rendering.

 

Think about it. These extra visualization services they are offering (and won that one client with) are likely not being pushed as an added service (for which extra can be billed), but as an added benefit of their services, the cost of which gets absorbed by Gensler (like any other overhead). They even said in the article that they plan on equipping every architect's desktop with this capability. Therefore one could expect that before long each person will be expected to generate renderings and animations as part of every job alongside the other required documents.

 

Meanwhile 'the client' is being conditioned to expect all these extra bells and whistles at no additional cost. Firms like Gensler get so excited about being able to produce in 5-minutes what used to take 5 hours that they foolishly make press releases trumpeting these new advancements to the world (forgetting that their clients are likely reading these publications too!). So what happens when the Gensler's try and squeeze extra fees from said clients for said CGI services? The client's got the architect by the balls because he knows the architect is fully capable of popping these things out in his sleep. Thus there is no reason under God's green earth why the client would or should pay one dime extra for such services!

 

These advancements, when openly proclaimed to the world via press pieces like this, are damaging to the architectural profession because it ups the ante on what is to be expected in a set of finished CDs.

 

Some of you are old enough to remember when desktop (computer) publishing started putting people out of work. Heck, it's so commonplace and interwoven into every computer, laptop and tablet these days that nobody even thinks of it as "desktop publishing" anymore. Similarly the web design template sites like WIX are putting the power of (basic) website design in the hands of the common man and eliminating the need for many web designers. What sites like WIX have done is de-value the skill necessary for creating that sort of content. The design savvy programmer/code writer will look at a WIX site and scoff, "PHFF! Look at this piece of #%&@* website design, blah, blah, blah" But the common man looks at it and says, "Perfect! This is exactly what I was looking for."

 

I can see the same scenario unfolding now with architectural CGI. Mark my words, within a decade you'll have the purists (us) barking about the 'art' of the illustration. While 90% of clients will be more than satisfied with the WIX's of CGI.

 

 

just my $0.02

Edited July 25, 2014 by renderhaus

[Tommy L]

Hi,

 

I agree with pretty much everything you said except for the statement of GPU rendering introducing a more 'gamey' look. Unless you're talking about true realtime GPU renderers (like Unreal, Unity, etc), the majority of GPU ray tracers out there execute very similar rendering algorithms to their offline CPU counterparts, so there shouldn't be any differences in the look.

 

Some of them do have faster rendering modes which are taking shortcuts (like that Octane video above, hence all this blue light leak) but these modes typically exist along with more accurate rendering options.

 

-Panos

 

yes, I was just using 'gamey' as an example. A popular and strealined toolset can lead the evolution of an aesthetic in a particular direction. Of course so can a wildly popular piece of work, such as the 3rd and the 7th.

I guess its a trait in many art-forms, but its particularly apparent in 3d rendering.

[komyali]

http://www.extremetech.com/extreme/175727-ibm-builds-graphene-chip-thats-10000-times-faster-using-standard-cmos-processes

[Jon Berntsen]

In direct response to the thread title, I would like to point out the fact that most of us rely on plugins that is not supported by GPU rendering. When I render a 10000x10000++ pixel rendering with proxies and stuff that in CPU mode (vray adv) might need me to break the render in different parts as well, in addition to the already mentioned plugin issue... Well, I just cannot see how GPU power (which is not as easily accessed/aquired as just buying or setting up another node IMO) can fulfill my needs with such a render in another 5 years.

 

If it wasn't for the need of big scale prints, things would be sligthly different. But these things will not change, which leads me to the next reason. The use of our works in huge scale print formats, combined with very many polygons, proxies, plugins and need for a lot of power. Advertising in public rooms, stands etc will not play a less role in the future. This will regulate itself, because if people stops using these sites for advertising, other people will see opportunities in less actors on that market, hence better response and increased value on those ads.

 

 

To the job thing, I want you to take the following into consideration:

1) We are fluent in advanced architecture techniques with the latest functions in photoshop (or nuke or aftereffects or another package) That would require most of architects to educate themselves in those technologies and their plugins.

2) Most of us spend a lot of time in photoshop adjusting curves, highlights/shadows, color balances on every single element of an illustration.

3) Architects have already very little time to do their todays architect tasks, so how in the world would they be able to add in time for hi-end visualization?

 

If we just keep our time spent below the time an architect would use, we will still win because our expertise. My guess is that we will see more inhouse illustrators again, because of the advantages it has. We see things differently than the architects, so we provide valuable questions to the architects even before it goes out of the house. I don't think it would become mainstream for architects, whose positions are anticipated to be threatened by computer replacement as well, to do another field's work with a much higher hour rate than ours. Now, I know the discussion is the sweet one-click magic render solution, but even if there is such a button, stuff needs to get prepared for print, people needs to be added for realism, light effects ++. This is a whole different arena than what an architect is supposed to have knowledge about or do.

 

That our clients needs are changing over time is just natural, and perhaps, as some others here say, animations will already be done mainstream in lumion or similar in a matter of few years, there are other aspects about it all telling me that we are the ones who will get a better day at work rather than work being taken from us. If the customer says he is satisfied with the first delivered image, if that becomes a norm, well then just make sure you are inhouse, then your task would be to make that specific image rather than the architect, who even works at a higher hourly rate.

Edited July 25, 2014 by chroma

[Michael J. Brown]

If we just keep our time spent below the time an architect would use, we will still win because our expertise... Now, I know the discussion is the sweet one-click magic render solution, but even if there is such a button, stuff needs to get prepared for print, people needs to be added for realism, light effects ++. This is a whole different arena than what an architect is supposed to have knowledge about or do...If the customer says he is satisfied with the first delivered image, if that becomes a norm, well then just make sure you are inhouse, then your task would be to make that specific image rather than the architect, who even works at a higher hourly rate.

 

I think the only solution to this dilemma (because there always have been and will continue to be a percentage of clients who are more than satisfied with the EASY-BUTTON solution) is to create a press release about the value of quality and artistry over quantity when it comes to architectural visualization. To have a write-up done in an even bigger, more mainstream industry publication than the one that story was run in would do a lot to educate clients.

Edited July 28, 2014 by renderhaus