Unreal Engine. How good for heavy pipeline? Pros cons

[Jon Berntsen]

Hi

 

I bet someone here is using Unreal Engine for creating competitive imagery. What would you say are the pros and cons? What about essential camera functions like tiltshift? Render elements? There is so much buzz about the render speed that I think people are forgetting what it really takes in a stable workflow to actually deliver on date, on budget, and with last time changes. What about xreffed files or light build times? Excessive? What about an entire team working on the same platform, is it practical? How about wage for new employees, will unreal people demand higher wage? Then we have max/vray which is also improving on render speed. Safe go-to, also for animations? What do you think?

[Arnold Sher]

I would love to know as we've been debating the same subject in our studio. Lots of our projects are really large and we heavily rely on ref's and proxies. What we've seen so far looks great but it is all for relatively small stuff and that is just not our reality. Most of our stuff is way above 10000 squares meters. So i will follow this discussion with great interest.

Hi

 

I bet someone here is using Unreal Engine for creating competitive imagery. What would you say are the pros and cons? What about essential camera functions like tiltshift? Render elements? There is so much buzz about the render speed that I think people are forgetting what it really takes in a stable workflow to actually deliver on date, on budget, and with last time changes. What about xreffed files or light build times? Excessive? What about an entire team working on the same platform, is it practical? How about wage for new employees, will unreal people demand higher wage? Then we have max/vray which is also improving on render speed. Safe go-to, also for animations? What do you think?

[Scott Schroeder]

If you are rendering in Unreal, there should be zero need for render elements and Photoshop. Either you get it all done in Unreal, or use Photoshop as an emergency use only. Unreal is powerful enough to do all of your color adjustments in the engine.

 

The biggest hurdle for us in Unreal is dealing with changes. The Max + Datasmith workflow is great, but it does get complicated with too much back and forth transfer from Max to Unreal as you make continual changes. The other big issue is how complicated Datasmith makes your materials in Unreal. A simple glossy plastic shader is way too complicated when Datasmith coverts it from Vray. Unreal's materials are pretty powerful on their own, but I think the need for Datasmith to convert Vray materials 1:1 makes anything you get almost too complicated to adjust in Unreal.

 

I was able in testing to import a 5,000 seat arena into Unreal and get all 5,000 rail clone seats and 5,000 forest pro 3d people proxies to come into Unreal with little issue. When using large numbers of Rail Clone or Forest Pro, the instantiate tools for the iToo tools are a God send. Importing to Unreal is smart enough to keep instances together, which helps at run time in Unreal.

 

Not too sure about teams working in Unreal, but obviously it is done with video games. I'm not sure if they use a version control software or not, but it can be done.

 

Light build times can be excessive if you watch a lot of the YouTube mumbo jumbo to get your precise noise free light maps. Taking an average of those settings helps. I can't speak for light build times too much as we use Unreal Swarm across 20 machines to make our maps build in an average of 30 minutes at full quality.

 

Unreal people shouldn't demand a higher wage because with the Datasmith workflow, 75% or more of your work is still done in Max and Vray. You are just essentially using Unreal to render.

 

We haven't used Unreal in production just yet. All of the stuff you see online, the person usually states they did in in 3 months. That's great. Try getting that done in 3 days.

[Eliot Blenkarne]

For us, being able to get render elements out of UE4 would be a godsend, and one thing holding back the wider adoption of it. Like has been mentioned, Datasmith-converted materials become too complex and cumbersome to be able to edit in real time, making them semi-redundant, so being able to quickly edit things in PS would be nice. Broader corrections are easy enough using the post-process volume. There is a plugin for camera correction, though I've not tried it.

 

I don't necessarily agree regarding wages, - what UE4 can do far outstrips a traditional viz workflow, and Datasmith represents a tiny amount of it. If someone is skilled in UE4, they're opening your company up to a pretty interesting variety of work, and if that's something valuable to your company, they should be paid accordingly. If you want to produce real-time VR with interactivity and slick menus, or a car configuration mechanism, or some form of HTML game - or all of these, whilst remaining linked to the viz world, it's hard to look past UE4. Again, it depends on what is valued in your firm, IMO highly-skilled specialists tend to get paid more than generalists. This applies to more than just UE4 though.

 

For mine, UE4's biggest weakness is in-fact what is perceived as its strength - it can do a lot of different things, but not necessarily that well, as nothing is as simple or well-developed as the traditional archviz workflow (like you hint at being your concern). Even lighting is unpredictable - you need only check out the Epic forums to see people battling to get consistency. Sometimes, things just break randomly and your lighting just stops working. Everything seems to be a case of roll the dice, try things out and see what happens, then make incremental changes. It remains quite a big jump - coming from someone using it full-time for nearly 3 years, I still struggle with it sometimes. For mine, those jobs that need quite a bit of post-pro work or camera matching, UE4 is still difficult to justify implementation. Powerful, but flawed!

[Jon Berntsen]

Thanks for chiming in your thoughts. So what I can read out of this, is less flexibility, and longer turnaround times, plus an extra export/import chain. The only good thing is the render times. Hmm.

 

In our studio, we are discussing testing Unreal for rendering for three reasons:

1) Rendering animations is an expensive hassle if just one small problem arises.

2) To get ready for the day when we can use expensive blueprint functionality in packages for WebVR and hit all the devices out there.

3) We want to be using the smartest tech to earn more money now and in the future

 

So is the answer perhaps having the entire team learning vray better, to render smart and with better render times? Should we wait for Chaos to launch even better render solutions for max? I've got a feeling that Unreal seems like an easy outcome for many people. Like, they don't got it first with vray the first time, so therefor they look at Unreal as a simple solution, because YouTube ... And you know, deciding workflows is really deeper than that, it's what I wrote in the initial post, delivering quality within time, and as well rigged for re-use, and network paths, without having to install a new engine version every month. What I say, is that I'm afraid that the time we allegedly save on rendering will be eaten up by bad resource, scene and asset management.

 

A few more questions:

Is the GUI feeling more and more intuitive along with experience? I've been blueprinting a lot in Unreal with the purpose of VR functionality lately, but it's almost that the interface is scaring me. Working with materials, assets, really not suited for network assets, right? Experienced too long path as well, for projects that we try to package from the network path. So the really big concern is the assets being saved in the file.

 

What I really see as the biggest pro, is not happening right now, but hopefully in the future, when WebVR gets so robust that you can package realtime stuff with great functionality for devices, and also do the same package with a few alterations to package for a full immersive experience with HMD's like Vive Pro 5 or 6 or something.

[dmitriikarpovich]

For me and my stuff, Datasmith doesn't work at all. I do not understand what happened to big arrays when you export, looks like they are not manageable anymore after the export done. The Generated UVs for simple CAD walls looks ok but for more sophisticated geometry it fails and there is no way to fix it.

Although once the setup is done taking pictures, tweaking colours, camera animation are the breeze.