How to get better at creating complex effects for shots?

[magneto]

Hi,

Personally I find there is a huge discrepancy between solving problems and creating complex effects for shots. What I mean is when I am asked to achieve a certain effect for a shot, I find myself lost, like I don't know what to do to best achieve the result. Even with reference, the constantly changing requirements from leads and supervisors and the inconsistencies between them doesn't help either.

If you ask me how to solve a particular problem, whether it's technical, mathematical, houdini specific, 3d specific, etc, I can show you 10 ways with different efficiency, speed, complexity, for example. So when I am asked to help with these kinds of problems, it's second nature to me.

But the larger nature of a shot effect often eludes me, leaving me scrambling for ideas. Because I don't have decades of experience like others, I don't know what would hold well for a particular effect for example. I have all these tools in Houdini but I don't really know what to use.

How can I get better at this? Are there any strategies, practices I can implement to help myself figure out solutions or is it "you either have it or you don't" kind of thing? Any help would be appreciated.

 

Thanks

[Guest tar]

25 minutes ago, magneto said:

How can I get better at this? Are there any strategies, practices I can implement to help myself figure out solutions or is it "you either have it or you don't" kind of thing? Any help would be appreciated.

The design part is more about story-telling than engineering.  Can you tell stories?

[ikoon]

Hello Ryan,
(disclaimer: I am just freelance motion designer, not from the vfx business)

My fiction is, that those shots, that look like heavy and complex simulation .... that those shots, for example smoke from car-braking-drifting going through propeller are made 99% of "fake volumetric velocities" generated by "smart curves" generated by "smart solvers" controlled by "smart rigs" with "smart fallofs".

So I think, that your potential is in creating those "smart intellectual and artdirectable fakes" and then using physical VFX solvers just as a "cherry on the cake" to make it look physical and generate those heavy data.

But it is just an idea for discussion, I may be totally wrong, unfortunately. Or I am just a naive fool stating the obvious, so sorry for wasting your time.

 

EDIT: As Marty said, I imagine, that if you have all those "smart fake handles" I was writing about ... you can then artdirect them into music, into camera view, into the right second (just before the door opens and the hero steps on the ground). Sorry for rambling.

Edited December 15, 2016 by ikoon

[magneto]

Thanks for replying but it doesn't help a lot

Also when I said complex fx, I mean even things like character getting ripped apart with the clothes, etc.

[sekow]

break it down. like every problem is better to disect it into smaller problems. start with to one you most comfy with. build up from there. 

identify whats important and whats not.
what comes first? (thats maybe the hardest part) most times the next step is pretty obvious, and I bet you will have a good picture of progression pretty soon sketch out in your mind.

thats why houdini is great at that kind of work. your effect can contain on bigass setup or, which is way better, is coming to live thru individiual blocks/modules.

 

Edited December 16, 2016 by sekow

[ikoon]

Ryan, ask more questions. Sometimes questions are the answers. Also I usually tend to find a parallel and solve the parallel it is not ideal, but it helps. So ... more questions and parallels and analogies may help you?

I mean questions and parallels in your attitude, but even in that character getting ripped ... it can have parallels too:

- more like a perforated plastic or paper (emphasise moments of tension and moments of disjoint)
- more like a filled plastic bag filled with bowels

Also parallels and emotions

- is it heroes body ripping? such a "parallel story" may be ... small dam on a hill above village, people trying to fix local cracks, being more and more tired of carrying those patches ... and those patches being flushed away
- does the viewer want the body to be ripped and enjoy that bastard suffering? make the ripping like erotic foreplay, something small here, something bigger there, look at the face (oh, parallels) kiss on face ( equals rip on face)

EDIT: Also articles like these may help you. I did not read this one, I just googled my usual problem, the "blue sky paralysis": http://99u.com/articles/6585/10-laws-of-productivity

Edited December 16, 2016 by ikoon

[Skybar]

Understand that almost every effect you see is something someone spent a lot of time on. And it's probably not the first iteration or workflow that you see. He started at one point, and it evolved over time to what you see on screen through just doing work. You can't expect to know from the start to a 100% how to do something - if that's your goal you'll go crazy.

When at first you get a task, you usually come up with an idea how to approach it right there. Start with that. What works, what doesnt? What do I need to change or add to bring me closer to the goal?

[Atom]

the constantly changing requirements from leads and supervisors...

Shots shouldn't have constantly changing requirements. Challenge your supervisors to be more clear and "lock down" what you are building. Let them know that vapid willy-nilly direction just cost more money. Cite the fall of major studios, like the one who made the life of pi, who based their business model on what was called the "open shot". The concept of the open shot is ridiculous and we should not repeat the mistakes of the past. With the open shot, directors pay one fee and then change their minds as often as the wind blows. In one of those shots the FX team struggled with adding in rain and complex fluid simulations only to have the director come in two weeks later and say "why the hell is there rain in that shot?". We know that is money down the drain yet the studio ate that cost, not the production. Try to avoid scope creep and let everyone on the team know that if it creeps too far that it becomes a new shot and requires a new budget.

Cost is your best defense against unreasonable demands. Always say yes to your client or supervisor yet present them with realistic cost and hours. Nothing takes 5 minutes. Print that out and post on the wall. If you lose a bid to a smaller shop who wants to under sell you then move on. I have created 1st drafts that were better than final deliverables provided to me from external vendors that the team wanted go with. When the results come back and it sucks no one says a word because they gambled and lost and just have to live with it or pay for it to be done right again.

[haggi]

As already mentioned, it is extremely useful to separate a complex shot into simpler elements. My preferred workflow is to keep everything as simple as possible and do an iterative approach and add detail only where it is really necessary.

• Find references and try to fix the look to avoid too much trial and error
• So try to seperate the complex shot into simpler elements
• Keep it as simple as possible, don't use a complex geometry to try out something
• Combine the elements to see if they fit together
• Keep in mind that a lot can be done (and has to be done) in compositing, most shots look like crap if you see the raw passes

If everything is in place, you can start to add detail. And if you can fake it instead of simulating something - fake it.

[rbowden]

3 hours ago, Atom said:

 

 

Shots shouldn't have constantly changing requirements. Challenge your supervisors to be more clear and "lock down" what you are building. Let them know that vapid willy-nilly direction just cost more money. Cite the fall of major studios, like the one who made the life of pi, who based their business model on what was called the "open shot". The concept of the open shot is ridiculous and we should not repeat the mistakes of the past. With the open shot, directors pay one fee and then change their minds as often as the wind blows. In one of those shots the FX team struggled with adding in rain and complex fluid simulations only to have the director come in two weeks later and say "why the hell is there rain in that shot?". We know that is money down the drain yet the studio ate that cost, not the production. Try to avoid scope creep and let everyone on the team know that if it creeps too far that it becomes a new shot and requires a new budget.

Cost is your best defense against unreasonable demands. Always say yes to your client or supervisor yet present them with realistic cost and hours. Nothing takes 5 minutes. Print that out and post on the wall. If you lose a bid to a smaller shop who wants to under sell you then move on. I have created 1st drafts that were better than final deliverables provided to me from external vendors that the team wanted go with. When the results come back and it sucks no one says a word because they gambled and lost and just have to live with it or pay for it to be done right again.

Shots shouldn't have constantly changing requirements but, unfortunately in the real world, they do. That is just the nature of the business unfortunately at a big studio. Sure, you can challenge your direct supervisor (which is more likely be the FX supe) but, they still answer to the VFX supe and then the VFX supe answers to the client based VFX supe sometimes, and then that VFX supe answers to the director. Everything goes through a chain of command and sometimes you get a person in that chain that has a different opinion of how the shot should go. Sometimes that direction is completely different from what the director wants and unfortunately the artists pay for that because most of the time, we have to make the changes last minute. Believe me, they know it costs more money and it really seems like it doesn't matter most of the time. 

Sure the "open shot" concept is bad but, unless you are making your own content, nothing is going to change. Most of the big places are still vendor studios and make alot of profit from it. It is easy to say move on if you lose a bid to a smaller shop but, imagine if that is happening constantly and you are the person that is trying to find work to keep all your artist employed...It is easy to turn into a yes person for the sake of keeping the company going.

As for the original post, I agree with everything Haggi said. Whenever I look at a shot, I break it down to the simplest element and build on top from there. It definitely can get overwhelming at times but, you just have to take a deep breath and go the route you know you can iterate pretty quickly.Your lighter (if you aren't lighting your own shots) and compositor are your best friends. I have seen compers turned what I thought was an awful looking pass into something amazing. Most people are visual so even if you get a flipbook out demonstrating your effect on a basic level, it can at least start a conversation. Starting a conversation and explaining what you have in mind is what you want, it helps paint a bigger picture for people most of the time.  

[stickman]

If you can get information that helps you understand what the artistic intention of the shot is, then you'll be in a much improved position when anticipating where the feedback will go, and hence which of the many solutions might be best to follow. For example, if a body is being torn apart - what is the target audience? (Perhaps it can't get to gory) Is this the end-all money shot that will give the show's climax sequence it's 'punch-in-the-gut'? These are somewhat general questions but hopefully point to the direction I mean to convey. If possible (as mentioned above, the chain of command can be long) I try to design my briefing, asking for adverbs/adjectives to describe the intended 'feeling', impact moments etc.

[Atom]

Quote

asking for adverbs/adjectives to describe the intended 'feeling', impact moments etc. 

This is a great habit to get into. I always take notes when getting direction and keywords for mood are often what I fall back on when I am stuck. "Oh yeah, it has to go from sad to happy".

[Marc]

I agree with a lot of what has been said here and I think the most important of them is to bear in mind the artistic intention of the shot and try and figure out what story the director is trying to tell with the effect. Effects are very rarely what you would see happen in the real world. They're something based in reality, but transformed through the lens of the camera (and the story the director is telling) to do something that should look real, but not behave 100% realistically. 

If you can figure out what the director is trying to do you will go a lot further than if you present something that is 100% correct from a technical point of view. That might have something to do with the changing notes/directions too. It's typically very difficult for a director to explain what he wants to see (the snow kicks should be happy!) and that's where the artistic interpretation comes in. 

[ikoon]

Marc, thank you very much. I remembered ... I have read this book .. Kees Dorst, Understanding Design. It contains small, but deep essays. I highly recommend it, it helped me very much.

[ikoon]

Here I found some PDF. For me it is soothing like heavenly music I dont remember which essays are the best, there are also essays about relations with investors, stake holders etc. Just give it a try:

https://docs.google.com/file/d/0Bzlese-wZI7WS2k2cTNfLVZwQ00/edit

EDIT: The linked PDF is just a small part of that book.

Edited December 20, 2016 by ikoon