Practical or CGI Green Lantern Suit?

I disagree 100%. You can definitely get a connected performance from a CG character. Blade Runner 2099, War for the Planet of the Apes, Avengers Endgame; all of these are great films with genuine emotional performances from CG characters. 2/3 have been nominated for an Oscar for VFX. All of these performances utilized CG characters with motion capture from humans on set. But it’s not like you can just take the motion capture data and place it onto a 3D model. It takes layers of muscles, bones, fur, etc; plus tweaks to the motion.

I also disagree that Thanos was dissociated from the scenes. Characters like him and Rocket interacted with others, eye lines were maintained, and the fact that Thanos has become one of the most well known villains of cinema history proves the performance worked. If it didn’t, we’d be making fun of it like the Scorpion King. I highly encourage you to watch this video (I linked it earlier in the thread), it’ll give you a lot more appreciation for the work and incredible performances that come out of CG and VFX:


We can agree to disagree. I didn’t think Thanos was very compelling at all.

As for Oscars, when the Academy gave the best picture award to Chariots of Fire over Kramer vs Kramer and All That Jazz, they lost all credibility as an objective measure of film. So f-the academy.

Actually if you pay very close attention, site lines are not always perfect including in the MCU and including with Thanos. Are they very difficult to detect, yes. But, they weren’t perfection.

I understand CGI, from its very early days. I helped with the render farms for Toy Story. There are a lot of basic rendering flaws in that film. Most people totally miss them. I saw them at the screening the day before the premiere.

I am not saying that vfx is bad. The people who do that work are true artists and craftspeople. I think film has gone to far on relying on VFX to makeup for often average storytelling, average writing, average directing, and average acting. There is this delusion that things are “supposed to be real”. The entire concept of stylistic approaches to any of those crafts has been mostly left behind.

Audiences by and large don’t seem to want (or are incapable, perhaps) of actually using their own imagination.

It’s not to say that Brolin’s performance as Thanos was bad, it certainly wasn’t. But, it left me whelmed at best. And I’ll be the first to admit, I’m a lousy audience, because when I watch live action film or theater all I look for are the flaws.

There have certainly been some vfx that I am highly impressed with. The thinning down of pre-soldier serum Steve Rodgers was a masterful use of vfx.

As your video pointed out using Jurassic Park it has its vfx flaws, but it advanced the story. It also didn’t rely on subtle, micro gestures for its cgi performances either.

And that is perhaps where vfx and their use often fall down. They are built on data and it is only as good as the data. You can raise the same eyebrow the same amount and yet it can have a multiplicity of expressions. Based on traits that are basically rooted in the subconscious and conscious of the actor. That immediate energy based on the characters intention. Sorry but that isn’t going to always render. An artist can take a decent guess at it, and they might guess right, but they as often, as not, guess wrong. That’s not the “fault” of the vfx artist, it’s not the fault of the actor. If anything, that falls to the director and/or the writer for not making the character’s intention stronger.

Perhaps the film industry needs to acknowledge the pitfalls, but to do that, they would have to be honest with themselves and their audience. That is something I don’t expect to see anytime soon. VFX is like music, someone can hear a song and it speaks to them, others can hear it and think this is really garbage. Both opinions are valid, and it doesn’t matter if more people think one way then another. Some folks think Thanos and Rocket were great characters, perhaps even a majority of people. They left me whelmed at best.

Who’s right and who’s wrong? That is a matter we disagree on.

Personally, I genuinely don’t care if they use practical or CGI for the suit, as long as it looks good and is believable.

1 Like

@OmniLad Don’t forget to add Gollum from The Lord of The Rings Trilogy to your group of Academy Award winning CGI characters. :wink:

Right, and we’ve come a LONG way since then. We can bring people back to life flawlessly (not going to dive into the ethics of that), and no one will notice. And I’m not talking about Star Wars.

But we should dive into the ethics of that. It is highly unethical in my book. It’s one of the few things I find actually offensive. If and only if, the actor gave consent in their will should it be allowed.

Did it ever occur to who’s doing that the actor, if alive, might have turned down the role.

It is like digging up their grave. A disgusting practice and something that without the actor’s consent, should be illegal.

1 Like

I’m also okay with it if the family gives the okay, too.

Deep Fakes are a whole new can of worms for sure.

1 Like

I agree, but it was off topic. I was just trying to point out how impressive and amazing VFX have come.

1 Like

What Marvel did with Captain Marvel would work I think.

1 Like

I want practical & most importantly, I want the white gloves! Those babies will pop!

1 Like

But they aren’t. Recreating Peter Cushing digitally. That was noticeably not a quality Peter Cushing performance. Same with others. You can tell they are vfx and the performances are flat.

What do you mean? Blade Runner 2099 and Logan had impeccable recreations of actors that most audiences didn’t realize. The only way you would know is if you knew of an actor’s passing or knew about the behind the scenes. Which most people didn’t.

It’s not off topic. There are ethics issues involved. (Not unlike human cloning) Just because we have a technology to do something, doesn’t mean we should use it in a specific way. It’s a violation of that person. If you want to use existing film stock, like they did Brando in Superman Returns. OK. I can see that, you are at least get the person’s performance. We bring back James Dean and Marilyn Monroe and give the starring roles in their own movies again, right? Have Monroe doing an NC-17 full frontal nudity seen. That’s fine, right? VFX is an art, but it is also a technology. As such, it has ethical ramifications. Ramifications that the industry is ignoring.

VFX haven’t gotten better, they merely gotten more “advanced”. Those advancements come with a cost. So please, don’t tell me that because something is “photorealistic” it is “better”. It’s flatter, more devoid of performance & in some cases dehumanizing.

Backgrounds are one thing, but, VFX in an attempt to recreate an actual performance sucks.

I think a little bit of both would do!

1 Like

I agree with you 100% but I was specifically talking about the advancements in technology. This thread wasn’t about the ethics so I didn’t want to muddle it down and mix it up. If this was talking about cg being used to bring people back to life, then we would discuss it. But it’s not, it’s about whether or not a cg suit is viable. I referenced the fact that we can bring people back from the dead as an example of how far CG has come from 2011 and that a CG suit is viable. If you would like to discuss the ethics, dm me so we don’t get in trouble for messing up the thread. Just know that I agree with you, so it wouldn’t really go anywhere.


But it hasn’t advanced. It’s become faster and cheaper. All the technologies used today existed 20+ years ago. Motion capture, texture, light source modeling, etc.

The only difference is that today we have higher speed computing and larger memory spaces. So thing are faster and therefore easier and more cost effective to do. (One could make the argument that creating rendering software that is more maspar (massively parallel) is more recent, but even that can be traced back to the mid ‘00.)

The reality is that you are simply seeing exponentially more polygons today. Rendering millions of polygons per frame vs thousands. You could do the same thing 20 years ago, but it would take 5+ years to render a film you can do a final render in a few days. It’s like saying we have big advances in AI today. We don’t. We merely have bigger, faster expert systems today because processing, memory & storage is exponentially faster, larger & cheaper. Expert Systems and actual artificial intelligence are not the same. It is mimicry that many people can’t tell the difference of, but for those that have been involved in or are truly interested in studying can see that the same flaws exist. Be that a suit, a person, a VFX whatever.


That’s just wrong. It wasn’t about rendering. It was just not possible to recreate the pores of human skin, or the way light reflects of an eyeball. It wasn’t possible, not because we couldn’t render it, but because the experimentation and research into how the human body interacts with the physical world hadn’t been done for the field of VFX. That’s why Dragonheart looks bad compared to the dragons in Game of Thrones. The knowledge of how light reflects on scales and various other things weren’t known. This is the last I’ll be responding to this topic in this thread. DM if you would like to continue the conversation, but it’s not fair to everyone else to keep spamming about something that has strayed too far from the original post.


I disagree this isn’t a spam of a thread. They often take twists and turns.

But since you are talking about eyeballs and pores. Yes, the dataset hadn’t been collected. But, that is all it is, a data set. It still has all the same issues (or fewer) than the Utah Teapot, which dates back to 1974. In 78, there wasn’t a dataset for pores because nobody had need of one. But they did creat The Mandrill in 1977 because it was a better way to test texture mapping. So all the fundamental building blocks have been around for 30-40 years. So the only thing that is new is new data sets.

Eyeballs have a great similarity to a lit lantern logo on a CGI suit. They both have texture and reflective and refractive properties and as such light must be calculated in terms of how light reflects and refracts off of not just the surface and the surface texture but the how reflective light off the cornea comes back through that textured surface and how the out going light through that “lens” effects the incoming light on the lens and which frequencies (and colors are boosted and which are cancelled.

This is the same issue that will effect a “lit” lantern on the suit because the suit at some layer has a reflective and refractive “lens” just like the eyeball. This issue affects any light that goes through a transparent (to some degree) texture and reflects (and refracts) when it hits the reflective surface and goes through the texture from the other direction, this “outgoing” texture is not necessarily the same as the “incoming” texture.

So a CGI suit will come down to the detail levels and properties the suit has. The outstanding question for that is…how much time & effort (and therefore cost) needs to be spent to do the CGI suit correctly?

Like anything, dating back to the “Utah Teapot”, the issue is creating the proper data set and is the cost of doing that the right way is a good business decision.

Certainly a CGI suit will be more expensive and time consuming than a practical suit. So the question here is not just acCGI or practical suit, but what is the value proposition in doing a CGI suit. Would folks who want a CGI suit be willing to forgo certain details such as how often GL is in the suit or the intensity/detail of the CGI constructs. It’s going to be a trade off in dollars somewhere in the process.

So which of those is more important?

Feel free to DM, I don’t want to deter people from joining the thread and giving their opinion on a practical vs CG suit because we’ve gone off on a tangent about the way CG has evolved and that’s what the thread has become. You’re typically pretty reasonable, Desade. I’m happy to explain my points in there.