Valve hands over its own movie-making tools to gamers | Hacker News Reader

Valve hands over its own movie-making tools to gamers

gizmag.com

96 pointsyitchelle28 comments

This is the sort of thing I was thinking of when reading the Kill Hollywood RFS (http://ycombinator.com/rfs9.html).

Who says you need a huge budget to bring a great science fiction story to life? Who says you have to dumb down the story to recoup that budget through a wider audience?

Once you can make movies, complete with special effects, on a home computer with a small team and a small budget, the variety and depth of content is bound to explode. I can't wait to watch all the interesting new stories that emerge from this type of technology.

Shameless plug, not only I perfectly agree. But I've been working on exactly that in the last few months :)

I can't think of a better "take the moral highroad" solution to the problems created by the MAFIAA and IP in general than to simply create our own content, independent of their little bubble of crap. And the awesome news is that there's a lot of room for improvement in current content creation tools. There's just so much we can do to reduce cost and time to produce multimedia content. We have so many technology that are barely being touched by current tools. Because they're stuck in their incumbent sluggish giant position.

And that's where startups come into play! We have no ties to old incumbent technologies. We're free to innovate and completely re-imagine how content is created. Look at what wordpress made for publishing websites. Today any random non tech person can say "hey, I just had an idea, let me make a website about it". Imagine if you could say the same for movies "hey, I just had an idea, let me just make a full AAA movie about it really quick". The Valve filmmaker is incredibly in the right direction. But it's not even at 1% of where we can go.

Do you want to kill Hollywood? Then build awesome tools that allows anyone to build their own content!

The one tool that I've been looking for the last few years has been a decent text-to-voice tool. I did do one series using some of the tools out there as of a couple years ago (http://www.youtube.com/watch?v=1oZ6TxrkrqE as one example), but the results were less than wonderful.

Use actors. It's cheaper and simpler to record the dialog first and then animate it. You don't need famous actors, there are lots of good actors who'll work for cheap.

A few problems there. First of all, cheap for a single actor may be fine. Once you actually have four or five characters, even cheap starts costing. Then you have the issue with cost of re-recording when you find that a scene that needs fixing either for a script change, or when you go to animate and find there was background noise in a take or a line was flubbed. Even when the acting is free (one of my first games which I gave away), the "cost" in time and effort compared to what a program could provide is still expensive.

Agree. That's a major issue to making movies on your own. It would be so much better to have efficient text-to-voice tools.

Plus, getting a "free" actor often means it's less than ideal, too. Being able to control the voice the way you want is clearly where we need to go.

when you go to animate and find there was background noise in a take or a line was flubbed

Speaking as a sound guy with 10 years experience, this doesn't happen when you treat me like an equal member of the team instead of trying to do it on the cheap.

the "cost" in time and effort compared to what a program could provide is still expensive.

Compared to what a technology-that-doesn't-exist could provide, everything is expensive.

I have always hoped that marketplaces such as iTunes, Google Play, Steam etc. are something that will only be necessary until consumers get used to buying stuff online.

A better future is one where each content producer would just publish meta data in an open format that others could redistribute. In the meta data there would be a link to a web service using a standard format for buying/downloading the content.

Publishers could provide affiliate programs as an incentive for creating innovative ways of distributing the content. There would still be room for iTunes as an aggregator or a service for payment/downloading, but we would not rely on a few gatekeepers for our content. By providing the meta data in an open format it will enable anyone to distribute content. We could have a "Hacker Movies" where people submit movies and vote them up or other possibilities that would be up to others to imagine.

It's worth noting that currently, the technology doesn't even exist to say "hey, I just had an idea, let me just write a full novel about it really quick". And making a feature film is orders of magnitude harder, and encompasses many, many more tasks, than writing a book. (I've done both.)

Having been working in this space for 15 years now, it's definitely a hard problem, and one encompassing multiple other hard problems within it. Even designing a tool allowing a person without experience to come up with camera angles and sequences - just cameras, nothing else - in a manner comparable to a Hollywood feature is very, very hard - I've seen multiple research projects fail at it.

None of which, incidentally, is intended to say anyone shouldn't work at it! If we can create a tool that gets anywhere near to that ideal, it'll be a fantastic advance for humanity as a whole.

> It's worth noting that currently, the technology doesn't even exist to say "hey, I just had an idea, let me just write a full novel about it really quick".

That's one of the first problems I've ever tried to solve, back when I was writing chat bots. Fortunately, building a robot that can write human readable text from structured data, is many order of magnitudes easier than building robots that does the opposite. Pattern recognition is what's keeping current chat bots so dumb. But we actually already have some really good bots that write text reports. You probably already read news/sports reports written by bots and never noticed.

I believe that's the way for future tools. Users give very basic structured data (who's the protagonist, what does he do, did he die at the end?). And robots transform that data into text, movie scripts or whatever.

> Even designing a tool allowing a person without experience to come up with camera angles and sequences

You don't need to do that. Bloggers don't need to understand HTML nor CSS. A good tool would give higher layers of abstractions to choose, which were designed internally by professionals. And the user shouldn't even need to know what good angles or lighting is.

I've heard suggestions like this before, but that's not to say that you won't succeed where others didn't! Looking forward to seeing what you come up with - if you fancy doing so, please do let me know! (info at strangecompany d o -t org)

The Blogger anology is flawed. Bloggers don't need to know HTML and CSS because they are judged by the quality of their medium, which is text. Perspective is essential for a visual medium, you can't abstract it out, you can only help and support.

It's going to be a bright future for creators and audiences when the physical and political barriers to entry come down.

I’m imagining a digital version of the entire Hollywood/ gaming worlds: online planning and development, digital movie/ game locations, acted in, tested, filmed all at the same time. Then films/ games can be distributed, and artists, developers, screenwriters, etc. can receive a passive income when a product containing their work is sold.

In the beginning we're going to see some amateur work, like the beginnings of the physical film and game industries. But as it starts to mature we'll have new specialists in direction, acting, writing and world design.

Think about what can do with elements of massively multiplayer online collaboration, LEAP, home motion/ model capture (Kinect 2.0).

The site:

http://sourcefilmmaker.com/

Wiki discussing the how tos:

http://sourcefilmmaker.wikia.com/wiki/Source_Filmmaker_Wiki

If I were Valve I'd distribute the films on Steam and take a small, transparent cut. I vaguely recall some movie being delivered via Steam already. It makes a lot of sense and I can't wait to see what comes out of it.

That would be Indie Game: The Movie http://buy.indiegamethemovie.com/

See also http://news.ycombinator.com/item?id=4168283 , where Source Filmmaker was also discussed.

Does it work? I clicked the link, filled in the survey, and don't seem to have an invite.

You have to first be accepted into their beta.

Next time you open Steam once they accept you, you'll see a message in the deals popup that comes up from time to time.

Sign puts you on a list to get an invite eventually.

anyone else thinks they'll be building a market around this too?

buying/selling models, scenarios, props... seems like another jackpot for Valve

That's definitely where I see this going. A market like that can make Valve a ton of money, but also help support a new generation of indie film makers.

Yeah, I expect so, but I think the real payoff for Valve will be facilitating monetization of the finished films.

Yep, they're a mass-market company, selling millions of units at sub-$50 prices. The bestselling resources will probably sell a few thousand copies, and most will probably only sell hundreds. It may be more cost-effective for them to focus entirely on the mass-market side, where they've proven they're extremely competent (among the best in their industry), and let others figure out the smaller volume stuff. But, maybe a marketplace would be cool/profitable. They have begun to move in that direction for third party game add-ons, though I think they're all free, so far, there's nothing stopping them from adding them to the store.

They already have a lot of experience as a distribution platform. I doubt the majority of Steam games even come close to a million units.

While it's not clear that there's enough market to extend that platform to commercialize their movie maker, it's not that far of a stretch from what they're already doing.

They already do. Tradable clothes, hats, guns are already part of the TF2 economy. You can bet that this has just inflated the prices of items that'll look good in one of these movies.

I'm pretty sure you can use the tradeable assets in your films without owning them. You only need to have them in your inventory if you want to use them in the game proper.

meet THE pyro, not "a pyro".

Reply on news.ycombinator.com