Episode 165 — Weta’s Compositing Lead Cameron Smith

 

Click here to listen on iTunes!

Get on the VIP insiders list!

Check out www.VFXRates.com.

Upload The Productive Artist e-book.

Allan McKay’s YouTube Channel.

Allan McKay’s Instagram.

 

Episode 165 — Weta’s Compositing Lead Cameron Smith

Hey, everyone!

This is Allan McKay. Welcome to Episode 165! I’m speaking to Cameron Smith, the Lead Compositor who has worked for ILM as well as other studios. I’m really excited about this one! Cameron is an old friend of mine. In this Episode, we backtrack through our careers a bit. Fifteen years is a long and successful career! Cameron shares his insight on how to achieve that. This is a really fun Episode.

Let’s dive in!

 

FIRST THINGS FIRST:

I. [00:40] Have you ever sent in your reel and wondered why you didn’t get the callback or what the reason was you didn’t get the job? Here is the thing: Most of us think that we can put our latest work on our reel, add some music — and get the job. A lot of us aren’t aware that the majority of reels sent to a studio are skipped through and sometimes never even watched in the first place.

Everything we’re taught about being an artist is wrong! Over the past 20 years of working for studios like ILM, Blur Studio, Ubisoft, I’ve built hundreds of teams and hired hundreds of artists — and reviewed thousands of reels! That’s why I decided to write a book from the perspective of someone who actually does the hiring. I want to:

  • Give you the formula to be the obvious candidate for the job;
  • Tell you how to build a reel and put it up on YouTube — that brings studios to you!

You can get this book for free right now! Whether you’re in design, film, tv or games, go to www.allanmckay.com/myreel!

II. [04:34] I’m releasing some new training based on the feature film Venom. I chose Venom based on a survey I ran. Do check it out! It will be available for the next few weeks: www.allanmckay.com/venom/.

If you want to sign up for my VIP Insiders List go to www.allanmckay.com/inside/.

 

WETA’S COMPOSITING LEAD CAMERON SMITH

Cameron Smith is an award-nominated visual effects artist who is also a Lead Compositor at Weta. He first studied computer science at RMIT University in Melbourne, Australia before transitioning into programing and visual effects. He first began doing visual effects for tv commercials and short films.

Over the course of his over a decade-long career, Cameron has worked for studios like Cutting Edge VFX, Fuel VFX, ILM and Weta, serving as a Compositing Lead, Supervisor and a Lighting Technical Director. Some of his projects include The Hobbit, Battleship, Furious 7, Man of Steel, Iron Man, King Kong, Avatar (for which he was nominated for a VES Award) and many more.

In this Episode, Cameron talks about his over a decade-long career as a visual effects artist, the benefits of programing, constantly learning new skills and the latest technology.

 

[05:29] Allan: Cameron, do you mind introducing yourself?

Cameron: Yes, I’m Cameron Smith. I’m a visual effects artist who works at Weta currently. I’m primarily a Lead Compositor, but I’ve done other jobs, like lighting, in the past. That’s who I am!

[05:56] Allan: I want to dive into that too. I didn’t realize you’ve done a lot of lighting. It’s cool that you’re doing both ends of it. Did you always want to be an artist, as a kid? Or did you want to be a ninja?

Cameron: I think everyone starts out wanting to be a ninja. One of the earliest memories I’ve had was rolling around on the floor (when I was 3 or 4, I think) and on tv there was an ad for these giant, four-legged robots in a snowfield. It was for The Empire Strikes Back. And I think probably fell in love with movies at that point in time. I never expected to work on them. I grew up in the middle of suburbs in Melbourne, Australia. All those movies were being made in Hollywood. I loved movies forever, but I went down the path of computer science. Half way through that, I realized computers weren’t really my thing. “I don’t want to do databases and stuff.” But during that time, I quite enjoyed graphics programming, and I fell in with the crowd that was doing short films and I thought that was a bit of fun. So I got a hold of After Effects from someone.

[08:23] Allan: So, where were you studying?

Cameron: I studied at RMIT [University]. I think the point of doing computer science was to learn to program so I could program video games. But the entire course was not at all based on anything that would help you do that. It was more for the military.

[09:05] Allan: so you decided to help out with your friends’ short films. Did you try to figure out what programs to use, so you could help out?

Cameron: Yeah, I guess during high school Babylon 5 came out. It’s a great story, but the CG is pretty primitive by now. That was made with LightWave. “Can I get that?” I got a copy of that and tried to figure it out.

[09:51] Allan: When you were dabbling with that, when did you realize you wanted to make it a career?

Cameron: Yeah, it was… I’m trying to think. It was toward the end of getting the degree. I started to do a bit of web design on the side. And again, from friends from the short film, we started to collaborate on different projects. We thought we’d try and do business with that. We did some late night tv commercials and it was actual paid work. That lead me to put together my reel and send that around Australia, on good old VHS. And I ended up working at Cutting Edge.

[12:02] Allan: I’m curious about that. Correct me if I’m wrong: Because you didn’t have much on your reel and were applying as a compositor, you had DVDs from something like Star Wars and you would add things or replace things. Is that correct?

Cameron: That is correct.

[12:52] Allan: I think that’s brilliant and resourceful that you were able to do that. Especially as a compositor, it’s so easy to make an excuse that you didn’t have anything to composite. You created your own material. 

Cameron: Basically, a friend of mine went to work on Star Wars: Episode 2 in Sydney. At some point during the shoot, some editor in the States released a fake Episode 2 trailer using footage from various films. That created a bit of a buzz and fooled all these people into thinking it was the real deal. I looked at it and realized I knew all those movies. It didn’t really convince me at all. I thought I would do my own fake Episode 2 trailer by grabbing shots from all these various movies, like that other guy had done. I could spruce things up. They were releasing still photos from Episode 2 (in pretty low res) and roto (Obi-Wan) off a plate from Episode 1 and comp him onto the photo from Episode 2. And I tried to be as elaborate and non-specific as I could. The last minute thing I threw in was a background from Episode 2, I think. I knew that Christopher Lee was going to be in Episode 2. So I rotoed him from Sleepy Hollow and I had a picture of him from Episode 2 and he had a beard. So I cut it out of the photo and tracked it onto his face. That was such a throwaway!

[16:50] Allan: I recall your writing some 3D elements into it. I thought it was brilliant. I’m constantly looking at footage from DVDs. I feel like the poor man’s YouTube was pulling stuff off DVDs.

Cameron: I was quite a challenge. I think it was all After Effects and LightWave; some camera tracks. I didn’t have the proper hardware to do it.

[18:17] Allan: Going to Cutting Edge, what was it like for you, going to work on feature films? (I remember it was around the time we got Macs.)

Cameron: It was quite scary. I was from Victoria. I had to move, I didn’t know anyone over there. It was time to do it! It was a nervous but an exciting time. They started me off on a 6-week trial, which I thought was pretty fair for someone with no real experience. I did the night shift.

[20:12] Allan: You came on but were able to move on from a junior artist pretty quickly though!

Cameron: What did bum me out about that shift, it was quite lonely. When I would get there, everyone would be playing Counter-Strike. I would come in and be by myself for the next 6 hours. I can’t remember is Shake was new for that time. Did you just start using Macs?

[21:11] Allan: No, we were using them for quite a while, since 2000 or 2001. Then Apple acquired it. We had licenses enough to cover just the compositors. 

Cameron: I’ve never used it. It was a pretty steep and rapid learning curve. I quite loved it. I still love node based compositing, it’s so much more flexible. Then we just got into it. I can’t remember if George of the Jungle 2 was running at that point or it was commercials that I was working on. I did my first 6 weeks and I transitioned to the day shift after, which made everything easier. And then it was one year of full on work on both George of the Jungle and tv commercial after a tv commercial. And looking back, I think it was the best way that could’ve gone. As brutal as the tv world was, it was such a good way to learn. You have to do it, get it done and move onto the next thing.

[23:20] Allan: I think starting in commercials is the best way to go. They will be rapid fire. You don’t have time for a nice pipeline. You really get the most growth out of that. By the end of the year, you’ve worked on 40-50 commercials.

Cameron: Yeah, I don’t miss commercials, but I did learn quickly at the time. You never got bored of looking at the same shot for 6 months. The turnaround was pretty rough.

[24:26] Allan: I just got a call about a certain Marvel movie that’s coming up. And I thought, “Eh, I don’t know.” The project is due in a week. One week on a Marvel film is going to be pretty brutal. The other thing you brought up is starting out and moving to a new city. A fear sets in for a moment, even though you’ve wanted to get it. It’s the reality and the doubt that kicks in.

Cameron: There is always going to be stuff like that. You just have to do it and see what happens. As long you don’t mess up or burn your bridges on the way — give yourself a bit of buffer on the way. It’s easier to do it when you’re 20, than 40.

[26:25] Allan: In terms of Weta, how many people are there from the Edge these days? It seemed like a lot of people ended up there.

Cameron: I think people came over, on their way to somewhere else. I think it’s 1,500 people here now.

[27:33] Allan: Going from Cutting Edge, you went to Weta next. Is that correct?

Cameron: That is correct. Around the year mark of my time at Cutting Edge, I started thinking what to do. I think Dave Clayton was the first to come over to Weta. I thought I would like that. Richard’s contacts got me a way in and the fact that they needed people.

[28:51] Allan: Going from a smaller studio — workflow wise and everything else — how much of a culture shock was to go into a bigger studio?

Cameron: Big, big changes! I can’t even remember if Cutting Edge had any real pipeline or tracking. At Weta, I went to roto painting and did a bit of comp as well. There would be 20 people just doing roto paint (not just 20 people doing the whole project). And then there would be managers, and VFX Sups and producers. There is an actual production tracking pages and internet pages and cool shots that let you seal the clips, and Shotgun. All these fancy pants reviewing stations! There were props from the movies on the wall. That’s another step up! There was more to learn about. I’ve never really worked on a film stock before. You have to deal with noise and de-noising.

[31:09] Allan: I think the most valuable thing going to a larger studio is getting exposure with amazing talent who have been pioneering some standards. Did you get a lot of growth those first couple of years?

Cameron: Yeah, definitely! There is such a talent pool of brain trusts of artists and amazing technicians. Everyone at Weta is really, really good. And they’re great people to work with! I’m the new guy who’s done George of the Jungle 2; and I’m sitting next to people who’ve done Starship Troopers and Raiders of the Lost Arc. Okay! I’m not really that special from a film point of view. But you can just ask them things. The whole time, everyone has been talented and helpful. If you don’t know something, you just ask. So it was great! It’s still a great learning place.

[32:03] Allan: What was your first project there? Was it Lord of the Rings: The Return of the King?

Cameron: It was. It was the last 3 months of the show, fully racing toward the deadline. It was like, “Okay, welcome. Here is your hotel room and off you go!”

[33:26] Allan: “Here is your hotel room you won’t be seeing quite that often.” I was in Berlin and I let them stay in my apartment [because] I’d been sleeping on the floor next to my desk. You’re getting a place for your friends to stay. Having that dream of having to work on Lord of the Rings — and actually getting to do it — is pretty cool!

Cameron: That story my friend likes to tell: When Fellowship was first announced and they released the first teaser that had Peter talking about the tech, we were watching that at my place, saying how cool it would be to work on it. And there I am 3 years later, working on shots that have Gollum. It’s quite cool! I still feel quite fortunate to make that jump. When that finished, I thought, “What now? What’s Weta doing?” Nobody knows. After that, there were even questions if Weta is still going to be a thing. That’s when I tried some work in Oakland. And then I went and worked at Fuel for a bit and did some ads, with Sam…

[35:22] Allan: There were a good bunch of guys there at 2005-2006. I would have been popping by the set. You’ve spent some time at Weta. What have been some of the more challenging projects and what specifically about them? 

Cameron: I think the most challenging slash highlights have been:

Avatar: I’ve spent a solid year on that on doing the floating mountain. That was challenging because it was photo-realistic full CG and the tech required to do it in Stereo that was pretty new We had a custom version of Shake that allowed us to do stereo. At the same time Peter Hillman was implementing the Deep Compositing Workflow. That’s been a huge advantage, for the last 10 years. We had to learn the simple things. It opened up all these new ways of working and how to comp things together. Just getting one’s head around that and how to use it effectively was a level-up on other people’s skills. I was also rendering clouds and expanding my skill sets with that. That was a big one.

[38:25] Allan: Just to segue for a second, to dive into deep compositing: Everyone I know who comes back from Weta and then goes back to Nuke, gets depressed. Do you want to talk about the benefits of working in Deep Flow?

Cameron: Well, the primary benefit I see is that nothing is baked in. In traditional comping, you have a character standing on the floor — that’s really easy. But if he’s standing behind a rock, either you need to hold out the character from the rock or from the background. That’s fine, but if you’re constantly having to re-render things because you’re changing animation, you’re wasting a lot of render cycles; whereas with Deep, you just render the elements separately. If the character’s arm animation changes, you wouldn’t have to re-render the background at all. But if it moved positions completely, the shadow would have changed and you would have to change the renders completely. You’ve saved a lot of render time. And there a lot of extra data that can go along with the renders. And that’s a big advantage with Deep. You can use it in a custom driven way, to recolor or relight things. So it’s the basic advantage and all the extra stuff that Weta put on top. It makes the files bigger, so only the big facilities can deal with it.

[42:04] Allan: What was the other challenging project?

Cameron: That was a little thing. I don’t know if you would’ve seen it. It’s that Disney logo that’s been at the front of every Disney movie for the last 10 years, or whatever. We first premiered it with Pirates of the Caribbean. That was a great thing to work on! It was such a talented team of people across the entire company. It all started with Michael Pangrazio who is the Art Director at Weta. He did this great concept art and all these fun little details that Disney in there.

[43:18] Allan: I would assume that piece would get a lot of scrutiny because it would represent Disney for quite a long time. 

Cameron: But also, there were fun things. Like the castle is an amalgamation of the various castles from different castles, all these different pieces. And then, in the background, there is a yacht. What’s that yacht? That’s Roy Disney’s yacht. And the train has a specific relevance. It was a hard project, a lot of late nights working with Michael, to get the colors just right. That was the other highlight of my time at Weta.

[44:23] Allan: You went to ILM as well. Were you working in SF? Having spent so much time at Weta, what was it like to make that shift?

Cameron: No, I ended up going to Singapore for that one. It was a bit of time of change, happening outside of Weta. My then girlfriend — now wife — went with me. We traveled a bit. ILM wise, it was interesting to compare pipelines and their way of doing things.

[45:26] Allan: What are you thoughts on that? Were you using Nuke there as well?

Cameron: I think they were in the later stages of their transition to Nuke. There were some things being rolled out that would have helped with rendering. It wasn’t fully featured as Weta’s implementation. I can’t remember what their production tracking system was. One of the things that struck me when I got there was their documentation. It had a lot of reference material and description of how things will look. Lots of elements! I quite enjoyed having the elements library. At the time, they were focused on what makes a shot look good vs. what makes it technically accurate. Weta was more about getting it right in the render and perfect from that point of view. ILM was more interested in making it look good.

[47:16] Allan: Who were the Sups you were dealing with?

Cameron: At ILM?

[47:26] Allan: I’m just curious. Depending on which Supervisor you’re dealing with, it totally changes. Like John Knoll (?) will totally disregard the shot and focus on the spec. It’s interesting to see the diversity.

Cameron: Yes, it’s definitely on the flavor of the Supervisor. On The Transformers, it would have been Scott Farrar on the San Francisco side. Battleship… I don’t remember! There’s been too long ago though, too much wine in between. But I think all the Sups I had focused on whatever looked good. What I did find there is that the workload was much lighter. I don’t know if the schedule was just lighter. I don’t know how it is now.

[47:57] Allan: I think from 2006-2007, ILM has gone through so many different growths. Just in general, for them to adapt, they’ve gone through many growth spurt, to see how they could compete with studios that were using out of the box software. In the end, the model that’s sustainable for them has been to have a global presence: London, Vancouver, San Francisco, Singapore. By being everywhere, they can take on anything; rather than charging more and being more selective. LucasFilm in Singapore outsourced a lot more. In 2011, it became ILM Singapore and began to bring in an international crowd of artists. For the US, it’s harder to get visas. But it’s easier to go work in Singapore.

Cameron: I think that sounds about right. When I was there, they were starting to move their pipeline development to Singapore, I think. So they were making lots of changes, and it was still in the old building.

[49:56] Allan: I love Singapore, but it’s so hot over there. If you want to take a photo, you’ve got about 3 seconds from the moment you take the lens cap off before it starts fogging up.

Cameron: You get used to it after a while. You only do your chores between [8:00] and [10:00] in the morning; or [5:00] and [7:00] at night. You stay inside during the day.

[51:32] Allan: You went back to Weta after ILM. What was it like returning? And what was the position you were taking on?

Cameron: It was a similar position. It was interesting, when I came back, the whole pipeline has been changed. All this new stuff has been implemented to make managing of assets easier. And there were all these new people, and they were doing more shows at once. And that happened in such a short time as well. I think they may have crewed up for The Adventures of Tintin. I came back and did some comp on Hobbit and Iron Man and Man of Steel. And that’s when I decided to try this lighting thing.

[53:13] Allan: Why did that intrigue you, and was it easy to switch departments?

Cameron: One of the things on Avatar, it’s an all 3D movie and an all render movie. A lot of the decisions are made in the render and that’s where a lot of the art direction decisions happen in simple drama sequences. It was partially driven by my desire to try art directing in an all-CG production and try something different. If you do the same thing for too long, it becomes repetitive. And I wanted to understand why it takes forever to get the new render. I learned pretty quickly why. It was a learning curve, a pretty steep one. I hadn’t really used Maya as a 3D rendering package.

[54:51] Allan: What were you rendering with?

Cameron: Initially, it was Renderman. And then for Hobbit 3, I might have starting using Manuka which is their in-house renderer. You get used it to using it one way and they you switch and you have to relearn it. I starting doing comping on some shows.

[55:45] Allan: I’m just curious, what did you do on Furious 7

Cameron: That was a lot of double work for Paul Walker, to finish the movie after his passing.

[56:03] Allan: What was it like when the job initially came in? Did you think that it would be hard?

Cameron: Yeah, it’s going to be hard, not your typical run of the mill. It’s going to take a lot of work from everyone: from shaders (from skin and how the blood moves under the skin, to expressions); to get it to match the reference plate and make it look a bit better — but still make it look like Paul Walker; to comping, to make sure to get it in there so it doesn’t look out of place when it cuts with the next shot that has Paul Walker in it. So it was pretty challenging, that one! Not much in the way of creativity, but satisfying when you get there.

[56:26] Allan: What was it like in reviews once you starting getting the shots to come together? Was it pretty mind blowing?

Cameron: You could never tell, even in the early renders it would be pretty spot-on. But then something could be wrong but you couldn’t figure out what it was. In the raw play, they would often use one or both of his brothers. The faces are different shapes. What looks good on the real version of his brother, does not look good on the CG version of Paul.

[58:24] Allan: Looking at something that has to look real, you could go down the wrong path a thousand times in terms of subtlety. I’m fascinated by that subject. It was nearly invisible to so many people, they didn’t even realize it. It’s pretty impressive that you were able to make it work, so people aren’t picking up on it. It’s not going to be one of those uncanny valley situation.

Cameron: That’s one of Weta’s strengths. They’re obviously a very good creature company: apes, dragons, all that sort of stuff. The Sup on that show was a very excellent technical guy. “Here is a render of Paul’s head. We need to tweak a bit of a shader.” It was very precise. You take on a show like this, it’s very challenging and risky, I guess. What if it didn’t work? But you work on a particular movie and built up the company’s skill set. It’s wise to take on these challenges when you’ve done all your homework. It’s a super challenge, and I think everyone did a fantastic job.

And there was a time limit as well. Everyone makes fun of Justice League and Superman’s mustache removal, but it was not given the time it could’ve had — to be perfect. With Furious, we were allowed to do our job properly. We’ve gotten heaps of reference material during the shoot. It was perfectly planned out.

[1:01:37] Allan: What was the timeframe for those Paul Walker shots once they came in?

Cameron: I don’t really remember the timeframe for everyone. I’m sure the client was like, “How are we going to finish this movie? Okay, we’ve got to do all these reshoots. Weta is going to do these shots with Paul. What do you need from us to do it properly?” And from that point on, we would’ve been working on their assets. I think everyone had a handful of shots they focused on. They were long shots, but we had to do it.

[1:02:48] Allan: That’s great! You mentioned Justice League. What did you do on that? You were a Comp Lead, is that right?

Cameron: Um, yes, the last two shows being a Tech Lead. I was looking after tech things, like testing templates for comp set-ups and ways to automate certain tasks; the tools we had in house, send stuff to stereo conversion company, particle set-ups. All sorts of things! It was more supporting the team, coming up with set-ups of templates and doing an odd shot here and then.

[1:04:15] Allan: I have to ask more about the sneaky mustache. That was Weta as well, correct?

Cameron: No.

[1:04:24] Allan: No?! Who was it?

Cameron: No idea!

[1:04:32] Allan: It’s just such a fascinating subject. The initial shot was done by a different studio, which was terribly bad. I’ll have to dig that up. The whole story is amazing: two studios battling over facial hair. One will pay the other to remove the mustache after.

Cameron: That’s right. Pick the worst option possible!

[1:05:23] Allan: I think it’s very clear that Paramount was doing Mission Impossible and making it difficult for Warner Bros. I’ve watched a lot of stuff on YouTube. I think that whole discussion is pretty amazing.

Cameron: And unfortunately for Justice League, everyone made such a thing of it beforehand. And then I saw the photo of his face, which was a wacky distorted version of it.

[1:06:41] Allan: That was from the beginning of the film, done by a different studio.  I think it was a rush pick-up job. It’s so much more obvious and yet it sets the tone. It’s distracting. I think it’s pretty fascinating that it’s no longer a gimmick to throw in a CG character. We’re getting closer!

Cameron: It’s good to see regular films become more adventurous with that. And it being seamless is cool.

[1:08:08] Allan: So what’s next for you? I didn’t realize it had been 15 years since we last caught up. Are you pretty settled in New Zealand?

Cameron: Yes, you know: two kids and a mortgage, all that sort of stuff. One thing that working in Singapore made pretty clear is that Weta is a pretty sweet place. You can live anywhere from 5 to 30 minutes away from Weta and the commute is super easy, the environment is quite nice, the weather is pretty nice. The people are cool. I’m enjoying the ease of being here. It’s easy to get home for dinner or leave for emergencies if the kids are being monsters. I wouldn’t be able to do that anywhere else, like London. It’s a 2-hour train ride.

[1:10:13] Allan: What’s your take on Wellington? It’s 50% American and every other nation is there. And everywhere is working at Weta.

Cameron: I don’t know what the breakdown of nationalities is these days. It used to be more American, but now it seems more diverse, especially with Weta’s decision to start training up locals. I like the diversity. There are all sorts of accents and sensibilities.   The city is good and friendly. Weta is a great company to have in the city; and how the city improved with cafes and restaurants, and a movie theatre being rejuvenated. And you have an easy access to skiing. I quite like it!

[1:13:10] Allan: That’s cool, man. Thanks for taking the time to chat! From a compositing or lighting perspective, is there anything new and exciting popping up?

Cameron: 

  • I reckon what you said about coding paying off. I’ve always been interested in reusing my coding background. In the last year, I’ve made a conscious decision to use Python more and more. It’s been fun and helpful to automate things and generate things. That applies to everything.
  • I’m really interested in what Unreal and Unity are doing, with their real time rendering. I look at demos of the features in Unreal and they throw lights in and spin it around, it’s real time. I’d like to be able to do that in Nuke or Maya. I think to me the biggest hurdle is the feedback loop and the time it takes to render. A shot I’m working on at the moment, I render out each part in 5 minutes. But it could still take 30 minutes before you see what it looks like. I’m interested in lots of things that I never have time to study.

[1:15:58] Allan: I totally forgot. Your having a background in programming, how you’re applying to what you’re doing these days. You can be creative with Nuke and build a strong workflow. 

Cameron: Took me a while to do it because I was tired or afraid. It’s really easy in Nuke to do stuff. You just search on Google something someone else has already done. It’s two lines to do it and it saves you 2 minutes.

[1:17:21] Allan: And those two minutes are two minutes of not having to do it every time, for the rest of your life. Which ends up being hours of time. That’s cool, man! Again, thanks for taking the time to chat. I haven’t been in Australia in 10 years. Very soon, I’m going to have to make my way down there. We’ll have to get some drinks. 

Cameron: I don’t think you’ll have any trouble having anyone to hang out with.

[1:18:20] Allan: I found a bar called Maya. Where can people find you online?

Cameron: I just have a randomly updated page at www.cameronsmith.info. I’m too busy to do anything about that too.

 

Thanks again to Cameron for taking the time to chat! I hope you’ve gathered a lot of valuable info from a very experienced artist.

I’ll be back next week! Until then —

Rock on!

 

INSTAGRAM