We are excited to be interviewing Jack Mathews, former Technical Lead Engineer on the first three Metroid Prime games. Jack discusses his time on the Metroid Prime Trilogy and highlights some interesting facts about their development.
I was one of the Technical Lead Engineers on the first three Metroid Prime games. Most of my contributions were in the core game engine, graphics, many of the visor effects, and data streaming.
We all knew we had something special when we showed the game at E3 2002 to universal acclaim (and surprise).
As for developing the sequels, on the engineering side at least, we just tried to push hardware as much as we could. For Prime 2, we really tried to go crazy with Dark Aether but had to end up pulling back on a lot of it for both technical and gameplay reasons. For Prime 3, we put most of the technical effort on trying a slightly more modern lighting model, having larger environments, and exploiting the Wii Remote’s controls.
On Prime 1 specifically, Retro was kind of this thing that was exploding all around us (this is all pretty well documented on the IGN nSider article). During the development of the game, Retro got cut in half, then cut in half again. Between that and all the leaks that were being reported at the time, we just tried to put our heads down and show everyone we could make a great game as a team. It was kind of a “fuck the haters” mentality, especially after the E3 redemption.
For me, it was maintaining 60FPS. This was not just about programming, but about discipline, both on Engineering and Art. The Metroid Prime series’ environment art was (and often still is) second-to-none, so it was a constant balance between keeping the game running well and still looking amazing. This meant making sure that every artist was running the game on the console, that QA had the tools to spot and surface frame rate issues, and honestly me just being a total asshole all the time about FPS.
SkyTown in MP3 was probably the most difficult out of all the areas in all the games to keep running at 60FPS. I remember the first time the concept was shown to me, I tried to get the area killed. We had to write a bunch of tools to support it and make the artists jump through a bunch of hoops. But even today, I’m still awestruck when I get to SkyTown and am super proud at what we all accomplished.
I understand his sentiment – he’s an eternal pessimist. That’s the thing about great creators (and he is one) – they see all the faults in their work and keep pushing to make the next thing better.
I don’t share the sentiment. I played Metroid Prime like 3 times in a row after we released it and have probably played it 9 or 10 times (a few times after Trilogy came out). I played Prime 2 a bit less 🙂 I played Corruption a bunch. My favorite game of all time was and is Super Metroid, and I love our games so much as well.
I feel like we could have built more, spent more, and made a cooler looking game on other consoles with more time. Would it have sold much more on more platforms? I’m not so sure about that any more.
A lot of Nintendo is built on the back of nostalgia. Just about anyone from my generation instantly lights up when you talk about Mario or Zelda (it’s the boy, right?). But when you talk about Metroid, it’s sort of a blank look of “oh, I kind of remember that.” Metroid, of course, has its fans and lovers, but I think it’s missing the mass market nostalgia that moves game sales. The easy answer is to look at the critical success but low sales and blame the platform, but I think that answer is a bit of a cop out. I think Metroid is a bit of a niche IP and that’s totally fine – we could use more big games that aren’t aimed at 90% of the market. But would going multiplatform be a big multiplier on sales? I doubt it.
My personal opinions? Prime 2 was too difficult, but it was masking a lot of other flaws. Being in the dark world was constantly stressful, the ammo mechanic was stressful, the dark world’s color palette was samey which led to environmental confusion and thus, stressful. The game simply made you anxious when you were playing it. It was a very difficult game to make because there were a lot of voices in how to follow up the first game, and I think you saw a studio and Nintendo, who newly owned Retro, trying to figure out how to really collaborate and follow up a critical darling. That came through in the end product.
Prime 3 has a lot of criticism leveled against it, but I feel like it is mostly due to the opening level of the game. Especially the linearity and cinematics – we spent so much time with all of that and the Hunters that it was feeling more Halo-like shooter and less Metroid. Once you get to Bryyo, the game really becomes a Metroid Prime game. And then when you get to SkyTown, I mean, holy crap… that area is just incredible for both gameplay and spectacle.
As for the motion controls, I so completely disagree – I loved our Wii controls and the Grapple Beam, and I feel like the game was unfairly judged for that. Especially when I go back and play Prime 1 on a GameCube controller, my hands pretty much seize up.
As for improvements? It’s hard because development is so iterative – many good things are built on the backs of failures. And many failures are the result of unmovable limitations. Like, for instance, I would have loved more differentiation between the light and dark worlds of Prime 2, but there was not time or money for it. We wanted to go the route of reusing environments between light and dark, because we could make a schedule work that way – we thought it would be great! More bang for the buck!
But of course, we could never find a technical solution to make the dark world variants different enough to be interesting, and it turns out that completely dark worlds are just not fun – the first concepts had Samus pretty much walking around with a spotlight emitting from her, and it was just impossible to play.
So, the artists had to do a ton of work to make the dark world variants work and, well, it was a ton more work than anyone had budgeted for and was completely unplanned. In fact, the Sanctuary Fortress was the only area built with this plan in mind, which is why both light and dark actually look visually distinct and amazing. But there was no time (immovable limitation) to really rebuild the rest of the game, but at least we got the Sanctuary Fortress. So what’s the solution? Plan better? Have less content? Be smarter? It’s hard to know, with the info we would have had then, what we reasonably could have done differently.
Well I do have one improvement – Prime 2 Multiplayer. That should have never happened 🙂
Well, I just feel that the game should have been either one or the other. Metroid Prime Hunters did this really well – it knew it was a multiplayer game and spent all its marbles there. When we started development on Prime 2, multiplayer was actually going to be the focus (I think we internally thought of the project as Metroid Prime 1.5). We were even going to have the ability to play as a Space Pirate and have things like wall grabs and such.
As we moved back towards a primarily single-player focus, we should have ditched multiplayer altogether. There was a ton of effort put forth to make multiplayer happen: we had to actually author third person Samus animations, we had to have support for multiple “players” in a game world, we needed to author all new lower quality effects, add game modes, et cetera et cetera. It’s all work that took quality and mindshare away from the single player.
Plus, as a Metroid fan, I just feel like single-player is how the IP should stay – seeing multiple people run around as Samus never felt right.
I don’t recall any code being written for them. There was a ridiculous amount of concept art for the game.
Probably the biggest throwaway I remember was Kraid. There were some prototypes for him, but it was pretty obvious we would run out of time to get him done.
I wish I remembered more. I just remember he was huge, that there were a bunch of jumping pieces, and that there was no way we were going to make it work in the time we had. I believe the Omega Pirate took his place, and I’m actually ok with that, because I thought that ended up being a pretty cool boss.
I don’t recall any code being written to support the concept, but I remember it being documented and thought through extensively. The inventory system as it was would not have needed significant changes to support the design, and while my recollection is hazy, I believe that the whole misunderstanding of the “bounty” in “bounty hunter” was discovered and talked through pretty quickly after we presented the game concept to Japan. I don’t know how much of the story was fleshed out at this point, Dark Samus or otherwise. I just remember that the original core concept revolved around Samus and her ship doing jobs around the galaxy to both get stronger.
I always had a great relationship with him. Any souring of that relationship with the studio would have happened after my departure.
I mean, the projects were all in different places and shut down for different reasons. As far as I remember:
* Car Combat was a concept that Japan just didn’t care for, even though at its cancellation it had more gameplay than the other projects combined.
* Football was deemed unnecessary after the game’s launch would miss the football season and EA had the market cornered anyway.
* RavenBlade was never going to come together. It was a concept way too big for its britches.
We couldn’t exactly bring gun-toting cars and football players into Metroid, so there wasn’t much recycling assets, but we did get wonderful team members who brought their expertise to the game. When the first major layoff happened and the Metroid team pretty much merged wholesale with the Car Combat team, development of the game really took off. Then when Raven Blade was cancelled and we picked up some of the former Football and Raven Blade guys, we ended up with the most talented engineering team I’ve ever worked with.
The sad truth is that the secret of Retro’s success story was starting with ~120 talented people, laying off almost half of them, then laying off almost half of THEM. This resulted in a team of top-notch people who REALLY had something to prove. There was a lot of emotional damage to unpack from that after Prime 1 shipped, and I applaud Nintendo for seeing that and making sure that the studio didn’t implode.
Strong female protagonist. Beautiful environment art. Amazing movement system. I would say it was less an influence from Prime itself insomuch as the people who put those elements into Prime. The Art Director and Game Director from the Prime games were the directors on ReCore.
Also, Mark loves logbooks.
My initial feeling was this weird, petty anger. Anger that Nintendo would have the gall to give Metroid Prime to someone else. I felt like Metroid Prime *was* Retro, so I reflexively thought of it as being a weird slap in the face, even though I hadn’t been part of Retro for like 8 years.
Soon after, it dawned on me that my attitude was totally childish and that Retro itself was just a weird unknown bunch of screw-ups when we GOT Metroid, so it may be nice to see what a fresh team with fresh eyes does with it. Now I’m just in “wait and see” mode. No matter what, though, I know that when it comes out I’ll blow the dust off the ol’ Switch and give it a whirl.
I doubt it. Some of my best friends in the world still work there, but the past is the past, and I doubt they’d have me back anyways.
I’ve been following the mod stuff off and on for years and I am so impressed. I also have to apologize to the modding community – I’m actually responsible for a ton of the file formats in the games – particularly the graphics assets, the PAK files, and I believe the area files. I cannot believe what people discovered without actual source, and I can assure all of you, I know how terrible the formats are and… sorry.
Wow, in answering this question, I really thought that the boost attack was still in Metroid Prime 2 Multiplayer. Looking it up, it appears we cut that that out and I think I remember why.
This attack was a multiplayer-only attack whereby you would lock onto someone, jump, then hold the morphball button. At that point, you would transform and launch into your opponent. I believe (and my recollection is super hazy) this was cut either because it would confusing that hitting the morph button would behave in this way, or because it could be nauseating to do a transition like this, or some combination of the two. It was a super fun attack though.
As for morph ball racing, I think we cut it just to cut down on the amount of work we’d have to do for multiplayer in general. I don’t recall the mode being very compelling.
As for the other stuff, I have no recollection of those.
I’d just like to say that working on the Metroid franchise has been the highlight of my career, and I’ll still be a fan until the end.
I left Armature last year to help support my wife with her business, so oddly enough, I’ve transitioned to being a food photographer most of the time. You can find her amazing food and styling work and me pressing a button at Love and Lemons.
I also helped my friends at Bluepoint Games on Shadow of the Colossus PS4.
As for me, you can find me on Twitter not posting all that much.
We wish Jack and his wife the very best of luck in their future endeavors!
© 2018 Darren Kerwin and Jack Mathews
Special thanks to RoyboyX
Interviewed on 7th January 2018