USG: You guys have talked a lot about authenticity, accuracy, on the technical level. Without getting too technical what sets this apart from just emulating the games or at the same time just going in and reprogramming them from scratch? Those are the two common approaches to dealing with these kinds of classics. Were you doing something different?
Frank: Porting from scratch, it's pretty necessary for later titles. You can't create a virtual environment from the PlayStation 2 or whatever. But for these older titles, it's a lot riskier than taking a simulation-based approach because porting, something's going to go wrong. There's been some recent, like the Silent Hill Collection. There were a lot of weird errors and stuff in that. They did a beautiful job otherwise, but those flaws that come out through the porting process are just so glaring and I don't think really works with the approach we wanted to take with this.
So, what we set out to do at Digital Eclipse, we didn't feel there was an equivalent to... like, scanning a film in 4K, is the analogy I like to use. You scan a film in 4K quality and you have that print down that's good for streaming, good enough for Blu-Ray, good enough for 4K Blu-Ray or whatever's next. You can make that one safe scan and it works, but with video games, there is no approach like that, because system architecture is always different. You can't port a game to a system and say it's safe, next time around, we'll just move it to the next hardware... It just doesn't work that way.
Rey: Even visually, from working on that older collection, once TV's progressed past 480P we're SOL. We had to do something else. So [the PS2] Mega Man Anniversary Collection, if we just ported or emulated the Anniversary Collection, it would look just like it did on the PS2, which isn't what it would look like on the NES, it's visually…. The one shot and you're done, doesn't work.
Frank: That was displaying 480i, so you have the blur, crop the sides, I don't want to talk bad about your previous project, Rey, but the approach we took with what we call the Eclipse Engine was, "Can we do something like scanning a film in 4K, but for games?" But there's no real magic bullet solution to that, but our approach to it is: We set up our Eclipse Engine, and we set up hardware simulation modules, and we convert using source elements provided by the publisher, — their original game to our format. So you have Mega Man running in our hardware module within the Eclipse Engine. The idea being, once it works in Eclipse, we forward Eclipse somewhere else, and it just works, for the most part.
Rey: There's UI stuff that happens with the games themselves.
Frank: But the games themselves are safe. The Eclipse Engine, at its core, is fairly simple and easily portable. The philosophy being, when PlayStation 5 comes around, Rey can call us and it won't be that hard to get these games running again.
Rey: So they don't have to port the game anymore, they just have to port the engine.
Frank: We think it's a solution. I don't know if it's the ultimate silver bullet solution for making sure games are preserved forever, but I don't think anyone's taken that particular approach.
USG: I'm sure cycle-perfect emulation would be ideal, but...
Frank: Do you know how far off we are from that? [Laughing]
Rey: NES and Super NES emulation is a hazy area that we're not really allowed to do, emulating games on a non-Nintendo system…
Frank: We take that approach at Digital Eclipse when no one has a problem with it. I think emulation is still a valid approach, but to your point, cycle-accurate is a phantom we'll be chasing for decades. Atari 2600 is cycle accurate to about 1 frame per second. That's how it looked. I'm talking about they did 3D models of the 6502 chip and tracing the electrons, that's the only way you're going to get actual cycle accuracy. We're nowhere near that [laughing], but given the modern hardware, we're really pushing it as far as we can to deliver a product that people can actually download and even play.
Rey: From the publisher standpoint, from Capcom, knowing that later down the road, we don't have to do it all again, the games themselves are going to be in a format that is easily updatable on a new system is great.
USG: So instead of the Eclipse Engine being an emulator running a ROM dump, you're basically feeding source code or ROM or whatever into something on your end, not on the user end, and that's putting out an executable for a specific platform.
Frank: That's about right. We have hardware-specific modules that we're running within our engines. I'm not intentionally being cagey with the approach we took with Mega Man, it's just really complicated. We did basically recompile it into a new format that runs in our engine.
USG: So you mentioned something about a hardware module, so the Eclipse Engine is not just NES, it has the potential to be…
Frank: It has the potential to be anything. But the announced product we have right now is a simulation of NES-like hardware. It's not really an NES module.
Rey: Capcom is really proud to be part of the first title in the initiative. Working with the local independent developers who I've personally worked with before in other incarnations, so it's really good to be working with them again.
Frank: Yeah, we've been working with Rey for a long damn time.
USG: So what learnings have you taken from those previous projects and applied to this one? Because it does seem like this one has a different ethos about it...
Rey: I think Digital Eclipse's focus is more on the technical standpoint and their focus on how they want to recreate the game. From my standpoint, it's been pretty much the same. On the tech side, that's where most of the learnings have come from. Being able to be accurate, be authentic. From the publisher's standpoint, it's always been celebrating whatever we're working with. In this case, it's Mega Man, and we're focusing on the good parts of it — what makes it so good. Why do people like Mega Man? Talking to folks like yourself that really understand it and have an appreciation for Mega Man itself, so, that's been the publisher point of it.
Frank: And everyone knows why we're doing this. And we're really fortunate that Rey and the rest of Capcom have been really supportive of our vision of what this project should be; we've all been on the same page. It's been a great project.
USG: Is there any particular reason that Mega Man is the initial/inaugural outing for this kind of tech or is it just happenstance?
Rey: We absolutely went through our entire library figuring out what we're going to do first. What's the best way to get this product out? And to be completely honest, we were talking about Capcom's 8-bit library as a package, and Mega Man makes sense. Don't get us wrong, Capcom loves Mega Man and other IPs as well. It's just the one that made the most sense. What do we want to work with? What are the fans, what is the largest audience really going to latch on to? Because if we did another one, say if we did a Disney IP, not everyone knows Little Mermaid, but kids nod at Mega Man. They go, "I know that guy." So Mega Man is just a bigger deal for us and totally makes sense from our perspective.
Frank: We're trying to sell this technology to other potential partners too, right? And we were really happy to come out with as big a splash as Mega Man.
Rey: He is our most important character. You're really familiar with the product itself so I don't think there's anything to say specifically about the games. I'm glad we're able to talk more really meta about the game itself. You have a deep understanding of the game itself, so it's good to talk more about they why's and how's.
USG: Well, that's what I've been curious about. Also, is the 3DS version going to have any special challenges with it, given that it's a less capable platform than PS4?
Frank: We're going to have second skin support and have some 3D elements. Otherwise, we'll talk about the 3DS more later on. Content-wise, it will be the same.