Polish games mag CD-Action (who, according to these videos, seem to have their own drink?) have posted an enormous John Carmack interview to the web, and we’ve reposted it below. In it Carmack chats about Rage, stepping away from graphics to work on player experience, the problems of contemporary developments, the challenges of modding in the “post-Doom 3 era”, and so forth. Thanks to Pat at VG247 for the heads up on this.
Related Stories:
There will always be modders capable enough to mod games as long as the developers do not go out of their way to prevent it. Anyone who thinks there are not modders out there every bit as capable as top game developers, absolutely including Carmack, is, in my opinion, quite wrong.
SDKs are a rare and wonderful gift from the devs. Simply releasing in-house tools without documentation, which would not require a huge time investment on their part, is a very god , but uncommon, thing. But even when devs do no such thing, modders find a way. Many modding communities have built their own tools in order to mod games with no dev-provided tools whatsoever and they’ve done a brilliant job of it. But it’s always a shame when devs don’t foster their modding communities. Time spent building the modding tools is time lost from doing the modding. Often that cost of entry is too high and games that would be great platforms for modding end up ignored.
Also 20 minutes in to this is like watching from 20 minutes into an episode of True Blood.
I have no idea what the hell is going on but I really want to.
Also how can you speak for 10 minutes each time someone asks you a question??!?
“So John you want to go for a pizza?”
“Well the spacial distortion ratio of the the blah blah blah blah blah blah blah…”
“Fuck you John… I’M OUT!”
They aren’t asking questions in between the videos. The only reason they’ve been split up is because of YouTube’s length limit. :P
@Guhndahb: Heh, yea, there are games that are modded without the offical tools ever having been released! Check out the custom Trespasser level “The Isle” over at Trescom (they’ll change hosts soon with IGN dropping support… bah, IGN :( ). The level was made with a hacked, community-created editor that allows to create entire new levels and even dinosaurs for the buggy, awesome, “Jurassic Park”-based game. So much for dedication and modders with patience…
Regarding the recentish changes in the mod scene, notably the fall off in number of releases from HL1 to HL2 or the increase in team size/delays, Adam Foster (of Minerva and now Valve fame) has been arguing this point for ages. Instead of following professional designers, building entire worlds with thousands of assets and ideas, consequently falling into the trap that Carmack suggests, Foster always pushed for high-focus realistic-ambition mods which had something more interesting to contribute than gun renders.
There are a couple of interviews about where he talks about this stuff, but I’m too tired to look. Any volunteers?
Nail on head there. Half the problem with mods nowadays is trying to replicate the sort of content and scope of a professional games studios who have 200 people on the payroll using nothing more than 20 blokes in their bedrooms doing it as a hobby. This was almost feasible back in the days of Half Life, but it just doesn’t work anymore.
I think the future of mods lies in smaller and more experimental stuff rather than these vast and sprawling TCs with 20 guns, 8 maps and all the other associated code and content.
The Doom 3 mod community failed because when it came out the multiplayer was barely playable except for a select few who had a system that could manage it and it was a horrific mangled mess. The HL2 community took off reasonably well because it had the CS player base to draw from – that’s certainly the reason we went for HL2 rather than anything else when it came down to doing Hidden.
But I also agree that there are now technical hurdles that get in the way for mod teams to do well. Part of that is the player base of games wanting stuff to be more shiny and professional looking – we released our first version using CSS assets and got ripped into for “just being a CS mod” by people who clearly never bothered to actually try the game. Look at the Living Legends, who have brought in a set of professional artists to do their mech models for them or Insurgency who got time in a mocap studio for their animations. The bar has been raised not only by the relevant tech level of the engines but also by the player base and the mod teams, which is why so many mods die on their arse these days having released a handful of ‘clay’ weapon renders.
Making content for a modern engine is a bastard, Source has the worlds worst tool chain for instance while a map in UT3 relies on about 90% mesh work to add detail and layout, all of which needs a normal map to be created for it or else it looks out of place. Creating a good normal map tends to require experience in a program like Mudbox or zBrush, which are whole new mysteries for an amateur to get their head round (but also basic requirements for artists wanting to get into the industry at this point).
Wow, that got a bit rambling, I’ll stop.
New drinking game idea…take a shot every time Carmack says “on there.” You’re gonna get messed up!
@vader
Why doesn’t RPS have their own cannabis strain! This is an outrage! I demand you make your own strain and put lots of THC in it!
I’m Polish me own self. I am really curious whether the Polish guys who interviewed Carmack understood a word he was saying, or were they just nodding and repeating “yes, yes” to keep him going ;)
Well i think its not just the tecnical hurdle but other stuff too
1. Flood of FPS. Nearly every month a new FPS gets released so there is simply no demand for another team deathmatch mod with an M16 and an american marine.
2. Short lived industry. You have to base your mod on a game that is still alive by the time you are done with your work. How many games outside HL2, UT and maybe a few others are still played by a lot of people 6 months after release?
3. No dev support. Where´s the little big planet of FPS where every noob can klick together a level or multiplayer mod in a few hours? Where are online asset databases everyone can use so no team has to ever do a crate or a spacemarine again by themselve and can only slap thogether assets already made. I mean how many more times do we have to moddel an M16. By now thousands of those moddels should exist why not simply bring them all in one place and let everyone use them.
Basically making it easier for the people is needed if the modscene is to survive and ever get a success like Counterstrike again.
@Schrodinger’s Lolcat
He makes good points. I was comparing Quake/2 era to now. Quake 2 didn’t have a massive amount of documentation until the community got hold of it. That’s partly due to the broad focus of PC gamers being on that single game, rather than split between many.
Major tools weren’t so necessary back in the Quake 2 days because there wasn’t such a hurdle for quality. Now, yes, being able to leap that raised hurdle easily is almost totally down to tool sets making increasingly complicated tasks easier.
Modding old games is more fun anyway. Except when it’s brain-meltingly frustrating (which is almost always).
I think it’s interesting to compare the Neverwinter Nights 2 modding with for example Half-Life 2 modding. NWN2 has 2D level design because it has no movement in the z-axis, its interiors are tile-based, and its exteriors are made with really user friendly terrain editing tools. It includes hundreds of objects and enemies for modders to use in their own work. And finally, the game has some really robust storytelling mechanics and tools – a great dialog editor, a really easy scripting language, good camera controls, etc.
My point being, I think because it’s so easy to create the world in-editor and set up a pretty complicated narrative, the focus in the NWN community is on modules with new stories rather than tons of new content. And since the tools are so powerful and user friendly, it’s way easier to create huge mods for NWN2 than to create even a small level for a game like Half-Life 2 or Doom 3.
And what Ging said.
Wow, there’s just an unholy amount of factors to what makes a game popular to mod.
I’d also say that the “vanilla”-ness of the base game is an issue. The more finely balanced/intricately designed the game is (i.e. the more special case/non-generic it is with behaviours) the easier it will be to “break” the game. If there are more general case things going on, then the game responds well to tweaks and additions which make use of those hooks.
firstly, about the lighting to me this is a clear stop gap between the classic raseriser in rage and the sparse oct-tree voxel tech he’ll be using to light his scenes in the future, rage doesn’t need too much dynamic lighting so it doesn’t support it, however id tech 6 will almost certainly support a raycasting-like lighting system.
about the mod stuff all he was saying was, he’s going to do what he always does, dump his source and let people fuck around with it. He was saying that the engine was too complicated for people to be able to mod with it without tools, he wasn’t saying that people wont be able to make mods with similarly complex games, he was saying that people will need tools to do it, which they will, the mod support for hl2 is down to the tools.