Comment by Arjuna | Hacker News Reader

Comment by Arjuna | Hacker News Reader

Wow, I'm just now seeing this news. Initially, I had that sinking feeling set in... I mean, like you, I have been impacted by his story, his games (not just the Wolfenstein/Doom/Quake franchises... I'm talking Commander Keen, boys and girls), his code, reading Masters of Doom, etc.

I can see my copy of Michael Abrash's Graphics Programming Black Book Special Edition sitting here, which was such a treat to read when it came out, because it has so many great chapters on the development of Quake and little stories about John's discoveries and thought processes throughout the development of the game.

But, then I thought... wait... this is a new beginning. I wrote about this previously, but, look for gaming to start heading in the direction of VR with technology like Oculus Rift. Also, with someone of the caliber of John Carmack involved (now totally focused on it because of the resignation announcement) with not only his passion and skill, but his ability to work with graphics hardware manufacturers and driver developers to effect change and garner the necessary support and backing, expect to see vibrant, compelling developments in this field.

In case you missed it, check this video out of John discussing some of his VR work. It is from E3 2012:

https://www.youtube.com/watch?v=NYa8kirsUfg

That momentary sinking feeling has faded away now... great things are ahead!

Frankly, he's too talented to keep making mediocre shooters for twitch gamers. Heck, 3D game engines are practically commodities nowadays. No idea if the Oculus Rift will succeed, but I love it when people on his level take chances. I watch the Rift project very carefully, and really see a lot of potential there. It might not end up being what we expect it to be in the end. We assume it'll be this great little game accessory like a Kinect. It has the potential to be much more.

Poorly optimized 3d engines are a commodity. Quake 3, Painkiller and Serious Sam 1 were a piece of great engineering. Also there was nothing mediocre in the twitch based era of id. Their decline began with Doom 3.

Yeah, Doom3 was really bad and unlike any of their previous shooters. And Rage was a piece of garbage. Not even technically impressive despite the mega-texture tech.

Doom 3's problem wasn't it's engine. It was just crap. No matter how careful you were to cover yourself and check each corner, you'd always cross a 'tripwire' of some kind and a section of wall would open up - something you could never check - and it would always been behind you. It was the laziest attempt at horror I've seen. Once that became apparent, sneaking around was pointless. Just barrel forward, because it doesn't matter what you do, there will always be an undetectable secret wall opening behind you.

Compare to Dead Space, where the things that came through walls did so at reasonable points - ventilation ducts or similar. You'd be on your guard around those spots, but a solid wall was never something with a 'gotcha!' compartment. Being wary was useful in that game.

Doom 3 could have had the sexiest engine in history, but it wouldn't have changed the poor gameplay design.

Sure, after a while Doom 3 became boring and self repeating, but hell, it was one scary motherfucker to play in the middle of the night with your headphones on. I remember playing Doom 3 on my original Xbox and I just couldn't do it many times, as it was so damn scary.

Most of this had to do with the excellent sound design though, and this scary effect kinda ended when you got the bigger weapons in the game.

I've tried to play it a couple times but the first half hour or so is walking around with no shooting and nothing happening and that's pretty unforgivable for an id shooter.

Even Carmack admitted that in retrospect the lack of radiosity contributed to the poor gameplay. He also lamented that most of the art was never seen because it was too dark.

So, a bit of a rant, I dislike the current game culture meme of "this engine is poorly optimized". I am tired of seeing the following statement: game x is poorly optimized because game y runs better on my box. This is akin to saying: I find apples delicious, Oranges are terrible. Get apples noobs.

I would argue that the fact that you can drop in any one of a half dozen solid engines based on your programming language of choice speaks to the fact that they are commodities. At this point, for the vast majority of cases, it makes as much sense to write a 3d game engine as it does to write your own web server.

You can do it, but you could spend the time doing something that directly relates to the product you want to sell instead.

I think you misunderstand that for a long time, Id Software has made a lot of money turning 3d engines into a commodity. For John Carmack, the emphasis has always been on developing new graphics technologies that advance the entire gaming industry. He's first and foremost an engineer. He may have done a significant amount of game design back in the day, but for at least a decade he's primarily been focused on pushing the hardware to its limits. Since Doom 3, Id's games have pretty much been tech-demos for their engines. That's not to say that Doom 3, Quake 4, and Rage weren't good games, its just that none of them are truly memorable experiences like their older games were.

There are only a handful of major players that are making meaningful contributions to the world of 3d engines. If companies like Epic, Id, Valve, and Crytek didn't continue to develop new 3d engines, they wouldn't be a commodity anymore.

Also, the Source, Unreal, Cry, and Id Tech 5 engines are still differentiated from one another enough to warrant all four of them existing.

In short, I don't think 3d engines are a "solved" problem. Of course an indie studio looking to publish its first 3d game should obviously use a readily available engine, but the major studios still have good reasons to roll their own.

I'd imagine that a lot of the need for new engines is driven by new hardware, and it seems that the video hardware has stopped evolving at a rapid pace. nVidia and AMD have been focused more on computation than graphics recently.

> Their decline began with Doom 3.

That was some 9 years ago, already...

Yep, this.

I'm extremely relieved that he's gone onto new things. As much as I loved and grew up on Quake, the company is still rehashing their greatest hits with little to no inventiveness or creativity.

I'd much rather see Carmack's talent applied to Oculus which I am incredibly excited about, rather than another extremely anodyne shooter. Gaming's moved on.

I am absolutely blown away by the video you linked. I recommend everyone here watch it to see John Carmack's brain at work. Just an example: he used his sensor experience from Armadillo Aerospace to calibrate the gyroscopes on the Occulus for significantly reduced display latency.

I'm buying one just because of this video.

I agree. Carmack's enthusiasm about this project is infectious. My heart rate increased at least 20 bpm by the end of that video, and I only understood fragments of what he was talking about.

Same feeling. Followed by, "Oh good, he's going to focus on Occulus."

If there's a man you want as CTO, it's Carmack. One of the all time great programmers working full-time on VR is the best possible outcome.

Hmm... I'm not convinced. I think you want john carmack as an engineer with a lot of freedom and resource and a limited amount of management responsibilities. He seems far better suited to being first among equals (first among equal engineers) than to being a technical business-decision maker. Give him the role of arguing for the right decisions rather than making them, and I think you'll get a lot better value for your money.

Can't say I'm too surprised by this. I mean, the guy's amazingly talented and has been working almost exclusively on technology for basically the same game for like 20 years. Granted, the hardware has changed considerably during that time, but I can see how that might become just a wee bit tiresome after that amount of time. Definitely looking forward to what he's able to do for VR though.

Wow that was one of the most informative interviews I've ever seen.

>https://www.youtube.com/watch?v=NYa8kirsUfg

Is Vinny using a wrench... on a hammer?

Reply on news.ycombinator.com