I love how Carmack is constantly pushing the industry forward.
He made raycasting engines when everyone else was still doing 2D stuff. He made fully 3D games when everyone else was still using raycasting. He built a game engine with completely dynamic lighting, when everyone else was still relying on lightmaps and some cheap hacks.
Arguably Duke Nukem 3D looked better than Quake, but polygonal was the future. Half-Life 2 looked better than Doom 3, but dynamic lighting was the future.
While everyone else keeps iterating current technology (often to astonishing results), Carmack does something new - something that may have a lot of drawbacks initially, but will be the future. I see the same happening in other industries - take Tesla vs. any other automobile industry out there for an example.
I haven't been this excited about a new piece of gaming hardware since I got the N64 for my birthday back in 1996. I'll buy it!
(Of course it's not only Carmack who is doing research in that department. This post isn't so much about praising him, as to praise innovative thinking.)
The UT and Q3A days were a classic time for competitive PC gaming! I loved them both but Q3 (specifically duelling) had an extra touch of class in my eyes.
TF2 is a weird one. I think it's undoubtedly a 10/10 game, but all the evolution since release has made the experience a bit "bloated", to use a normal software term. It's almost as if the game has had too much support for its own good.
I can see where you are coming from, but those extra stuff does add depth to the game once you learn all the new weapons. Granted, this is coming from somebody who plays a lot of TF2, but the "too many options" problem goes away certainly after 30 hours of playing, and often much less.
When I tried playing it, it was a very confusing experience. The built in tutorial was very simple, and the actual game didn't resemble it much at all. I gave up.
I think what has multiplied the confusion factor in the years since it was released (can you believe the game came out in 2007?) is the... culture that has sprung up around it, fueled by a unique feedback loop between fans and Valve. There's this whole strange world of hats, crafting, backstories, inside jokes, Australia, and hats that has coalesced, making it about as culturally daunting for a newcomer as an established Blizzard game.
This feedback loop has evolved what was once billed as a "class-based team action game" to, and I quote Valve's official PR, "America's favorite war-themed hat simulator." I'm not kidding.
At the same time, I can get on and play for 20 minutes with my friend in a different city, and we have a great time on public servers. If you want to get really into it, you can. I haven't spent any additional money on it since I bought it in 2007, and it's still a riot.
Sure, there's some cultural stuff that you'll have to get used to (and will become apparent) as you play more, but it's not nearly as much of a commitment as, say, WoW. A newbie (to the game, not to FPSes) can pick it up and play immediately at a fairly decent level. At the same time, there's a rich culture and stories and little things you can discover, if you care enough to spend the time finding them. If not, you can just blow some stuff up and go about your day.
I actually still play UT99 daily! There's still a fairly competitive community where we hold tournaments and stuff.
Plus, I'm developing something (long-awaited) for UT99 called NewNet (similar to Lotus's UT2004 UTComp mod) that eliminates all movement lag and simulates zero ping server-side. And at some point I'd love to create something similar to QuakeLive (Quake via the browser, with tons of stats tracking and stuff) for UT99.
For a 13 year old game, it's still more fun than most games out there if you ignore the trolls.
If anyone is interested in our little community, it's mostly on irc.globalgamers.net within the following channels: #tdmpickups, #iPug, #mlut, and #mlig. Basically you join pickup games (aka pugs) - 4v4, 5v5, etc. - using an IRC bot that keeps track of everything, captains choose teams, select a server, everyone enters, ready-up, and may the best fraggers win. And of course hilarity ensues on IRC afterwards. ;)
I misread your comment several times before I consulted Wikipedia to find some clarifications. It might be the case that others need the same clarifications as me, so I might as well post them here:
* Quake was released June 1996
* Quake II was released November 1997
* Unreal was released May 1998
So, the Quake "at that time" is Quake II. Just to clarify :)
"Make something fun" needs no clarification, though. That one needs agreement ;)
I agree with every bit you wrote about John Carmack's contribution to gaming hardware and computer graphics.
But it should also be noted that both these topics are not that important in the gaming industry as they used to be when Carmack started. Graphics have become "good enough" for 3D in 2004 or so, and much earlier for 2D.
The bottlenecks in todays video games are in Artificial Intelligence (for simulations), game mechanics and story telling (which is on a all time low if you ask me).
The latest Game from Carmack I know of, Rage, was just more or less the same as every other "AAA" shooter released within the last 6 years.
Yes; that's why I like Indie games more now; they are more creative with game mechanics + story telling. I think the current next-gen hardware is not enough of a leap to make enough difference; if you just take the Unreal engine it'll all look nice now. Complex AI or actual realistic human + animal movement of body/face (no, currently they are not) would require a lot more advances in hardware/software. For a big jump in AI I would say memristors would be something. But after HP shouting that they would be selling them last year, the hype died down :)
> Arguably Duke Nukem 3D looked better than Quake, but polygonal was the future. Half-Life 2 looked better than Doom 3, but dynamic lighting was the future.
Maybe so but almost 10 years later and half-life 2 is still more enjoyable than doom3. Quake1 and quake3 are the only id games that are remembered not only for their tech advancements (and of course, commander keen).
Now if we had had source engine facial and body animations combined with doom3 engine from the beginning (and of course the much better story-writing of hl2):
http://www.nofrag.com/2004/nov/24/15016/
I don't know, maybe. In my mind doom1 was too much of a "superwolfenstein", only 2 players max., no immersive ambiance (that is subjective of course). But it's a given fact doom1 was a major stepstone. As far as I am concerned I don't remember a lot of good moments with doom1 (on the other hand Q1 MP did :) Maybe I was playing too much console games at that time though, or maybe replaying kirandia or flashback for the 23d time.
Disclaimer: I always disliked dn3D, didn't like the mood of it.
Maybe it's an age thing? How old were you when it came out?
That game changed my life. It was the most immersive, scariest thing ever. It was 4 players (not 2), and we had endless LANs and modem games. We made maps, there were whole map-making communities, whole BBSes only talking about Doom and Doom 2 and Heretic and other games based on it.. It was the biggest thing ever, even bigger than Quake in many ways because there was a lot fewer similar games at the time. I played through the single player so many times that I can revisit most maps in my mind 15+ years later...
Even the book Masters of Doom is mostly about how that game was created, rather than Quake or other id games that came before or after because Doom truly was the game changer.
I recently replayed Doom 2 with a couple of friends, and for us this is definitely a legendary game. I have lots of fond memories of Doom 1 as well. While I also spent a lot of time with Duke Nukem 3D, I never found it as enjoyable as Doom. I'd put Doom in the list before Quake.
Wolfenstein 3D should also be there.
Luckily for me, you are the one making the sweeping statement, so I only have to provide a counter-example: I remember Doom and Wolfenstein for the enjoyment they gave me. You are too quick to dismiss them :)
(And just as a matter of factual correction; Doom 1 and 2 both supported four players over IPX)
Exactly. First I didn't understand why MegaTexture was implemented in a production-ready game (Rage) because it made lots of texture look like a decade old, but I guess Carmack sees the future in it.
Rage was a huge shame (tech wise). They needed to be able to ship more data to make the tech really shine. The game was 3 DVDs as it was but it needed to something more like 8 or 9 DVDs.
It just feels like such a shame for the graphics to look bad for want of the ability to ship data to the customer.
Hm. I only played the PS3 version, but was actually impressed by the graphics. Not only does it look great, it runs smooth as butter, especially when moving. I don't think the Rage engine was ever built for still shots, but in movement, it think it is gorgeous.
I don't know how they did texture compression for Rage, but the pc version was 24gb. Skyrim came out a month later and was 7gb. They had "some" reasonable texture quality on character meshes and such, but a lot of the world textures were pixelated garbage - I have no idea where they had the game become so bloated it was 24gb like that.
Reminds me of the Force Unleashed, which was another game that took up 25gb for no reason whatsoever, when the texture quality and game size were no larger than War for Cybertron that took up 8gb.
It has to be their texture compression in use, but they should figure that out early on! (note, I have no idea what I am talking about).
Rage uses a new rendering approach in which every surface has a unique texture. In a normal game, textures are reused all over the place. The texture compression Rage uses is actually very good, it's just that the data set is massive.
Their raw data set is much, much larger than what they shipped. Carmack has given out figures like 300gb per level.
In order to cull that down to a game that can fit in 24gb, they had to drop a huge amount of detail. The way they did it was quite interesting as well.
They get telemetry on clients playing the game, and then they build a weighted set of textures at each mip that were actually loaded by real clients. Then they take the highest weighted 24GB of them, and that is what they ship.
Why are you so opposed to this idea? I think it would work well if implemented properly. It's similar to how games like GTA IV work. They don't load a level, they load the immediate area you're looking at. As you drive around, the new maps are streamed from the DVD. Other games have played with loading a small first level, and as you're working through the first level the rest of the game streams into cache behind the scenes.
It would be nice to get a game on a 4GB DVD and have the rest of the data get pulled down as you're playing. Guild Wars did this. While you're in one zone, the next zone is being downloaded in the background. It would only work for geographical regions which don't have caps on Internet traffic (for example, Americans on Comcast, but not Canadians on Rogers), but it'd be a nice option.
I haven't looked into MegaTexture much, but from what I can tell it doesn't repeat terrain textures.
In most modern games, smaller textures are tiled over larger areas, allowing for lots of detail for less memory at the expense of variation. Most do something to mix it up (mixtures of textures, adding random variations), but it ultimately constrains what can be done.
MegaTexture seems to solve the problem of using a single large texture by streaming only the parts needed, but requires much more data on the disk to match the detail level of tiling.
You can't really compare it to skyrim though. Skyrim doesn't use mega texturing (or virtual texturing). It repeats textures all over the place.
In rage, there is a huge texture for everything.
IIRC, the uncompressed textures performed a total of 1.5TB or something like that, so I would say that they did an awesome job with the compression.
Not really. The lighting in Half Life 2 was based on static lightmaps. They were very fancy lightmaps, which enabled them to use techniques such as normal mapping or specular reflections, which are normally incompatible with lightmaps, but they were static lightmaps nonetheless.
It has some dynamic lighting effects, the same way Quake did them (modifying the lightmaps). It could also do projected shadows and self-shadowed objects using shadow maps, but not on world geometry.
I believe that the current source engine has some kind of dynamic lights. Either that, or Portal 2 did a fantastic job of faking them.
Still, that was 2011. Doom 3 came out in 2004, and exclusively used dynamic lighting, without a static lightmap in sight.
I knew someone working in VR about 20 years ago, and the big problem then was latency - if there's too much delay between when your head moves and the scene is rendered at the new orientation, it just doesn't work. IIRC the eyes are sensitive, in this respect, to delays of less than a movie frame's 1/24 sec.
Carmack is both very concerned and very knowledgeable about latency in rendering, so it seems likely that he's solved this problem... but I haven't seen it explicitly addressed in anything I've read about his recent work. It might help his advocacy if he explained why this tremendously exciting thing failed last time, and that he's solved it, to overcome the "VR winter".
These videos were more interesting than the OP. Hearing him geek out about all the things he's tried to make it work well makes me really excited for the stuff that's in the pipeline.
I found John Siracusa's summary of the march-of-progress on Hypercritical insightful. Along the lines of:
> Processors can never be fast enough, memory can never be big enough. If you take your favourite device 20 years in to the future and try to convince people that it had been "good enough" to settle down, they'll rightfully laugh in your face.
---
Then again, it's a double-edged sword if increasingly advanced ways to bombard you with distraction, or make trash jingoistic war games, come along for the ride.
The 1990's were fifteen years ago. Since then Carmack has spent 60 hours per week writing code for every conceivable console, PC, and mobile platform. Most people in his position would have moved into management by this point, he chose not to. As a result he has more expertise on more topics than just about anyone else in the gaming industry. And more than most in computer science generally. If he has an opinion on something, its because he knows what he's talking about.
I'm not saying he doesn't know what he's talking about, he just never liked consoles, he was always a PC guy and if he were truly "excited" about the nextgen that's like hearing Armstrong will quit riding bikes and be a F1 pilot.
And I'm saying, your opinion is stuck in the 1990's. John Carmack was very excited about both the xbox 360 and the Nintendo Wii. He's just not excited about the next generation because, in comparison to the most recent generation, he believes we're at diminishing returns for having more and more powerful 3d graphics. Which is exactly why he's taking VR headset development into his own hands.
Not yet, it was planned to go live 1.5 weeks ago, but some "good news" delayed it by 2 weeks, so it should probably go live this week. And the good news might be some great hardware partner(s).
It doesn't sound as if anything has been done about the VR sickness problem. Wearing bulky screens in front of your eyes probably doesn't have much of a future, but I think that augmented reality using something like the EyeTap device has a much bigger future which extends significantly beyond gaming.
My understanding is that the sickness was in large part due to the latency (Solved by Carmack). Beyond that it's just a matter of getting the perspective correct (Also solved by carmack).
It occurs if there's a sustained mismatch between your vestibular sense and visual odometry. That's ok if the game only involves looking around in a fixed position, but if the avatar is walking/flying then you're likely to get VR sickness, no matter what the frame rate is.
Apparently, VR users do habituate to the effects after a few days, in the same manner as space sickness, but in the intervening period there can be some unpleasantness.
And if I were as wealthy and respected as Carmack in the industry (and as sharp) I would also be building rockets, supporting Linux gaming and similarly quixotic but awesome enterprises.
It's a hack, it's not meant to be a broad commercial product. It's not much different than the test engines he builds for experiment. You're indeed being unnecessarily cynical.
I'm not all that excited about John Carmack's antiquated view of the video game market. While there are companies releasing swaths of titles in different genres, he's got... Doom 3. Yay, whatever.
Most people will not spend $500 on a peripheral. You can get some hardcore gamers to do it, but unless the VR gear also includes a wireless headset, you're shit out of luck. So now you're looking at close to $600 bucks a unit. Good luck with that. I'd rather buy a 48 inch TV that I can use when I'm not gaming.
John Carmack has written 3D engines widely used in commercial games for decades now (most recently Rage - whatever its demerits as a game, the technology is cutting edge and the game is definitely not antique). Whether or not you liked the content in Doom 3 has very little to do with Carmack's very technical, back-end work.
When the 3dfx boards came out, they cost something like $500. But it still opened up a big market. I don't see why you are assuming that a wireless headset is necessary and will not be included. Buy a TV if you want, all you've shown is that you are not in the targeted market at this early stage. (Probably because there is no way to provide that kind of hardware at lower cost, particularly without more traction)
Carmack is certainly wealthy enough already, I would assume that his interest in VR headsets is that he is a geek who wants to move the state of the art forward (not too different from the reasons for doing Armadillo Aerospace)
I don't see why you are assuming that a wireless headset is necessary and will not be included.
Wireless VR is very important. Getting tangled in cables while you're physically thrashing around sucks. Trust me (owner of original Virtual iGlasses).
Well, the kit cost $500. Once it's in the hands of a company that can mass produce, the price could be substantially lower. And with console manufacturers, there's always the possibility they'll sell hardware at a loss to move software. The point being we don't know the final end price yet.
The real concern there is the Big Corp will compromise the design just enough to make it uninteresting.
Video games market IS antiquated. Crappy gamepad controllers, requirement to use TVs, hardware upgrades lock for 5-6 years, dumbed down games (as a result of gamepad use).
Sure, it has billion marketing budgets, easy to start using equipment, but technologically it is antiquated.
He made raycasting engines when everyone else was still doing 2D stuff. He made fully 3D games when everyone else was still using raycasting. He built a game engine with completely dynamic lighting, when everyone else was still relying on lightmaps and some cheap hacks.
Arguably Duke Nukem 3D looked better than Quake, but polygonal was the future. Half-Life 2 looked better than Doom 3, but dynamic lighting was the future.
While everyone else keeps iterating current technology (often to astonishing results), Carmack does something new - something that may have a lot of drawbacks initially, but will be the future. I see the same happening in other industries - take Tesla vs. any other automobile industry out there for an example.
I haven't been this excited about a new piece of gaming hardware since I got the N64 for my birthday back in 1996. I'll buy it!
(Of course it's not only Carmack who is doing research in that department. This post isn't so much about praising him, as to praise innovative thinking.)