JC Keynote talks consoles

[AlNom]

Did JC really say this Titiano?



If so I in no way can believe this.

Well if he's using code that was written for OoO processors on an in-order processor(Xenon), there's your problem. Of course OoO code is going to run faster on an OoO processor compared to an in-order.

Don't tell me Carmack is pulling an Anandtech....

"If you just take code designed for an x86 that's running on a pentium or an athlon or something, and you run it on either of the PPCs for these new consoles, it'll run at about half the speed of a modern state-of-the-art system, and that's because they're in-order processors; they're not out-of-order execution or speculative..."

In the same thread. *sigh*.....

 

[pc999]

I really dint agree with him about a few thinghs from what is say.

About AI he is compeletely off way, people do notice when a AI is scrippted, if not at first time second or third they will (unless they are very stupid), this is the kind of thing that really add replay vakue (it is the reason why Halo is famous IMO, and even Halo ido a have a AI that should be very poor in comparition to what they should be to do if AI already is a prymary focus).

Physics could have a big replay value IMO, and probably they dont need to be 100% acurrate o be a very good addition to games, if that can unlesh some power to other things.

Gfx IMO are done ate least from a gameplay prespective, once we have things like UE3 and WarDevi I am sold , I really doubt that there is anything (gameplay) that can be added by gfx, I do expect to start see even more unrealistc games.
But about HD I compeletely agree with him for first gen it will be nice have t as a bonus but I dont see I they should have less power to play, if they want, insted of doing it at 480p.

I think their vision add very little to gameplay, that more things like new interfaces, interactions or context sensitive situations (to new forms of interaction it the game) will be the future of good games.

At least with this consoles (hint at REV).

 

[London Geezer]

Ok ok let me try!

If you run code optimised for a PPC, it will run like ass on an x86 class PC!!




If you try to run an Xbox game on PS2 it won't work at all!! RIDICULOUS!!

Am i the only one thinking that these days Carmack is more of a PR guy than a coder guy? As such, i'd take whatever he says with some grounded black pepper. Besides, his last "work of art" was Doom3. Enough said. I think there are some myths that should just drop dead. Metaphorically obviously, i'm not wishing anyone's death.

 

[Richard]

At first I could see why someone reading the completely butchered transcripts could misinterpret JC's keynote, but now that the full video is available I'm in awe at how so many misinterpretations continue to propagate.

I'm also a bit taken aback at the many "The almighty Carmack loves my favourite console I won't be able to purchase for x months therefor I rule" attitudes. But don't mind me...

 

[SanGreal]

I roughly transcribed his comments on procedural synthesis and multiplatform development, which I thought were interesting

There is a fallacy which has been made over and over again which is being made yet again on this console generation and that’s that procedural synthesis is going to be worth a damn. People have been making this argument forever. That this is going to be how we use all of this great CPU power, that we are going to synthesize our graphics. It just never works out that way. Over and over and over again. The strategy is bet on data rather than sophisticated calculations on there. Its won over and over again. You basically want to unleash your artists and designers more and more. You don’t want to have your programmer trying to design something in an algorithm.

If you want to do the best on all platforms, you would unfortunately probably try to program towards the Sony cell model, which is isolated worker threads that work on small little nuggets of data rather than kind of pier threads because you can take threads like that and run them on the 360. You won’t be able to get as many of them, but you can still run.. You’ve got 3 processors with 2 threads or 3 cores with 2 threads so you can go ahead and make a game that has half a dozen little worker threads that run on the cell processor there and run as just threads on the 360 and a lot of the PC specs will have HT, the processor is already twice as fast. If you just let the threads run it’ll probably work out ok on the PC, although the OS scheduling might be a little dodgy for that which might actually be something that Microsoft improves in Longhorn.

That’s kind of an unfortunate thing that that would be the best development strategy to go there because it’s lot easier to do a better job if you followed the pier thread model that you would have on the 360 but then you’re going to have pain and suffering porting to the cell. I’m not completely sure which way we’re going to go. The plan of record is that it’s going to go the Microsoft model where we have the game renderer running as two primary threads and then we have targets of opportunity for render surface optimization, physics work going on in the space processor or threads which will be amenable moving to the cell. But its not clear yet how much the hand feeding of the graphic processor on the renderer, how well we’re going to be able to move that to a cell processor. That’s probably going to be a little bit more of an issue because the graphics interface on the PS3 is a little bit more heavyweight. You’re closer to the metal on the Microsoft platform and we do expect to have a little bit lower driver overhead. People that program directly on the PC as the first target are going to have a significantly more painful time, though it’ll essentially be like porting a game like we did on the xbox in doom. Lots of pain and suffering. You take a game that is designed for 2ghz or something and try to run it on an 800mhz-ish processor you have to make a lot of changes and improvements to get it cut down like that. That is one of the real motivators for why we’re trying to move some of our development to the consoles, to sort of make those decisions earlier.

I didn't go back and fix the grammar or anything, sorry in advance.

 

[mckmas8808]

Am i the only one thinking that these days Carmack is more of a PR guy than a coder guy? As such, i'd take whatever he says with some grounded black pepper. Besides, his last "work of art" was Doom3. Enough said. I think there are some myths that should just drop dead. Metaphorically obviously, i'm not wishing anyone's death.

You feel exactly like me london. I'm sorry guys buy some of the things he is saying is just plain wrong. No ifs, ands, or buts about it.

What about a game like Motorstorm. I'm not saying that it is 100% real right now. That's not my point. My point is wouldn't physics play a huge part in the mudd since in that video. Wouldn't 1000s of pieces of mud have to be calculated in real-time to get it to stick to the side of cars, trucks, 4 wheelers, and dirt bikes the way it does?

Look at the way it streaks on the windshield when the wipers are turned on. If SCEE can get this done right the way they displayed at E3 wouldn't that put Carmack's theory about physics at a 0?

 

[CNCAddict]

WOW, I've been reading this forum for almost 2yrs and never felt compelled to write...until now. I don't think I've ever heard JC try to spin anything, but that feeling ended yesterday. How he can possibly believe physics won't have a MAJOR affect on gameplay is beyond my comprehension. I lust after games that contain fully destructable envoroments.

Imagine a pirate game where the cannon balls leave gaping holes in the sides of the ship. Water might rush in if the void is low enough, or if up high it could be a point of entry for the enemy, or maybe a "sniper" position.

I played Doom3 for about 1 hour before moving on to something more "interactive". I got WAY more enjoyment from Painkiller since the physics model is very well done. He also makes excuses about how hard it will be devoloping games like this since you can't predict every scenario. This may be true, but only if the AI is held down to JC's level.

The linear style of gameplay characterized by the Doom series is dying a rather fast death. Am I alone here, or have any of you out there played around with "Garry's mod" and seen how a pretty fun game can be had when there really is no game at all, only imagination.

 

[London Geezer]

Mckmas, Will you let Motorstorm and Killzone drop already!!

Anyway, yes it's obvious that there are many ways thousands of objects moving at the same time can make some games better, not only in the look but also in the gameplay.

He's got this thing about super-shading very few polygons, and i just don't get it. Nature, reality is made of MANY things moving at the same time. That's how we perceive living environments.

Making a few things look realistic on the surface - blocky surface i might add - is not going to win him any awards. He likes Silent Hill and Doom3 kind of scenarios but i'm sorry i'm not that limited and it seems a lot more people aren't either. Thanks god Epic's shown what can be done with some forward looking attitude this time around.

 

[mckmas8808]

Will you ever let Motorstorm and Killzone drop!!

Anyway, yes it's obvious that there are many ways thousands of objects moving at the same time can make some games better, not only in the look but also in the gameplay.

Put it this way, if SCEE comes through on these games I will never let drop. Why should I? Hundreds of indiviual pieces of mud flying at you from different angles is a great example of who physics could affect gameplay.

This guy seems to be living in a box. A closed world even. I like people who a visionaries. People who have a huge vision that's different from what we are use to and chase it. The guys at Epic seem to have this vision while keeping the overall cost realively low. Why doesn't ID?

 

[pc999]

You dont need to go futhers in physics, just put every rocket (etc...), being affected by the wing (variable force and direction, from both wind and you), you should notice a big diference in gameplay from that once that now you would need to calculate one more thing, and that would be a big plus for a more tatical gamers like me.

Or in others genre you can think a new world of puzle games (3D) tht modern CPUs could do (even desktop ones that can already do 600 moving objects).

 

[one]

He's a graphics guy. I think he'd prefer if games were always graphics focussed. But as he says himself, now there are "physics guys" coming along, raising the bar for everyone, forcing him/devs to spend more time with physics software. I get the feeling that's not what he wants to be doing.

But... but... he loves building a rocket, doesn't he!!!???

id has only developped arcade-ish games so far, I think his priority on graphics is just his personal preference for computer games. He once said games interaction should be simple, in the wake of Nintendo's Iwata. AFAICS it's the reason Doom III sucked

BTW, isn't shader technology one form of procedural synthesis?

 

[SanGreal]

I also found his comments on HDTV interesting

(again, very rough transcription)

Microsoft’s got this big push that I’m somewhat at odds with them about, about minimum framebuffer rendering resolutions on the 360. Its not completely clear how that pans out, but they essentially require all games to render at HDTV resolution. That may not be exactly the right decision, where if you’ve got the option of doing better rendering technology at less pixels with higher AA rates, that seems like a perfectly sensible thing that someone might want to do. Having a blanket though must render at 720p or something, probably not... some marketing person came up with that and decided it was needed. Which is another one of those things that I hate about the console environment, that you get some marketing person making that decision and everyone has to abide by it. Not clear yet exactly how that turns out. Obviously things like quake 4 runs good at the higher resolutions but in the next generation rendering technology there are some things like if it comes down to per-pixel depth buffered atmospherics at a lower resolution, I’d rather take that then rendering the same thing at a higher resolution. But I’ll be finding out in the next 6 months or so what I actually can extract from the hardware.

I wonder if there is a chance we'll see games in the future with more effects at SDTV resolutions. While it seems something like that would be against Microsoft's certification requirements, there have already been games announced that don't have the same experience on HDTV and SDTV. FM2006, for example, only supports split screen matches (split between the match and information -- not multiplayer) on HDTVs.

I find the comments particularly interesting coming from a PC developer, since his PC games are likely designed with higher resoltuions in mind.

 

[jvd]

, it certainly is one of the odder comments (and frankly isn't true - there's no GPU on the PC market as powerful as RSX or Xenos

Well last i check the rsx and xenos weren't on the market yet . By november at i will have thier tri card r520 on the market and by the time rsx comes out in marchish of 2006 i'm sure both ati and nvidia will have put out refreshes or modified cards , if not be reading thier new chips like the r600 and whatever replaces the g70 .

So his statement can very well be true . Of course these set ups on the pcs will cost thousands but that is the strength of the pc

Quote:

The next-gen consoles are about as powerful as current high-end pc's, but their cpu power is current PC processor

If so I in no way can believe this.

don't forget current highend cpus are now dual core . a dual core athlon 64 may very well come pretty close to a x360 cpu and a cell in realworld tasks . We really haven't seen any games specificly targeting dual core cpus

Originally Posted by mckmas8808
I couldn't disagree with him more. The guys from Epic stated that with a CPU like CELL like 15,000 pieces could be calculated at once compared to around 500 today. How can physics the way he is describing not help gameplay?

isn't the ppu add in card for pcs supposed to do 50000 pieces ?

 

[Saem]

I think Carmack's point is simple, physics and AI will improve a fair bit, but even with that, games will be, for a long while, dominated by graphics, because ultimately, that's the vast majority of how the user receives their sensory input, with sound being a distance second -- sound just doesn't get the same amount of love.

Moreover, some of the research Halo programmers were doing, they found that the user didn't really clue into the AI unless the action was exaggerated or very obvious.

Improved physics will be neat, but that won't be worth a damn unless the rendering can clearly represent it.

Lastly, don't forget, light and shadow are probably the most powerful mood setters, Carmack's lighting obsession stems from movie making wisdom.

 

[mckmas8808]

Well then if that the case he can take his light and shadows with great graphics and let the game animate bad.

 

[Colourless]

I think Carmacks point about Physics is this. If you want to do it 'all' you had better do it all 'correctly' and that is going to take more power than these consoles will have. Tumbling boxes is one thing, completely destructable environments where everything acts correctly is a huge huge problem

 

[blakjedi]

I dunno... high tech does not a game make... unless you can represent a liquid correctly graphically does it matter that its physics are accurate? i dunno, in many ways we accept what we see or not and keep playing or not.


A game like wave race would improve dramatically with realistic moving water. However the best water I remember are the sewer levels of Batman Returns CD on Sega CD and the first Nintendo waverace titles... how many "advanced physics" titles on water have come out since??

 

[chachi]

Anyway, yes it's obvious that there are many ways thousands of objects moving at the same time can make some games better, not only in the look but also in the gameplay.

Bolded for emphasis, accurate physics might not make the gaming experience better for the user. For example, in a certain offroad racing game that someone likes to talk about so much, would the game be more enjoyable if there was so much mud kicked up on your windshield that the wipers couldn't clear it fast enough and you had to try to see through the mess on your screen? How about if you had to drive at a realistic speed or go over the side of the mountain and die? No. But it'd be more realistic. Huzzah. What Carmack is saying is - where's the greatest payoff for our time and resources, in something the user can't see and probably won't appreciate much, or in the graphics that everybody cares so much about?

AI is important of course, but it's a delicate balance because you don't want to make the game too smart or users will get frustrated. Realistically if you had 20 people trying to kill you in a maze and they were coordinated you would die every time (or near enough it wouldn't matter). People don't want brilliant AI, they want challenge but 9/10 they'd probably rather see someone (else) get blown up impressively than sneak up behind you and slit your throat. Game over...? Whuh? Also to be considered of course is that you could spend a lot of resources making the game opponents smart and independent but if the graphics suffered then all you'd hear is "wallguy" type comments and you'd lose your shirt. Where would *you* spend the resources if it was your money?

One thing about processor performance, don't make the mistake and think because his primary platform has been the PC that he's talking x86, he's talking generic C code. The thing is that the PC processors have evolved to run that unoptimized code very well, to get the same performance from the XCPU and Cell you're going to have to spend time profiling and optimizing the code. It's not like the x86 processors don't benefit from loop unrolling and the like as well, it's just they don't usually need it. That's inconvenient but nothing that can't be worked around given enough time and money. The hype for these processors always talks about the cumulative theoretical performance and how much more that is than the competition but in a non-SIMD single-threaded task they'll probably get spanked pretty badly by a top of the line x86 processor. It's up to the developers to work around that.

I think Carmack is probably more correct on this than a lot of people want to believe.

 

[pegisys]

I agree with chachi, I don't think he was trying to knock the consoles or anything like that, he was just bringing people back down to earth abit, expectations are to high, the new consoles are going to be great but they are not going to be the end all solution

What about a game like Motorstorm. I'm not saying that it is 100% real right now. That's not my point. My point is wouldn't physics play a huge part in the mudd since in that video. Wouldn't 1000s of pieces of mud have to be calculated in real-time to get it to stick to the side of cars, trucks, 4 wheelers, and dirt bikes the way it does?

he spoke on that as well, thats just like the grass swaying in the wind, or more rock rolling down a hill, yeah it looks good but it doesn't really add anything to game play, even if the mud wasn't there the game would be the same

and about the todays high end pc having simular performance I kinda agree if you are talking about dual athlon fx-57's and 2 7800gxt's, and since in order code will probably run better on an out of order cpu, then out of order code will run on a in order cpu, and the fact that more memory can be more inportant than cpu power in some cases

 

[jvd]

You guys hae to remember that graphics will allways be ahead simply because it has hardware dedicated to it .

The cpu has to do physics , ai , graphics and other things . So physics and a.i will allways be behind .

Perhaps in the next generation the one after this upcoming one . The consoles will be designed with physics acceleration (if it takes off on the pc side). Deviding these tasks up are the only ways to jump foward