34 Replies. 2 pages. Viewing page 1.
< Newer [ 1 2 ] Older >
 |
| 34. |
Re: No subject |
Dec 20, 2005, 08:25 |
Bill Nakano |
|
|
How does more natural hair and/or cloth improve GAMEPLAY?
More like something else to ramp up production budgets so that only EA can afford to make games. WHere's HAVOK stand with all this?
|
|
|
|
|
|
|
|
| |
|
| 33. |
Re: No subject |
Dec 19, 2005, 23:43 |
The Truth |
|
|
you come back to the problem of having to support both those that have a PPU and those that don't, and physics isn't really like rendering, you can't really have the equivalent of LOD,
Actually you can. Things like cloth simulation, hair, grass, leaves etc, can all be switched on and off (An immediate example that springs to mind is EQ@, where you can turn cloth simulation on and off). When its off it doesent really look that different too normal games, but with it on it enhances the game.
The thing is running proper physics effects on the game isnt really noticable until you then remove it, its a natural progression, and its only when you go back to older games that you see what its like without being able to interact with the environment and without being able to watch it naturally.
Watching that fluid demo and the other two was pretty impressive, its a shame they dont link them together, to show the plane crashing in, hitting the ball, watching it knock over the crane and then smashing the water thing open. imagine being on the ground shooting at baddies and then that happens, youve only got a few seconds to get to the door, but damn it would be cool.
Heres the thing, if it would improve the game play, via more natural hair, cloth etc then ill buy one.
MMX you say isnt used, yet all chips now support it, so even though at the time of introduction there wasent any mass surge toward it, im pretty sure most titles now use it, you may not notice it, but then its just one more thing that layers onto the game, allowing us to play near real games as opposed to Nada 3.
just my 0.02 cents.
Seriously guys lighten up on the negativity.
------
$1 tax for the national debt? |
|
|
|
|
|
|
|
|
------
Diablo & Diablo 2 for the DS, it makes sense Blizzard! |
|
|
|
|
| |
|
| 32. |
Re: No subject |
Dec 19, 2005, 23:43 |
Semantics |
|
|
don't you want to upgrade your $450 physics card twice a year?
|
|
|
|
|
|
|
|
| |
|
| 31. |
No subject |
Dec 19, 2005, 23:25 |
Masa |
|
|
Adding a physics card is pretty neat sounding, but does it have to be an entire card? Can't they intergrate this thing into the mobo at some point?
|
|
|
|
|
|
|
|
| |
|
| 30. |
Re: No subject |
Dec 19, 2005, 22:48 |
MMORPGHoD |
|
|
Which came first, the physics enhanced chicken software or the physics card egg?
The first steps need to be taken regardless and people need to see what a dedicated processor core for physics can give you. I think current client/server game solutions painfully show the current limitations of client calculations for physics. After a couple revisions, this could be the answer. Let's face it, we need something that's going to raise the bar for gaming and break the same old cookie cutter game sequels. I just paid $450 for a video card that's probably a few times better than my 9800. Count me in on being a $250 physics card guniea pig.
|
|
|
|
|
|
|
|
| |
|
| 29. |
Re: No subject |
Dec 19, 2005, 22:44 |
Rilcon |
|
|
I remember a good many games back in the day that required a 3DFX card to run, back when Voodoo 1 cards were just being introduced. And man, that card was well worth it.
|
|
|
|
|
|
|
|
| |
|
| 28. |
No subject |
Dec 19, 2005, 22:18 |
Tracer |
|
|
We won't see the Phys-X PPU used to its full potential until games start REQUIRING a PPU in order to run. It's the same way it was with video cards back in the day.
|
|
|
|
|
|
|
|
| |
|
| 27. |
Re: No subject |
Dec 19, 2005, 22:15 |
Speckled Jim |
|
|
There are a few problems with multiplayer as I see it. Fine, you can have a PPU enabled server, and then everyone gets the benefit of the heavy lifting done server side, but that does mean all the client side effects are inconsequential, because they have to be, or the game is not in sync.
Given that people tend to turn off high end detail in online games to improve their frame rate, there is really only justification in the servers being PPU enabled.
Now for single player, you come back to the problem of having to support both those that have a PPU and those that don't, and physics isn't really like rendering, you can't really have the equivalent of LOD, you have to run it all or you end up with weird stuff like boxes that float in mid air until the player comes in range, and then they suddenly fall to the floor, etc.
So given the need to run physics constantly and support those without a PPU, how can you have a compelling PPU demonstration game. It's either just fluff (loads of crap to knock over, that you leave out if no PPU is there), or it runs like a dog if you have a load of really useful physics stuff, but no PPU.
Seems like in the end you'd really have to produce a game that was PPU only, if you really wanted it to show its stuff.
I think I've worn out my P and U keys.
This comment was edited on Dec 19, 22:16. |
|
|
|
|
|
|
|
| |
|
| 26. |
Re: No subject |
Dec 19, 2005, 21:40 |
SquirrelZero |
|
|
In a server/client situation only the server would benefit with important game objects like boxes and barrels, things that need to be the same on the client and server. There would be a significant server benefit, though -- in our games, the server basically handles all the physics simulation and only replicates the location and rotation of the object to the client, while the client interpolates between each update. The main function of the PhysX processor on the client would be for things like fluid dynamics, particle effects, and cloth simulation, which we'll be using a lot of, and aren't done on the server. Those with PhysX will get much nicer effects (and speedier, too) and those without out it will get standard, toned-down effects. In the client's case it's mainly for visual effect: faster and more accurate ragdoll simulation, cooler-looking blood thanks to fluid dynamics, realistic water effects, realistic fire, smoke, etc.
|
|
|
|
|
|
|
|
| |
|
| 25. |
No subject |
Dec 19, 2005, 21:25 |
sponge |
|
|
None of these threads seem to bring up one huge problem.
What do you do about multiplayer? PhysiX card only servers? Requiring the card to play multiplayer? Dumb it down so normal CPUs can handle it? You can't give players two different physics models, so something has to be done. I have a feeling it'll be falling back to simple physics in multiplayer, putting a huge stake in the heart of the product.
It's very cool, however some of the problems come time for implementation bother me.
|
|
|
|
|
|
|
|
| |
|
| 24. |
Re: No subject |
Dec 19, 2005, 20:28 |
Halsy |
|
|
I have no doubt it's porbably a cool little gadget, but I remember when Intel came out with MMX and how everyone was supposed to jump on board with that - and never did. And countless other cool next-gen hardware and software that was long on promise of everyone utilizing it and short on actualy delivery. Most any of the games that ever came out supporting such things sucked ass, and were little more than glorified tech demos.
Tom has the right of it, show me some GOOD titles being made, and I don't mean just 1 or 2, but several dozen - that would justify the cost of this thing. The videos you linked didn't really impress me at all.
"Many defendents get religion when they want mercy from the court. When you see me coming out for Jesus, you'll know they really have something on me."
- HST |
|
|
|
|
|
|
|
|
"And then, suddenly and without warning, it turned into a real-life case of hungry, hungry hippos."
- Stephen Colbert |
|
|
|
|
| |
|
| 23. |
Re: No subject |
Dec 19, 2005, 19:09 |
SquirrelZero |
|
|
Why would anyone want to buy a physics add-in card when we have a future of upcoming duel core up to potentially quad core cpus that can be useful for much more than just physics, and act as dedicated physics processor in games. Sure maybe the proprietary card is faster, but it's also too specific for my tastes.
This is a very common argument, but one that isn't necessarily valid. For example, in high speed current-gen processors (the top-end 4ghz/4000+ ones) you'll max out at only a couple hundred rigid bodies and joints before the physics processing is consuming too much CPU and causing performance issues. With a PhysX processor, you can have up to about 6000 rigid bodies and joints at once, all active, with very little CPU usage, and those are prototype specs which I expect to see increasing when driver builds are more stable. To be able to adequately handle 6000 rigid bodies/joints, standard CPU processors would most likely need to be 10 times as powerful as they are today. That's at least a decade away.
People used to wonder what the need for a 3D accelerator would be. With processor speeds reaching 200mhz and more back at their inception, they thought the CPU would be able to handle everything. Today we know better. No amount of CPU would be able to do what a 3D graphics card does.
The best argument against them, I think, is that you won't really NEED one. Like you said, they're very specific in what they can do, and no game will be made (at least right away) that absolutely requires one. However, people have already been finding interesting applications for the PPU, like speeding up math operations in 3D graphics programs and other non-standard uses. I for one am very excited to see what else people can do with these things.
As for cost, I'd rather not speculate. It's not really my place.
|
|
|
|
|
|
|
|
| |
|
| 22. |
Re: No subject |
Dec 19, 2005, 19:09 |
Tom |
|
|
DrEvil: Read your post again, but imagine that you're back in the 90's and talking about a 3D accelerator add-in card instead of a physics card. Then fast forward a little bit and think about Quake vs. GLQuake. Then fast forward to today and consider what the market is like for 3D accelerators. See why these guys are making the attempt?
And sure, multi-core CPUs could help with this stuff, but in the end no matter how many cores you have, they're still general-purpose. Ya, they have some instruction sets like SSE and 3DNow that can help a lot, but they're still general-purpose. Special-purpose hardware can be extremely potent.
The real problem with these things is something you touched on: making games support them in a way that is cool enough to make people plunk down the cash and buy them, but of course the game could not REQUIRE it. Just as no 3D game in the early days of 3D acceleration could REQUIRE a 3Dfx card. What they need are dramatic examples of things that simply can't be done with general-purpose hardware. And so far, that fluid dynamics video is the only thing I've seen that comes close to being such an example. And even it is quite far from being a real game showing a real benefit from the technology.
Personally I'd like to see these things get incorporated into graphics cards or motherboards as a $5-10 chip. But if it takes a generation or two of standalone cards to get there, I think they have a shot. Maybe an outside shot.
|
|
|
|
|
|
|
|
| |
|
| 21. |
Re: No subject |
Dec 19, 2005, 18:51 |
DrEvil |
|
|
Why would anyone want to buy a physics add-in card when we have a future of upcoming duel core up to potentially quad core cpus that can be useful for much more than just physics, and act as dedicated physics processor in games. Sure maybe the proprietary card is faster, but it's also too specific for my tastes.
We'll see how it goes I guess. I could maybe go for a $50 max physics addin card later on when support is higher.
Anyone know what these AGEIA cards are rumored to cost initially?
Edit: Problem with these things is IMO it's going to be hard to penetrate the market with them. Nobody in their right mind would make a game that absolutely requires them, so they will likely always have to use the power on something that those without the accelerator card can do without, probably something not terribly gameplay effecting.
This comment was edited on Dec 19, 18:53. |
|
|
|
|
|
|
|
| |
|
| 20. |
Re: MOM |
Dec 19, 2005, 18:46 |
Tom |
|
|
Thanks for the Airtight links. Now I've finally seen something that looks like it could be worthy of a dedicated physics processor: the fluid dynamics simulation. That actually looked pretty good. The rest of the Airtight videos were ok, but really I don't see why they would need a dedicated processor. And those Ageia videos are just plain awful. Good thing they got Airtight to start making good looking videos. 
I still won't be convinced enough to buy a dedicated card with one of these until I see videos of a real game with and without the PhysX being used...
|
|
|
|
|
|
|
|
| |
|
| 19. |
Re: No subject |
Dec 19, 2005, 18:42 |
Bhruic |
|
|
Man, every time I hear the term 'Action RPG', I want to puke.
|
|
|
|
|
|
|
|
| |
|
| 18. |
Re: No subject |
Dec 19, 2005, 17:25 |
Speckled Jim |
|
|
Regardless of whether it gets much support at first, I'd be very unlikely to pick up a first generation card. It's likely to be rather underwhelming.
As for those people thinking they are going to be getting totally destructible environments, dream on. Can't see that happening any time soon.
I'm rather inclined to think you'll just be seeing a lot of extra unnecessary piles of boxes to be knocked over, or a lot of extra wavy grass for those that have the PPU.
|
|
|
|
|
|
|
|
| |
|
| 17. |
Re: No subject |
Dec 19, 2005, 16:41 |
SquirrelZero |
|
|
But it already is supported. Unreal Engine 3 has full hardware PhysX support, and numerous upcoming titles use UE3. That's only one game engine. I know of about 2 dozen Reality Engine titles that are also adapting it. This is excellent support, for example, in comparison to the amount of support the first 3D accelerator had.
|
|
|
|
|
|
|
|
| |
|
| 16. |
Re: No subject |
Dec 19, 2005, 16:40 |
nin |
|
|
I'd rather see the major video card manufacturers just add a physics unit to their video cards. That way you won't have wasted money on a piece of hardware that may or may not ever be supported.
Good lord! Doesn't your video card run hot enough? Shit, we'll have to water cool them...
--------------------------------------------------------------
GW: Nilaar Madalla, lvl 20 R/Mo / Tolyl Nor, lvl 20 E/Mo / Xylos Gath, lvl 13 W/Mo
http://www.goldfrapp.co.uk/ |
|
|
|
|
|
|
|
|
RollinThundr Apr 17, 2013, 12:25: Eh really tossing stuff like that in there only to get your panties all bunched up. If you really want to call that trolling sure.
Mr. Tact Apr 17, 2013, 12:33: Pretty sure that's the definition of trolling... |
|
|
|
|
| |
|
| 15. |
No subject |
Dec 19, 2005, 16:35 |
nightfend |
|
|
I'd rather see the major video card manufacturers just add a physics unit to their video cards. That way you won't have wasted money on a piece of hardware that may or may not ever be supported.
|
|
|
|
|
|
|
|
| |
34 Replies. 2 pages. Viewing page 1.
< Newer [ 1 2 ] Older >
|
|
Twitter 
