Which physics engine can do a holodeck?

Status
Not open for further replies.

konst

Weaksauce
Joined
Dec 4, 2006
Messages
83
ok seriously ... my question is which cards have sdks that can do real physics and mathematical calculations since these new cards are supercomputer class.
 
Joined
Oct 26, 2005
Messages
2,340
konst said:
ok seriously ... my question is which cards have sdks that can do real physics and mathematical calculations since these new cards are supercomputer class.
Although the definition of a supercomputer is vague, it certainly doesn't include $200 consumer hardware.

You'll have to elaborate a bit on "real physics". If you were semi-serious about the holodeck thing, and you're talking about an accurate real-time simulation of the real world, then no, a PhysX can't pull it off. Not even a $300 million supercomputer comes close. If you're talking about physics calculations "real" enough for scientific/engineering applications, then the hardware is certainly capable (though how fast it can do it is another question entirely), but the software definitely isn't written with that in mind - games just don't require that kind of accuracy. I'd say you'd have to code it yourself.
 

konst

Weaksauce
Joined
Dec 4, 2006
Messages
83
LuminaryJanitor said:
Although the definition of a supercomputer is vague, it certainly doesn't include $200 consumer hardware.

You'll have to elaborate a bit on "real physics". If you were semi-serious about the holodeck thing, and you're talking about an accurate real-time simulation of the real world, then no, a PhysX can't pull it off. Not even a $300 million supercomputer comes close. If you're talking about physics calculations "real" enough for scientific/engineering applications, then the hardware is certainly capable (though how fast it can do it is another question entirely), but the software definitely isn't written with that in mind - games just don't require that kind of accuracy. I'd say you'd have to code it yourself.

the 8800GTX can do 500 GFLOPS. That certainly would qualify as a supercomputer a few years ago.
 

kent

2[H]4U
Joined
Apr 8, 2001
Messages
2,537
When the 1GHz Athlon (SlotA) debuted, MaximumPC said "This thing is a freakin' super computer!"
 

drizzt81

[H]F Junkie
Joined
Jan 21, 2004
Messages
12,361
I think the most difficult part about the Holodeck is generating the objects. I understand that theoretically we can convert energy into mass, but I somehow think that doing this both forward and reverse while people are interacting with it, is rather difficult.

In short: No physics card can do a holodeck.
 

konst

Weaksauce
Joined
Dec 4, 2006
Messages
83
drizzt81 said:
I think the most difficult part about the Holodeck is generating the objects. I understand that theoretically we can convert energy into mass, but I somehow think that doing this both forward and reverse while people are interacting with it, is rather difficult.

In short: No physics card can do a holodeck.

Actually a holodeck is supposed to work by interaacting magnetic fields. Considering that all the interaction with object in our lifes are through the electromagnetic field the holodeck idea isn't too far fetched.
 

pigpen

2[H]4U
Joined
Aug 17, 2004
Messages
2,408
I've been waiting patiently for 20 years for Wonka-vision. And i'm still wondering when these scientists are going to get off their collective ass and make me small enough to fit in a purse.
 

SuperGee

Gawd
Joined
Jun 3, 2006
Messages
551
konst said:
ok seriously ... my question is which cards have sdks that can do real physics and mathematical calculations since these new cards are supercomputer class.

There isn't hardware out with that computational power and the technology gaps to make a interactive virtual world where you are in it interacting.

So no API suport that because why, Because there so much more detail in computational formulas wich would put hardware out now to a crawl of the overhead.
API are developt with current hardware in mind.
API for PPU and GPU will support features do-able with that hardware.
Each next generation the posibility grows and with that some new feature become possible and API are start to support it.

So I don't expect a API to look that far in the future.
 

konst

Weaksauce
Joined
Dec 4, 2006
Messages
83
SuperGee said:
There isn't hardware out with that computational power and the technology gaps to make a interactive virtual world where you are in it interacting.

So no API suport that because why, Because there so much more detail in computational formulas wich would put hardware out now to a crawl of the overhead.
API are developt with current hardware in mind.
API for PPU and GPU will support features do-able with that hardware.
Each next generation the posibility grows and with that some new feature become possible and API are start to support it.

So I don't expect a API to look that far in the future.

I was just kidding about the holodeck but seriously API and current programming models have trouble keeping up with even our current level of research like global warming simulations, QCD, String Theory, ...

API's wont work in the future. they'll be as antiquated as a bicycle is to the space shuttle
 

drizzt81

[H]F Junkie
Joined
Jan 21, 2004
Messages
12,361
konst said:
I was just kidding about the holodeck but seriously API and current programming models have trouble keeping up with even our current level of research like global warming simulations, QCD, String Theory, ...

API's wont work in the future. they'll be as antiquated as a bicycle is to the space shuttle
You are aware that many API are designed to give "application programmers" a black-box experience. The models and details of the execution may change, but the interface will stay the same.

Yes, APIs age and get replaced (DX10) but that doesn't make the "antiquated". As we all know time is limited, so we need to settle for evolutionary changes to simulation models, rather than suddenly having it all.

Of course the current GAMING physics simulation is significantly behind the physics that is being researched in universities. There are many reasons for this, one of them being that it is pretty much impossible to do real-time physics simulation of anything that has a "reasonable" size. Look at the Folding@Home presentations, they are hoping that soon enough they will be able to simulate microseconds. If you think about what a large project it is and that they are simulating tiny pieces of matter, I really do not understand why you are complaining about the inaccuracy of PPUs.

They are made for gaming and if games get closer to approximating newtonian physics, I will be a happy camper.

see slides:
http://hardforum.com/showthread.php?t=1104427&highlight=presentation
 

este

Limp Gawd
Joined
Oct 10, 2006
Messages
233
I can't believe anyone would seriously reply to the op.

I feel like s loser for being on [h] today.

ps> drizzt81 is right, you don't seem to know what an API is.
 

jimmyb

2[H]4U
Joined
May 24, 2006
Messages
3,165
konst said:
Considering that all the interaction with object in our lifes are through the electromagnetic field the holodeck idea isn't too far fetched.
Unless you've come up with a unified field theory, and consolidated everything under electromagnetism (which would be an odd choice seeing as how it already means something), this is certainly not true

@drizzt81 - Didn't you know abstraction is on its way out? In the future progammers will have to understand everything they work with completely.
 

Brent_Justice

Moderator
Joined
Apr 17, 2000
Messages
17,755
The holodeck system on Voyager uses molecule size magnetic "bubbles" that the computer can operate on in any dimension. It can literally make up any form it needs to. Don't forget that the transporters and replicators are also tied into the systems, it is based on the same principals. For example when you create a meal on the holodeck it is a real replicated meal that you can eat and it stays with you when you leave the holodeck. More detail on how they work is I believe explained in the episode "The Phage".

Voyager used the Federation's latest holography technology at the time of its creation, superior to what the Enterprise NCC-1701D had, though I'm not sure how sophisticated the 1701E is in terms of holography, but I would venture to guess just as updated as Voyager. I am sure by the time Voyager got home 7 years later the technology has improved even further. We do know holograms such as the Mark I are used to scrub warp plasma conduits and mine for dilithium in space, I am sure they have created other specific holograms for specific tasks.
 

Martyr

2[H]4U
Joined
Jan 6, 2006
Messages
3,653
holodeck has 3 important phases of matter.

first is purely holographic. this what you see, backgrounds, far off places, sky, etc etc.

second is holomatter. this is that magnetic such and such. controlled by forcefields and imaged by holography, this is anything you touch, interact with, etc.

third is real matter. this is created on the fly by the replicators and beamed in place of the holomatter, should one decide to do something like eat it.
 

konst

Weaksauce
Joined
Dec 4, 2006
Messages
83
este said:
I can't believe anyone would seriously reply to the op.

I feel like s loser for being on [h] today.

ps> drizzt81 is right, you don't seem to know what an API is.

I know what an API is. I was just saying programming in the future will hardly be recognizable by todays standards.
 

nhusby

Limp Gawd
Joined
May 1, 2006
Messages
303
konst said:
ok seriously ... my question is which cards have sdks that can do real physics and mathematical calculations since these new cards are supercomputer class.

I've heard that if chuck norris licks your nVidia 8800 it can do a holodeck...
 
Joined
Oct 26, 2005
Messages
2,340
konst said:
the 8800GTX can do 500 GFLOPS. That certainly would qualify as a supercomputer a few years ago.
My TI-83 calculator would've qualified as a supercomputer some time in the past. That certainly doesn't make it one today. "Supercomputer" simply means one of the fastest computers in the world, and no piece of comsumer hardware will ever qualify.

konst said:
API and current programming models have trouble keeping up with even our current level of research like global warming simulations, QCD, String Theory...
I don't see what APIs have to do with this at all. As far as I'm aware, modelling weather systems or quantum interactions involves nothing more than obscene amounts of number crunching. You don't even need API calls for this. The only thing which could possibly be "having trouble keeping up" is raw processing power.

What exactly are you expecting to come along and replace APIs?
 

G'ßöö

[H]ard|Gawd
Joined
Aug 11, 2004
Messages
1,456
Which physics engine can do a holodeck? if you knew anything, you'd know there is no ppu needed for a holodeck
 

HOCP4ME

2[H]4U
Joined
Jul 1, 2005
Messages
2,959
On the subject of holodeck:

It can do four things:
-Synthesize a real object out of molecular "packets". This object could be carried outside of the holodeck.
-Arrange the molecular pakcets into the shape of an object and hold them together by electromagnetic fields. The object would fall apart when taken out of the holodeck.
-Use magnetic fields alone to create the illusion that the object exists when it really doesn't.
-Project the image of an object that doesn't need to be interacted with but still needs to be seen.

The holodeck automatically chooses the option that would work best in each given situation.

The real breakthrough will come, however, when we really can convert energy to any type of matter at will. Power such a device with a source of energy such as a fusion reactor, and we would have all the oil, natural gass, raw materials, food, and water we would ever need.
 

Glow

Supreme [H]ardness
Joined
Jul 20, 2002
Messages
4,834
If hardware mods and OCing was starting to be cool, it all just got sent back a few steps in this thread.
 

SuperGee

Gawd
Joined
Jun 3, 2006
Messages
551
G'ßöö said:
Which physics engine can do a holodeck? if you knew anything, you'd know there is no ppu needed for a holodeck
I think one PPU would be a water drop on a very hot plate.
The Physics power needed would sucseed the most powerfull supercomputer today.
Plus also for rendering there is a lot of computational power needed.

And some break trou technology to replace monitor keybord mouse User interaction
and making the simulation real in a specifick room to be able to interact with.
That a big gap because would that be even possible.

Every next gen games can use more Physics and dedicated hardware will give that a boost.
 

MartinX

One Hour Martinizing While You Wait
Joined
Jan 23, 2003
Messages
7,187
pigpen said:
I've been waiting patiently for 20 years for Wonka-vision. And i'm still wondering when these scientists are going to get off their collective ass and make me small enough to fit in a purse.

I suspect that developing the technology to create very large purses would be more cost effective.
 

Rabid Badger

2[H]4U
Joined
Mar 9, 2006
Messages
2,093
HOCP4ME said:
The real breakthrough will come, however, when we really can convert energy to any type of matter at will. Power such a device with a source of energy such as a fusion reactor, and we would have all the oil, natural gass, raw materials, food, and water we would ever need.

The Universal Constructor in Deus Ex needed 4 fusion reactors to power it. ;)

Also, if you have a fusion reactor, why would you need to sythesize fossil fuels?
 

HOCP4ME

2[H]4U
Joined
Jul 1, 2005
Messages
2,959
Rabid Badger said:
The Universal Constructor in Deus Ex needed 4 fusion reactors to power it. ;)

Also, if you have a fusion reactor, why would you need to sythesize fossil fuels?

Realistically, by the time a universal constructor is invented, a hydrogen infrastructure will already be in place, so you wouldn't need fossil fuels.

But if the device came as a sudden breakthrough in about fifty years (won't happen, I know), it would be used to temporarily resolve the oil crisis while a hydrogen infrastructure was being built.

Hey, I think this topic has regained it's "cool factor". Too bad it's not about hardware and overclocking anymore.
 
Joined
Jun 24, 2005
Messages
2,993
konst said:
the 8800GTX can do 500 GFLOPS. That certainly would qualify as a supercomputer a few years ago.

Yeah, I suppose so. This whole notion of "Look, it's as fast a a Supercomputer! (circa 1985)!" is pretty ridiculous, IMHO. Is there *any* such card/CPU/PC that can qualify as a Supercomputer by TODAY'S standards? No, there isn't. Only a (get this) SUPERCOMPUTER today qualifies.

Thank you, Sony, for this irrelevant, useless non-metric you've fooled people into believing is legitimate.
 

FLECOM

Modder(ator) & [H]ardest Folder Evar
Staff member
Joined
Jun 27, 2001
Messages
15,696
it dosent matter, no computer has at least a laser 9 output... laser 7s will only get you float screens... you need laser 9 for full out holodecks... the playstation 69 will support the laser 9, but you need a laser 9 cable, which will set you back a few grand...
 

ThreeDee

[H]F Junkie
Joined
Sep 5, 2001
Messages
10,943
FLECOM said:
it dosent matter, no computer has at least a laser 9 output... laser 7s will only get you float screens... you need laser 9 for full out holodecks... the playstation 69 will support the laser 9, but you need a laser 9 cable, which will set you back a few grand...
yeah ..that's all fine and dandy ..but the real question is .."Will we be able to fold on it?!"

:)

[F]old|[H]ard
 

twoeyes

Limp Gawd
Joined
Aug 15, 2006
Messages
243
HOCP4ME said:
Realistically, by the time a universal constructor is invented, a hydrogen infrastructure will already be in place, so you wouldn't need fossil fuels.

But if the device came as a sudden breakthrough in about fifty years (won't happen, I know), it would be used to temporarily resolve the oil crisis while a hydrogen infrastructure was being built.

Hey, I think this topic has regained it's "cool factor". Too bad it's not about hardware and overclocking anymore.


Yeah and all we need is more fossil fuels polluting the atmosphere. How about use the thing to create super cheap and CLEAN running cars by scrapping the old gas automobiles? We have enough gas as it is anyway its just the oil companies keeping prices high and demand high by stifling research, bring those prices down and then you'll loose support of the consumer for such a hydrogen infrastructure (which they will probably do as a last ditch effort while they switch to alternative fuel productions with their saved capitol), then the whole thing just wastes.
 
Status
Not open for further replies.
Top