Sign in with
Sign up | Sign in
Your question

Physics card price

Last response: in Video Games
Share
May 22, 2005 10:50:44 AM

Archived from groups: comp.sys.ibm.pc.games.action (More info?)

http://www.xbitlabs.com/news/multimedia/display/2005052...

"The cards based on the new AGEIA PhysX processor will start sampling
in Q3 2005, and when they appear in retail in Q4 their price is
supposed to be between $249 and $299"

They are really going to fly off the shelves at that price aren't
they!
--
Andrew, contact via interpleb.blogspot.com
Help make Usenet a better place: English is read downwards,
please don't top post. Trim replies to quote only relevant text.
Check groups.google.com before asking an obvious question.

More about : physics card price

May 22, 2005 11:07:59 AM

Archived from groups: comp.sys.ibm.pc.games.action (More info?)

I really wonder at the use of these, in particular with the advent of
dual-core cpus.

If you can make a system call and have the physics off-loaded to this card,
you could just as easily have the system call look for the presence of a 2nd
cpu and hand off the task as another thread.

rms
Anonymous
May 22, 2005 4:25:27 PM

Archived from groups: comp.sys.ibm.pc.games.action (More info?)

On Sun, 22 May 2005 06:50:44 GMT, Andrew <spamtrap@localhost.> wrote:

>http://www.xbitlabs.com/news/multimedia/display/2005052...
>
>"The cards based on the new AGEIA PhysX processor will start sampling
>in Q3 2005, and when they appear in retail in Q4 their price is
>supposed to be between $249 and $299"
>
>They are really going to fly off the shelves at that price aren't
>they!

More to the point, if I went out and got one today what games are actually
going to work with it?
Related resources
May 22, 2005 4:32:11 PM

Archived from groups: comp.sys.ibm.pc.games.action (More info?)

On Sun, 22 May 2005 12:25:27 GMT, Memnoch
<memnoch@nospampleaseimbritish.ntlworld.com> wrote:

>More to the point, if I went out and got one today what games are actually
>going to work with it?

There are a few games penciled in for supporting it, but I think the
key issue is that no game (for the forseable future) will require it
so it will only add a bit of fluff to a game, not affect the core
gameplay. I really can't see a point of such a card, it stands no
chance of becoming mainstream, especially at that price point.
--
Andrew, contact via interpleb.blogspot.com
Help make Usenet a better place: English is read downwards,
please don't top post. Trim replies to quote only relevant text.
Check groups.google.com before asking an obvious question.
Anonymous
May 22, 2005 7:11:36 PM

Archived from groups: comp.sys.ibm.pc.games.action (More info?)

"rms" <rsquires@flashREMOVE.net> wrote in message
news:jdWje.3838$VS6.1255@newssvr19.news.prodigy.com...
>I really wonder at the use of these, in particular with the advent of
>dual-core cpus.
>
> If you can make a system call and have the physics off-loaded to this
> card, you could just as easily have the system call look for the presence
> of a 2nd cpu and hand off the task as another thread.
>
> rms
>

I could see graphics cards possibly adding a physics CPU (a PPU?). I can't
imagine one physics processor being that much more advantageous over
another, like with graphics. Other than how many particles it can calculate,
I guess. But we will see how it pans out. I thought "3D" graphics cards were
just a gimmick. Until I played Descent 2 with my first Monster 3D card....
drool.

However, I can see how not only does a gamer have to worry about a
motherboard, CPU, and graphics card, but what physics card to get too.
yikes. So just when you think things are getting better with onboard LAN,
USB, and sound, you are now filling up your PCI(e) slots with more stuff.

Unless someone comes up with a way to calculate physics of particles that
revolutionizes the world, all I can see it needing is pure horsepower which
will determine how many particles can be calculated.
Anonymous
May 22, 2005 7:58:37 PM

Archived from groups: comp.sys.ibm.pc.games.action (More info?)

On Sun, 22 May 2005 06:50:44 GMT, Andrew <spamtrap@localhost.> wrote:

>http://www.xbitlabs.com/news/multimedia/display/2005052...
>
>"The cards based on the new AGEIA PhysX processor will start sampling
>in Q3 2005, and when they appear in retail in Q4 their price is
>supposed to be between $249 and $299"
>
>They are really going to fly off the shelves at that price aren't
>they!
>--

DEAD-DUCK....

Developers will use the processing power of the dual-core CPU,
not program for a minimal-sales peripheral. Anyway, where are
the extra slots in a PCIe MB not potentially taken up by avid
extreme gamers ( the only people that might buy this card)
with extra graphics cards (PCIe) and/or audio/media processing
cards (PCI) ? The dual-core CPU takes up no extra space !!

John Lewis

>Andrew, contact via interpleb.blogspot.com
>Help make Usenet a better place: English is read downwards,
>please don't top post. Trim replies to quote only relevant text.
>Check groups.google.com before asking an obvious question.
Anonymous
May 23, 2005 2:45:12 AM

Archived from groups: comp.sys.ibm.pc.games.action (More info?)

x-no-archive: yes

what is the biggest complain pc gamers have with the obvious
exception of steam?
its high cost in pc games

what is one of the biggest challenges in pc games besides
fighting steam piracy and consoles?
controlling the escalation of prices in pc games

we can't let pc games become a elitist expensive for a minority
type of entertainment
pc games must be accessible, affordable and democratic

so its obvious my answer is NO, a big NO to the introduction
of proprietary specific dedicated hardware to deal with physics
processing in pc games
physics processing must be done by the CPU like it has been done
since the beginning
if CPU are not enough powerfull to deal with complex physics
then its simple... we must wait until they are

--
post made in a steam-free computer
i said "NO" to valve and steam
Anonymous
May 23, 2005 3:51:57 AM

Archived from groups: comp.sys.ibm.pc.games.action (More info?)

What about latency? Sending physics parameters across the PCIe bus then
getting the results back is going to be fantastically slow relative to
having a dual core system with dual FPU's. I predict this is going to be a
stillborn tech.

--
Remove nospam to email
"Andrew" <spamtrap@localhost.> wrote in message
news:7qa091lhsg4fkrqrrrogmaaj2c7k4q2e16@4ax.com...
> http://www.xbitlabs.com/news/multimedia/display/2005052...
>
> "The cards based on the new AGEIA PhysX processor will start sampling
> in Q3 2005, and when they appear in retail in Q4 their price is
> supposed to be between $249 and $299"
>
> They are really going to fly off the shelves at that price aren't
> they!
> --
> Andrew, contact via interpleb.blogspot.com
> Help make Usenet a better place: English is read downwards,
> please don't top post. Trim replies to quote only relevant text.
> Check groups.google.com before asking an obvious question.
Anonymous
May 23, 2005 5:42:28 AM

Archived from groups: comp.sys.ibm.pc.games.action (More info?)

On 22/5/05 10:25 PM, Memnoch wrote:

>>http://www.xbitlabs.com/news/multimedia/display/2005052...

> More to the point, if I went out and got one today what games are actually
> going to work with it?

You could say that about any advance in graphics technology starting
with the Voodoo.

The physics processor might well be able to do the physics faster than a
general-purpose CPU. Certainly a graphics CPU does that job better than
a general-purpose CPU does.

My concern is that the job of physics probably involves a lot of
decision-making that doesn't occur when colouring triangles.
Anonymous
May 23, 2005 8:51:31 AM

Archived from groups: comp.sys.ibm.pc.games.action (More info?)

In article <4290aaf7.2029329@news.verizon.net>,
John Lewis <john.dsl@verizon.net> wrote:
> Developers will use the processing power of the dual-core CPU,
> not program for a minimal-sales peripheral. Anyway, where are

Unless one of the major consoles were to integrate a dedicated physics
chip. In that case, a lot of developers might program for it.

Rick R.
Anonymous
May 23, 2005 12:11:47 PM

Archived from groups: comp.sys.ibm.pc.games.action (More info?)

On Mon, 23 May 2005 04:51:31 +0000 (UTC), rickr@is.rice.edu (Rick
Russell) wrote:

>In article <4290aaf7.2029329@news.verizon.net>,
>John Lewis <john.dsl@verizon.net> wrote:
>> Developers will use the processing power of the dual-core CPU,
>> not program for a minimal-sales peripheral. Anyway, where are
>
>Unless one of the major consoles were to integrate a dedicated physics
>chip.

Provided it cost < $1, since the next-gen consoles will be already
sold as loss-leaders for at least a year. Anyway, with the multithread
and parallel processing power that is claimed for both the Xbox360
and the PS3, emulation of that physics chip should be trivial in those
consoles and with little or no effect on visible performance. A
technology brought to market too late.

John Lewis

> In that case, a lot of developers might program for it.
>
>Rick R.
May 23, 2005 8:56:31 PM

Archived from groups: comp.sys.ibm.pc.games.action (More info?)

"Sweeney: Well, Unreal Engine 2 just runs single-threaded so you don't get a
significant benefit from it but in Unreal Engine 3 you'll be able to do
rendering and animation updates and physics in multiple threads so, I
wouldn't say it would double, but it'll increase performance significantly"

This spells the death of this addon card, imho.
From
http://forums.beyondunreal.com/showthread.php?p=1818720...

rms
Anonymous
May 23, 2005 11:26:27 PM

Archived from groups: comp.sys.ibm.pc.games.action (More info?)

On Mon, 23 May 2005 01:42:28 +1000, Richard Cavell <richardcavell@mail.com>
wrote:

>On 22/5/05 10:25 PM, Memnoch wrote:
>
>>>http://www.xbitlabs.com/news/multimedia/display/2005052...
>
>> More to the point, if I went out and got one today what games are actually
>> going to work with it?
>
>You could say that about any advance in graphics technology starting
>with the Voodoo.

I was one of those early adopters. Remember that room you could wander around
in with the different shaped objects just laying around? And when the 3dfx
patch was released for Tomb Raider? Knocked you sideways back then!

>The physics processor might well be able to do the physics faster than a
>general-purpose CPU. Certainly a graphics CPU does that job better than
>a general-purpose CPU does.
>
>My concern is that the job of physics probably involves a lot of
>decision-making that doesn't occur when colouring triangles.

True, but I think it will be a less obvious thing than when you bought your
first 3D accelerator.
Anonymous
May 25, 2005 2:03:53 PM

Archived from groups: comp.sys.ibm.pc.games.action (More info?)

Alot of PC games are going to be graphics intensive. So far, very few PC
games are physics intensive. That says to me the generalized CPU is a
better deal. For the 300 dollars a physics card costs, I could put some of
that money towards upgrading with a new motherboard and CPU.
!