AGEIA PhysX products pricing disclosed

Holy shit. Well, you know where they can shove that chip then.
 
:rolleyes: As if you people WONT buy them.

Gamers pay $250-300 to get 10-15fps increases ALL THE TIME. This promises to make an even bigger difference, or to allow current frame rates, with MUCH greater detail. The only thing that will stop these from being a huge success will be support in games. If they make it work with Source and the Unreal engine, they will be HUGE. That is if it does what it says it does.

ANd according to their website, Epic is already onboard with them, so expect to see support for the next Unreal if this thing actually works.

This would make car racing sims able to me MUCH more realistic too.



Of course this is all depending on it ACTUALLY DOING what it promises lol, and we all know how that often goes....
 
Davenow said:
Of course this is all depending on it ACTUALLY DOING what it promises lol, and we all know how that often goes....

And developers actually supporting such a device.
 
its going to be a wait and see thing for me, it all depends what the developers do with it. imo i think its about 100$ too expensive for a first gen product , they should get weaker/cheaper versions out to give us a idea about what they can do with the cards then sell us better versions kinda like what they do with video cards. 300$ is a prety big chunk of money for me to spend on a product that i dont realy know what it will do. i waited a long time before i got a 3dfx card back in the day because honestly i wasn't sure it would do anything for me and the cards were so damn expensive. i can kinda see the same thing happening for this because of price.
 
Honestly i think this Ageia is going to be used in more than just games. If they had stock i would surely buy it.
 
Thier news sounds alot like what we at nvnews got from them in the interview... wonder why thier source is N/A ...
 
why dont they intergrated these buggers into the graphics cards? they could make shitloads of $$, every card would have one.
 
ryanrule said:
why dont they intergrated these buggers into the graphics cards? they could make shitloads of $$, every card would have one.

This has already been discussed. The basic reason is that physics happens much earlier in the pipeline than final rendering.
 
ill buy one around the time ut2007 comes out. by then my comp should be upgraded :x
i dont see much use for it just this minute. 250-300 isnt cheap :(
 
ryanrule said:
then put em right on the gaming mobo's

And when motherboards catch up to the high pricing of current video cards, there will be a riot!


 
uzor said:
And when motherboards catch up to the high pricing of current video cards, there will be a riot!


if they sells tons of them, they can bring the price down.
 
It will fail most likely if it isn't under hundred dollars. The demand will be really low and it will be pulled then someone like what nvidia or ati will buy them out like nvidia did with 3dfx and use the technology for future gpus. ;)
 
Personally I think that these cards should be in the $25-$75 price range (the same range as decent audio cards). I can't see how they are any more complicated than a good EAX4 card.

Edit: Why not bundle the PPU and the soundcard together? The both have to work off of aspects of the environment.
 
Keep in mind that the Voodoo 1 retailed for $299 back in the day. It was in basically the same boat as this. A second card to do something that the CPU does "just fine" already.

 
i agree, who's running the show over there? do they think they can find a niche in a niche market crowded with $5-600 video cards? AND right before the new models are due to be announced?

Don't get me wrong, its a great idea. Maybe they need to move some of those heavy hitters from R&D to market research. If the cost is in the fab, then thats one thing, although I can't imagine thats the case. If the cost is in the licensing....they're doomed.

Good luck.
 
mastercheeze said:
Don't get me wrong, its a great idea. Maybe they need to move some of those heavy hitters from R&D to market research. If the cost is in the fab, then thats one thing, although I can't imagine thats the case. If the cost is in the licensing....they're doomed.

Good luck.

I'm sure the cost at this point is mostly in R&D, especially since this is basically the first of its kind. I would imagine internally that this chip is similar in nature to a graphics card; i.e., many parallel processing units. If it can do what they say it can do (several times the performance of an A64 X2 in physics calculations) then it must be a fairly complex chip. I'm sure after the first generation, they will come down in price though.

 
uzor said:
Keep in mind that the Voodoo 1 retailed for $299 back in the day. It was in basically the same boat as this. A second card to do something that the CPU does "just fine" already.


So that means in 5-6 years we will pay 400-500usd for the new PPU :p
 
HaVoC said:
So that means in 5-6 years we will pay 400-500usd for the new PPU :p
Ahh, but by then, we'll all be hooked and won't be able to do without anymore. :D

 
Aye. Eventually there will be Programmable-Physics models, but it will take the industry 3 years to fully adopt this as standard, at which point, we will have a horde of untechnical savy users wondering why their PPU4mx is unsupported, though, when asking for tech support will state "I have a PPU4, why doesn't it work?".

Furthermore, there will be high-end versions that can shit bone-projection interpolation for breakfast, and benchmark kiddies will gloat over how their 600 dollar PPU makes ragdolls look MUCH more realistic, because there are no more "jerkies", forcing hardware reviewers to release video files for benchmarks to illustrate just how much (little) of a difference 32x rotator-cuff de-splinalizing makes.

Just gimme a mid-range physics proc, for now. It is untested, and all I really want right now are more dynamic sparks, rubble, and bones. Once we get some programmable pipes, and can REALLY start using that fun Geo-mod tech Red-faction had, then I will be throwing down 300 dollars for a PPU, but no sooner.
 
Kiggles said:
Aye. Eventually there will be Programmable-Physics models, but it will take the industry 3 years to fully adopt this as standard, at which point, we will have a horde of untechnical savy users wondering why their PPU4mx is unsupported, though, when asking for tech support will state "I have a PPU4, why doesn't it work?".

Furthermore, there will be high-end versions that can shit bone-projection interpolation for breakfast, and benchmark kiddies will gloat over how their 600 dollar PPU makes ragdolls look MUCH more realistic, because there are no more "jerkies", forcing hardware reviewers to release video files for benchmarks to illustrate just how much (little) of a difference 32x rotator-cuff de-splinalizing makes.

Just gimme a mid-range physics proc, for now. It is untested, and all I really want right now are more dynamic sparks, rubble, and bones. Once we get some programmable pipes, and can REALLY start using that fun Geo-mod tech Red-faction had, then I will be throwing down 300 dollars for a PPU, but no sooner.

Its back to the chicken and the egg. Without people buying the first revision there won't be a reason for them to create a new generation of PPU's. If you build it, will they come?
 
phatmatt said:


I'll buy it.. if they make patches for the current games to use it, or come out with some new ones. If they release this thing before any of the games can actually use it. Whats the point of it?

Either way it looks like a good idea. My only real gripe is it going to pci express? or pci? or what. Im not sure I have enough room for another card that is probably going to be a lot like a graphics card in size. And what happens if you have an SLI setup with 2 video cards and a seperate sound card? Where is this thing gonna go?
 
I think within 3 to 6 months of the intial release they will release a version for about $125 to $150 with slower a/o less RAM and lower clockspeed. This will be the version that most people who end up buying it in the first year will actually buy.

The reason I think they are releasing what will turn out to be the high-end card first is for the enticement factor.... First they release an expensive uber-card that can do everything they promise even if the drivers need some work and almost no games support it. Then they give you a few months to see the new visuals, for the drivers to get fixed, and for benches in various games to come out so you really want one. But of course few people buy it because their aren't enough games to justify the high price. They still WANT it, but few people actually have it yet. So then when they release the version for about $100 less it suddenly seems really cheap, and with the better drivers it's now running almost as fast as the expensive one was the first couple months, and so tons of people run out and get one.

Another great benefit for them is that their is a certain segment of the market who must always have the newest thing for their PC's, but who would have settled for the cheaper one if they came out at the same time. With the lag they will sell all those people the higher price one and so have higher total sales of those cards for the first year than they would have otherwise as well.

A final benefit for them is that the slow sales early on will allow them to work out the manufacturing kinks so they never have a shortage of PPU's. It will also give them a huge pool of cores that work at the cheaper versions lower core speed, but that won't work quite right at the high-end speed. So this way there won't be a shortage of high-end cards early on to give them a bad reputation, and their will be tons of cards for the "real" release when they start shipping the "normal" cards, so there won't be a shortage at that time either.

It's pretty standard marketing for many new tech items... Sell the full featured one first and don't mention a cheaper one is on the way, then sell the cheaper version later to maximize profits. This doesn't work for everything, DVD's for instance. With them you sell the cheaper one then release the "special edition" to get people to re-buy the same movie. But for products like this it's the perfect method.
 
arentol said:
This doesn't work for everything, DVD's for instance. With them you sell the cheaper one then release the "special edition" to get people to re-buy the same movie.


Still pissed about having two copies of LOTR:FOTR. :mad:

I was ready for their scheames when they released the next two movies though. :cool:
 
S1nF1xx said:
Still pissed about having two copies of LOTR:FOTR. :mad:

I was ready for their scheames when they released the next two movies though. :cool:

Yeah, I got lucky and noticed about 3 months before the regular version came out that an extended edition was coming out a few months later, so I never got caught in that trap.

I knew plenty of people that bought both version of all 3 movies though, they just didn't care. What's funny though is that I have read the books over 20 times, and am a serious fan, and yet I was willing to wait for the extended editions while my friends who had never read the books had no patience for them at all.
 
Kiggles said:
Aye. Eventually there will be Programmable-Physics models, but it will take the industry 3 years to fully adopt this as standard, at which point, we will have a horde of untechnical savy users wondering why their PPU4mx is unsupported, though, when asking for tech support will state "I have a PPU4, why doesn't it work?".

Furthermore, there will be high-end versions that can shit bone-projection interpolation for breakfast, and benchmark kiddies will gloat over how their 600 dollar PPU makes ragdolls look MUCH more realistic, because there are no more "jerkies", forcing hardware reviewers to release video files for benchmarks to illustrate just how much (little) of a difference 32x rotator-cuff de-splinalizing makes.

Just gimme a mid-range physics proc, for now. It is untested, and all I really want right now are more dynamic sparks, rubble, and bones. Once we get some programmable pipes, and can REALLY start using that fun Geo-mod tech Red-faction had, then I will be throwing down 300 dollars for a PPU, but no sooner.
red factio was the shit. that made multiplayer hella fun. no more fucking camping, ill blow a hole in your protective wall.
 
ryanrule said:
red factio was the shit. that made multiplayer hella fun. no more fucking camping, ill blow a hole in your protective wall.

The glass house demo was really cool but I never ended up buying the game (although I have a long sleeve tshirt for it).
 
Ryland said:
The glass house demo was really cool but I never ended up buying the game (although I have a long sleeve tshirt for it).
Dammit! You just reminded me how much fun that demo was! :( *starts downloading it again* I would buy the full game if the demo wasn't already more entertaining.
 
I am getting severe Deju Vu here. Didn't we already have this conversation. Or did I completely lose track of what thread I am in :)

Highlights from other thread
1) It's expensive. If you can't afford it, or don't want to...don't buy it.
2) It will come down in price. Everything comes out expensive then get's cheaper
3) People will buy them. There are two types of people. The type that say I want one. then the people that ask...how much.
4) Games will start to use it. Some already do. In theory anything using the NovedeX physics engine will be able to use it very easily.
 
aringail said:
I am getting severe Deju Vu here. Didn't we already have this conversation. Or did I completely lose track of what thread I am in :)

Highlights from other thread
1) It's expensive. If you can't afford it, or don't want to...don't buy it.
2) It will come down in price. Everything comes out expensive then get's cheaper
3) People will buy them. There are two types of people. The type that say I want one. then the people that ask...how much.
4) Games will start to use it. Some already do. In theory anything using the NovedeX physics engine will be able to use it very easily.
mj gives me deja vu
 
Screw physics, where the hell is my darkness accelerator for Doom3?!
 
Number of games that take advantage of a one year old graphic card: 3

Each game took 3-4 years to make.....


So I guess we wont be seeing any good games until 07 that will actually take advantage of this lol...even then, we'll probably see one or two. :p
 
Unless some of the game companies release a patch to take advantage of this. What a PPU really needs is a common spec like DirectX.
 
Anyone wonder how Dual Core CPU's are going to effect this? Can a single "extra" CPU by itself replicate the power of a ageia card?
 
People keep saying wait for dual cores, just get anoter CPU, etc.

Have any of them actually read the articles?

Normal games can handle the physics and collision detection for maybe 200 or so rigid bodies, while this chip can handle some 32,000. Fluid dynamics get a similar double-order-of-magnitude performance boost, as the chip can calculate some 40,000 to 50,000 particles for fluid dynamics.
Now obviously a single CPU has other things to do besides physics, but still, its not even fucking close. A couple hundred compared to thrity-two to fifty thousand? Somehow I have a feeling your going to be spending slightly more on CPU's then you would be buying a $250 PhysX card.
 
All this company needs to do is release PhysXmark 2006 and half the PC gaming community would buy it...just to watch some demos and OC thier $300 physics chip to get the highest benchmark. Helps them justify thier purchase :D Hey, it worked for high end SLi didn't it? :p
 
Circuitbreaker8 said:
All this company needs to do is release PhysXmark 2006 and half the PC gaming community would buy it...just to watch some demos and OC thier $300 physics chip to get the highest benchmark. Helps them justify thier purchase :D Hey, it worked for high end SLi didn't it? :p
ROFL!!!!!!!!

Well said!!!!! And so Freaking True!!!
 
Back
Top