New Mafia II PhysX ON/OFF video.

[GaiaHunter]

What do you think would be more productive then?
Complaining about how terrible current PhysX content is, or supporting it to further it's advancement?

Maybe you should start a thread asking what people would like to see done in physX and why they are disappointed with physX, a la Starcraft 2 AA?

But actually many people already answered.

They want to see effects they haven't seen before.

They want those effects while having decent frame rates.

They don't want to see effects they have seen in the past done by another method and then be told "but now these effects use more realistic methods instead of a script".

Who cares?

I know illusionism is not real but if it is done properly it can be entertaining.

Now about the supporting part.

Does NVIDIA need money to develop physX?

I keep reading they have money in the bank, so if physX isn't advancing it isn't because NVIDIA lack of resources (and if NVIDIA has a lack of financial resources there is a thing called investment loans where someone lends saved money with an interest to some company/individual that will invest it and then repay the loan and pay the interest).

So what does NVIDIA need?

100% marketshare?

That is for the market to decide and the market decides what to buy based on current products, not on future products (that is the future market).

Now that we discounted the silly ways to support NVIDIA physX, the only productive way for a potential consumer to support it is to tell NVIDIA what he would like physX to be and what he dislikes about physX.

And that is what most of us here are doing.

Sorry if we can't make it more clear.

So what is left is the developers of games.

It is their fault (or maybe physX simply isn't good enough and doesn't allow the developers to go beyond what is seen ) that there are no games where physX plays a central compelling role.

 

[Keysplayr]

Maybe you should start a thread asking what people would like to see done in physX and why they are disappointed with physX, a la Starcraft 2 AA?

But actually many people already answered.

They want to see effects they haven't seen before.

They want those effects while having decent frame rates.

They don't want to see effects they have seen in the past done by another method and then be told "but now these effects use more realistic methods instead of a script".

Who cares?

I know illusionism is not real but if it is done properly it can be entertaining.

Now about the supporting part.

Does NVIDIA need money to develop physX?

I keep reading they have money in the bank, so if physX isn't advancing it isn't because NVIDIA lack of resources (and if NVIDIA has a lack of financial resources there is a thing called investment loans where someone lends saved money with an interest to some company/individual that will invest it and then repay the loan and pay the interest).

So what does NVIDIA need?

100% marketshare?

That is for the market to decide and the market decides what to buy based on current products, not on future products (that is the future market).

Now that we discounted the silly ways to support NVIDIA physX, the only productive way for a potential consumer to support it is to tell NVIDIA what he would like physX to be and what he dislikes about physX.

And that is what most of us here are doing.

Sorry if we can't make it more clear.

So what is left is the developers of games.

It is their fault (or maybe physX simply isn't good enough and doesn't allow the developers to go beyond what is seen ) that there are no games where physX plays a central compelling role.

So in essence, because of what PhysX offers presently, we'd all be better off without? I get the message that we all want to see it advance, but at the same time, I get the message that some want to see it die. And in representing this cause, constantly complain about it. We'll, if they want it to advance, they have a funny way of showing it. Gaia, maybe you should start a thread as you suggested, asking what people want to see from PhsyX.
I know people have mentioned it but mainly in scattered locations. If it's all in one place, maybe there could be something that comes of it.

 

[biostud]

When has good gameplay ever been dictated by technical specifications.

It's like saying a DX11 game is better than a DX9 game.

I'm all for new technology as long as it doesn't create a monopoly. I would think that software developers would start making their own opencl/directcompute physics engine once there's enough dx11 adopters.

 

[MagickMan]

Might want to tone down the language there MagickMan, mods have been trying to clean up the Video forum cussing. Although I totally agree with the sentiment on standardization.

I put characters in the correct places to protect the "innocent".

 

[Wreckage]

we'd have a decent physics standard right now that everyone could benefit from.

We already have a great physics standard that everyone can benefit from. Just because AMD cards are too crippled to run it, does not prevent you from buying the correct card to run the best hardware physics on the market.

Maybe you could direct your anger at AMD for holding back gaming technology. :hmm:

 

[evolucion8]

We already have a great physics standard that everyone can benefit from. Just because AMD cards are too crippled to run it, does not prevent you from buying the correct card to run the best hardware physics on the market.

Yeah, there's so much GPU accelerated PhysX tittles, that not even in a lifetime I could play them all, and since you don't know a thing about GPU architectures, that's only pure speculation and FUD.

Maybe you could direct your anger at AMD for holding back gaming technology. :hmm:

Yeah, like being first in the DX11 market with a complete SKU's from top to bottom and being first adopting a great multi display solution for the masses (Yeah Matrox was first but can you see lots of members using it here?, its not practical) and DX10.1, making Tessellation part of the DX11 along with Gather4 and 3Dc known as BC3 and BC6 compression schemes, being first with the latest in the manufacturing process, I wonder why nVidia is sponsoring DX9 tittles? Could it be because they want to hold back gaming technology? DX11 is already available :awe:

 

[Wreckage]

Yeah, like being first in the DX11 market with a complete SKU's from top to bottom and being first adopting a great multi display solution and DX10.1, making Tessellation part of the DX11 along with Gather4 and 3Dc known as BC3 and BC6 compression schemes, being first with the latest in the manufacturing process, I wonder why nVidia is sponsoring DX9 tittles? Could it be because they want to hold back gaming technology? DX11 is already available :awe:

DirectX is a proprietary API owned by Microsoft. So much of your speech should be thanking Microsoft I guess.

NVIDIA has DirectX 11 cards, does AMD have a physics solution? :hmm:

 

[MagickMan]

We already have a great physics standard that everyone can benefit from. Just because AMD cards are too crippled to run it, does not prevent you from buying the correct card to run the best hardware physics on the market.

Maybe you could direct your anger at AMD for holding back gaming technology. :hmm:

Wow, [redacted]. Who knew?

Personal attacks/insults are not acceptable.

Moderator Idontcare

 

[evolucion8]

DirectX is a proprietary API owned by Microsoft. So much of your speech should be thanking Microsoft I guess.

NVIDIA has DirectX 11 cards, does AMD have a physics solution? :hmm:

DirectX API is made by Microsoft by the help of hardware makers like nVidia and AMD, thanks to AMD, DirectX 11 has features that are supported on both hardware vendors like Tessellation and BC3 and BC6 compression technology, Microsoft owns DX11 but doesn't have the hardware to run it do they?

Phisics is great when its implemented well, but in its currently form, PhysX is a joke, does nVidia owns the majority of the DX11 market? Does nVidia have a complete DX11 line up from top to bottom? Does nVidia have the fastest single card in the market? Does nVidia have the leadership in performance per watt discipline? Well, that makes PhysX even more irrelevant than already is, who's gonna risk to code a propietary API like PhysX for a hardware vendor that owns less than 11% DX11 market share?

 

[Wreckage]

Flame-baiting is not acceptable. Inciting a flame-war is not beneficial to the community.

I expect better, I know you can do better.

Moderator Idontcare

 

[MrK6]

Man, some of the arguments here are really grasping for anything. If you feel the need to come on here and apologize for NVIDIA or try to talk down AMD instead, you do realize that just cements the complaints against NVIDIA further, right? I mean, how poorly does a company have to be doing that they pay/bribe/fool people into posting in tech communities because they can't sell products by the product's merit alone? Maybe if NVIDIA spent less time on spin and marketing and more time actually developing their supposed "solutions," we wouldn't be having much of this discussion in the first place.

I'm hoping Microsoft throws a hardware physics solution in the next DX, as that's probably the only way we're going to see it adopted by the masses. NVIDIA doesn't have the marketshare or presence to push their own solution and it's really just coming out a mess (kind of embarrassing, really).

 

[Will Robinson]

We already have a great physics standard that everyone can benefit from. Just because AMD cards are too crippled to run it, does not prevent you from buying the correct card to run the best hardware physics on the market.

Maybe you could direct your anger at AMD for holding back gaming technology. :hmm:

There's nothing great about PhysX...at very best it's alright.
Pimping it here constantly when its not much more than a work in progress just highlights its limitations and creates unrealistic expectations that NVDA has yet to deliver on.

 

[evolucion8]

Flame-baiting is not acceptable. Inciting a flame-war is not beneficial to the community.

I expect better, I know you can do better.

Moderator Idontcare

 

[Wreckage]

[redacted by Moderator Idontcare]

Wait who has the fastest chip right now? It sure is not AMD. As for a CPU, how much profit is AMD making from CPUs? LOL.

NVIDIA should just buy Valve and make all future source games NVIDIA only. It would settle this once and for all and we could finally move forward in the gaming market. Although I predict AMD will be bankrupt by 2012 so that may settle it just fine.

I'm done being baited by the AMD propaganda machine. They have their work cut out for them as they have to do this every single time a PhysX title comes out and there are more on the way.


Thread-derail is not acceptable.

Moderator Idontcare

 

[GaiaHunter]

So in essence, because of what PhysX offers presently, we'd all be better off without?

I think the reaction of a decent portion of the consumer market to the current implementation of physX will be "meh".

I get the message that we all want to see it advance, but at the same time, I get the message that some want to see it die.

You know what the problem with physX is Keys.

You've stated it - " [...]or supporting it to further it's advancement?"

It needs the consumer support in advance, as in NVIDIA needs to have huge market share, essentially a monopoly, to make physX something more than it is.

But the consumer doesn't care about NVIDIA or AMD. They want several providers, the more the merrier, but they don't care if it is NVIDIA or XPTO or KJDLF.

Did you see many people buying the Ageia PPU because of its future promises?

Would you buy a car today from X brand because they promise that in the future they will release a flying car if you support them today?

And in representing this cause, constantly complain about it. We'll, if they want it to advance, they have a funny way of showing it. Gaia, maybe you should start a thread as you suggested, asking what people want to see from PhsyX.

Do they have a cause or simply they share their view of why physX isn't a priority when buying a new GPU?

Of course many, myself included, have difficulties in seeing a future for physX (or physX as in something compelling) considering the pre-requisites it needs to be more relevant. Some even see it as a deterrent to have a physics solution that everyone could work with.

 

[tviceman]

It says quite a bit when a company doesn't even have faith in its own products.

You see, this is the difference between you and I. I can favor a particular product or company, but still maintain a bias. I can use the word "if" in appropriate instances, because all the facts aren't known and grand assumptions turn out to be wrong, it just demonstrates how quick they are to bash something they don't personally like no matter if it's right or wrong.

So like in this particular case, I've pointed out something with Nvidia and physx that I don't like, but since I don't know for sure what the reasoning or actual situation is, I didn't jump to uninformed conclusions. I stated that *if* that's actually how it was meant to be implemented and there is no plans to change it, then it's unfortunate. Driver releases and patches come out regularly, so I'll be patient and withhold judgment.

But once again, unlike you, I don't bashed lesser preferred companies or products at every opportunity and I do in fact regularly say good stuff about both sides.

 

[tviceman]

I'm hoping Microsoft throws a hardware physics solution in the next DX, as that's probably the only way we're going to see it adopted by the masses.

And this is my biggest complaint about AMD. They have great products right now but they aren't doing a damn thing to differentiate HOW the products are being used. If AMD had their own gpu-accelerated physics solution out on the market for as long as Nvidia has done gpu-physx, we'd probably have 3 times as many games doing gpy-physics and we would also have a standard being utilized in today's games that would run on either vendor.

 

[MagickMan]

Wait who has the fastest chip right now? It sure is not AMD. As for a CPU, how much profit is AMD making from CPUs? LOL.

NVIDIA should just buy Valve and make all future source games NVIDIA only. It would settle this once and for all and we could finally move forward in the gaming market. Although I predict AMD will be bankrupt by 2012 so that may settle it just fine.

I'm done being baited by the AMD propaganda machine. They have their work cut out for them as they have to do this every single time a PhysX title comes out and there are more on the way.

[redacted]

Personal attacks and insults are not acceptable.

Moderator Idontcare

 

[GaiaHunter]

If AMD had their own gpu-accelerated physics solution out on the market for as long as Nvidia has done gpu-physx, we'd probably have 3 times as many games doing gpy-physics and we would also have a standard being utilized in today's games that would run on either vendor.

I doubt it.

Lets put ourselves in the game developers shoes.

We have a given market for our game.

GPU PhysX by definition will require a reasonable modern ring and most likely one with 2 GPUs - 1 for the rendering and 1 for the physics (and I'm saying this based on the effects we have today and still cause a GTX 480 to take a decent hit in performance).

We face 2 problems.

1) The market with the required GPU power is small - how many people have something like a GTX 280 + 9600GT for physX?

2) Not only that high-end market is small by definition it is also split close to 50-50 between AMD and NVIDIA.

We want to sell loads of copies of our game. Are we going to spend our resources to create a game that has brilliant physics effects that are the foundation of our game gameplay and target a tiny market or are we going to invest in other areas, like story, multiplayer, etc, that can potentially target a much broader costumer base?

Even a standard physic API might not enough if the hardware requirements are too high.

 

[tviceman]

I doubt it.

Lets put ourselves in the game developers shoes.

We have a given market for our game.

GPU PhysX by definition will require a reasonable modern ring and most likely one with 2 GPUs - 1 for the rendering and 1 for the physics (and I'm saying this based on the effects we have today and still cause a GTX 480 to take a decent hit in performance).

We face 2 problems.

1) The market with the required GPU power is small - how many people have something like a GTX 280 + 9600GT for physX?

2) Not only that high-end market is small by definition it is also split close to 50-50 between AMD and NVIDIA.

We want to sell loads of copies of our game. Are we going to spend our resources to create a game that has brilliant physics effects that are the foundation of our game gameplay and target a tiny market or are we going to invest in other areas, like story, multiplayer, etc, that can potentially target a much broader costumer base?

Even a standard physic API might not enough if the hardware requirements are too high.

Nobody "needs" a dedicated physx card. Remove the NPC cloth effects and Mafia II runs great on a gtx260 with physx on high.

The joy of gpu-physx is that it's scales to the amount of physics being computed. Mirror's Edge and Batman had great effects and ran well on less than top-of-the-line Nvidia cards. And, once again, using the little check mark box with the "off" word next to it, corresponding to added-gpu affects alleviates problems for low end hardware.

The rest of what you reason is conjecture that we just happen to disagree on and can't prove our arguments either way.

 

[Barfo]

NVIDIA should just buy Valve and make all future source games NVIDIA only. It would settle this once and for all and we could finally move forward in the gaming market. Although I predict AMD will be bankrupt by 2012 so that may settle it just fine.

Sure, a monopoly would be great to drive forward the market, especially with a company as sleazy as nvidia.

 

[SirPauly]

And this is my biggest complaint about AMD. They have great products right now but they aren't doing a damn thing to differentiate HOW the products are being used. If AMD had their own gpu-accelerated physics solution out on the market for as long as Nvidia has done gpu-physx, we'd probably have 3 times as many games doing gpy-physics and we would also have a standard being utilized in today's games that would run on either vendor.

I thought EyeFinity was a step in the right direction and very welcomed. Also the pro-active nature of trying to get 10.1 and 11 content in gaming titles was as well.

 

[evolucion8]

Wait who has the fastest chip right now? It sure is not AMD. As for a CPU, how much profit is AMD making from CPUs? LOL.

NVIDIA should just buy Valve and make all future source games NVIDIA only. It would settle this once and for all and we could finally move forward in the gaming market. Although I predict AMD will be bankrupt by 2012 so that may settle it just fine.

I'm done being baited by the AMD propaganda machine. They have their work cut out for them as they have to do this every single time a PhysX title comes out and there are more on the way.

AFAIK you don't buy chips, you buy videocards and the HD 5970 has a commanding lead against the GTX 480. AMD's profits regarding CPU's are higher than nVidia's CPU market (Which are just like their chipset and Tegra market, none!) Do you really want a sleazy company as nVidia to own the entire gaming market? For real? [redacted]


Personal attacks and insults are not acceptable.

Moderator Idontcare

 

[Cerb]

We already have a great physics standard that everyone can benefit from. Just because AMD cards are too crippled to run it, does not prevent you from buying the correct card to run the best hardware physics on the market.

Maybe you could direct your anger at AMD for holding back gaming technology. :hmm:

What has nVidia's Physix done for a game, that Half-life 2 didn't do, on far lesser hardware? So far, nothing. A handful of games actually look noticeably better (Batman: AA, FI), that's about it. They get slowed down by it, though. meh.

Meanwhile, CryEngine 3 is featuring a multithreaded cross-platform physics engine, which is not bolted-on, and that can be used to enhance both gameplay and eye candy, at the same time, with green or red cards (oh, and it's not stuck using plain x87). That's what advancement looks like. Future games using future Epic and Crytek engines will have the possibility to really make physics in the game work with you, both against and for (I'm not sure if Croteam has any physics up their sleeves, or not).

Imagine, FI, instead of pretty cloth and such BS, you could have, say, MW6, where you could interact with individual trees of a jungle or forest environment, or have fully destructible building for cover, whose failure modes and their consequences would be plausible and intuitively predictable (generalized structure data, that turns into objects to be handled by the physics engine as collisions/damage occur), with damage dealing/taking actually being based on the material, its inertia, and approximate impact area, with the mech's own structure factoring in; such that piloting a Nova or a Shadowcat would be far different than how far left and right you could aim (yes, I am jonesing for the MW 5/reset...even though I'll miss my beloved CERML/PPC/Flamer Shadowcat ). Such tactical roles the environment could play would also make it easier to balance out the weight classes for multiplayer (3 & 4, FI, got it where 45-80 were pretty balanced, IMO). That is what I want to see from game physics--a non-static game world--and nVidia isn't going that way. Luckily, others are.

Physics simply cannot be extra, and also matter to the game. What we are disappointed about are the resources going into a feature that doesn't give us any better gameplay than we had ten years ago, and gives no better actual game physics than we had six years ago (HL2), in the best cases (games where PhysX is always in use).

 

[Piotrsama]

I don't know if somebody already mentioned, but the video posted by the OP reminded me of the heaven benchmark: DirectX 11 Vs DirectX 10 BS.