Radeon R9 290X Priced at $730 at Newegg

[SiliconWars]

But that's the thing, this is not outside the box thinking. It's a reintroduction of glide, an introduction of a mechanism to lock you into their products. Every company wants to do this if they could, there's no fresh thinking in this. Though I will admit, they were very clever in bringing it about, if it works.

The limitations in the PC gaming architecture you mentioned it earlier? That's the ability to use peripherals from different vendors, based on the vendors designing their products around a standard. Amd is flipping this around by trying to create a standard based on their proprietary designs. They are not advancing the gaming experience, the are advancing the AMD gaming experience. That's a big difference. If it works, that's great for AMD, it was a bold move. But don't make it into something altruistic or noble, it's vendor lock in.

You are right to be worried about it, as most people who care about truly open standards would be.

But there is another way to look at it, and one that is much more likely to come true. AMD has long struggled against Nvidia's money and marketing, and relations with Intel. This is their chance to flip the balance in their favour. They have caught Intel and Nvidia completely unawares, they have a huge lead here that they can manipulate to their great advantage...but you can be sure as anything that Nvidia and Intel will do *everything* in their power to prevent it from happening.

The realistic best-case scenario is that AMD becomes the gamers champion in the hearts and minds of most people, gets to 60-70% market share and doesn't go bankrupt.

Back Mantle with everything you can, and don't worry about AMD hegemony. They need this, that's why they did it in the first place.

 

[Pariah]

Do you feel the chains? The cold steel cage you are in? Or maybe you think you are outside and the rest of the world is in the one giant cage? That is what Microsoft is doing, they keep you in cage and feed you with a little spoon.

There are certain areas where a monopoly is good. Desktop OS is one of them. Gaming API for PC's is another one. If Nvidia comes out with a competing proprietary API, how exactly is that supposed to benefit the PC gamer? One of the major benefits of gaming on the PC is that you can play any game regardless of who makes the parts in your PC. No one wants to have to buy 2 different video cards to be able to play every game available on a PC.

 

[SiliconWars]

There are certain areas where a monopoly is good. Desktop OS is one of them.

Yeah...gl with that one.

 

[golem]

Why do I care which company it is, if they allow me to play the games I enjoy, in the best experience? And yes, it is outside of the box thinking, regardless of it being done before. Glide supported games kicked ass compared to OpenGL/directx. And nobody said altruism or nobility is the motivating factor, those are your words.

You don't have to. But just be aware if it's too successful, the video card industry goes from a duopoly to a monopoly. It's been a while since AMD has been dominant in anything, but if you think back to the Amd x2 days, they charged an arm and a leg for their cpus.

No it's not, trying to implement a direct to metal api is not new thinking. It just that no one has had the clout of providing the cpu/gpu to all the dominant consoles in the past. The fact that AMD was able to this is the impressive part.

You're used the terms addressing the limitations of pc gaming and efforts towards advancing the gaming experience. To me, at least, they sound like you were trying to make this project seem noble or for the greater good of gaming. Sorry if that wasn't your intention.

 

[golem]

They have caught Intel and Nvidia completely unawares, they have a huge lead here that they can manipulate to their great advantage...but you can be sure as anything that Nvidia and Intel will do *everything* in their power to prevent it from happening.

This I totally agree with. It was a very impressive and bold move to land the contracts to all three consoles. This was pretty brilliant.

They also caught Microsoft offguard, if this is very successful, it weakens directx

 

[Silverforce11]

Here's a hypothetical for all the naysayers out there: If you are a big dev studio and your focus is on the lucrative console market, how would you best entice your audience to pick your game over some other similar titles? Gameplay and graphics superiority, no doubt.

Given the anaemic GPU grunt on the XBONE and slightly better PS4, to extract the best performance out of them to deliver awesome graphics, your studio absolutely needs to go full to the metal (as usual) as well as any further GCN-specific optimizations. PERIOD.

That is where Mantle comes in, it enables GCN-specific functions as well as optimizations. You want to be the fools not using it while your competitors enable more vibrant and fluid graphics? Note how rapid the public statement from Activision, requiring the future use of Mantle after DICE's announcement.

 

[Deasnutz]

You don't have to. But just be aware if it's too successful, the video card industry goes from a duopoly to a monopoly. It's been a while since AMD has been dominant in anything, but if you think back to the Amd x2 days, they charged an arm and a leg for their cpus.

No it's not, trying to implement a direct to metal api is not new thinking. It just that no one has had the clout of providing the cpu/gpu to all the dominant consoles in the past. The fact that AMD was able to this is the impressive part.

You're used the terms addressing the limitations of pc gaming and efforts towards advancing the gaming experience. To me, at least, they sound like you were trying to make this project seem noble or for the greater good of gaming. Sorry if that wasn't your intention.

Yes, I may have generalized a bit much. But you have to admit, giving existing customers the option to gain a potentially higher value out of already purchased hardware is a bit nice, and pretty out of the norm.

 

[RussianSensation]

But they are basically addressing the "limitations" of the pc gaming by introducing the next glide. If they succeed, they basically lock you into purchasing their cards because Mantle doesn't benefit any other video card architectures but their own. So you get faster speed but only on their cards, if this isn't in the interest of selling you a card, then what is it?

This argument is no different than buying NV cards for 10 years because they had a strong developer relations program and solid drivers. API vs. drivers makes no difference to a gamer. What they care about is how well a GPU runs their games. The means of how the GPU achieves this (API, drivers, game patches, black magic) is irrelevant to the consumer. They want a plug & play solution.

You are also assuming that:

1) NV won't release its own API for games in the future
2) Mantle API will bring such dramatic increases in performance that in no way could NV match AMD's cards with brute force
3) Developers will actually embrace Mantle and many games will use it.

None of these may even be true.

If I buy a GPU from NV/AMD, I really don't care if they use 10 different programs in the background to optimize my gaming performance as long as my game runs well and my PC isn't bogged down. If architecture specific API improves performance, then so be it.

No one stopped NV from going after the console design wins. They chose not to themselves. So now NV owners are complaining that AMD's API is taking advantage of GCN architecture specifically because AMD has PS4/XB1 + GCN dGPUs for years to come? Whose fault is that? What difference does it make if "AMD specific driver improves performance for AMD cards only" or if "AMD's specific API improves performance for AMD GCN cards only". No difference to the PC gamer. Since DICE will send Mantle update seamlessly, it's almost like a patch for BF4 players.

This is akin to complaining that it's unfair that NV's cards run "insert Game ABCD" because the game is more optimized for Kepler architecture and/or NV's driver team did a better job than AMD's in this title.

The other thing is people keep missing is that AMD needs a boost in GPU market share and profitability a lot more than NV. If AMD regains market share to 50/50, then it's a lot better for us gamers since the GPU market will be a lot more competitive. What if in 4-5 years, NV and AMD will have their own APIs and all of our PC games run 20-30% faster than without AMD/NV API?

What if we get:

Game ABCD
Under DX
AMD no API = 90%
NV no API = 100%

Game ABCD
Under specific API
AMD = 120%
NV = 130%

I'd much rather have NV and AMD start introducing their own APIs if we get a 20-30% boost in more PC games over using DX. Of course this is all conjecture but without an introduction of a new lower-level API, we'll never find out just how inefficient DX really was.

 

[golem]

Yes, I may have generalized a bit much. But you have to admit, giving existing customers the option to gain a potentially higher value out of already purchased hardware is a bit nice, and pretty out of the norm.

Yup, if this works, it's great for future AMD video card owners and recent current owners. My past video cards have probably tended towards 70/30 Nvidia/AMD at a guestimate. Nvidia's shortcomings are less annoying to me than AMD's shortcomings. But if this takes off and Nvidia doesn't have a counter, I'll easily choose AMD going forward.

 

[SiliconWars]

Yes, I may have generalized a bit much. But you have to admit, giving existing customers the option to gain a potentially higher value out of already purchased hardware is a bit nice, and pretty out of the norm.

That's not all that unusual now with AMD - remember the free games that they gave to buyers after purchase? What was it, FarCry Blood Dragon and another one?

Anyone who has a GCN card must be wetting themselves in anticipation of Mantle lol. My girlfriend is getting this card and I'll be going for a new one based on careful deliberation. Probably a 280 or 280X will be more than enough for my needs...

 

[Pariah]

If I buy a GPU from NV/AMD, I really don't care if they use 10 different programs in the background to optimize my gaming performance as long as my game runs well and my PC isn't bogged down. If architecture specific API improves performance, then so be it.

True, none of us do, or should care how the optimizations happen. However, Mantle could turn into something much more damning to the PC gaming industry. If it sucks, and we "only" get 10-20% speed improvements to optimized games, then it is just a nice bonus to AMD owners. If on the other hand, we see 40-50% or more improvements, than the gaming experience will be fundamentally better on AMD hardware, and NVidia, who is locked out of being able to implement Mantle, will be forced to come up with their own Mantle competitor. At this point, the optimizations will no longer be transparent to the user, and we will see a fracturing of the market. Suddenly, the cute slogans that NVidia and AMD use for games optimized for their hardware, won't mean a slight boost in framerates, it will mean do I have to buy a 2nd card if I want to play a game that is not optimized for the current card I have. We should all care if that is going to happen.

 

[Pariah]

What if in 4-5 years, NV and AMD will have their own APIs and all of our PC games run 20-30% faster than without AMD/NV API?

That's not likely to happen. Developers don't like having to code for multiple API's. Where is Glide now? Where is OpenGL in the gaming world? I'd feel much better about the Mantle announcement, if Nvidia was supplying GPU's for either the XBOne or the PS4. Then there would be a competitive balance. One console for each, plus their PC market share. As it stands now, even if NVidia comes up with a competing API, it's unlikely any developer would be interested unless it was WAY faster than Mantle, which isn't going to happen. From a developer standpoint, using an API that will work on both major consoles is a huge boon. Being compatible with a major PC GPU supplier is simply a bonus. Any competing NVidia API would have to coded knowing, it will only work on PC's equipped with certain NVidia GPU's. Who would be interested in that?

 

[Elfear]

I'm excited about Mantle's potential but it's not much of a selling point until we see some actual results.

Bs. Its sweet talk that sounds right. But like everyone here you will not buy an 780 or titan untill you see what mantle brings.

Also for the reason that if mantle delivers nv will lower their prices.

This pretending its not here. And yet particepate and downplaying its significance pages over pages right now is just not the reality happening.

Mantle is here. It might not deliver in december. But right now its very much here.

**looks at sig, looks at quote, looks at sig, looks at quote...**

As a current GCN owner, why would I be irrationally downplaying Mantle?

Why would you assume I'm planning on buying a Titan or 780 when I say I wouldn't buy a 290X based on Mantle at this point in time. The concept of Mantle sounds awesome IMO but until I see some actual results, it's not much of a selling point. If I end up buying a 290X or two, it will be because of performance not the hopes that Mantle will take off one day.

 

[golem]

This argument is no different than buying NV cards for 10 years because they had a strong developer relations program and solid drivers. API vs. drivers makes no difference to a gamer. What they care about is how well a GPU runs their games. The means of how the GPU achieves this (API, drivers, game patches, black magic) is irrelevant to the consumer. They want a plug & play solution.

To the gamer it makes no difference between API and driver. To the companies making the cards it does. There's always the hope that at least with bad drivers, you can get them up to snuff and at least equal if not exceed the competition.

If the competition is using a direct to metal API (and it doesn't suck) while your using a more abstract higher level API, there's almost no chance you can achieve performance parity unless you have some other sustainable competitive advantage.

Nvidia was asleep at the wheel when the didn't land at least one of the consoles, that's the only reason AMD can even try introducing their own API. If it Mantle is successful, what developer in their right mind would code for Mantle, DirectX and a new API that Nvidia came up with? PC ports are usually bad because developers don't want to spend the resources to develope pc specific code, and that's for the entire pc gaming market. If they have mantle for AMD and DirectX for everyone else, the chance of them also coding for an Nvidia specific API is 0, without payments from Nvidia.

If Mantle is all it's cracked up to be, it won't be 50/50 it will be slowly (or maybe not so slowly) be just AMD. Nvidia will either go bankrupt or leave the market, just like with CPU chipsets.

 

[stahlhart]

The last post even remotely on topic was four pages back. Thread closed.