The AMD Mantle Thread

[ASM-coder]

I just now saw this story from very earlier today. I don't think it has already been sited.

http://www.legitreviews.com/amd-man...to-40-percent-performance-gains-in-bf4_134873

"AMD is working overtime this week trying to get the first Mantle driver out for gamers. We were told last night that they were working on another build of it and that build would be done this morning. Once it passes internal testing it should be available publicly."

"... notes for the AMD Catalyst 13.35 Beta drivers that should be out today or tomorrow if there are no major issues found with the build that was created overnight."

I'm hoping this delay isn't because of a major issue.

I also think that site got there hands slapped for suggesting earlier today(in a different article) that Nvidia owners got nothing in the latest BF4 patch. There were a number of performance fixes for everyone.

 

[nenforcer]

Oh, you mean like extreme/retarded levels of tessellation?

http://techreport.com/review/21404/crysis-2-tessellation-too-much-of-a-good-thing/2

Crysis 2 DX11 has the worlds most highly tesselated concrete barrier.

 

[Despoiler]

http://techreport.com/review/21404/crysis-2-tessellation-too-much-of-a-good-thing/2

Crysis 2 DX11 has the worlds most highly tesselated concrete barrier.

All hail our concrete barrier overlord!

 

[cytg111]

Only if you want to play the 0.01 percent of games that support mantle.

Wow. Just wow.

 

[chimaxi83]

Wow. Just wow.

Heh, I know. Many of the responses here are hardly surprising, with matching silly company bias. Oh well!

 

[Gikaseixas]

Heh, I know. Many of the responses here are hardly surprising, with matching silly company bias. Oh well!

And this is why many here are trying so hard to downplay the improvements shown so far. Childish behavior unfortunately. I want to see major sites reviewing Mantle with different setups to come to a full conclusion but taking in consideration what we have so far it's actually good.

 

[wand3r3r]

I'll just guess Mantle is serving better than AMD could have planned.

There is a ton of hype, and this thread is full of trolls desperate to downplay (and some of the reverse) everything about it. AMD has made a product that has kept people captivated since the announcement. Watch NV try replicate it and suddenly the people who have trolled this whole thread will be praising it as a gift to humanity. AMD has finally figured out how to market themselves and keep their name in the spotlight for a long time. It's almost like they have outdone themselves on this one and had a viral campaign, ironically due to fanboys (of both sides) desperate for certain outcomes.

Personally I'm excited to see what this will do for 290x crossfire, it looks promising. I'll wait until their are reviews out although I'll try test it a bit myself but if it's between 20-60% that is absolutely amazing in my opinion. Crossfire already scales a bit better but now if the mantle optimizations work anywhere even remotely near the leaks, it will wipe the floor even with mid range cards. It will be interesting to see e.g. 280x crossfire vs. 780/ti. My expectations are not on 60% since it's clearly a scenario that has been presented by AMD and therefore is probably a best case example.

 

[NTMBK]

It is pretty clear that a lot of people in this thread are downplaying Mantle because they are under the spell of that green eyed monster, jealousy... or as our Spanish friends say, envidia.

 

[Abwx]

Posted by HFR reviewer....and a HFR forum member..

During the QA process, we identified an installation issue in AMD Catalyst 14.1 that renders it unsuitable for distribution for testing in your labs. We are testing a new build overnight and will update you on its status by mid-afternoon Eastern Standard Time on January 31. We apologize for the delay and appreciate your patience and understanding.

 

[cytg111]

I'll just guess Mantle is serving better than AMD could have planned.

There is a ton of hype, and this thread is full of trolls desperate to downplay (and some of the reverse) everything about it. AMD has made a product that has kept people captivated since the announcement. ...
Personally I'm excited to see what this will do for 290x crossfire, it looks promising. I'll wait until their are reviews out although I'll try test it a bit myself but if it's between 20-60% that is absolutely amazing in my opinion. Crossfire already scales a bit better but now if the mantle optimizations work anywhere even remotely near the leaks...

With the utter failpath of bulldozer 1, 2, 3 and 4, AMD sort of needs this one, this and HSA, to take off. The console wins, mantle and hsa. Looking at a company on life support, well, nice play. (is this Rory?)
If reviews are good for 290x crossfire and a vanilla i7 haswell, I could see my self selling my 780. (and going dual cards again since voodoo2)

But the dust hasnt settled yet, it can still dive into failoblivion.

 

[redzo]

http://techreport.com/review/21404/crysis-2-tessellation-too-much-of-a-good-thing/2

Crysis 2 DX11 has the worlds most highly tesselated concrete barrier.

It could mean:

1. Someone at crytek forgot about optimizing the concrete barrier.
or
2. The performance impact of tesselating only the concrete barrier is so insignificant that they tessellated the hell out of it without giving a damn.

 

[Carfax83]

With the utter failpath of bulldozer 1, 2, 3 and 4, AMD sort of needs this one, this and HSA, to take off.

For the sake of competition, I don't want AMD to fail. If AMD fails, then NVidia will be the sole supplier of discrete graphics cards and life will suck hard.

That said, I cannot see a path in which Mantle gains a foothold in the industry. It has nothing to do with the virtues of the technology, which are probably going to be substantial. It has to do with the severe drawbacks, and the fact that Microsoft will not be standing still.

As it is right now, Mantle could NEVER replace DirectX, as it has no backward compatibility, nor cross architectural compatibility. But if it cannot replace DirectX, what good is it?

Does AMD believe developers will be content to develop games for two code paths ad infinitum, especially when Mantle only works on a small segment of the overall PC hardware market?

And Microsoft will inevitably come out with a faster, lighter and more threaded version of Direct3D that offers similar performance to Mantle, but with none of the drawbacks.. They've most likely already begun to work on it, as their developer blog hinted last year.

So Mantle will get it's time in the Sun and give AMD some much needed exposure hopefully sell some more cards, but that time is also very limited.

 

[cytg111]

And Microsoft will inevitably come out with a faster, lighter and more threaded version of Direct3D that offers similar performance to Mantle, but with none of the drawbacks.. They've most likely already begun to work on it, as their developer blog hinted last year..

- An even bigger win right there! (And problary would not have been put in motion without the notion of a little competition?)

 

[Fanatical Meat]

Wow. Just wow.

No World of Warcraft does not support mantle at this point.

Ha I couldn't resist....

 

[el etro]

And Microsoft will inevitably come out with a faster, lighter and more threaded version of Direct3D that offers similar performance to Mantle, but with none of the drawbacks.. They've most likely already begun to work on it, as their developer blog hinted last year.

How they will make DirectX backwards compatible with everything and faster than Mantle?



-------------------------------


Man there's a lot of thread crapping over here. Mantle bashers need more convincing arguments.

 

[frozentundra123456]

For the sake of competition, I don't want AMD to fail. If AMD fails, then NVidia will be the sole supplier of discrete graphics cards and life will suck hard.

That said, I cannot see a path in which Mantle gains a foothold in the industry. It has nothing to do with the virtues of the technology, which are probably going to be substantial. It has to do with the severe drawbacks, and the fact that Microsoft will not be standing still.

As it is right now, Mantle could NEVER replace DirectX, as it has no backward compatibility, nor cross architectural compatibility. But if it cannot replace DirectX, what good is it?

Does AMD believe developers will be content to develop games for two code paths ad infinitum, especially when Mantle only works on a small segment of the overall PC hardware market?

And Microsoft will inevitably come out with a faster, lighter and more threaded version of Direct3D that offers similar performance to Mantle, but with none of the drawbacks.. They've most likely already begun to work on it, as their developer blog hinted last year.

So Mantle will get it's time in the Sun and give AMD some much needed exposure hopefully sell some more cards, but that time is also very limited.

Well said. It is ironic that simply acknowledging the serious drawbacks of mantle somehow makes one an nvidia fanboy.

 

[ICDP]

It is ironic that simply acknowledging the serious drawbacks of mantle somehow makes one an nvidia fanboy.

This is indeed a problem. Mantle does have drawback, such as GCN only and low number of games supported (so far). If Mantle does give improvements and innovates then why are Nvidia fans not waiting with baited breath to see what it delivers? Then if it does genuinely bring improvements to PC gaming, they can say "we want this" to Nvidia. Instead we get people claiming DX is perfectly fine, despite the fact that PC game developers have already stated it is most definitely not.

The thing is that many of the people slating Mantle are doing so purely because it is an AMD initiative and they want it to be a failure.

 

[sontin]

So, you mean like with FreeSync?

Right...

BTW: What should we see? I see no improvement for an user who wants to buy a new GPU. Mantle is only interessting if you care about AMD's CPUs. Everyone with an Intel processor need not to care about Mantle.
So when you increase the GPU workload - more details, MSAA/SSAA, GPU-Physics, 3D, 4k etc. - then you have no improvement.

 

[caswow]

So, you mean like with FreeSync?

Right...

BTW: What should we see? I see no improvement for an user who wants to buy a new GPU. Mantle is only interessting if you care about AMD's CPUs. Everyone with an Intel processor need not to care about Mantle.
So when you increase the GPU workload - more details, MSAA/SSAA, GPU-Physics, 3D, 4k etc. - then you have no improvement.

yea why do you even bother? you will never get an amd card so why are you so obsessed with mantle,freesync or everything else amd related?

 

[ICDP]

So, you mean like with FreeSync?

Right...

BTW: What should we see? I see no improvement for an user who wants to buy a new GPU. Mantle is only interessting if you care about AMD's CPUs. Everyone with an Intel processor need not to care about Mantle.
So when you increase the GPU workload - more details, MSAA/SSAA, GPU-Physics, 3D, 4k etc. - then you have no improvement.

See what I mean, totally makes stuff up to suit the agenda that mantle = suck. There has been results released showing a high end Intel CPU with 2x R9 290X in crossfire gaining 58% performance. Have you elected to ignore that result? Clearly this alone shows that there is potential there for enthusiasts who run multi-GPU, as multi GPU usually runs into CPU wall long before the GPU wall.

While I will believe it when I see it, I am at least willing to wait for the actual real world results before declaring mantle as worthy or not.

 

[dacostafilipe]

How they will make DirectX backwards compatible with everything and faster than Mantle?

They could split Direct3D in two.

DirectShell, the top layer compatible with most hardware, similar to todays version.

DirectCore would be the low level layer, similar to Mantle. It would allow vendor specific extensions and/or extensions injection by the driver. DirectShell would be build on top of DirectCore.

Vendors would be able to provide special versions of DirectShell optimised for a specific hardware type.

Yes, this would be a nightmare to support, but hey ...

 

[redzo]

There is a ton of hype, and this thread is full of trolls desperate to downplay (and some of the reverse) everything about it. AMD has made a product that has kept people captivated since the announcement. Watch NV try replicate it and suddenly the people who have trolled this whole thread will be praising it as a gift to humanity. AMD has finally figured out how to market themselves and keep their name in the spotlight for a long time. It's almost like they have outdone themselves on this one and had a viral campaign, ironically due to fanboys (of both sides) desperate for certain outcomes.

Mantle's main drawback is that it is a vendor specific windows 3d api. This is a purely rational claim and I don't think that it should be considered trolling.

 

[cytg111]

So, you mean like with FreeSync?

Right...

BTW: What should we see? I see no improvement for an user who wants to buy a new GPU. Mantle is only interessting if you care about AMD's CPUs. Everyone with an Intel processor need not to care about Mantle.
So when you increase the GPU workload - more details, MSAA/SSAA, GPU-Physics, 3D, 4k etc. - then you have no improvement.

FUD and lack understanding. At best.

 

[NTMBK]

So, you mean like with FreeSync?

Right...

BTW: What should we see? I see no improvement for an user who wants to buy a new GPU. Mantle is only interessting if you care about AMD's CPUs. Everyone with an Intel processor need not to care about Mantle.
So when you increase the GPU workload - more details, MSAA/SSAA, GPU-Physics, 3D, 4k etc. - then you have no improvement.

The Freesync thread is that way ---->

As for "it only improves AMD processors"- sorry, but this is complete and utter nonsense. The largest improvement in framerates, a 58% improvement, came from the combination of an Intel 3970X and Crossfire 290X. Go get some new talking points.

 

[NTMBK]

They could split Direct3D in two.

DirectShell, the top layer compatible with most hardware, similar to todays version.

DirectCore would be the low level layer, similar to Mantle. It would allow vendor specific extensions and/or extensions injection by the driver. DirectShell would be build on top of DirectCore.

Vendors would be able to provide special versions of DirectShell optimised for a specific hardware type.

Yes, this would be a nightmare to support, but hey ...

To be honest I would quite like Microsoft to make Mantle an official "DirectShell"-style API, with support for multiple GPU vendors. Mantle already uses Microsoft's HLSL to write shaders- why not make it an official part of DirectX 12, and get NVidia and Intel on board too?