DirectX vs Mantle current and long term

[Will Robinson]

BFG you are a good guy n all but please... NVAPI ?
Its nothing like Mantle.

Also:

Replaced drives that cost 10x less per GB?

No...how SSDs ridicule HD drives for speed.

 

[ShintaiDK]

You know everyone keep bringing up AMD paying EA/DICE for BF4 exclusive support. But everyone does it, i dont think they should be knock on for that.

And look at all the BF4 codes that AMD cards and upcoming AMD CPUs are getting for free.

I am pretty sure EA/DICE did not give those away for free. Makes you wonder where that money that AMD gave them went too.

And the amount of Mantle games for 2014 can be counted on one hand. Its no different than the PhysX part. Pay or it wont be used.

 

[ICDP]

Exactly like hardware PhysX - lower level API, exposes extra performance, only works with one vendor with specific cards, and only in rare sponsored titles.

If someone isn't consistent with Mantle and hardware PhysX, they have a vendor bias. An impartial person either likes or dislikes both because they're fundamentally the same thing.

Replaced drives that cost 10x less per GB?

You mean NVAPI, which has been around for at least a year?

PhysX is not a low level API and it most certainly does not expose extra performance. It is a middleware that works with DirectX and in all cases where GPU PhsysX is enabled performance is reduced significantly on Nvidia hardware.

 

[RaulF]

And the amount of Mantle games for 2014 can be counted on one hand. Its no different than the PhysX part. Pay or it wont be used.

I am not disagreeing with you on that part.

 

[dacostafilipe]

I sense a dark force hovering over the "close thread" button ... maybe we can defeat it by going back to topic?

@ Carfax83

In DX11, rendering commands (aka batch/draw call) are still not thread safe.

That's why Dan call DX's attempt to MT a "failure". I would call it a "not there yet".

Let's hope that Microsoft will give us more information about the future of DX, but as of now, the Mantle architecture look more promising from a technical point of view.

Will this be enough to make it a success? I don't think it is, but this will certainly help/force the other API's to rethink how they will evolve.

 

[frozentundra123456]

BFG you are a good guy n all but please... NVAPI ?
Its nothing like Mantle.

Also: No...how SSDs ridicule HD drives for speed.

Hard drives vs SSD is a totally irrelevant analogy. How many games can you play on a SSD vs a HDD? exactly the same number. How many games can you play on Mantle vs DX: one (if you have the right hardware) vs thousands. End of analogy.

 

[stahlhart]

I sense a dark force hovering over the "close thread" button ... maybe we can defeat it by going back to topic?

You would accomplish a lot more by exercising some self-restraint in this forum and not feeding trolls.

They keep returning, because you keep feeding them. They know you're going to, because you can't help yourself. You're absolutely going to have the last word in that argument, and you never come to the realization that you're never going to get it.

So, for a change, why not try -- not feeding them?

Just a suggestion.

-- stahlhart

 

[realibrad]

If Mantle gets MS to update DX, its a success.
If Mantle gets picked up by game devs, its a success.

I don't see Nvidia just sitting by while AMD attempts to get a performance boost for their cards. Based off of that assumption, Nvidia can create their own type of Mantle, which isint likely in my opinion. They could also push MS to update DX so that Nvidia also gets a boost.

I don't see Nvidia making their own Mantle, as devs have already said they would not want to work in DX, Mantle, and Nvidia. Nvidia has its new toy in G-Sync that so far has been pretty well received. Chasing AMD with its own low level API would be kinda silly. I really think Nvidia will push to have DX updated, and I cant see AMD fighting that, as it would mean that AMD wouldnt have to "sponsor" games to get Mantle in.

 

[dacostafilipe]

...

Really?

This forum just hit a low for me ... gl hf!

 

[AnandThenMan]

If someone isn't consistent with Mantle and hardware PhysX, they have a vendor bias. An impartial person either likes or dislikes both because they're fundamentally the same thing.

The only thing PhysX and Mantle have in common is they are hardware specific. Everything else is not comparable.

You mean NVAPI, which has been around for at least a year?

NVAPI is almost nothing like Mantle.

 

[yulgrhet]

Everybody should read this. It's from Dan Baker who made Lore Engine for Civ5.

Is importance: "However, we'd like to point out that our Direct3D performance is absolutely outstanding, relative to what is expected. We have spent a huge amount of time optimizing around D3D, and we feel we are actually pretty biased in D3D’s favor. Mantle, on the other hand, we've spent far less time with and currently have only pretty basic optimizations. But Mantle is such an elegant API that it still dwarfs our D3D performance."

What is happens when spend much more time with Mantle?

 

[ragnar323]

So you want your own special thread to bitch about mantle in? You don't even use mantle. If you don't plan on buying GCN hardware, why do you even care? I assure you, nobody is gonna come into your house and take your Geforce cards, and replace them with those awful AMD things.

Time wasted by game developers on implementing mantle is time wasted on not developing the game imo. Even nVidia users should care about mantle.

 

[yulgrhet]

Its quite simple. Mantle is not universal, hence never get a chance. Its economics.

How makes sense when Mantle just new and Beta still?

 

[realibrad]

Time wasted by game developers on implementing mantle is time wasted on not developing the game imo. Even nVidia users should care about mantle.

You are making the presupposition that Mantle is a waste of time. Your logic also implies that working on anything not mainstream is a waste. This would mean working on new tech is pointless.

It's not about time wasted, its about trade offs. "If" it turns out that Mantle helps the industry as becoming a standard that gives less overhead, then it would mean devs have more processing power they can use.

It currently appears that AMD is offsetting the development costs for the studios by paying them to use Mantle.

 

[yulgrhet]

You would accomplish a lot more by exercising some self-restraint in this forum and not feeding trolls.

They keep returning, because you keep feeding them. They know you're going to, because you can't help yourself. You're absolutely going to have the last word in that argument, and you never come to the realization that you're never going to get it.

So, for a change, why not try -- not feeding them?

Just a suggestion.

-- stahlhart

If know trolls why not to ban them?

 

[ShintaiDK]

How makes sense when Mantle just new and Beta still?

Mantle needs to work on nVidia and Intel. If the answer is never. Then it cant be universal. And since its very hardware specific (and closed)and there is no intention for any of the 2 other to support it. The answer is never. Its AMD losing graphics and CPU marketshare. But being the small player with roughly 1/3 of the dGPU segment and below 1/5th of the IGP. Maybe more like 1/6th.

 

[Stormflux]

Speaking of long term. What can AMD do to make Mantle more widespread?

I only mostly know what has been linked to on this forum.
Moneyhatting is always an option but.. with what? Specific Games? Would it be more prudent for AMD to say go after Valve, whose SteamOS and SteamBoxes are perfect for this? That would mean supporting Linux. How reliant is Mantle on the Windows Platform?

Rather than games, go after Engines? Dice is onboard with Frostbite, there's still Source, Cry, Unreal, Unity, RED, Fox, Panta Rei, dozens other I can't think of.

How open and readily available is the API? Would it be completely terrible for other graphic chip makers to adopt it? I know it's still a beta but I'm really wanting to hear about future plans on Mantle. This tech is still new and the advantages are clear so I can't help but want to know more.

 

[dguy6789]

If know trolls why not to ban them?

Don't you know how VC&G works? They ban people who tell trolls to quit acting like they're 13 and to try and have an intelligent conversation about a topic we all are interested in. They never ban the actual trolls.

 

[f1sherman]

Speaking of long term. What can AMD do to make Mantle more widespread?

I only mostly know what has been linked to on this forum.
Moneyhatting is always an option but.. with what? Specific Games?

Every single new game that they can hold to,
CPU hogs like ARMA 2/3, notoriously single threaded STALKERs, air sims, all kinds of MMORPGS..

How about World Of Warcraft?
I doubt Blizz would even consider this, but what if AMD makes them an offer that cant be refused?

It always boils down to money

 

[Spjut]

TC's Civ5 benchmark for the HD 7970 seems to be from when the HD 7000 series was released. The HD 7000 series got big performance improvements as the drivers matured.

This more recent test by Anandtech shows that Nvidia still has the advantage in Civ5, but it's less than it was before
http://www.anandtech.com/show/6774/nvidias-geforce-gtx-titan-part-2-titans-performance-unveiled/13

In the case of Civ5, it's also possible that there were alot more driver optimizations done than just adding support for DX11 multithreaded rendering.

IMO, considering the few confirmed games supporting it, you're giving DX11 multithreaded rendering far too much credit. It's a warning sign when there are so many games that DON'T support it.


Regarding support for Mantle, Frostbite 3 supporting it is a good start. If AMD were to get it supported in other major engines, like the Unreal Engine, it would immediately get widespread.

I sure as hell will be extremely happy if the next DirectX and OpenGL even can get remotely close to closing the gap with Mantle, but that's probably happening first with Windows 9. And it'll probably require new GPUs as well. I hope DirectX 12 will be backwards compatible with current cards though, like how the DX11 API brought improvements to DX10.x as well

 

[yulgrhet]

Mantle needs to work on nVidia and Intel. If the answer is never. Then it cant be universal. And since its very hardware specific (and closed)and there is no intention for any of the 2 other to support it. The answer is never. Its AMD losing graphics and CPU marketshare. But being the small player with roughly 1/3 of the dGPU segment and below 1/5th of the IGP. Maybe more like 1/6th.

What you sayings I see no proofs or links to.

Johan Andersson saying Mantle work on Intel and Nvidia. Johan gaming god here. What is proof not work? How you know Intel and Nvidia support Mantle 'never'?

Am thinking you speak from derriere.

 

[ShintaiDK]

Speaking of long term. What can AMD do to make Mantle more widespread?

I only mostly know what has been linked to on this forum.
Moneyhatting is always an option but.. with what? Specific Games? Would it be more prudent for AMD to say go after Valve, whose SteamOS and SteamBoxes are perfect for this? That would mean supporting Linux. How reliant is Mantle on the Windows Platform?

Rather than games, go after Engines? Dice is onboard with Frostbite, there's still Source, Cry, Unreal, Unity, RED, Fox, Panta Rei, dozens other I can't think of.

How open and readily available is the API? Would it be completely terrible for other graphic chip makers to adopt it? I know it's still a beta but I'm really wanting to hear about future plans on Mantle. This tech is still new and the advantages are clear so I can't help but want to know more.

The problem is, all those will still have to support DX/OpenGL. And every single dev would ask why should they support Mantle with extra cost, dev time, QA etc on their own.

Mantle is not in consoles either. Intel and nVidia is not supporting it. If Mantle is to even have the remote chance of hope. They would need either Intel or nVidia aboard. And I think we can pretty much exclude nVidia. And I dont think Intel would be interested either to save AMDs API.

 

[yulgrhet]

Mantle is not in consoles either. Intel and nVidia is not supporting it. If Mantle is to even have the remote chance of hope. They would need either Intel or nVidia aboard. And I think we can pretty much exclude nVidia. And I dont think Intel would be interested either to save AMDs API.

If Mantle being way AMD catch up Intel CPU why not Mantle being way Intel catch up AMD GPU?

 

[parvadomus]

Everybody should read this. It's from Dan Baker who made Lore Engine for Civ5.

Amazing. Now I understand how brillant Mantle is from a HW vendor perspective. It is just umbeatable in terms of drawcalls, and it leaves a lot of optimizations @ dev hands.
The GPU acts like a processing unit that executes everything its thrown at it, no more graphics pipeline, no more buffers, no more textures. The GPU just processes what it receives.

The best part is that Mantle reads HLSL shader code... and Mantle should work under linux, so it should be rather easy to port Mantle games to Linux. Just a minor changes, rebuild and bingo!

 

[Bateluer]

Of course we now start getting the old, "anything that shows AMD/Mantle in a good light" must be debunked BS.

Thats how the Internet and fanboys work. If something shows the brand you dislike in a positive light, its debunked, pointless, and trash. See it a lot in the MD&G form with some of the WP/Android/iOS discussions. If it shows Android positively and the commenter prefers iOS, its pointless trash. If it shows WP in a positive light and the commenter prefers Android, its pointless trash. If the commenter prefers iOS, everything else is pointless trash and can't compete.