Firaxis discusses why Mantle is revolutionary for their work with Civilization: Beyond Earth

[LtMatt]

What does Mantle Buy You?

Simply put, Mantle is the most advanced and powerful graphics API in existence. It provides essentially the same feature set as DX11 or OpenGL, and does so at considerably lower runtime cost.
The conventional wisdom in real-time rendering is that batches, or “draw calls” are expensive. On the PC, with current APIs, this notion is firmly rooted in fact. This is a problem that has plagued engine and driver design since at least the DX9 era, and a large body of real-time rendering tradecraft is motivated by it (instancing, state sorting, texture atlasing, texture arrays, “uber-shaders”, to name a few). Civilization, it turns out, requires a significant amount of rendering to generate our view of the world, and that in turn means we are required to make many, many more draw calls than you might expect.. Our birds’ eye view of the world means that we have a lot more “stuff” on screen than is typical, and our UI (a rich source of draw calls) is considerably more complex than the average.
Mantle changes things by working at a lower level than its competitors. Much of the work that drivers used to do on an application’s behalf is now the responsibility of the game engine. This means that the Mantle API is able to be backed by a very small, simple driver, which is thus considerably faster. It also means that this work, which must still be done, is done by someone with considerably more information. Because the engine knows exactly what it will do and how it will do it, it is able to make design decisions that drivers could not.
Besides being more efficient, core per core, Mantle also enables fully parallel draw submission (this has been attempted before, but never with the same degree of success). Until now, the CPU work of processing the draw calls could only by executed on one CPU core. By removing this limitation, Mantle allows us to spread the load across multiple cores and finish it that much faster.
All of this means that Mantle has, quite literally, reduced the cost of a draw call by an order of magnitude. This is an amazing technical achievement and difficult for us to exaggerate the importance of this savings. It is a disruptive technical development which will have far-reaching implications for PC gaming. It will alter the dynamics of the market. It will re-write portions of the real-time rendering book. It will change the design of future APIs and engines and greatly enhance their capabilities.

Why did the developers of the upcoming Sid Meier’s Civilization: Beyond Earth decide to optimize the game for the AMD Mantle API and what are the benefits of using the API?

Mantle is an extremely powerful graphics API. It works very efficiently at a much lower level than the game engine and has fully parallel draw submission, which lets us spread the work around multiple cores. For a game like Civilization which renders a significant number of things on the screen, this has reduced the cost of a draw call by an order of magnitude. This is an amazing improvement in performance for us.

The AMD Mantle website states that “Mantle makes game development easier.” If this is so, how does Mantle and the Mantle API make game development easier for developers?

The Mantle API is able to be backed with a much simpler, faster driver. The increased efficiency means that we no longer need to add as much complexity to our rendering pipeline to achieve the required level of performance. Mantle’s simplicity also means that it will probably be more stable over the long run. We anticipate that the bug rate for Mantle to be lower than previous systems we worked with.

Full Article & Source
http://www.firaxis.com/?/S=5845faee...eb5f6cb11/blog/single/why-we-went-with-mantle
http://blog.programmableweb.com/201...hat-will-utilize-new-amd-mantle-graphics-api/
http://www.amd.com/en-us/innovations/software-technologies/mantle#overview

 

[Marine-RX179]

In before the "oh not another Mantle thread" and "but the next Civ game is (going to be) crap"

Joke aside, it is good to see that games series that actually has suffer poor performance due to CPU bottleneck for decade is jumping on board Mantle. Wish more strategy games and mmos even to adapt Mantle as well, as they desperately need it.

I still don't believe dx12 is going to be all that (nevermind it won't be out till end of 2015 or later as well), with it most likely being prioritise in the order of Xbone, tablets and lastly PC. But then again Xbone is essentially a PC...a PC with AMD graphic GCN based at that as well, so people that are thinking dx12 is going to benefit Nvidia more than AMD...I just dunno...

I wonder if Microsoft's decision to make dx12 supporting Xbone and Windows tablets as well rather than solely for PC/laptops actually pushed developers into jumping onboard the Mantle wagon.

 

[mmj_uk]

AMD need industry wide support not just half a dozen developers payed to add support for and promote it, it's hard to see it becoming an industry standard with so few GPU's supported, especially with DX12 around the corner.

 

[shankly1985]

AMD need industry wide support not just half a dozen developers payed to add support for and promote it, it's hard to see it becoming an industry standard with so few GPU's supported, especially with DX12 around the corner.

Wouldn't call two years around the corner tbf. I do agree with more GPU support, but I dont think it will come with past GPU's and only future GPU will support Mantle.

 

[jh30uk]

2 years?

Beta year end and early access before that, is mantle out of alpha/beta?

 

[Final8y]

Wouldn't call two years around the corner tbf. I do agree with more GPU support, but I dont think it will come with past GPU's and only future GPU will support Mantle.

I think most will be happy of just having the option whenever its available while others just keep bellyaching.

 

[Orangey]

TW series could benefit from this as well, ain't they CPU limited badly?

 

[TheRealDeal]

2 years?

Beta year end and early access before that, is mantle out of alpha/beta?

The first game is said to be out at the end of 2015. So yea for us gamers its more just over a year and a half. I assume beta and early access is for developers not for us.

 

[Sugarhell]

http://www.firaxis.com/?/S=5845faee...eb5f6cb11/blog/single/why-we-went-with-mantle

 

[LtMatt]

http://www.firaxis.com/?/S=5845faee...eb5f6cb11/blog/single/why-we-went-with-mantle

Nice one Sugar will add that to the OP.

 

[Marine-RX179]

The first game is said to be out at the end of 2015. So yea for us gamers its more just over a year and a half.

So only a year and half before Windows 7 gamers are forced to abandon an OS they are perfectly happy with and jump onto Windows 8.1 instead (if Windows 9 not out by then)...

 

[TheRealDeal]

So only a year and half before Windows 7 gamers are forced to abandon an OS they are perfectly happy with and jump onto Windows 8.1 instead (if Windows 9 not out by then)...

Yea i am using windows 8 atm with a free program called pokki that installs a start menu and forces it to boot into desktop. I see no real difference using win 8 from 7 now.

 

[humbug]

DX12 exists only in Microsoft Public Relations Noise. after 10 years why are Microsoft only now apparently making a better API.

The worlds largest Software Company got humiliated with Software from a Hardware Company a fraction of its size.

Will DX12 come to realise the hype? or will it simply be Better than DX11 but still not as good as Mantle.

Despite the fact that the XB1 has an AMD GCN GPU Microsoft have to make DX12 work on Nvidia, Intel, Qualcomm and AMD.
DX12 will contain a lot of Fat.

Developers are still going for Mantle, i think they already know what to expect from DX12.

Microsoft are making a lot of noise to try and keep their monopoly relevant, and Developers are having none of it.

 

[mmj_uk]

Hype or not it's the best most people can hope for if Mantle continues to be GCN exclusive only. Mantle would also contain a lot of fat if the compatibility wasn't so limited.

Developers are going for Mantle but is because they really like it or because AMD is paying them? it seems a lot of hassle to go through for such a small userbase if they aren't geting financial incentives.

 

[Marine-RX179]

Will DX12 come to realise the hype? or will it simply be Better than DX11 but still not as good as Mantle.

Despite the fact that the XB1 has an AMD GCN GPU Microsoft have to make DX12 work on Nvidia, Intel, Qualcomm and AMD.
DX12 will contain a lot of Fat.

Developers are still going for Mantle, i think they already know what to expect from DX12.

Microsoft are making a lot of noise to try and keep their monopoly relevant, and Developers are having none of it.

That's one reason I see why I've see some reviewers expressed their worries that dx12 could potentially become a matter of "jacks of all trade, master of none". I mean I don't blame them; look at GFWL when M$ tried to lump Xbox and PC gaming together...then we have Windows 8 with M$ trying to lump tablets/touchscreen users and PC keyboard and mouse users together. Now with Windows 8.1, it is going backward looking more and more like Windows 7 with each update.

IMO if anything, rather than hyping up and praising dx12 before even seeing it, PC gamers should sign a petition to M$ to keep dx12 for PC, and for everything else, work on something else to support them separately. If I have to pay for upgrading to new window version for the sake of dx12, I want no performance compromise made for the sake of benefiting other platforms that are irrelevant to me- we've seen launch of Window 8 with aim to support more things than one at the same time, and that did not went down well.

 

[humbug]

Hype or not it's the best most people can hope for if Mantle continues to be GCN exclusive only. Mantle would also contain a lot of fat if the compatibility wasn't so limited.

Developers are going for Mantle but is because they really like it or because AMD is paying them? it seems a lot of hassle to go through for such a small userbase if they aren't geting financial incentives.

AMD paid DICE $10m to help develop Mantle, But they can't afford to keep paying for people to use it.


I think this quote answers your question.

The AMD Mantle website states that “Mantle makes game development easier.” If this is so, how does Mantle and the Mantle API make game development easier for developers?

The Mantle API is able to be backed with a much simpler, faster driver. The increased efficiency means that we no longer need to add as much complexity to our rendering pipeline to achieve the required level of performance. Mantle’s simplicity also means that it will probably be more stable over the long run. We anticipate that the bug rate for Mantle to be lower than previous systems we worked with.

This has been said many times by other Developers, Mantle is very simple to implement and use.

I also think CryTech are now going to help AMD stitch it into Linux, another poke in the eye to Microsoft.

 

[humbug]

That's one reason I see why I've see some reviewers expressed their worries that dx12 could potentially become a matter of "jacks of all trade, master of none". I mean I don't blame them; look at GFWL when M$ tried to lump Xbox and PC gaming together...then we have Windows 8 with M$ trying to lump tablets/touchscreen users and PC keyboard and mouse users together. Now with Windows 8.1, it just look more and more like Windows 7 without much of a difference other than some slight performance advantage.

IMO if anything, rather than hyping up and praising dx12 before even seeing it, PC gamers should sign a petition to M$ to keep dx12 for PC, and for everything else, work on something else to support them separately. If I have to pay for upgrading to new window version for the sake of dx12, I want no performance compromise made for the sake of benefiting other platforms that are irrelevant to me- we've seen launch of Window 8 with aim to support more things than one at the same time, and that did not went down well.

DirectX, or at least DX11 is very difficult, time consuming and clumsy to use, its expensive Developing games for DirectX

Mantle is nice, simple and cheap, what i would like to see is for Nvidia to make their own Mantle.

Yes that is then 2 API's, but both of them put together is probably easier and cheaper than using DX alone, i'm not convinced DX12 will fix the problems inherent to it.

There is no need for Microsoft and DX, AMD have proven they can make a much better one without Microsoft, i'm sure Nvidia could too.

 

[Phixsator]

Does it really matter if AMD is paying for the support or if devs are adopting mantle freely as long as it benefits the consumer?

 

[bru]

The increased efficiency means that we no longer need to add as much complexity to our rendering pipeline to achieve the required level of performance.

That line is a bit worrying, are they saying that they wont be doing the full rendering pipeline that DirectX will need?
Or as I hope that they are saying, 'this extra rendering pipeline that we are doing is less complex than the one for DirectX that we already have.'

With reference to what Humbug was saying, yes Microsoft have made a fair amount of noise over DirectX 12, will it live up to the hype?.....no it wont but in the same breath Mantle hasn't exactly lived up to the hype either, seven months after it was announced and it is still only available in 2 games both of which it was added after launch.

At least we know that DirectX12 will work with all titles in the future, as it would be suicide to launch a game without support for the most used graphics API. Unless you really think that DirectX 12 wont be just as popular as all the other DirectX's before it?

 

[Kaapstad]

With 4K gaming becoming common in the near future means that the Civ game developers really have very little choice but to use Mantle if they are going to bring out newer versions of the game with even bigger map sizes.

I am a big fan of Civ 5 and play on huge maps with extreme settings, unfortunately this means that todays hardware is pushed to it's very limits and it is only just possible to finish a game before huge graphics problems set in. I have some saved games where I have built huge empires that just won't load at all without losing half the tile improvements - this is a known bug that the game Devs have never been able to deal with. Hopefully now with Mantle on board Civ 6 will not have this problem.