The AMD Mantle Thread

[HisDivineOrder]

Game developers don't agree, and is the major reason Mantle was created in the first place.

SOME game developers don't agree. Many more do agree. Tim Sweeney and Carmack both seemed very opposed to the idea. They're not exactly uninformed and before you say that was at an nVidia event, they also had the guy you worship as a god for building a Mantle version of BF4 there and he seemed to imply that he saw Mantle as a means to get Microsoft to improve DirectX.

All in all, that's not the most rousing endorsement. Even your own press release from AMD with Eidos people didn't have them directly announce a single game running Mantle. They suggested it was a great technology, but that's hardly the ringing endorsement a full-on Thief 4 running Mantle announcement would have been. They had no problem announcing Thief 4 integrating TrueAudio, but nothing except vague praise for the technology for Mantle. No game integration announcement. Hmmm...

Considering that Star Citizen will also run all the latest nVidia technology, the magnitude of it also running Mantle is dulled somewhat.

And Oxide is a new developer. A startup that needs capital. Ahem, AMD has a little spending money left from their 8 million reasons they gave EA to support Mantle with Frostbyte (which btw just lost two franchises from the division that uses Frostbyte (C&C which was cancelled and Need for Speed which is moving over from now on to the sports division which uses Ignite and not Frostbyte as its engine of choice)). I'm sure Oxide could use the fast capital to do some great stuff. Thus, an endorsement for a technology they won't even use next year and maybe not even the year after is a great way to get some quick cash for next to no commitment in the short term. By the time their game is close to release, you're likely to see Mantle be another forgotten AMD fad like TruForm, Stream, GPU acceleration of Havoc, or temporal AA.

Like I said, the big names in engine development meanwhile are either saying, "Wtf?!?" like Sweeney or Carmack (who you'd think would be the first in line to be begging for something like this if droves of developers were screaming for it) or suggesting it's just a clever way to light a fire under MS and DirectX's butt. Or the OpenGL committee's butt. Whichever butt. A butt will be inhaling smoke in this world.

Eh. The most damning aspect of all this is that AMD's "big" announcement around Mantle was three middling announcements. One from Eidos that didn't even confirm a single game supporting Mantle, one from a studio in dire need of startup money, and one that's supporting everything high end on PC which is not even a full game and may not be for quite a while.

This is the HUGE news they have to show in this press release. I mean, I read that and expected Crytek or Unity support. I got... that? I actually thought, "That's it?"

From the way Sweeney was railing on it, I doubt Unreal's supporting it any time soon. Crytek seems more likely to given their help in building the Ruby demo, connection to Star Citizen, and in general their tie to AMD over the last year with Crysis 3, but ehhhh... what is Crytek doing in the near term? Is there a Crysis I missed? Homefront 2, right? That's been under development a long time and I doubt it could be given a Mantle add-on at the last second that did much for it...

Plus, the most damning thing of all is that AMD doesn't just have to get engines supporting Mantle. They also have to convince publishers and then developers that adding in Mantle code in addition to the DirectX code they will absolutely have to use is going to be worthwhile. They're going to have to convince publishers that won't even authorize patches for buggy games and developers who don't bother to add 1080p+ resolutions, high resolution textures, FOV sliders, bindable keys, or proper PC online functionality to support a second graphics API through QA, support, patches, and updates. All for... "better graphics performance" for a group of AMD owners that include either 7xxx series discrete cards or greater (from the last two years) or APU's that are as yet unreleased (Kaveri). Mantle won't benefit anything AMD older than that. Not 4xxx series, not 5xxx series, not 6xxx series, or any desktop or notebook conventional APU yet released by AMD (besides Temash and Kabini, though they're low power tablet APU's). Not to mention Mantle won't support Intel graphics or nVidia graphics of any kind.

So these publishers and developers who don't even bother to add in useful functionality to improve PC ports in rather important ways that should barely cost any time or money to add in... are suddenly going to throw in support for a low level API they have to spend comparatively more money to use for... a gain to a very small group of users?

In so many ways, Mantle seems like a really tough sell. Hence why I think the DICE guy is fine with supporting Mantle as a way to light a fire under Microsoft's and the OpenGL committee's collective butts. Because that's all it is likely to have a shot of doing.

 

[desprado]

Good question. A fact is, they explicitely said they could cut cost. They could say anything else but thats what was drawn into attention. So how?

Here is my take why:
Programming for mantle and get okey performance is extremely cheap. Is far far more simple than dx. Especially for the heavy classes/layers on the lower level on the gfx side.
If you dont demand the last level of performance on dx you save a ton of proramming time.
The total calculation of the above is saved cost in this case for the content side.

Add to the above that if nv want better dx perf they can just pay for it. And to the degree they want. Its their business.

But it simply shows the benefit of mantle to dx.
Its both cheaper ro program for.
You get better performance for less.
Porting is a walk in the park when the job is done.

Go imagine you were a developer. I would take that route everyday. Who wouldnt want mantle instead of dx. Dx is the old world focusing on hardware and lots of programming ressources for the low level to get good performance. Be it pc or consoles. For the gaming market that is bad use of ressources. Why should they care.

The story that dx was one api is off. Its a complicated, heavy and fragmented huge pile of rocks remisent of the ibm pc era.

Time goes on. The situation with similar pc and consoles changed it.
Its ready to die of age.

Here is big fact which some people are getting hard to understand that AMD dont rule the gpu market or CPU
Second Fact It is not Nvidia that has to pay for DX but it will be AMD that has to pay for Mantle because Nvidia has around 60% of overall PC user and Intel 5 to 7% and if dev make Mantle first priority than Dev will be only targeting 30% of PC user.

 

[desprado]

SOME game developers don't agree. Many more do agree. Tim Sweeney and Carmack both seemed very opposed to the idea. They're not exactly uninformed and before you say that was at an nVidia event, they also had the guy you worship as a god for building a Mantle version of BF4 there and he seemed to imply that he saw Mantle as a means to get Microsoft to improve DirectX.

All in all, that's not the most rousing endorsement. Even your own press release from AMD with Eidos people didn't have them directly announce a single game running Mantle. They suggested it was a great technology, but that's hardly the ringing endorsement a full-on Thief 4 running Mantle announcement would have been.

Knowing that Star Citizen will also run all the latest nVidia technology kinda dulls the magnitude of it also running Mantle.

And Oxide is a new developer. A startup that needs capital. Ahem, AMD has a little spending money left from their 8 million reasons they gave EA to support Mantle with Frostbyte (which btw just lost two franchises from the division that uses Frostbyte (C&C which was cancelled and Need for Speed which is moving over to the sports division which uses Ignite instead)). I'm sure Oxide could use the capital to do some great stuff. Thus, an endorsement for a technology they won't even use next year and maybe not even the year after is a great way to get some quick cash for next to no commitment.

Like I said, the big names in engine development meanwhile are either saying, "Wtf?!?" like Sweeney or Carmack (who you'd think would be the first in line to be begging for something like this if droves of developers were screaming for it) or suggesting it's just a clever way to light a fire under MS and DirectX's butt. Or the OpenGL committee's butt. Whichever butt. A butt will be inhaling smoke in this world.

Eh. The most damning aspect of all this is that AMD's "big" announcement around Mantle was three middling announcements. One from Eidos that didn't even confirm a single game supporting Mantle, one from a studio in dire need of startup money, and one that's supporting everything high end on PC which is not even a full game and may not be for quite a while.

This is the HUGE news they have to show in this press release. I mean, I read that and expected Crytek or Unity support. I got... that? I actually thought, "That's it?"

From the way Sweeney was railing on it, I doubt Unreal's supporting it any time soon. Crytek seems more likely to given their help in building the Ruby demo, connection to Star Citizen, and in general their tie to AMD over the last year with Crysis 3, but ehhhh... what is Crytek doing in the near term? Is there a Crysis I missed? Homefront 2, right? That's been under development a long time and I doubt it could be given a Mantle add-on at the last second that did much for it...

Plus, the most damning thing of all is that AMD doesn't just have to get engines supporting Mantle. They also have to convince publishers and then developers that adding in Mantle code in addition to the DirectX code they will absolutely have to use is going to be worthwhile. They're going to have to convince publishers that won't even authorize patches for buggy games and developers who don't bother to add 1080p+ resolutions, high resolution textures, FOV sliders, bindable keys, or proper PC online functionality to support a second graphics API through QA, support, patches, and updates. All for... "better graphics performance" for a group of AMD owners that include either 7xxx series discrete cards or greater (from the last two years) or APU's that are as yet unreleased (Kaveri). Mantle won't benefit anything AMD older than that. Not 4xxx series, not 5xxx series, not 6xxx series, or any desktop or notebook conventional APU yet released by AMD (besides Temash and Kabini, though they're low power tablet APU's). Not to mention Mantle won't support Intel graphics or nVidia graphics of any kind.

So these publishers and developers who don't even bother to add in useful functionality to improve PC ports in rather important ways that should barely cost any time or money to add in... are suddenly going to throw in support for a low level API they have to spend comparatively more money to use for... a gain to a very small group of users?

In so many ways, Mantle seems like a really tough sell. Hence why I think the DICE guy is fine supporting Mantle as a way to light a fire under Microsoft and the OpenGL committee's butts. Because that's all it has a shot of doing.

Even if it has 20% of gain in performance still dev will not go for it beacuse AMD dont have market or majority.

 

[3DVagabond]

Even if it has 20% of gain in performance still dev will not go for it beacuse AMD dont have market or majority.

Are you saying that it must have more than 20% to even be worth it? Makes sense to me.

 

[Saylick]

Even if it has 20% of gain in performance still dev will not go for it beacuse AMD dont have market or majority.

I think the logic here is backwards. Mantle should boost sales for developers who use it vs. if they don't.

Let's say that the minimum target for enjoying a new game is 30 FPS @ high settings. Let's say that out of every 10 people who have an AMD GPU with GCN, only 3 are capable of playing this game at 30 FPS @ high settings, chiefly because most people don't have fancy graphics cards to begin with. In other words, only 30% of AMD GPU owners are capable of hitting 30 FPS @ high settings.

If Mantle gives 20% more FPS, then all of a sudden the people who were originally only capable of hitting 25 FPS @ high settings w/o Mantle are now able to hit 30 FPS @ high settings w/ Mantle. Consequently, you see more people being able to pass the minimum requirements for said game and thus you see increases in sales. This might translate into 1 or 2 more people out of the original 10.

In theory, Mantle should help out those with weaker hardware more than those who have powerful hardware. My example supports this line of thinking as the original 3 people who could play the game at 30 FPS @ high are now producing 36 FPS. Either way, they would have enjoyed the game regardless of Mantle. Better yet, they may potentially be able to kick up the settings some more and enjoy more eye candy. Either way, it's a win.

Long story short, if Mantle offers enough of a performance increase to generate more sales than is lost in developing the game to support Mantle, it will be worth it.

 

[desprado]

Are you saying that it must have more than 20% to even be worth it? Makes sense to me.

of Course it has to be if it only 15% than it is major fail because u can also do that by driver.

Yup they dont have market or the majority of PC user.They have to convince not just Dev but Pc user to shift from Nvidia to AMD.

 

[desprado]

I think the logic here is backwards. Mantle should boost sales for developers who use it vs. if they don't.

Let's say that the minimum target for enjoying a new game is 30 FPS @ high settings. Let's say that out of every 10 people who have an AMD GPU with GCN, only 3 are capable of playing this game at 30 FPS @ high settings, chiefly because most people don't have fancy graphics cards to begin with. In other words, only 30% of AMD GPU owners are capable of hitting 30 FPS @ high settings.

If Mantle gives 20% more FPS, then all of a sudden the people who were originally only capable of hitting 25 FPS @ high settings w/o Mantle are now able to hit 30 FPS @ high settings w/ Mantle. Consequently, you see more people being able to pass the minimum requirements for said game and thus you see increases in sales. This might translate into 1 or 2 more people out of the original 10.

In theory, Mantle should help out those with weaker hardware more than those who have powerful hardware. My example supports this line of thinking as the original 3 people who could play the game at 30 FPS @ high are now producing 36 FPS. Either way, they would have enjoyed the game regardless of Mantle. Better yet, they may potentially be able to kick up the settings some more and enjoy more eye candy. Either way, it's a win.

Long story short, if Mantle offers enough of a performance increase to generate more sales than is lost in developing the game to support Mantle, it will be worth it.

u are not getting but here some user are saying that soon DX will die it is rubbish.
For example DICE start using only Mantle as some member are saying than only 20% of PC user will be entertained not all AMD use because many people still have HD 6xxx and HD 5xxx.It will not be downlink for Nvidia but mostly for Dev.If they starting doing this than Bankruptcy will be very near.
Some member said Nvidia will have to pay for DX hell it other way around Dev will for Majority of Pc user not minority.AMD is paying for mantle and still have to for mantle for each game that is a simple fact.

 

[Saylick]

u are not getting but here some user are saying that soon DX will die it is rubbish.
For example DICE start using only Mantle as some member are saying than only 20% of PC user will be entertained not all AMD use because many people still have HD 6xxx and HD 5xxx.It will not be downlink for Nvidia but mostly for Dev.If they starting doing this than Bankruptcy will be very near.
Some member said Nvidia will have to pay for DX hell it other way around Dev will for Majority of Pc user not minority.AMD is paying for mantle and still have to for mantle for each game that is a simple fact.

Kind of hard to understand what you are saying but I'll try...

I'm not saying that DX is going to die. As a matter of fact, I think that Mantle will simply just coexist with DX and OpenGL.

In other words, if you have an AMD GPU, you will be either running DX code or Mantle code, depending on whether or not the game supports Mantle. If it does, you run Mantle code. If it doesn't, you can always fall back on DX or OpenGL. Simple as that. If you own an Nvidia GPU, your only options are DX or OpenGL. Nothing changes if you own an Nvidia GPU.

With that said, I do not believe DICE will ONLY use Mantle. Just because AMD is supporting DICE in developing and integrating Mantle with Frostbite doesn't mean they will only code their games in Mantle. Such a move would be idiotic for DICE as you would be limiting your game to consumers who have Mantle-capable GPUs. DICE would be shooting themselves in the foot if they only coded in Mantle. DICE loses money, AMD loses money, and consumers lose because only a handful of people would be able to play this game.

As for AMD going bankrupt, it comes down to whether or not the following occurs:
1) AMD pays a developer to integrate Mantle into their game.
2) Developer makes game with Mantle, which means GCN GPUs run faster in this game, which allows more gamers to enjoy this game because they can run it on weaker hardware than before.
3) Developer makes more $ because more people can play this game and thus they buy this game.
4) Other developers begin wanting to get involved with Mantle because they want to make more money as well.
5) More and more games start to support Mantle (along with DX/OpenGL of course).
6) Because so many new games support Mantle, there is more incentive to buy an AMD card because a lot of new games will benefit from a Mantle-supported card.
7) AMD gets more sales and thus makes more money.

If this doesn't play out, then yes, Mantle will be a bust. The difficulty lies in the fact that AMD will have to invest time and money in the beginning with the hopes of generating more sales in the long run. We will have to wait a bit to see how this will play out. Hindsight 20-20; you will have your answer in 2015.

 

[VulgarDisplay]

Why do people continue to think Carmack and Sweeney are relevant to PC games? They aren't.

Smart guys, but that does not change the fact that hey are washed up, and incapable of releasing quality PC titles.

 

[desprado]

Kind of hard to understand what you are saying but I'll try...

I'm not saying that DX is going to die. As a matter of fact, I think that Mantle will simply just coexist with DX and OpenGL.

In other words, if you have an AMD GPU, you will be either running DX code or Mantle code, depending on whether or not the game supports Mantle. If it does, you run Mantle code. If it doesn't, you can always fall back on DX or OpenGL. Simple as that. If you own an Nvidia GPU, your only options are DX or OpenGL. Nothing changes if you own an Nvidia GPU.

With that said, I do not believe DICE will ONLY use Mantle. Just because AMD is supporting DICE in developing and integrating Mantle with Frostbite doesn't mean they will only code their games in Mantle. Such a move would be idiotic for DICE as you would be limiting your game to consumers who have Mantle-capable GPUs. DICE would be shooting themselves in the foot if they only coded in Mantle. DICE loses money, AMD loses money, and consumers lose because only a handful of people would be able to play this game.

As for AMD going bankrupt, it comes down to whether or not the following occurs:
1) AMD pays a developer to integrate Mantle into their game.
2) Developer makes game with Mantle, which means GCN GPUs run faster in this game, which allows more gamers to enjoy this game because they can run it on weaker hardware than before.
3) Developer makes more $ because more people can play this game and thus they buy this game.
4) Other developers begin wanting to get involved with Mantle because they want to make more money as well.
5) More and more games start to support Mantle (along with DX/OpenGL of course).
6) Because so many new games support Mantle, there is more incentive to buy an AMD card because a lot of new games will benefit from a Mantle-supported card.
7) AMD gets more sales and thus makes more money.

If this doesn't play out, then yes, Mantle will be a bust. The difficulty lies in the fact that AMD will have to invest time and money in the beginning with the hopes of generating more sales in the long run. We will have to wait a bit to see how this will play out. Hindsight 20-20; you will have your answer in 2015.

Bro u still dont get i am not saying that AMD will go bankrupt i am saying Dev will get in loss to target minority as making it first priority.If did majority of User and Pc market mantle will be a total waste.

 

[Noctifer616]

Bro u still dont get i am not saying that AMD will go bankrupt i am saying Dev will get in loss to target minority as making it first priority.If did majority of User and Pc market mantle will be a total waste.

Last time I checked BF4 had DirectX before it had Mantle. So, do you mind explaining how exactly are developers prioritizing Mantle over DirectX or OpenGL?

Also, if it's true that Mantle only needs to be supported by the game engine, and support doesn't need to be added to each game individually, then there is very little cost in adding Mantle support into a big number of games since a few engine have the majority of games on the market (aka Frostbite 3 will be used in more than 10 games, Mantle has to be added to the engine, not the game.).

 

[Saylick]

Bro u still dont get i am not saying that AMD will go bankrupt i am saying Dev will get in loss to target minority as making it first priority.If did majority of User and Pc market mantle will be a total waste.

Well, keep in mind that developers will get financial backing from AMD if they go with Mantle. Secondly, like I said previously, developers will not be giving their undivided attention to Mantle. DX and Mantle will coexist. More importantly, the quality of the DX code shouldn't be different than if they werent going to support Mantle. Again, a key point of Mantle is that you are able to port a good amount of console code to the PC, so in theory the Mantle version should take less time to develop than the DX path.

Long story short, developers WILL develop for DX, in the same way they have always done it, post-Mantle. The only thing that changes is that they can choose to do a Mantle version as well in addition to the standard DX version. This will take extra time (but not as much as the DX version) and it will cost money, but that's why AMD will help with money.

 

[Erenhardt]

I'm reading through AMD AMA. I know, straight from AMDs mouth bal, bla, bla... anyway:

Real numbers regarding Mantle will be published at the AMD Developer Summit running November 11-13 in San Jose. I will say that we are not undertaking such a mammoth effort to yield 3-4% performance--that would be a waste of time.

http://www.tomshardware.co.uk/forum/id-1863987/official-amd-radeon-representatives.html#11877195

 

[desprado]

Well, keep in mind that developers will get financial backing from AMD if they go with Mantle. Secondly, like I said previously, developers will not be giving their undivided attention to Mantle. DX and Mantle will coexist. More importantly, the quality of the DX code shouldn't be different than if they werent going to support Mantle. Again, a key point of Mantle is that you are able to port a good amount of console code to the PC, so in theory the Mantle version should take less time to develop than the DX path.

Long story short, developers WILL develop for DX, in the same way they have always done it, post-Mantle. The only thing that changes is that they can choose to do a Mantle version as well in addition to the standard DX version. This will take extra time (but not as much as the DX version) and it will cost money, but that's why AMD will help with money.

yes correct that i am trying to say but some member are so hyped and excited that they forget the fact that Mantle will use DX.So it is not a game changer for MS or any API.

 

[Saylick]

yes correct that i am trying to say but some member are so hyped and excited that they forget the fact that Mantle will use DX.So it is not a game changer for MS or any API.

Whether or not it becomes a game changer depends on how much of a boost Mantle provides.

If Mantle gave you twice FPS than the DX version, would you say that it's a game changer?

 

[desprado]

Whether or not it becomes a game changer depends on how much of a boost Mantle provides.

If Mantle gave you twice FPS than the DX version, would you say that it's a game changer?

If it had give twice FPS than the DX version than AMD would not pay every developer for using it and for each game.Mostly it will be only 20% and it is not magic and it is depended on Dx.

 

[Saylick]

If it had give twice FPS than the DX version than AMD would not pay every developer for using it and for each game.Mostly it will be only 20% and it is not magic and it is depended on Dx.

Good point. If Mantle was THAT good, I wouldn't be surprised if things flipped and developers started paying AMD to use it.

20% isn't bad though. Something is better than nothing.

 

[krumme]

Here is big fact which some people are getting hard to understand that AMD dont rule the gpu market or CPU
Second Fact It is not Nvidia that has to pay for DX but it will be AMD that has to pay for Mantle because Nvidia has around 60% of overall PC user and Intel 5 to 7% and if dev make Mantle first priority than Dev will be only targeting 30% of PC user.

100% console. And thats the difference.

 

[krumme]

Why do people continue to think Carmack and Sweeney are relevant to PC games? They aren't.

Smart guys, but that does not change the fact that hey are washed up, and incapable of releasing quality PC titles.

Everytime someone take old dinosaur like carmack here there is probably new talented developers having a laugh or get irritated. Come on we are past quake3 and the new young guys at eg Dice is rocking so hard. They think cross platform from higend machines to tablets and phones while at the same time giving an engine that gives games a superiority to eg unreal that is more like two generations ahead.

Doom.

 

[krumme]

With consoles and mantle amd have over 70% of the market for the comming games. Who wants to put ressources to finetune dx for last bit of perf then?

 

[krumme]

Does amd have to pay developers to program for mantle consoles?

Do you expect developers to program only for the dx side in the slow xbox?

How about bf4 on the consoles without the mantle layers and drivers amd provide?

Saying no to mantle is the same as not selling for the consoles. Its unavoidable. Its already happening in big numbers what matters is how close is the console programming to pc mantle eg how much is compatible and not only similar.

Who voice up for support have nothing to do with that. Thats only marketing, relation caring and pr.

Who want to confront jhh saying we use mantle? He is going to explode in his tshirt.

 

[desprado]

With consoles and mantle amd have over 70% of the market for the comming games. Who wants to put ressources to finetune dx for last bit of perf then?

I dont what u are saying but Multiplatform games now are first made on PC for example Watchdog,Thief,Witcher 3,BF4,AC 4,COD Ghost,Titan Fall e.t.c has beed developed on PC so what are talking about.

Xbox360 had AMD GPU though and Multi platform were ported from Xbox 360 not PS3 and did that change the fact or AMD pc market no.
Plz Mantle is not used in Console and sorry but MS and Sony will never use mantle as said in there statement.

 

[Noctifer616]

Plz Mantle is not used in Console and sorry but MS and Sony will never use mantle as said in there statement.

That doesn't mean that they have nothing in common with Mantle.

If you have two different low level API's for the same architecture, there has to be some kind of common code. Also, both the PS4 and XBONE API's were developed with the help from AMD.

Here is the statement from AMD that confirms what I say:

http://community.amd.com/community/...0/17/the-four-core-principles-of-amd-s-mantle

In short, Mantle is a new and better way to bring the code developers are already writing for next-generation consoles to life on the PC. It achieves this by being similar to, and often compatible with, the code they are already writing for those platforms.

 

[ShintaiDK]

Xbox One uses DirectX and will be a direct copy to PC with DirectX. Mantle however, more work from more or less scratch. The PS4 uses OpenGL, again, not Mantle or close to Mantle either. Its not long ago some people claimed Mantle was the low level API in consoles. And we all know how that vent.

 

[desprado]

Xbox One uses DirectX and will be a direct copy to PC with DirectX. Mantle however, more work from more or less scratch. The PS4 uses OpenGL, again, not Mantle or close to Mantle either. Its not long ago some people claimed Mantle was the low level API in consoles. And we all know how that vent.

And some people dont now that Mantle will also use Dx as well so it is end of story for being independent API.Now even PS4 is also using Dx11 versions.I remember sony use to bash Dx and now they are using it.