• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD Halts Optimizations for Mantle API

Mantle still used for in house development.

Vulcan uses Mantle as a starting point for API development.

Future Mantle modules should be easy ports to Vulcan due to this.

Mantle forces development of DX12.


We all benefit from this. It's a bit **** when I see people posting crap about AMD and Mantle when they quite frankly shepherded us into a new age of gaming. Free performance upgrades for all, not just their own customers.


BTW, from the OP..

"DirectX 12 will feature a super-efficient command-buffer that scales across any number of CPU cores"

From what I have seen DX12 only seems to scale to 6 cores. Perhaps this is just a cap in development that could be opened up later or perhaps being flooded with extra power the GPU becomes a true bottleneck again? Anyone have an answer to this?
 
Last edited:
OpenGL Next was just a working name for the project. Vulkan shares nothing with opengl. its underlying systems are mostly Mantle while the interface is SpirV.



I am talking about the way mantle was designed. It was not made to specifically work only on GCN. AMD just never got to the point where they gave out an SDK for other IHV's to use and make a mantle driver. And so for political reasons AMD ended up making Mantle part of Vulkan. etc.

Rubbish, AMD design a low level API for GCN, nothing more. If other peple had to make massive changes to get it to work on different hardware then that functionality was never in Mantle.


Volkan is the evolution of OpenGL, which was why it was claled pen GL Next, the clue is in the anme. The first shading langiage support is going to be the OpenGL shading langauge (GLSL).

Vulkan has used Mantle source code but it is now very different to mantle, Khronos has said that themselves.
 
Mantle still used for in house development.

Vulcan uses Mantle as a starting point for API development.

Future Mantle modules should be easy ports to Vulcan due to this.

Mantle forces development of DX12.


We all benefit from this. It's a bit **** when I see people posting crap about AMD and Mantle when they quite frankly shepherded us into a new age of gaming. Free performance upgrades for all, not just their own customers.

Some corrections are needed here:

mantle code was donated to the Volkan project to expedite development. Volkan existed long before Mantle did, by a different name, openGl next. The codes is very different now, just ask the creators, Khronos.

Mantle did not force the developed of DX12, DX12 was in development before Mantle existed.



We do benefit from AMD contributing Mantle to project Vulkan, and we benefit from what AM,D Nvidia, Apples, Qualcomm, Intel, Valve, Epic, Oxide, ARM and many others have contributed to Vulkan and DX12.

It's a bit **** when I see people posting crap about AMD and Mantle when they think that AMD alone shepherded us into a new age of gaming.
 
And there it is, anti Nvidia bile.


Gameworks!



Some people assume this is the case,

The other possibility is Microsoft would have included it in Windows 10 anyway to shift more copied of the new OS.


Or not, there could have never been a DX12. MS doesn't have any incentive to throw money at DX12. But if there were a true lowlevel api that freed ppl from the windows ecosystem, now that would be a reason to combat a new api.
 
Last edited:
DX12 may well have been in development before Mantle but would it have gotten to us in the same form without Mantle? Personally, I think not. I hadn't heard of Low level APIs before Mantle other than a lot of chatter about why consoles perform as well as they do on very basic hardware.

Also OpenGL next would have been nothing like Vulcan without the development of Mantle and AMDs open source mentality.

I really think that AMD got the ball moving with this but I'd be happy to look over any material that says otherwise if you care to link me to any resources?
 
Gameworks!






Or not, there could have never been a DX12. MS doesn't have any incentive to throw money at DX12. But if there were a true lowlevel api that freed ppl from the windows ecosystem, now that would be a reason to combat a new api.

MS have a lot of incentive to throw money at DX12, they have a games console and a PC platform to use it on and have a shared API. Otherwise they might find them selves losing gamers to a PS4, of a SteamOS box, or an Xbox1 (that would still require a fast low level API). Valve was putting in a lot of effort in OpenGL and GlNext,
 
DX12 may well have been in development before Mantle but would it have gotten to us in the same form without Mantle? Personally, I think not. I hadn't heard of Low level APIs before Mantle other than a lot of chatter about why consoles perform as well as they do on very basic hardware.

Also OpenGL next would have been nothing like Vulcan without the development of Mantle and AMDs open source mentality.

I really think that AMD got the ball moving with this but I'd be happy to look over any material that says otherwise if you care to link me to any resources?


Yet again, NVidia, Microsoft, Intel, ATI and others were talking about a low level multi-threaded APi resembling DX12 back in 2005. Low level APIs are absolutely nothing new, all graphics APIs used to be low level and every manufacturer had its own incarnation. Nvidia had its own low level API, as did 3DFX, Renditon, S3, Trident, 3DLabs, Matrox.



OpenGL Next was planned to be very similar in the main concepts of Mantle: drop legacy support, provide low level access, multi-threaded buffer queues, and an API design from the ground up around programmable GPUs. That is why they called it GL Next, next generation Open GL.

Mantle simply provided a code base that already implemented many of the design ideas and paradigms, saving time getting the project moving faster. for that AMD deserve a big thanks but don't for 1 second think that Vulkan wouldn't exist or would be radically different.
 
Yet again, NVidia, Microsoft, Intel, ATI and others were talking about a low level multi-threaded APi resembling DX12 back in 2005. Low level APIs are absolutely nothing new, all graphics APIs used to be low level and every manufacturer had its own incarnation. Nvidia had its own low level API, as did 3DFX, Renditon, S3, Trident, 3DLabs, Matrox.



OpenGL Next was planned to be very similar in the main concepts of Mantle: drop legacy support, provide low level access, multi-threaded buffer queues, and an API design from the ground up around programmable GPUs. That is why they called it GL Next, next generation Open GL.

Mantle simply provided a code base that already implemented many of the design ideas and paradigms, saving time getting the project moving faster. for that AMD deserve a big thanks but don't for 1 second think that Vulkan wouldn't exist or would be radically different.

Talk is cheap. They had 8 years to produce before mantle came around yet nothing happened. Mantle comes and we have action within a year. Yea nothing to do with Amd forcing there hand.
 
[sarcasm mode] AMD created Mantle, AMD did this, AMD did that, AMD did everything. [/sarcasm mode]

Of course the reality of things is probably more a case of Dice created mantle and AMD then stuck their name on it, pretty much the same way they get outsiders to create all their software.

Maybe AMD did write Mantle all by themselves, maybe they did kickstart DirectX 12, maybe Vulkan is a direct use of the entire Mantle code. Of course the truth will probably never be known. :)
 
The reason it only worked on GCN parts was becasue all of the 6000 series and bellow still had Fixed function hardware in their VLIW architecture. GCN was amd's first general compute architecture with full memory control etc.

And its not a matter that it didnt work on intel or nvidia hardware. AMD just didn't get to the point where they released the SDK for others to write drivers for it.

ITs not that the api itself could not be used on different hardware. just that they never released it to other vendors.

Didn't AMD also say that they didn't know if Mantle would offer any benefits to GPUs with architectures other than GCN? Which suggests to me it was written for GCN.

As for Mantle pushing Microsoft to make DX12, could you not use the same logic to deduce that Nvidia had a programmable GPU architecture before AMD thus pushing AMD to come up with GCN? Without which AMD wouldn't have come up with Mantle.
So Nvidia are indirectly responsible for Mantle and by extension Vulkan and DX12!
 
Competition is what drives companies to do better and to try and beat their rivals, if they don't... then they lose customers and thus profits/money so it isn't too far fetched to say that mantle was the kick up the backside that MC needed for a low level API like DX 12.

Same logic applies to every thing in the tech. world.

And I agree with someone further up, until mantle, I hadn't read much of, if any low level API talk, certainly not to the same extent as mantle anyway.
 
Last edited:
Gameworks!






Or not, there could have never been a DX12. MS doesn't have any incentive to throw money at DX12. But if there were a true lowlevel api that freed ppl from the windows ecosystem, now that would be a reason to combat a new api.

Even if Microsoft don't give a sausage about gaming they still had plenty of incentive to include DX12 in Windows 10. Microsofts biggest problem of the last few years has been trying to find ways to get people to upgrade their old versions of Windows, DX12 is a good incentive for people to make the upgrade.
 
Even if Microsoft don't give a sausage about gaming they still had plenty of incentive to include DX12 in Windows 10. Microsofts biggest problem of the last few years has been trying to find ways to get people to upgrade their old versions of Windows, DX12 is a good incentive for people to make the upgrade.

What Microsoft want is for PC gamers to move to their consoles so they can charge us a recurring fee for the privilege to play the games.

They only pretend to care about PC gaming now because they no longer have the API monopoly to hold PC gaming back, if they didn't react to Mantle / Vulcan MS would start dying a slow but sure death on the enthusiast PC markets.

They had more than 10 years of low level API's from consoles to care about PC and never did.

And don't take it from me, Game developers have been saying the same thing for a lot of years, long before Mantle.
 
Last edited:
Didn't AMD also say that they didn't know if Mantle would offer any benefits to GPUs with architectures other than GCN? Which suggests to me it was written for GCN.

As for Mantle pushing Microsoft to make DX12, could you not use the same logic to deduce that Nvidia had a programmable GPU architecture before AMD thus pushing AMD to come up with GCN? Without which AMD wouldn't have come up with Mantle.
So Nvidia are indirectly responsible for Mantle and by extension Vulkan and DX12!

GPU's were already heading towards general compute architectures. Just nvidia got there 2 years before AMD when fermi dropped. :P

which is why fermi can support DX 12 driver overhead reductions.
 
Talk is cheap. They had 8 years to produce before mantle came around yet nothing happened. Mantle comes and we have action within a year. Yea nothing to do with Amd forcing there hand.

Exactly. APIs like this were justified when dual core CPUs hit the market a decade ago. There was no hype or competition for anyone to do anything before AMD drew everyone's attention to it. Mantle opened up a floodgate of demand.
 
Exactly. APIs like this were justified when dual core CPUs hit the market a decade ago. There was no hype or competition for anyone to do anything before AMD drew everyone's attention to it. Mantle opened up a floodgate of demand.

And 8 years ago Microsoft, Nvidia,Intel, AMD started talking of multi-threaded DX. Back then there were more important changes and the single threaded nature didn't have a serious impact back then.

DX12 was in development before Mantle was ever publicly announced.
 
GPU's were already heading towards general compute architectures. Just nvidia got there 2 years before AMD when fermi dropped. :P

which is why fermi can support DX 12 driver overhead reductions.

Graphics APIs were already going towards multi-threaded lower-level with reduced overheads. Just AMD got there a few months before Microsoft made a public announcement and no one bothered following OpenGL Next news or Apple Metal for tat matter.
 
Back
Top Bottom