Our reader ‘Russell Collins’ has informed about an interesting interview of MaximumPC with Oxide Games about Mantle. When asked about Mantle and Nvidia cards, Oxide Games’ Dan Baker claimed that Nvidia will have to adapt – one way or another – if they want to stay relevant.
Now before jumping to early conclusions, this does not mean that Nvidia will have to support Mantle. Baker’s statement means that if Nvidia does not support Mantle, a new universal API has to be made to take advantage of what Mantle is currently doing.
“Mantle does detract from other platforms, and we are already seeing a big dialogue in terms of what future APIs should look like. Mantle is kind of the disruptive technology that gets everyone rethinking things. Whether this means a new version of OpenGL, or a new version of D3D, we can’t say. But it is clear that they will have to adapt if they want to stay relevant. Some of us have been screaming for change for years. The arguments we got in the past were: 1) it couldn’t be faster 2) it would be too hard to use, and 3) we have enough performance, so more isn’t useful. We wanted to show that all three of these things are provably false.”
A lot of people believe that OpenGL is capable of delivering performance similar to Mantle via some extensions, however Baker believes that OpenGL has essentially all the core problems of Direct 3D.
“OpenGL has essentially all the core problems of D3D, except that one can add extensions to it. The main difference between OpenGL and D3D is that D3D made an attempt to be threaded, and failed, whereas OpenGL has not yet attempted it. One question is whether it is worth building a new API or making a bunch of extensions to an existing API. You can get some mileage out of making extensions, but extensions can’t bridge things like being able to use multiple CPUs. Also, at some point it’s cleaner and easier just to hit the reset button, rather than throw yet another feature in a fairly big API. Believe it or not, Mantle is actually easier to support than OpenGL. OpenGL has many unobvious pitfalls and traps, whereas Mantle really doesn’t. “
It will be interesting to see whether other developers will embrace Mantle or not. For instance, Flying Wild Hog told us that although Mantle looks interesting, its usage – at least for now – is quite limited as it’s GCN only.
We already know that DICE wants Mantle to be multi-vendor, so it will be interesting to see what the team behind this new API will achieve. 2014 will be an interesting year for Mantle, and we’ll be sure to keep you posted!
John is the founder and Editor in Chief at DSOGaming. He is a PC gaming fan and highly supports the modding and indie communities. Before creating DSOGaming, John worked on numerous gaming websites. While he is a die-hard PC gamer, his gaming roots can be found on consoles. John loved – and still does – the 16-bit consoles, and considers SNES to be one of the best consoles. Still, the PC platform won him over consoles. That was mainly due to 3DFX and its iconic dedicated 3D accelerator graphics card, Voodoo 2. John has also written a higher degree thesis on the “The Evolution of PC graphics cards.”
Contact: Email
Being using it since it first releases for BF4 and I can see its strengths. From minimum 45 to go minimum 65 its amazing. Being able to enjoy Shanghai and Paracel Storm on my 7950 is amazing. At this early stages it shows huge potential. Its not even optimized for my card yet. For me its more stable than dx11.
Its for Us users of AMD/ATI GPU & CPU on mine x6 1100BE 3.83Ghz and XFX 7970 GHz in 2k + AA i have smooth gameplay 🙂 from 1920:1440 no AA (all Ultra) i’d like to see it in Crysis 3
Star Citizen looks like a beast, any help is welcome, looking forward to see what Mantle can do when it’s up and running properly.
Someone downvoted Star Citizen, Mantle? Some proper plebs use this site lol.
This site is lacking a sense of humour, the articles are making the fanboys worse.
We’re getting old Sean, let us end it now (:
Or it could mean that Mantle becomes irrelevant because NVIDIA/Intel don’t care and they’re the biggest GPU/Integrated chipset providers to the market. I think NVIDIA have more of a problem with Mantle because it’s from their competitor and AMD already out of the gate have an advantage from it.
Mantle is exciting but I have to wonder what happened when AMD supposedly invited NVIDIA or Intel to the table. It could go either way, Mantle could die without the support of Intel and NVIDIA no matter how much devs like it, Microsoft made DirectX relevant because Windows was the only OS around on PCs, they integrated it into Windows, you can’t use that leverage with Mantle.
I become famous and this is how you treat the article? Tis tis
If NVIDIA start losing money on their GPU side due to AMD/Mantle then Oxide might have some weight behind what they claim, but it’s just hot air from a company that wants to become relevant via it’s engine, which is using Mantle.
Seems like a lot of posturing from Oxide to me, on behalf of AMD.
I think that is what we want as consumers though right? I am an Nvidia fan, but Nvidia losing money forces their hand to innovate by accepting Mantle or bringing something new to the conversation. The climate in the gaming industry is perfect for this to happen. More Devs self publishing, Mantle is making waves, Steam OS is causing chatter. People are ready for change.
As long as AMD cards can mine, they won’t be leaving the market any time soon. When gpu mining is past-tense, that would be interesting to view.
Maybe in the sense on what we’re aware of what they’ve presented to us thus far. But that’s not to say that they don’t know they’re shit, or they have a massive understanding of all of this.
Good point. I makes more of a case now that Mantle was only designed to push the conversation forward.
John Carmack says OpenGL can now do what Mantle can do, Oxide doesn’t agree. I’d like to see some OpenGL benchmarks doing the same just to prove a point.
I have a hard time with anyone disagreeing with Carmack. lol
Yes because Carmack is consistent in what he says and has been saying. Who’s Oxide again? 🙂
Id’s John Carmack has warned against developers leaning to heavily on hopes for APIs to magically solve slow-running engines. Despite the softwares being developed to allow developers to have a more direct interface with graphics cards, he said that out of the box “any API, even if you made all the API overhead vanish, it doesn’t make that much of a difference.”
http://www.pcgamesn.com/john-carmack-any-api-even-if-you-magically-made-all-api-overhead-vanish-it-doesn-t-make-much-difference
But it makes a difference and the numbers are here. He is an Nvidia fanboy for sure. He created an engine (Rage) that was running like crap on AMD cards and it didn’t look that good. He is overrated. He stacked in the past. Look all his work. After Doom nothing. Doom for me was phenomenal for its time.
Nope, the numbers backup what Carmack said. There are very small gains on systems which are CPU or GPU limited, the API overhead is where exactly?
Gains on mid-range systems are no better than a driver optimisation and the biggest gains come from expensive top end hardware.
Its not for every engine. For RTS engines it can do very well even with a beasty system. And my 8320 at 4.3 with a 7950 at 1000 is not a powerful system? We saw such big improvements in hardware over the years but the software lucks behind. The software must compliments the hardware. Check the crossfire improvements with an i7. And its not only fps. Frosbite implement there custom multigpu implementation and the numbers speak for them self’s.
Kyle from Hardocp:
Since it allows low-level access to GCN GPUs, it supports, well, anything a programmer wants to implement. The shaders on the GPU are agnostic, they do whatever the programmer tells them to. Mantle allows access to them in more bare metal way than DX does, there is no HAL (Hardware Extraction Layer) to go through. So it isn’t a matter of support of features, it’s a matter of what the programmer decides to do, and the door is open to any possibility. You can’t compare “feature sets” between DX and Mantle, because Mantle operates in a very different way than DX. It allows the programmer to use those shaders to do any 3D effect they want. You won’t see AMD specific features removed, and you won’t see 3D effects removed with Mantle, it can improve these things in performance and IQ potentially. Something like Tessellation really has the potential to be done better with Mantle since it will know exactly how the GPU its running on works, instead of having to go through DX.
If the overhead is small, then why I get huge boost in fps, especially in the minimums (about 50%)? Gains are there, it just depends under what conditions are you searching/testing for them.
I’d like to see more games with more benchmarks. Battlefield 4 does utilise the CPU cores better than most games.
I would say that’s even a bigger plus for Mantle really. 🙂 Under Mantle the video card is better utilized (higher temps and much higher vRAM usage). There is one thing to use 99% of the gpu and another to use it efficiently, because it’s about the same usage, but highly different fps. Also the game is more responsive, less input lag.
In the end, to say the difference is to small, it’s not accurate at all.
Bf4 is using all my core 8 in total @ 70% in Mantle.
It does in Direct X as well that’s the point, other games just don’t initialise the cores very well and don’t even use more than 4 cores *cough AC4 Black Flag *cough.
My next CPU will probably be an AMD FX 6300.
The efficiency is the point. Go for the 8320.
Far from overrated. He is a visionary and innovator. That is why he is at Occulus now. ID stopped innovating, so he moved to where he could innovate. And your right Doom was phenomenal for it’s time. And Doom 3 and Rage. And as far as an Nvidia fanboy, let me ask you this, if you were a developer and you asked AMD for some technical assistance fitting their architecture to your game engine and the reply was ” all the answers you need are in the documentation ” and you call Nvidia for the same assistance and they say ” Sure we can help with that. We are sending some people to help you until you get off the ground “, who would you be talking about more? Someone who told you to go screw, or who helps you out? John explained this years ago as being his reason for adopting Nvidia tech.
Exactly right, it’s what I’ve been saying about PhysX and why game devs support it in their games, they get the tech support and software that comes with it, with DirectX they probably have to go read docs on Microsoft’s website.
this subject is making me think of my old family station wagon. Everyone loved it. Had pretty decent constuction. comfy seats. However after a few years things started changing in the automotive industry. 2 things stand out. Airbags and the change in the air conditioning standard. We converted to those things( at a heft price. I was stubborn. I loved that car. lol) after a while nickle and dime-ing the old beast, it got to the point that not everything could be retro fitted to and it became more costly to keep it then get a new car. That is where we are with APIs and OSes. Time to move on and stop treating APIs like Mr. Potato Head and his bucket of parts!!! We need a better potato, not just different parts! lol
Actually, the guy said you can get away with using OpenGL, but only up to some point. Even so, you need to write specific extensions for each party, it’s not just general stuff in OpenGL and you get the same benefits. If you take into consideration the statement that Mantle it’s easier to use than Opengl and dx, probably some individual APIs for nVIDIA and AMD doesn’t look that SF and roll back to general api for backwards compatibility and Intel. After all, they are all writing the code already for a lot of apis.
PS: An “authority” can be wrong from time to time.
Yes but OpenGL is used on consoles, smart phones and other OS’s other than Windows so the easier to write for is a matter of platforms now not just Microsoft’s walled garden.
In spite of that, there are only major D3D games out there at the moment. Leaving Rage aside, I can’t recall of any other game on opengl. Moreover, there are differences between Rage and oxide’s demo (FPS, relatively small setting vs. huge scale, lots of AIs), Oxide’s engine seems to be in league all of it’s own – even compared to the rest of the engines that are on market today.
Worth mentioning, would be the fact that in today’s environments, it’s not too difficult to support multiple APIs, like Baker said. So having Mantle for AMD, X for nVIDIA and opengl as standard or DX, wouldn’t be such a difficult thing to do and of course, not all games would need such love. It all depends on the complexity of the world. However, it would mean that games that DO require extra horsepower, would run much, much better on AMD and nVIDIA (also Intel if they have the HW and SW up to speed) than on DX or OpenGL.
you never know, valve said their games run better on opengl than dx so they made steamos and so far even their games run better on windows.
Party because of the Linux kernel though.
Great article. Where I don’t think Nvidia will adopt Mantle, I think Nvidia does have a responsibility to bring something significant to the table. With thier new partnership with Intel, Nvidia could put some serious pressure on Microsoft to revamp / reboot Direct X. Really, Microsoft is in a tough position right now with gamers and this could start turning the ship around for them. How much more does MS have to see before it wakes up??? If you think about it, Nvidia, Microsoft and Intel are giants, but are all in danger of irrelevance. Times are ah changin’!
Exactly, if it takes Mantle coming along to shake things up and getting a more efficient DX, then we all win. Competition is a great thing, now if only AMD could challenge Intel in the chips dept, they’re making a start with HSA I guess.
who’s oxide games ? lol
atleast release a god damn game and then talk every day
They have a game engine that wowed a lot of journalists.
That’s something…I guess.
it was sarcasm dude, i downloaded starswarm (tech demo) in steam when it came out
They are hyping mantle because their game engine uses it for Star Swarm.
So let me ask this, if Mantle really takes off, how do you think Microsoft and Nvidia will respond? Will MS dig in and just offer another layer of crap to DX to try and pacify us? Will NVAPI take off into something more??
Here is there answer tiled resources. http://blogs.windows.com/windows/b/appbuilder/archive/2013/10/14/raising-the-bar-with-direct3d.aspx
Cryofreeze me and wake me up in 2-3 year where games do not suck like they do now and use mantle for better perfomance.
Oxide Games
NVIDIA, you need to adapt or become irrelevant.
NVIDIA
Who are you?
Oxide Games
We made this cool engine that can render a billizon things on screen with AMD’s Mantle.
NVIDIA
Who are you again?
Oxide Games
We made our game engine faster with Mantle.
NVIDIA
We have lots of software, support and goodies for people like you, come and play with our PhysX.
Oxide Games
No, AMD is paying us a lot to say you’re should adapt or become irrelevant because our engine uses Mantle.
NVIDIA
Resistance is futile, we have John Carmack and Tim Sweeny, they’re more relevant than you will ever be and they make game engines, best in the business.
Oxide Games
Well you just wait, we’ll get more people to make us relevant.
NVIDIA
Who are you again?
(You guys voting this down are too serious, lighten up)
“billizon things on screen” :))
it turns out it’s only 5,000 starship on screen at max, they said up to 10,000 ! but where is other 5,000 ? they said 10,000 if it’s something new, i can play totalwar with 40,000 or even twice more units on the screen not 2D space map with 4types of starship but with 3D enviorment,grass,trees,fires (lots of alpha effects) more than 10types of unites with Diffrent armor and weapon types,motion captured animations,buldings, voices, even scenarios and tactics yet they saying 10,000 unite if no one knows thinks it’s for the first time in history. those people with thier mouths
He’s too serious and trying to make something out of nothing.
We are Nvidia and we can create the The world’s greatest virtual concrete slab just to make AMD look bad.
http://techreport.com/review/21404/crysis-2-tessellation-too-much-of-a-good-thing/2
Or the fact that AMD’s tessellation performance isn’t good. It’s worth noting that Crysis 2 tessellation is pre-tessellated and imported into CE3 at the time and if you use LOD tessellation then it’s not as issue because it works like LOD models.
AMD can do the same with OpenCL effects, ATI did it with pixel shader 2.0, on NVIDIA cards PS 2.0 performed bad.
First off, I thought Mantle does NOT require GCN and will thus work with NVIDIA cards?
Regardless, Oxide Games is stupid. NVIDIA is more than relevant…in fact, it’s AMD that has to play catch-up.
I don’t think NVidia will adapt mantle
When something better comes along it always wins. That’s why we have firewire ports on all of our computers and all of our sound files are in FLAC, while USB and MP3 have all but disappeared.
Oxide games can go suck AMD D some more… They might as well just call themselves AMD’s new head of PR department. Who gives a damn what they say.
So we are stuck with dx and opengl and the software responsible of making graphics doesn’t matter? Lets go back to dx9 then. He is so wrong. Why the developers want to remove the overhead then? And Mantle is not only to make the fps go up. The programmers need to know how their engine is running and have full control and not NVIDIA and AMD with their optimization a month or a year later. Yesterday NVIDIA released a driver that improve performance for 8 games that most of them are 6 to 1 year old. We have to wait for the drivers to give us performance. And most of the are coming to late, after we have done with the games. The developers must be responsible for all this without the middle man, the drivers.