DirectX, OpenGL May Soon Allow Low-Level Hardware Access

Page 3 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.


So I was correct in thinking that you didn't want to let this rest in the PM's, which is contrary to this comment of yours! :lol:


 


Actually it was an attempt to assist in lessening further misinformation. Bloodroses75 statement was correct. Your attempt to correct him was misguided, although I can see where you are coming from. Although you are right, Sony and MS will not be using mantle, he did not state anywhere that Mantle was going to be used on either console. He was merely stating what mantle is, and that is a low-level api similar to that used by PS4 and Xbox to have direct control over the hardware to lessen the need for the CPU (which the fx series is a bottleneck). Sony or MS would have no use for Mantle as its the developers who need to opimize their game for a certain API and not a manufacturer of consoles. Not to mention the API's between the Xbox and the PS4 are different due to the embedded SRAM used by the Xbox.

The comment from the other thread has no place here. Why you are dragging it over here as if there's any relation, other than me and you discussing technology, is something I don't quite follow.
 


You ought to read the text that is in bold and then re read read it.
"bloodroses75 said:
People forget to realize that both the Xbone and PS4 use AMD exclusive graphics, so Mantle still has it's place on those platforms."

That implies that Mantle is going to be used on the consoles and you agree with me that it isn't, so what is your issue and why are you trolling for an argument?
 


An argument is not bad by default. Its typically a discussion. I am not angry, nor is this conversation heated. I did actually misinterpret his phrasing, as Mantle will make it easier to port games over to the GCN architecture (Sony and MS API's are geared towards GCN/PIledriver architecture) I did also state that you were correct that Mantle will not be used on the consoles themselves. Although his statement, if he did in fact mean it would be used on PS4/Xbox instead of being used in conjunction with in the porting process, is false on the premise that it will not be used on the consoles themselves.

However, I'm still not sure how this relates to either of our other discussions.
 

It relates in that you are quoting my posts and saying that I'm somehow both wrong and right at the same time and because you say that you somehow know what my opinion of Mantle is even though I have never expressed an opinion on Mantle.
 
You apparently did not read my last post. I admitted I misread/misinterpreted what he stated in his post. I'm not sure how that renders you both right and wrong.

I did actually misinterpret his phrasing, as Mantle will make it easier to port games over to the GCN architecture (Sony and MS API's are geared towards GCN/PIledriver architecture) I did also state that you were correct that Mantle will not be used on the consoles themselves.

I admitted I was wrong, how else are you going to interpret this?

Although his statement, if he did in fact mean it would be used on PS4/Xbox instead of being used in conjunction with in the porting process, is false on the premise that it will not be used on the consoles themselves.

My explanation of how I thought I original read his comment. Although yes, mantle will not be used on the consoles themselves, that's not how I read his original quote, and I explained based on that assumption. Not too sure why you are still taking it as an attack...
 


Deferred contexts actually are not as great as they're supposed to be. I guess there's a lot of "gotchas", making it hard to realize anywhere nears it's max performance.
 
For those claiming D3D is not optimized for multi-threaded workloads, that is simply not true (D3D 11+, display device drivers must support free threading). It was implemented in 2007 (and released in 2008)...

Also, there is already a low level implementation of Direct3D called Direct 3D 11.x (implemented in 2012 - before Mantle). And it is coming to desktop (see the link) - not as is, of course - a layer of abstraction has to be added to support multiple GPU vendors.

Oh really? Can you be more specific? When I use an infinite for loop in several threads with priority set to maximum to simulate a CPU bottleneck, deferred contexts boosts performance by ~22% (on average).
 


They were already awake or it wouldn't be IN OpenGL already working NOW. Mantle is still a beta AFAIK. So I guess you could say Khronos beat AMD/Mantle right? 😉 They only comment Directx stuff is coming, while opengl is already here.
 


You have to admit that it is funny reading all the Mantle fanboi comments though?
 


I'm sorry, I must have missed it...When did AMD share Mantle with Nvidia? It only works on GCN so far, and AMD has merely CLAIMED mantle can work on other stuff, but so far they've shared NOTHING. On top of that it only works on SOME GCN cards right now, and they still have to optimize the rest.

However, NV will license Kepler to AMD if AMD would like to make some keplers (or anyone else that wants to use kepler IP). They started that IP sharing stuff last year. Anyone who wants kepler IP merely has to pay for it supposedly. I wouldn't want to own IMG.L stock as that is how they make the puny profits they do (clearly not much being made on all the gpus in iphones/ipad/ipod etc). Now NV will be competing vs. anyone using powervr stuff, which at some point should increase the bottom line. I'm sure a few will sign up as kepler mobile has all the features of a full blown desktop gpu. Other SOC makers will have trouble keeping up with NV now IMHO. All game devs already know kepler inside and out and will take that experience right over to mobile. All other SOC makers/gpu makers (img.l etc) have ZERO game dev experience (no dev knows their hardware) and ZERO driver experience aimed at gaming.

If AMD can get a SOC out for phones/tablets etc they will have the same advantage as NV over the rest of the soc field. I wish Seattle was aimed at mobile instead of servers. I hope they get their mobile one out soon but I haven't seen anything about it.
 
Don't be silly Somebodyspecial, AMD will share Mantle as soon as they finish making it. Why? Because it serves them MUCH better if everyone uses it than if only they do. Remember that Mantle is an API, and Nvidia have no API, so Nvidia is not competition in that space. Assuming it does what it says no the tin, uptake will be fast, and naturally Nvidia will never see the speed boost that GCN will as Mantle is built specifically for that chip design... but it will still see significant boosts, forcing Nvidia to actually have to build a faster card than AMD just to keep up and allowing AMD bigger margins on similar performing models until Nvidia can redesign Mantle orientated chips instead of OGL/DX orientated chips.

This all naturally assumes that Mantle is reliable and hugely faster than DX/OGL. Fast enough that gaes are coming out that completely revolutionise gaming graphics. With the Carmack quotes and the 30,000 draw calls vs 3,000 DX draw calls, that may well be the case, but it's hard to separate HUMA and Mantle advantages so far.
 


What is the point? Mantle only works on GCN cards and Nvidia are not going to change their architecture any time soon (have you seen what Maxwell can do?) and as Mantle doesn't even work on older AMD cards why should anyone believe that Mantle will work on Nvidia cards?
 
What will happen is that DirectX and Opengl will use a modular API system so that if your hardware doesn't currently have low level API it can still access DirectX or Opengl high level for output. Thus being said that it'll still be up to AMD and NVidia to release drivers containing the low level API components for OpenGL and Microsoft to use.They'll probably use Mantle's subroutines too.
 


No. Mantle only CURRENTLY works on GCN cards. Whilst we're talking about limitations that will change, Mantle only currently runs under 1 game so far.

As for why Nvidia will support Mantle, in case you didn't read the second part of my post, Mantle is proposing that it will be able to do 10x the rendering that DX could in the same time due to its much more direct approach. If it's only 2x the rendering in the end product that will still force Nvidia to adopt it as there will be a whole string of very shiny new games that will not run at max on anything but Mantle, suddenly taking something like a Titan and making it look like a 8600GT next to AMDs current generation.

Mantle claims to have 10x the draw call ability and hUMA. If it has either, Mantle only games or games that use Mantle only to generate the 'Ultra' graphics settings will start appearing. Nvidia needs the flexibility (to play Mantle only games) and the ability to keep up with AMD.


And @Mupples. That's called a HAL. DX already is a HAL. It is bypassable, but the whole issue with bypassing a HAL is that there's no such thing as a universal command structure beneath DX on a card as every generation of card speaks a different language even before we get to the differences between AMD and Nvidia (think of it being like they all have different instruction sets and DX is a Java virtual machine on top of them that allows DX applications to run no matter the stuff underneath).

Mantle is proposing the equivalent of a runtime 'C++' for Nvidia which is at the same time a runtime 'Assembly' for it's own cards. If Nvidia take the carrot, they can make their next generation of card as fast or faster than AMD's under Mantle.
 


Unless you have some hard evidence to back up your claims then it's just speculation and pipe dreams, the simple fact is that at the moment Mantle does not work on anything other than GCN cards and you saying anything to the contrary isn't going to change that.
 


Lest we forget that Android only ran on one phone at one point.
 


And what does that have to do with Mantle? :heink:
 


Everything. The launch strategy is the same (flagship product till they iron out the kinks then SDK so the others can join the fray), the integration strategy is the same (device manufacturers (in Androids case Linux and Linux drivers) expose calls to engine), both started off with one chipset until things were solid then released that solid product to the public. Both have sizeable advantages over their competitors in the ability that developers can squeeze out of them, and both are released by companies with specifically anti-restrictive policies that have a history of opening up everything to the public (See OpenCL, see x64)

Mantle will largely be ignored if AMD do not open it up to Nvidia, so Mantle's survival depends on it. No platform specific API really survives in the face of platform Agnostic ones, as eventually the Agnostics will just assimilate the functionality, even if it is done in software.
 


Android was starting out in a new market and as such didn't have any competition, Mantle is trying to change an established market so they are not the same thing at all IMO.
 


Wow, so the iPhone didn't come out a year before Android had the G1 to compete with it? So Mobile Phones weren't an established market that both were trying to change? So Symbian was not an already established and very mature OS that ran some very advanced Java applications on the early smartphones like the Nokia N9x as well as the iPhone being a competing product with a near identical feature set, superior software but the simple failure of not being platform agnostic.

Whether or not you have the lateral capacity to see the parallels is neither here nor there, they most certainly exist.
 


The Smartphone market is nowhere near as old as the Ye Olde Mobile phone market and there are distinct differences between the two but if in your mind they are one and the same then so be it.
 


Wait, wasn't your argument that the market was a new market? Maybe you should just give up on trying to use market maturity as your defence. The smartphone market was the phone market in the same way that the Ultrabook market was the Laptop market when the first Ultrabook came out all on it's lonesome.

For the same reason the Low Level API market (containing only Mantle) is currently the Graphics API market full stop. Should there be a bunch of competing offerings brought forward, then the market will separate from Graphics APIs to Low Level (DXLL,OGLLL, Mantle, maybe NVAPI soon) and High Level (current and future normal DX, OGL) the very same way Laptops spliced into DTRs, Standard Sizes, Netbooks and Ultrabooks and the same way Mobile phones spliced into Smartphones and dumb phones.

Standard market mechanics. There's no such thing as an emerging market of one.
 


I was pointing out that Android was geared for a different type of phone that has only emerged recently but if you want to twist things around then that's up to you, I'm done responding to you.
 
Status
Not open for further replies.