GDC 15: AMD Mantle Might Be Dead as We Know It: No Public SDK Planned

Subject: Graphics Cards | March 2, 2015 - 02:31 PM |
Tagged: sdk, Mantle, dx12, API, amd

The Game Developers Conference is San Francisco starts today and you can expect to see more information about DirectX 12 than you could ever possibly want, so be prepared. But what about the original low-level API, AMD Mantle. Utilized in Battlefield 4, Thief and integrated into the Crytek engine (announced last year), announced with the release of the Radeon R9 290X/290, Mantle was truly the instigator that pushed Microsoft into moving DX12's development along at a faster pace.

Since DX12's announcement, AMD has claimed that Mantle would live on, bringing performance advantages to AMD GPUs and would act as the sounding board for new API features for AMD and game development partners. And, as was always trumpeted since the very beginning of Mantle, it would become an open API, available for all once it outgrew the beta phase that it (still) resides in.

View Full Size

Something might have changed there.

A post over on the AMD Gaming blog from Robert Hallock has some news about Mantle to share as GDC begins. First, the good news:

AMD is a company that fundamentally believes in technologies unfettered by restrictive contracts, licensing fees, vendor lock-ins or other arbitrary hurdles to solving the big challenges in graphics and computing. Mantle was destined to follow suit, and it does so today as we proudly announce that the 450-page programming guide and API reference for Mantle will be available this month (March, 2015) at www.amd.com/mantle.
 
This documentation will provide developers with a detailed look at the capabilities we’ve implemented and the design decisions we made, and we hope it will stimulate more discussion that leads to even better graphics API standards in the months and years ahead.

That's great! We will finally be able to read about the API and how it functions, getting access to the detailed information we have wanted from the beginning. But then there is this portion:

AMD’s game development partners have similarly started to shift their focus, so it follows that 2015 will be a transitional year for Mantle. Our loyal customers are naturally curious what this transition might entail, and we wanted to share some thoughts with you on where we will be taking Mantle next:

AMD will continue to support our trusted partners that have committed to Mantle in future projects, like Battlefield™ Hardline, with all the resources at our disposal.

  1. Mantle’s definition of “open” must widen. It already has, in fact. This vital effort has replaced our intention to release a public Mantle SDK, and you will learn the facts on Thursday, March 5 at GDC 2015.
     
  2. Mantle must take on new capabilities and evolve beyond mastery of the draw call. It will continue to serve AMD as a graphics innovation platform available to select partners with custom needs.
     
  3. The Mantle SDK also remains available to partners who register in this co-development and evaluation program. However, if you are a developer interested in Mantle "1.0" functionality, we suggest that you focus your attention on DirectX® 12 or GLnext.

Essentially, AMD's Mantle API in it's "1.0" form is at the end of its life, only supported for current partners and the publicly available SDK will never be posted. Honestly, at this point, this isn't so much of a let down as it is a necessity. DX12 and GLnext have already superseded Mantle in terms of market share and mind share with developers and any more work AMD put into getting devs on-board with Mantle is wasted effort.

View Full Size

Battlefield 4 is likely to be the only major title to use AMD Mantle

AMD claims to have future plans for Mantle though it will continue to be available only to select partners with "custom needs." I would imagine this would expand outside the world games but could also mean game consoles could be the target, where developers are only concerned with AMD GPU hardware.

So - from our perspective, Mantle as we know is pretty much gone. It served its purpose, making NVIDIA and Microsoft pay attention to the CPU bottlenecks in DX11, but it appears the dream was a bit bigger than the product could become. AMD shouldn't be chastised because of this shift nor for its lofty goals that we kind-of-always knew were too steep a hill to climb. Just revel in the news that pours from GDC this week about DX12.

Source: AMD

March 2, 2015 | 02:54 PM - Posted by Anonymous (not verified)

DING DONG THE WITCH(MANTLE) IS DEAD

March 4, 2015 | 05:19 AM - Posted by Anonymous (not verified)

Mantle is more alive than ever. It's just called Vulkan now. ;)

March 2, 2015 | 02:57 PM - Posted by H1tman_Actua1

Pfff it was never alive. Destined to fail from the start. AMD!

March 2, 2015 | 03:00 PM - Posted by Anonymous (not verified)

completely missing the part about it driving other APIs (MS and Nvidia) forward.

March 2, 2015 | 06:42 PM - Posted by arbiter

Can't prove 100% that mantle is why DX12 has closer to metal. Only claim to that was AMD announced mantle before MS announced DX which is the only claim to be the push. Given History of AMD and their marketing lies I would give benifit that MS was doing closer to metal well before this.

March 2, 2015 | 08:46 PM - Posted by Anonymous (not verified)

M$'s Graphics API may not have all that mantle will provide, for sure Both Nvidia, an AMD are the creators of their respective GPU hardware, so what gives M$ more inside advantage over Nvidia's, or AMD's knowledge of their own products and IP. For sure Mantle has all of AMD's product Knowledge available in its API that AMD chooses to share, same goes for Nvidia with its driver/API IP. The complete Internal APIs both companies may have, to counter any M$ market tactics are withheld. Certainly M$ exerts an inordinate amount of control in the PC/Laptop market, over both AMD and Nvidia, But M$ is not given the blueprints to everything. Don't You think for one New York Nanosecond that either AMD, or Nvidia provide M$ with the complete workings of their proprietary hardware/software IP!

So by what magic is M$ knowledge of AMD's, and Nvidia's proprietary IP better than the ones who create the hardware? Mantle was made by the maker of the GPU hardware, next one would expect you to make the claim that M$ can do a better job with drivers/API for Nvidia's GPU hardware/drivers than Nvidia could. How do you figure M$ is so omnipotent! Without AMD's and NVidia's help DX-Anything would not even exist. DX12 is just another API, and for sure AMD, and Nvidia know their own IP, and exactly how the their metal works. You think Nvidia's driver code mafia would allow M$ a look at the real books!

Are you serious, M$ DX12 is better than mantle, and takes no influence from mantle! Follow the call stack, it ends up at ether AMD's, or Nvidia's proprietary drivers for their respective GPU hardware, and those drivers drive the metal. OMG, your lack of reasoning in this matter is egregious.

March 2, 2015 | 08:50 PM - Posted by arbiter

I guess the FACT that Nvidia has been working with MS on DX12 yet AMD seems to missed the bus probably in favor of their own API which no shock failed.

March 2, 2015 | 09:49 PM - Posted by Anonymous (not verified)

They both have been working with M$, and neither of the two Discrete GPU makers gives M$ the keys to the city, and what does a graphics API have to do with stock prices. You have no idea how or what extra wrappers the DX12 software stack are sent through, once the calls work their way down past the DX API and into the GPU companies' proprietary software stack. And what about the Firmware on the respective GPU companies discrete GPU PCI cards, running that proprietary firmware and whatever other proprietary code calls, that are hidden, and abstracted from even the OS on the motherboard, M$ has no earthly knowledge of that. Mantle was never intended to supplant DX12, OpenGL, or any other API. You Know with the right bus mastering/DMA the code can be passed from the GPU's internal memory directly to the CPU, and the OS is none the wiser. Do you think M$ puts on its special soothsayer turban, and divines the inner workings of AMD's or Nvidia's proprietary knowledge. Your egregious lack of Knowledge is once again demonstrated, with that little stock market post hoc fallacy.

Has any GPU hardware, from any company, ever had in hardware a full implementation of any graphics API's full implementation of all the features that are implemented in software, It takes a good while for features to be adopted in hardware, and it is always an ongoing process. I'm still waiting for dedicated ray tracing hardware on every GPU, one company has some ray tracing functionality implemented in hardware, but it will take time to reach all companies GPU products.

Give it a rest, take that fanboism down to the local Football/whatever match, where everything is simple enough to wrap your limited understanding around. But in the complexities of technology, and business markets, Things are much to complicated for you, and other Fanbois.

OMG AMD missing the BUS, more like M$ needing help, from both Nvidia, and AMD, if they don't want another OS to make more inroads into M$'s already shrinking market. Forget what is said publically for public consumption by M$, AMD, and Nvidia, and reason for yourself.

March 2, 2015 | 10:04 PM - Posted by Anonymous (not verified)

Strike the stock market section, Its "no shock failed". Ha ha, I'm in need of a better brain, and some sleep! I'll admit that was my mistake! that's bad!

March 3, 2015 | 12:27 AM - Posted by Anonymous (not verified)

Marketing lies, From a "modern profession" that traces its ancestry back to the snake oil salesman, and original sin(for you religious types). The term Marketing is synonymous with Lying in most folks' dictionary, are you mad because you got fooled. In the technology industry of today, where the billions are made, liars, cheats, hucksters, and such abound, cue the obligatory Obi Wan, Mos Eisley clip. AMD marketing got ma mama, and I's madder den ah hive of hornets!! But Sir Nvidia marketing says 4, and only had high bandwidth to 3.5, That's got to be the pot calling out the kettle, and telling him that his pants are on fire, right next to the kettle on the stove, the pot on the other burner with his pants on fire.

April 22, 2016 | 11:34 AM - Posted by Anonymous (not verified)

yeah there are still 4 gigs of memory on the card though. much better than when AMD claimed 8 cores and were off by 4. the whole lag past 3.5gb is BS anyways it was MARKETING from AMD if you can believe that. Ask any 970 owner including me when the card goes past 3.5GB it is fine there is no difference in game play hence why Nvidia released it like this. It was a big fucking brain wash.

March 2, 2015 | 04:07 PM - Posted by Anonymous (not verified)

It caused a major shift in the industry that will benefit all gamers, by showing an API that WORKED and gave real performance benefits. Now all of us will reap the rewards.

March 2, 2015 | 06:28 PM - Posted by JohnGR

AMD is killing it because it succeeded. It did it's job and AMD didn't even had to try hard.

March 2, 2015 | 07:31 PM - Posted by arbiter

how did it succeed? it was a complete udder failure, funny how they went proprietary route when they claim to be all for standards.

March 2, 2015 | 08:36 PM - Posted by Anonymous (not verified)

Without Mantle i wouldn't be able to play BF4 above 60 fps. When bf4 first came out, i had 45 fps max and average was around 35. When Mantle support came out i have 75+ fps, most of time over 100, with average 60. So, i dunno how do you mean it's fail, except from game developers point of view, where only few release games with Mantle support. But from aspect of performance boost, it was big win. And maybe it speed up MS to release dx12.

March 3, 2015 | 03:00 AM - Posted by JohnGR

Answer me this, if you can.
Who benefits the most from DX12?
1) Nvidia
2) AMD
3) Intel

As for proprietary route, with Mantle probably inside DX12 and OpenGL next, it is more open standard than you want to admit.

March 3, 2015 | 07:16 PM - Posted by Sihastru

Without solid information about new GPU designs from the major players in the industry... as of today, nVIDIA is the only company that has GPUs that fully support DX12. They are, and count them, GTX 960, GTX 970 and GTX 980.

Yep, only three GPUs today support DX12 fully. All the other GPUs that are said to support DX12 from AMD, Intel or nVIDIA, will only support subsets of DX12 functionality ("feature levels" as they call them).

So nVIDIA has the upper hand, as it seems.

March 4, 2015 | 04:28 PM - Posted by JohnGR

300 series is around the corner. You can't decide who has the upper hand today, considering that DX12 is not yet available to the public. In a race it's not important who will take position first, but who will run faster.

March 5, 2015 | 06:23 PM - Posted by Sihastru

Yes, about that. Faster won't be the defining attribute for the 300 series.

Titan X strongly hints that the GM200 based GTX 980Ti (I vote GTX 985, but I think it will have a new series name) will be a GTX980 + 50% GPU. It will have 50% more shaders, 50% more memory bandwith, 50% more memory (6GB, but feel free to make the "last 0.5GB" joke). Since Titan X only needs 6+8 pin power, that means GM200 will also be efficient, so clockspeeds will be comparable to the ones of the current generation.

So the 300 series from AMD needs to be 65% to 75% "faster" then the 290X in order to actually be faster then the GM200 top dog.

I hate to say "impossible" but it really is.

April 22, 2016 | 11:36 AM - Posted by Anonymous (not verified)

Na pretty sure the turtle wins bro... so no not who runs faster.

March 2, 2015 | 03:02 PM - Posted by Anonymous (not verified)

As long as OpenGL Next has Mantle like features, then Mantle has done its Job!

March 2, 2015 | 09:35 PM - Posted by Anonymous (not verified)

It is a positive because if you were around in the 90s when there was OpenGL, 3DFX Glide, Direct3D and several other competing APIs you would know fighting M$ to the bitter end will just end in you running out of cash propping up your API. Many of those companies are gone or are alive only on paper. AMD made the right call here by forcing the Microsoft to react. Now AMD has had several wins lately by creating industry standards for HBM GDDR5 3D stacked memory with Hynix and adopted by JEDEC. Nvidia attempted to create it's own version and lost. They also have an industry standard for Adaptive-Sync accepted by VESA.

March 2, 2015 | 03:08 PM - Posted by Searching4Sasquatch (not verified)

Amusing to see AMD fanboys see this as a positive. NVIDIA has had better DX11 performance and is already showing a huge lead in DX12 performance. Instead of focusing on Mantle, AMD should have spent some trying getting their DX code out of the garbage.

March 2, 2015 | 03:10 PM - Posted by Anonymous (not verified)

Spoken like a true Nvidiot fanboy!

March 2, 2015 | 03:32 PM - Posted by Anonymous (not verified)

You're right, ignore the troll or uninformed ignorant dumbass.
FYI DX12 was in development way before AMD even started on Mantle idiot fanboy. Google it!

March 2, 2015 | 06:33 PM - Posted by JohnGR

And how was the little poor incompetent AMD able to come out with Mantle before DX12 that was in development way before? Maybe because someone convinced MS to keep DX12 in the drawer?
As a true green fanboy you can't thank enough AMD in public, but I am pretty sure you are doing it already in private.

March 2, 2015 | 06:44 PM - Posted by arbiter

MS was working on DX for 3-4 years BEFORE mantle was announced. AMD had a small window to do something with mantle, they only had to hit a date size of an ocean but didn't even hit the planet it was on.

March 2, 2015 | 10:22 PM - Posted by Anonymous (not verified)

Without GPUs there would be no DX*.*, so which came first the DX or the GPU. M$ sure had no knowledge before on AMD's, or Nvidia's, or Matrox,...GPUs. AMD has Mantle M$ better get it incorporated into DX, or be behind with AMD's product line, Same go for Khronos, get up to date with any thing AMD has, better keep up graphics API code base managers, proprietary, and open code base managers/associations. DX was 3-4 years before mantle, but DX was not even in existence before there were GPUs, and DX12 is not even here yet, and GPUs are have a different added feature set than, the DX had 3-4 years ago.

March 3, 2015 | 03:06 AM - Posted by JohnGR

So AMD managed to do something in 1-2 years that Microsoft is trying to do for 5-6 years? That's more likely like hitting the ocean on a small planet while shooting from the near solar system.

March 3, 2015 | 06:58 PM - Posted by Sihastru

AMD needs to cater to just a couple of GPU designs using a single GCN architecture. There was no provision for the past or the future.

Microsoft needs to cater to an entire industry, it's past, it's present and it's future (backward/present/forward compatibility).

One of these is much more complicated then the other.

March 4, 2015 | 04:36 PM - Posted by JohnGR

If you have an API in development for years and only 3 companies that need to support it, it's pretty easy. The GPU manufacturers will have to follow your guides in their hardware architectures. Also you forget Vulkan. It doesn't seems for Khronos to took so many years to create it.

March 4, 2015 | 06:19 PM - Posted by Sihastru

You're oversimplifying a very complex process. If it were "easy" to write an industry wide API everyone and their grandmother would have one.

Mantle was "easy" indeed, and I explained why. DX12 is a few orders of magnitude more difficult than that.

March 2, 2015 | 09:36 PM - Posted by Anonymous (not verified)

It is a positive because if you were around in the 90s when there was OpenGL, 3DFX Glide, Direct3D and several other competing APIs you would know fighting M$ to the bitter end will just end in you running out of cash propping up your API. Many of those companies are gone or are alive only on paper. AMD made the right call here by forcing the Microsoft to react. Now AMD has had several wins lately by creating industry standards for HBM GDDR5 3D stacked memory with Hynix and adopted by JEDEC. Nvidia attempted to create it's own version and lost. They also have an industry standard for Adaptive-Sync accepted by VESA.

April 22, 2016 | 11:37 AM - Posted by Anonymous (not verified)

AMD needs to quit the 2 market shit drop CPU or drop GPU and focus on making one the correct way before selling 2 products that are mediocre.

March 2, 2015 | 03:56 PM - Posted by Anonymous (not verified)

Getting some pop corn

March 2, 2015 | 03:58 PM - Posted by Anonymous (not verified)

Mantle served well, but is no longer needed.

March 2, 2015 | 04:08 PM - Posted by Daniel Nielsen (not verified)

Indeed.

March 2, 2015 | 10:42 PM - Posted by Anonymous (not verified)

Internally AMD will keep Mantle, AMD's gaming partners will still have access to Mantle's updates, should they want to do something when M$ fails to rapidly enough adopt the new into DX12. Nvidia has the same thing with their internal unnamed API, and will keep it handy, should M$ get lazy updating for Nvidia's features. None of these companies will let M$ lay around anymore, and Steam OS will have its adopters, and the Khronos group will need to keep up.
Internally both AMD and Nvidia, have internal Driver/Graphics API skunkworks projects, they always have had them, how is the hardware going to be tested without some from of internal graphics/driver API, hell they even simulate the hardware in software/FPGA before even the first batch of engineering samples come off the fab line.

March 3, 2015 | 12:50 PM - Posted by Martin Trautvetter

I doubt AMD has the resources or conviction to keep a separate API going semi-internally.

They're not even competing in their core markets any more, no need for extra drains on whatever resources they still have.

March 2, 2015 | 04:09 PM - Posted by Anonymous (not verified)

i wonder what the nvidia fanboys are going to do once they know the only full mantle .... ejem dx12 capable cards are amd.

jejej 4 years in the making and no nvidia card will support it, nFail does it again

March 2, 2015 | 04:26 PM - Posted by Anonymous (not verified)

Joke's on you, AMD can't support DX12 because they can't even do Rasterizer Ordered Views in GCN 1.0, 1.1 and 1.2 while Nvidia Maxwell fully supports it in hardware and even exposes it in OpenGL extensions.

Enjoy buying a new card if you are this clueless and a moron.

March 2, 2015 | 06:47 PM - Posted by arbiter

Yea, DX12 has new graphic techs in that NONE of the dx11 cards support. Only full DX12 card is Nvidia gtx900 series at the moment. Reason Nvidia has the leg up on that is they were working using the Market STANDARD. Irony how AMD fans complain when nvidia doesn't use standards and AMD does yet this time AMD was one that went proprietary.

March 3, 2015 | 04:38 AM - Posted by remon (not verified)

Or is it because AMD hasn't yet released the 9xx equivalent? I'm sure that's not hard to grasp.

March 2, 2015 | 11:03 PM - Posted by Anonymous (not verified)

So AMD is behind there, can it be implemented in software, and what is the performance penalty for any software implementation? If I were both AMD and Nvidia I would put one CPU core on die with the GPU just for accelerating new features, maybe even some FPGA/other acceleration resources, on die or package/interposer, hell AMD has some future planes to put CPU/other processing power directly on the memory chips. Nvidia is looking at things too. For sure some On die/package with the GPU, CPU power connected by a fat HBM style memory BUS could allow both AMD and Nvidia to have the flexibility to supplement/accelerate DX, other graphics APIs, new features before any dedicated functional blocks are baked into the next generation GPUs, or product update.

March 2, 2015 | 11:18 PM - Posted by Anonymous (not verified)

Show me any graphics API where any GPU maker had the full feature set of that graphics API implemented in hardware.
There are a lot of extended features that need to be implemented in hardware, and the APIs have built-in methods for adding, enumerating, linking, and delegating features in software, including new extended features being added, its called extensibility. It's best that GPU makers start including CPU/Other accelerators on die/package with the GPU to help with the new stuff being added all of the time. Better both AMD and Nvidia add some on the fly hardware extensibility, via dedicated CPU accelerators/whatever with some CPU/whatever(On DIE/Package) processing capability to keep up with graphics APIs.

March 2, 2015 | 04:12 PM - Posted by Anonymous (not verified)

Mantle is glNext a.k.a. Vulkan.

The name ist not coincidental, obv.

Vulkan is the SDK. Get it? It's really not that complicated.

March 2, 2015 | 05:17 PM - Posted by Anonymous (not verified)

Poor Spock, all this illogical Nvidia/AMD fanboi posturing, like a bunch of Klingons. Romulans and Cardassians are a bit more cerebral! Let's get a military draft going, tied to these rabid fanbois, ok you maggot/s! time to walk point! OR just give them a red shirt duty on a Star Trek away team/landing party.

March 2, 2015 | 09:01 PM - Posted by Anonymous (not verified)

No Mr. Smartass. Vulkan is "german" for volcano which is a rupture in the earth mantle-> an open mantle.

Open Mantle-> OGL Vulkan

March 2, 2015 | 11:47 PM - Posted by Anonymous (not verified)

So its the German spelling of volcano derived from the Latin Vulcanus. So who is the Vulcanus now! And what about that Mantle, does Nvidia have a Graphics API called DiMaggio. Those Vulcans are a logical bunch, what about the bog standard Fanboi, I say lets get them in a different draft.

And do you know why AMD used the Mantle Metaphor, has something to do with layers of software under the OS, and the hardware abstraction LAYERS below, and how they are implemented with respect to the software that abstracts the hardware, via the software layers that abstract the hardware. Drivers on the Metal, graphics APIs above. Oh those layers of abstraction, look at some OS/system software diagrams, they do have those layers in the classical computer science diagrammatical representation of a computing systems OS/software hierarchy. Ever wonder why the call the deepest OS functional part the Kernel.

March 2, 2015 | 04:35 PM - Posted by Anonymous (not verified)

What does "super-seceded" mean?

March 2, 2015 | 04:40 PM - Posted by Ryan Shrout

lol whoops fixing!

March 2, 2015 | 09:43 PM - Posted by Anonymous (not verified)

Civil War II: The Super Secession.

March 2, 2015 | 04:43 PM - Posted by ZoA (not verified)

“DX12 and GLnext have already super-seceded Mantle in terms of market share... “ - Ryan Shrout March 2, 2015. when not a single peace of commercial software in either DX12 or GLnexst was published, in fact GLnexst was not even publicly demoed.

And Ryan' employees wonder why people think their boss is dumb.

Ryan's nonsense aside I'm disappointed mantle will not go with public SDK. Now I'll have to buy Win10 and I hate giving money to M$.

Also by the sound of it my guess is AMD will probably make mantle 2.0 in to some sort of specialty API for graphic and computing in professional applications. Meh :(

March 2, 2015 | 05:23 PM - Posted by Ryan Shrout

Let me know in 4 days how many games are known to use DX12 vs known to use Mantle.

March 2, 2015 | 05:26 PM - Posted by pdjblum

Wish I was dumb enough to have been able to found and run a successful enthusiast site that has enough resources to employ people. To me, entrepreneurs are the lifeblood of this country and the ones who should be admired.

Not that I don't have issues with Ryan, as he continually annoys me by supporting exorbitant prices for both the faux flagships gpus's and even more exorbitant prices for the flagships, when they are finally released. This paradigm is great for Nvidia margins, but not for us.

March 2, 2015 | 06:49 PM - Posted by arbiter

Your an idiot, only companies that USED mantle was ones that were paid to use it. The day DX12 was announced mantle was put on life support and AMD finally pulled the plug on it.

March 2, 2015 | 08:57 PM - Posted by Anonymous (not verified)

see the above reply to one of your posts, you egregious fool.

March 3, 2015 | 12:00 AM - Posted by Anonymous (not verified)

Your(sic) an idiot

I never get sick of people embarrassing themselves like this.

March 2, 2015 | 04:50 PM - Posted by Anonymous (not verified)

AMD will keep mantle, it just will not make the SDK public, lots of pressure from the Big OS players, for sure, and the Khronos group has access. Mantle will be AMD's in house Ace in the hole, should they need to keep/make the their GPU feature set available, dew to laziness on the part of the DX12, and Khronos developers. And Keeping an SDK private, is not the same thing as keeping an API private.

"The Mantle SDK also remains available to partners who register in this co-development and evaluation program. However, if you are a developer interested in Mantle "1.0" functionality, we suggest that you focus your attention on DirectX® 12 or GLnext."

OK so Mantle "1.0" is out there, but this statement in no way explicitly states that there will be no Mantle 2.0/*.*, it just implies that maybe all of the Mantle "1.0" feature set is included in DX12, and OpenGL Next!

Only AMD knows for sure, and they are not saying anything definite about any future Mantle like API/SDK. All GPU makers have their own in house testing/development APIs.

March 2, 2015 | 06:56 PM - Posted by arbiter

Re look at the mantle announcement, I remember AMD making the claim "they said MS had nothing in works for next DirectX"

Can look at that statement as they knew there was DX coming and LIED their butt's off? There is No damn way they didn't know MS wasn't working on it, they are one the major GPU makers. Mantle I believe is nothing more then a PR stunt and AMD had no intention to ever release SDK from the start.

March 2, 2015 | 08:56 PM - Posted by Anonymous (not verified)

See the above reply to one of your posts, you egregious fool!

March 3, 2015 | 12:25 AM - Posted by Anonymous (not verified)

You are the only correct person in this thread.

March 2, 2015 | 07:24 PM - Posted by Anonymous (not verified)

Mantle's job was to give Microsoft's DX a kick up the ass, mission achieved, it can now enjoy it's retirement in some AMD basement

March 2, 2015 | 08:15 PM - Posted by pdjblum

Wish the same could be said about gsync.

March 2, 2015 | 09:43 PM - Posted by TopJoe (not verified)

I can't understand why grown ass men or women go to sites and argue about the different companies that are trying to take every penny they can from you. I don't complain because I have used both companies, so it doesn't matter too much to me.

But I have to admit this is always entertaining when people from both sides come out throwing body blows about whatever. It's fun to read.

March 3, 2015 | 12:33 AM - Posted by BSODEnthusiast (not verified)

The articles are up to date and designed as a fast paced scrollings, it will be gone and lost in sequential system quickly along with your unprofessional posted words. You can comment anonymously, hiding your ass if you might be wrong. No moderation, which is good, I prefer sites that condone freedom of speech. All the writers are nice people, they won't write nasty things back and they prefer to only respond to good logical comments.

Yeah, anyway, guys (PCPer), you wrote nice articles and should get credits more. I especially appreciate the quick test-and-publish articles such as EVO 840 slows down, Gsync flickering, etc. Add an article rating and put a popular archive links to that empty spot on the mainpage.

March 3, 2015 | 01:15 AM - Posted by TopJoe (not verified)

So says the guy hiding his ass behind his keyboard. There are quite a bit non logical comments stated here and other sites. I just found it interesting that people argue about these companies of who is better than the other. There was nothing nasty about what i said, just stating what I keep reading.

Dude get a grip and stop being so uptight about someone stating their opinion just as you are. They are just that, opinions. They are not meant to be professional because once again they are just opinions.

March 3, 2015 | 03:49 AM - Posted by BSODenthusiast (not verified)

Oh, sorry, you misunderstood me. I agreed with your first post and do enjoy reading the banters. I didn't meant to reply to you specifically and should have started with "That's because this site is a PC enthusiast' trolls heaven..."

March 3, 2015 | 04:18 AM - Posted by TopJoe (not verified)

No problem bro, I understand.

March 2, 2015 | 10:13 PM - Posted by praack

wow amazing the fans lying in wait here, if i remember correctly usually there are parts of a new DX release that require a hardware upgrade. the OS upgrade is usually to just upgrade the OS_ though if anyone is still running XP as their daily build they are a bit off.

7 is solid, 8 not so much - 10 is looking pretty good so far

but even with DX 12 and 10 - i won't be buying a new graphics card right away so much of the uptick will pass me by

As for AMD and Mantle- most people agree MS was dragging their heels a bit- waiting for the next OS to come out- in the end- it looks like they only announced it a few months earlier anyway

March 3, 2015 | 12:57 AM - Posted by Anonymous (not verified)

7, and Steam And OpenGL Next, or whatever they call it to impress the slackjaws. Really Khronos group, you are an Industry Open Standards group, no need for any fancy marketing naming jive.

7 because it doesn't pipe my private desktop file searches through the BING-cloud, and get all metrics gathering on me. 7 because it's paid for, and the UI is not so TIFKAM/Tiled F-upped, and still those tile are there lurking under the surface of win 10. 7 because I do not want that slippery 10 slope to win 365, pay to play OS as a service, no money no funny, cloud and services closed ecosystem. Steam OS after 7's 2020 EOL, and even before, in dual boot, with 7 running any legacy stuff, not already on Steam OS. 7 locked down in a VM, and restricted from direct internet access for any after 2020 continuing legacy needs.

March 3, 2015 | 01:22 AM - Posted by loccothan (not verified)

0_o So AMD finally do The Best Thing in History.
Now we Have DX12/oGL Next and that's a Next-gen Mantle and Full Hardware Radeon Utilisation on every API essential to Gamers.
Well Done AMD/ATI. Thats Open :D

March 3, 2015 | 02:19 AM - Posted by ThorAxe

Mantle changed nothing that DX12 wasn't already planned to do.

It takes more than a year or so to develop an API.

March 3, 2015 | 02:41 AM - Posted by Anonymous (not verified)

can you say the same for mantle ? they had the idea but just copy_cating/ writing codes is a lot easier than innovation .
yes M$ was working with NV but it doesn't mean they had the idea

March 3, 2015 | 05:13 AM - Posted by Anonymous (not verified)

Everyone knew that Mantle is Glide but I didn't expect it fail too soon.

March 4, 2015 | 05:23 AM - Posted by Anonymous (not verified)

Nothing failed. Mantle succeeded. That's why there will be Vulkan and DX12. Obviously some people just don't understand the intention of Mantle.

March 3, 2015 | 08:22 AM - Posted by CrisisHawk

So the feature that swayed my decision to get a 290 over a 780 at the start of last year is effectively dead...Well played, AMD...Well played.

March 3, 2015 | 08:24 AM - Posted by CrisisHawk

woops, double posted

March 3, 2015 | 12:46 PM - Posted by Martin Trautvetter

I hope this finally means an end to any more Star Swarm demos.

March 3, 2015 | 02:01 PM - Posted by ObsidianBlade (not verified)

Maybe they are going to use Mantle 2.0 with LiquidVR

March 4, 2015 | 09:51 AM - Posted by DocACE911 (not verified)

Does anyone feel sorry for AMD? They try so hard but can/will never have the money to compete with Intel or, for the most part, NVIDIA.

This whole mantle idea was really a step forward for gaming but is lost with DX12 just a few months later.

One must ask, why did AMD pour precious resources/money into Mantle when they knew the timetable (and features of DX12)? The narrative that Microsoft somehow invented DX12 post mantle is simply not supported by the data/complexity of the project/rumors.

I wish them all the best. Just sad to see money that could have gone into better 3xx parts thrown at making battlefield 4 run 6 FPS faster on an i7 part. I really hope the 3xx parts are not 28nm...

March 4, 2015 | 02:27 PM - Posted by Anonymous (not verified)

Thank you, AMD for paving this way again. Thank you for being open and close to dev as possible. Thank you also for the commitment in VR.

You don't have to be a genious to see when a company tries its best to be correct to ... well everyone. Perhaps up to a certain point - after all, money has to be made.

I will be getting two of those R300 flagship cards when they come out, even if the new Maxwells are a bit better than them. The "Red side" has deserved it!

March 5, 2015 | 12:41 AM - Posted by DocACE911 (not verified)

Are you serious? You would buy a R3xx if it was not as good since the "red side" deserves it??

NVIDIA nor AMD are charities. While I hate to see AMD becoming non-competitive on the desktop CPU and less on the GPU, why donate your money to a better cause?

Lets hope R3xx is a leap forward like the 290 vs 780 at a better price point.

until then, please donate your money toward cancer research :)

March 5, 2015 | 02:00 PM - Posted by AMD CEOnonymous (not verified)

AMD - lame duck (copy&paste) company in the world... tztztz AMD BSOD driver error / install problems / Omega Alpha phase II driver summer 2015 - error AMD CCC Catalyst crash. OpenCL & property AMD Mantle rules the world - in all HPC Supercomputer of the world with HSA and hUMA revolutionary AMD Software - and in all (underpowered AMD) NextGen Game Console!

AMD - lame duck (copy&paste) company - so WTF ! AMD have 100% marketshare of AMD... AMD Marked Leader...

In real world - AMD go bankruptcy with under 20% markedshares and best 'AMD Fixer 6 for dummys' professional AMD PR campaign ever - go AMD!

March 16, 2015 | 12:13 PM - Posted by P0ci (not verified)

Despite me being Mr King of fanboism with AMD I will admit Mantle literally sucks.

When I tryed it with BF4 I had mantle64.dll crashing every 30 min. Gave up on it and sacfriced the extra performance which it did indeed give like it claimed. It just was never stable.

Ill take good ol Open GL any day of the week

February 17, 2016 | 12:41 PM - Posted by Anonymous (not verified)

It's so dead that Nvidia's Vulkan decided to adopt it?

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Lines and paragraphs break automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd> <blockquote><p><br>
  • Web page addresses and e-mail addresses turn into links automatically.

More information about formatting options

CAPTCHA
This question is for testing whether you are a human visitor and to prevent automated spam submissions.