(DSOG via MaxPC) Nvidia Finally Officially Speaks About AMD’s Mantle

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

DarkKnightDude

Senior member
Mar 10, 2011
981
44
91
They also mentioned that they have 300 Phd or some master mind in graphic technology which are just working for gameworks.

Yeah... look at the massive benefits of those 300 PHDs with gameworks in Ubisoft's games.

Marketing at its finest.
 
Feb 19, 2009
10,457
10
76
Yes. I wish we could just talk about the subject, but we'll have all the same posts we've had in the past over and over.

Not exactly because the original Mantle thread was full of crap with people claiming it will make next to no difference, especially those who say high end rigs won't benefit.

Well we know that's false, even with top of the line i7, in multi-GPU setups (the real high-end), in MP we knows it makes a massive difference in raw FPS, minimum FPS and very stable and low latency frame times.

So we KNOW it works very well even in FPS which is generally regarded as GPU bottlenecked. The point really is whether DX12 will obsolete Mantle. I think it will, an open API with the major feature of releasing CPU bottlenecks, what else does Mantle bring? Nothing extra at the moment and with AMD showing no sign of pushing it further.

There was a recent interview IIRC where AMD admit basically DX12 is Mantle and that they "may" add extra features to Mantle to keep it relevant once DX12 is out. That's PR terms for "forget about it". Another thing is the vibes from studios that sign up for the Mantle beta, they mostly hype its main feature as "a stepping stone to DX12" or "getting in early to prepare for DX12".

It looks & smells like once DX12 saturates (hopefully MS won't do something stupid like lock it to Win9), Mantle is dead.
 
Aug 11, 2008
10,451
642
126
First, I was not really impressed by the nVidia guys. They basically said nothing except that they are not going to use or say anything good about mantle. Duh, it is an api developed by their arch rival. What do you expect them to say: "Mantle is great, I wish we had developed it"?

On the other hand, all the claims about the openness of mantle are contradictory at best. First lets look at the current situation, not some hypothetical future scenario which has not taken place yet. Is mantle open now, has AMD released the source code? No. Apparently Intel even asked for the code and was denied. I have no problem with this, because mantle was developed by AMD, so why should they release it to Intel. What I have a problem with is with the claims of openness.

But, AMD claims, it is not ready, we cant release the code until it is ready. Why not? It is "ready" enough to have been implemented in BF4 six months ago, and subsequently for Plants vs Zombies (strange) and Thief. And AMD is bragging up how many developers are on board and how many games using mantle are in development. Again, I have no problem with AMD trying to get mantle implemented in a many games as possible, but that hardly is consistent with the claim that mantle is not "complete" enough to be utilized.

If AMD really wants mantle to be "open" they should have released the code already so that competitors could look at it and if interested, start developing drivers to utilize it, which obviously would have to be modified later, but at least some of the work would have been done already. So mantle at this point in time, is a proprietary API that runs on GCN hardware only, gives a value add to AMD dgpus, and makes AMD cpus somewhat more competitive. Anything more is just PR.
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
Not exactly because the original Mantle thread was full of crap with people claiming it will make next to no difference, especially those who say high end rigs won't benefit.

Well we know that's false, even with top of the line i7, in multi-GPU setups (the real high-end), in MP we knows it makes a massive difference in raw FPS, minimum FPS and very stable and low latency frame times.

So we KNOW it works very well even in FPS which is generally regarded as GPU bottlenecked. The point really is whether DX12 will obsolete Mantle. I think it will, an open API with the major feature of releasing CPU bottlenecks, what else does Mantle bring? Nothing extra at the moment and with AMD showing no sign of pushing it further.

There was a recent interview IIRC where AMD admit basically DX12 is Mantle and that they "may" add extra features to Mantle to keep it relevant once DX12 is out. That's PR terms for "forget about it". Another thing is the vibes from studios that sign up for the Mantle beta, they mostly hype its main feature as "a stepping stone to DX12" or "getting in early to prepare for DX12".

It looks & smells like once DX12 saturates (hopefully MS won't do something stupid like lock it to Win9), Mantle is dead.

You touched on the biggest problem with DX, msft decides when to add features. We have basically been on the same feature set for 5 years now. Sure we've DX11.1 and 11.2, but they are so minor that they aren't even supported fully by the IHV's or game devs. With Mantle a hardware feature is developed and it can be added to the API at the same time. Also, the only reason msft is cooperating with the devs and IHV's now with DX12 is because they got competition from Mantle and they are scared of losing their stranglehold on PC gaming. They're still going to make us buy a new OS if we want DX12. That's the other issue that's really annoying with DX.

Mantle is still brand new. It's started off real well. We'll see if it maintains momentum. If it does, by the time DX12 comes along it could easily stick. Especially if it has the advantage of running on Win7/8-8.1 and DX12 requires a new OS. It will be in a position where it runs on virtually everyone's Windows PC and DX12 won't run on any out of the gate. There's also the possibility Mantle will expand beyond Windows. That will never happen with DX.
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
First, I was not really impressed by the nVidia guys. They basically said nothing except that they are not going to use or say anything good about mantle. Duh, it is an api developed by their arch rival. What do you expect them to say: "Mantle is great, I wish we had developed it"?

On the other hand, all the claims about the openness of mantle are contradictory at best. First lets look at the current situation, not some hypothetical future scenario which has not taken place yet. Is mantle open now, has AMD released the source code? No. Apparently Intel even asked for the code and was denied. I have no problem with this, because mantle was developed by AMD, so why should they release it to Intel. What I have a problem with is with the claims of openness.

But, AMD claims, it is not ready, we cant release the code until it is ready. Why not? It is "ready" enough to have been implemented in BF4 six months ago, and subsequently for Plants vs Zombies (strange) and Thief. And AMD is bragging up how many developers are on board and how many games using mantle are in development. Again, I have no problem with AMD trying to get mantle implemented in a many games as possible, but that hardly is consistent with the claim that mantle is not "complete" enough to be utilized.

If AMD really wants mantle to be "open" they should have released the code already so that competitors could look at it and if interested, start developing drivers to utilize it, which obviously would have to be modified later, but at least some of the work would have been done already. So mantle at this point in time, is a proprietary API that runs on GCN hardware only, gives a value add to AMD dgpus, and makes AMD cpus somewhat more competitive. Anything more is just PR.

Something else that's been talked about dozens of times.

Mantle is still in development. At first AMD only worked with a very small select set of devs. Guys who could help build and shape mantle and AMD could support. It's far enough along now that AMD is opening it up to 40, or so, devs. Again, a number that they can handle the support for. When it's mature enough that they can offer it to everyone and handle supporting it for everyone, they will. The plan is before the end of this year. At that time Intel can have access with everyone else.

I guarantee you that if AMD had given Intel access to Mantle months before they gave it to nVidia (not that they have any reason to as I've already explained), the crap would fly. I would think that nVidia could possibly claim anti competitive practices on AMD.
 

SlowSpyder

Lifer
Jan 12, 2005
17,305
1,001
126
Intel asked AMD for access to mantle and was told it was not available at this time. Why not give them access? One could guess it would make AMD look even worse with their cpus. DX12 is the future because Intel supports it.

Mantle is still in beta. I believe I've read somewhere that AMD will open access to some others once Mantle is out of beta. So far only software developers are on board, as far as I know. Once it is out of beta we'll see if hardware manufacturers want jump on board or not, and if AMD allows it or not.


Over 83% of the sold graphics capable hardware cant run Mantle. But they can run DX12.

Then you can hate and dislike any other company as much as you like. But it still doesnt change that fact. Not to mention how AMD tapdances around the "Open API" BS they claimed.

Yea, but what percentage of hardware that is bought for gaming can run Mantle? That's the important question, not the whole 83%. I have a few aunts that couldn't care less if they had a Titan Z or Intel IGP, as long as it is quiet and runs Facebook games and Solitaire, that's all they care about. I imagine a large percentage, probably a large majority of the Intel IGP sales fall into that category (as well as some of AMD's APU sales). But for gamers who buy gaming hardware, I wonder what percentage of those sales can run Mantle?
 

ShintaiDK

Lifer
Apr 22, 2012
20,378
145
106
So what you say in short is, that AMD have no intention of sharing Mantle. And its been nothing but a PR spin to claim otherwise, while attacking the competition for closed standards. And thats assuming that Mantle could even run on any uarch besides GCN.

If AMD had any intention of getting Mantle to be widely adopted. They would have shared it. Even Qualcomm is on the D12 train.
 

ShintaiDK

Lifer
Apr 22, 2012
20,378
145
106
Yea, but what percentage of hardware that is bought for gaming can run Mantle? That's the important question, not the whole 83%. I have a few aunts that couldn't care less if they had a Titan Z or Intel IGP, as long as it is quiet and runs Facebook games and Solitaire, that's all they care about. I imagine a large percentage, probably a large majority of the Intel IGP sales fall into that category (as well as some of AMD's APU sales). But for gamers who buy gaming hardware, I wonder what percentage of those sales can run Mantle?

Even if we only look on recent dGPU sales only and dont exclude those cards that is still not GCN. Then we talk about less than 35% of new sales. And that would be an utopian number for Mantle.

AMD made sure Mantle would never succeed.
 

ShintaiDK

Lifer
Apr 22, 2012
20,378
145
106
But thats the endresult isnt it. Windows 9 with DX12 isnt so far away anymore. And AMD have shown no intentions that its willing or capable of sharing Mantle. Perhaps because it may turn out that Mantle only works on very specific GCN hardware. May also explain why AMD noted that there would always be a DX path, should future hardware not support Mantle. (Read new uarch.)

And it turns back to the article, where nVidia is completely right about Mantle.
 
Last edited:

SlowSpyder

Lifer
Jan 12, 2005
17,305
1,001
126
Even if we only look on recent dGPU sales only and dont exclude those cards that is still not GCN. Then we talk about less than 35% of new sales. And that would be an utopian number for Mantle.

AMD made sure Mantle would never succeed.


Include APU's, too. I wonder what percentage of systems could run Crysis at even medium settings with a smooth frame rate when it was released? Or why did Watchdogs bother with a 3GB+ texture setting? What percentage of systems do you think can utilize that smoothly? Yet they were made and are / were successful products.

Mantle may very well fail, maybe, maybe not. But developers seem to be on board and games are coming out that use it. The industry buzz seems very positive (which is probably somewhat helped along with money from AMD, though I do think some of the excitement is indeed genuine).

I think Mantle's future depends on how good DX12 is. If DX12 can provide the same benefits and run on AMD, Intel, and Nvidia, then why would developers bother with Mantle after that point? If DX12 isn't as good and Mantle already has a real foothold in the industry, then I can see if being used for some years yet, again depending on the next DX iterations.
 
Last edited:

Mand

Senior member
Jan 13, 2014
664
0
0

You may not be saying that, but AMD certainly did. Huddy said, very clearly, that AMD was going to have sole control over Mantle for the express purpose of using it to better optimize their hardware.

That's the very definition of a closed, proprietary product.
 

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
Precisely.

More specifically.

A true "open" software specification is held by a consortium of partners who collaborate on trying to achieve the best end result. True "open" software is modifiable and enhance-able by ANY MEMBER of that consortium of partners. This was brought up in the interview. AMD has no intention of doing this. So there are two problems here. Mantle requires GCN architecture, and even AMD's own non GCN GPUs cannot support Mantle. So someone please tell me how competing non GCN GPUs can support Mantle if AMD's own non GCN GPUs cannot. MUCH MORE IMPORTANTLY. AMD has expressly stated that Mantle is their own so it is not truly open by any proper definition. Open means that there is a standards board with many partners collaborating on the end result and implementation. AMD will not do this: AMD stated that Mantle will always, always be controlled by AMD and no one else. So it is not surprising that NV would not be interested. NV is not developing a GCN GPU and they would not support an API that isn't truly open - Mantle is not "open", it will always be under control by AMD and AMD alone.
 
Last edited:
Aug 11, 2008
10,451
642
126
Something else that's been talked about dozens of times.

Mantle is still in development. At first AMD only worked with a very small select set of devs. Guys who could help build and shape mantle and AMD could support. It's far enough along now that AMD is opening it up to 40, or so, devs. Again, a number that they can handle the support for. When it's mature enough that they can offer it to everyone and handle supporting it for everyone, they will. The plan is before the end of this year. At that time Intel can have access with everyone else.

I guarantee you that if AMD had given Intel access to Mantle months before they gave it to nVidia (not that they have any reason to as I've already explained), the crap would fly. I would think that nVidia could possibly claim anti competitive practices on AMD.

Just reinforces my point. If it is truly "open" they should have made it available to both nVidia and Intel and their would have been no reason for nvidia to feel slighted.
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
You may not be saying that, but AMD certainly did. Huddy said, very clearly, that AMD was going to have sole control over Mantle for the express purpose of using it to better optimize their hardware.

That's the very definition of a closed, proprietary product.

AMD said that anyone would be free to write a driver to run Mantle on their hardware if they wish. No license needed. The source code will be freely available to anyone who wanted it. If that's not good enough for nVidia, or you, then fine, don't use it. The reason you've given though has nothing to do with it's performance potential or any advantages it might offer, which is why nVidia claimed they won't use it. I don't know why anyone is even arguing this angle. It's not what nVidia said when they were asked.

So nVidia is officially out of the picture as far as Mantle goes. Now we wait and see what Intel does. They at least seem to be approaching it with an open mind. Apparently they are willing to look and see if it can benefit their customers or not.
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
Precisely.

More specifically.

A true "open" software specification is held by a consortium of partners who collaborate on trying to achieve the best end result. True "open" software is modifiable and enhance-able by ANY MEMBER of that consortium of partners. This was brought up in the interview. AMD has no intention of doing this. So there are two problems here. Mantle requires GCN architecture, and even AMD's own non GCN GPUs cannot support Mantle. So someone please tell me how competing non GCN GPUs can support Mantle if AMD's own non GCN GPUs cannot. MUCH MORE IMPORTANTLY. AMD has expressly stated that Mantle is their own so it is not truly open by any proper definition. Open means that there is a standards board with many partners collaborating on the end result and implementation. AMD will not do this: AMD stated that Mantle will always, always be controlled by AMD and no one else. So it is not surprising that NV would not be interested. NV is not developing a GCN GPU and they would not support an API that isn't truly open - Mantle is not "open", it will always be under control by AMD and AMD alone.

None of this is the reason nVidia gave for not being interested in Mantle. Of course people are free to continually use it as a straw man instead of addressing what nVidia really said.

It's 2:00AM here. Good night folks.
 

ShintaiDK

Lifer
Apr 22, 2012
20,378
145
106
AMD said that anyone would be free to write a driver to run Mantle on their hardware if they wish. No license needed. The source code will be freely available to anyone who wanted it. If that's not good enough for nVidia, or you, then fine, don't use it. The reason you've given though has nothing to do with it's performance potential or any advantages it might offer, which is why nVidia claimed they won't use it. I don't know why anyone is even arguing this angle. It's not what nVidia said when they were asked.

So nVidia is officially out of the picture as far as Mantle goes. Now we wait and see what Intel does. They at least seem to be approaching it with an open mind. Apparently they are willing to look and see if it can benefit their customers or not.

Intel just called AMDs bluff. Thats what that was about.

And the BS about anyone can just write a driver is so far out.
 

SniperWulf

Golden Member
Dec 11, 1999
1,563
6
81
This whole podcast would have been better with just Tom Petersen talking. He should have left the other guy at the office.
 

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
This whole podcast would have been better with just Tom Petersen talking. He should have left the other guy at the office.

If anything it should have been the opposite. I find the knowledge of actual software developers (which rev IS a software developer) to be much more reality based than anything a marketing scumbag would say. Marketing guys really don't have a clue and will lie through their teeth. Perhaps Tom is that way, and certainly on the AMD side with clowns like Huddy and Hallock who can't write software but can talk up a storm in a marketing press release. And they both have done their full share of lying. I'd place much more value on the guys in the field doing the work. The actual software engineers. Not the marketing clowns who don't know what the hell they're talking about and will freely lie and probably lie so much they believe themselves. It's much more insightful.
 
Last edited:

Creig

Diamond Member
Oct 9, 1999
5,171
13
81
So what you say in short is, that AMD have no intention of sharing Mantle. And its been nothing but a PR spin to claim otherwise, while attacking the competition for closed standards. And thats assuming that Mantle could even run on any uarch besides GCN.

If AMD had any intention of getting Mantle to be widely adopted. They would have shared it. Even Qualcomm is on the D12 train.
Wrong.

"I know that Intel have approached us for access to the Mantle interfaces, et cetera," Huddy said. " And right now, we've said, give us a month or two, this is a closed beta, and we'll go into the 1.0 [public release] phase sometime this year, which is less than five months if you count forward from June. They have asked for access, and we will give it to them when we open this up, and we'll give it to anyone who wants to participate in this."

http://www.pcworld.com/article/2365909/intel-approached-amd-about-access-to-mantle.html
 

Mand

Senior member
Jan 13, 2014
664
0
0
AMD said that anyone would be free to write a driver to run Mantle on their hardware if they wish.

Ha, that's rich. That's a marketing scam, pure and simple. Not only does saying "hey you can make a driver too!" not make it open, not even all of AMD's own hardware can run Mantle. It requires GCN. If it requires GCN...wait for it...nobody else can use it. Unless AMD specifically changes Mantle to allow it. But...again...wait for it...if they specifically change Mantle to allow someone else to use it, they could theoretically change it back and screw a competitor over. Nvidia can't take that risk, especially when there's literally zero reward given DX12.

They may have said it, to get people like you to believe that AMD is the white knight of the graphics hardware world, and it seems to have worked. It has no basis in reality.

Mantle is not open, and it does not make a company evil to refuse to put its own products at the mercy of a direct competitor.

In a parallel universe, if Mantle had come out and Microsoft just threw up its hands and said "Guess what guys, no more DX!" then yeah, Mantle might have supplanted it and become an open standard. But it's not going to do that, because AMD is choosing to not make it open. And because AMD is exerting full control, their competitors will not use it. Period.
 
Last edited:

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
Ha, that's rich. That's a marketing scam, pure and simple. Not only does saying "hey you can make a driver too!" not make it open, not even all of AMD's own hardware can run Mantle. It requires GCN. If it requires GCN...wait for it...nobody else can use it.

They may have said it, to get people like you to believe that AMD is the white knight of the graphics hardware world, and it seems to have worked. It has no basis in reality.

Mantle is not open, and it does not make a company evil to refuse to put its own products at the mercy of a direct competitor.

Even in some fantasy land where it DID work on non GCN hardware, how the hell is an open standard controlled by one company? True open standards are consortium based. All members are free to input their own enhancements and additions. Will AMD do this? They said, NO, they will not. So it would be most interesting if Mantle doesn't work on non AMD GCN hardware, that by some weird fantasy land magic that it will work on competing non GCN hardware. How? For a 0% performance increase? Intel and NV can't have any input into changes in Mantle? And anyone acts surprised that NV isn't interested. Give me a break. Not having a consortium of members that can modify the API - That by definition is not open. Marketing scam, like you said.

Besides which, it doesn't even matter. On high end hardware the differences are usually not astounding. If NV wanted to create their own to the metal API, they could. But DX12 will be the standard for Windows based games, with all parties being able to have INPUT in changes and enhancements. Mantle will not allow this. This is the key difference. And it doesn't even matter. If NV wanted, they could simply make their own "to the metal" API, but they haven't.
 
Last edited:
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |