AMD's Richard Huddy on the state of PC graphics, Mantle 2 and APUs

Page 4 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

BrightCandle

Diamond Member
Mar 15, 2007
4,762
0
76
If AMD wants Mantle to be open then it needs to be run by a standards board. All the IHVs have an equal stake in its future development so that no one of them can add features that would hurt their competitor. More than that an amount of backwards compatibility on future cards is necessary. That is what open means.

AMD can not claim its open while they are sole say on what gets added and when. The reason is AMD has enormous potential to hurt its competitors. It develops a hardware feature in secret, tells only a few key developers about it, then the hardware, driver and games release at the same time. This is actually the advantage of control buddy is talking about, the ability to fully utilise all the features. The problem is on day 1 Nvidia/Intel won't even have a driver that can run the game. Worse than that they will be attempting to do the feature in software, with hardware cycles as they are its probably 2 years before they can release a card that can utilise that feature.

I can't stress what a terrible situation that would be, how crippled the competitors could be by this. So until its actually open its a proprietary API that AMD controls and that means the other hardware developers are better off not going anywhere near it, because it creates a serious problem with competition.
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
If AMD wants Mantle to be open then it needs to be run by a standards board. All the IHVs have an equal stake in its future development so that no one of them can add features that would hurt their competitor. More than that an amount of backwards compatibility on future cards is necessary. That is what open means.

AMD can not claim its open while they are sole say on what gets added and when. The reason is AMD has enormous potential to hurt its competitors. It develops a hardware feature in secret, tells only a few key developers about it, then the hardware, driver and games release at the same time. This is actually the advantage of control buddy is talking about, the ability to fully utilise all the features. The problem is on day 1 Nvidia/Intel won't even have a driver that can run the game. Worse than that they will be attempting to do the feature in software, with hardware cycles as they are its probably 2 years before they can release a card that can utilise that feature.

I can't stress what a terrible situation that would be, how crippled the competitors could be by this. So until its actually open its a proprietary API that AMD controls and that means the other hardware developers are better off not going anywhere near it, because it creates a serious problem with competition.

DX is proprietary too. What you are asking AMD to do to satisfy you is never going to happen. They will make everything available for others to use it if they decide to, but they aren't going to simply hand it over to a committee. We already have an API that's managed that way.

With everything "open" by AMD definition any shens they could decide to pull will be right there for everyone to see.
 

monstercameron

Diamond Member
Feb 12, 2013
3,818
1
0
If AMD wants Mantle to be open then it needs to be run by a standards board. All the IHVs have an equal stake in its future development so that no one of them can add features that would hurt their competitor. More than that an amount of backwards compatibility on future cards is necessary. That is what open means.

AMD can not claim its open while they are sole say on what gets added and when. The reason is AMD has enormous potential to hurt its competitors. It develops a hardware feature in secret, tells only a few key developers about it, then the hardware, driver and games release at the same time. This is actually the advantage of control buddy is talking about, the ability to fully utilise all the features. The problem is on day 1 Nvidia/Intel won't even have a driver that can run the game. Worse than that they will be attempting to do the feature in software, with hardware cycles as they are its probably 2 years before they can release a card that can utilise that feature.

I can't stress what a terrible situation that would be, how crippled the competitors could be by this. So until its actually open its a proprietary API that AMD controls and that means the other hardware developers are better off not going anywhere near it, because it creates a serious problem with competition.

don't conflate open source/proprietary and having a single IHV controlling the spec but I do agree that AMD would need to lessen the risk for other parties.
 

Noctifer616

Senior member
Nov 5, 2013
380
0
76
If AMD wants Mantle to be open then it needs to be run by a standards board. All the IHVs have an equal stake in its future development so that no one of them can add features that would hurt their competitor. More than that an amount of backwards compatibility on future cards is necessary. That is what open means.

AMD can not claim its open while they are sole say on what gets added and when. The reason is AMD has enormous potential to hurt its competitors. It develops a hardware feature in secret, tells only a few key developers about it, then the hardware, driver and games release at the same time. This is actually the advantage of control buddy is talking about, the ability to fully utilise all the features. The problem is on day 1 Nvidia/Intel won't even have a driver that can run the game. Worse than that they will be attempting to do the feature in software, with hardware cycles as they are its probably 2 years before they can release a card that can utilise that feature.

I can't stress what a terrible situation that would be, how crippled the competitors could be by this. So until its actually open its a proprietary API that AMD controls and that means the other hardware developers are better off not going anywhere near it, because it creates a serious problem with competition.

So Mantle will be no different than DirectX in that regard?

Also, if I remember correctly, Mantle does support extensions and it could be possible to expose certain new hardware features trough extensions even though there is no native support in the API (going from stuff AMD people said).
 

Genx87

Lifer
Apr 8, 2002
41,091
513
126
DX is proprietary too. What you are asking AMD to do to satisfy you is never going to happen. They will make everything available for others to use it if they decide to, but they aren't going to simply hand it over to a committee. We already have an API that's managed that way.

With everything "open" by AMD definition any shens they could decide to pull will be right there for everyone to see.

Of course DX is proprietary as well. The difference is DX is proprietary to an OS. Mantle is proprietary to hardware. Where would you as a consumer want a proprietary standard?
 

PPB

Golden Member
Jul 5, 2013
1,118
168
106
Of course DX is proprietary as well. The difference is DX is proprietary to an OS. Mantle is proprietary to hardware. Where would you as a consumer want a proprietary standard?

Well, you can only run that OS in an specific hardware ISA, so DX is by extension also propietary to hardware.
 

PPB

Golden Member
Jul 5, 2013
1,118
168
106
That is a real stretch.

It is not, as the end result is the same: DX can run on specific hardware because of the OS lock in, and Mantle can run on specific hardware, because of the feature set needed to make it run.
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
That is a cop out. I'm sure you are A-OK with PhysX then right?

Why? Is nVidia releasing the source code and allowing AMD to implement it on their hardware with no licensing or fees? I'm not a big fan of *PhysX as it's been used, but in that situation I might.



*We of course are talking about GPU PhysX. Before we have people chiming in that PhysX also runs on the CPU.
 

Genx87

Lifer
Apr 8, 2002
41,091
513
126
It is not, as the end result is the same: DX can run on specific hardware because of the OS lock in, and Mantle can run on specific hardware, because of the feature set needed to make it run.

Right except the market we are talking about everybody is using that hardware base. Now we are discussing whether within that sphere is it better for us to be locked in at the software level(DX) or hardware level(AMD).

I think what you are saying is a real stretch to make these two equivalent.
 

Genx87

Lifer
Apr 8, 2002
41,091
513
126
Why? Is nVidia releasing the source code and allowing AMD to implement it on their hardware with no licensing or fees? I'm not a big fan of *PhysX as it's been used, but in that situation I might.



*We of course are talking about GPU PhysX. Before we have people chiming in that PhysX also runs on the CPU.

Claiming "I want the best API" is a cop out. Do you prefer proprietary standards on the software or hardware levels?
 
Last edited:

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
Huddy literally said that they will never optimize Mantle for other hardware and that they will have the last word for everything but on the other hand they demand access to nVidia's Gameworks libaries and that nVidia must optimize those for them, too.

Hypocrite.

That is an interesting point. AMD is not out to save the world with mantle, they're trying to make money like everyone else but they're just not good at making money, they're doing all the wrong things (marketing focused instead of software focused) to make money. Anyway, It is interesting how AMD demands gameworks access but NV and intel cannot add Mantle features for their hardware. Hilarious. Hypocrite indeed. Besides which, there are TOO many things that don't add up. MS demonstrated DX12 on NV hardware with NV beta drivers. How hilarious is that if people here claim DX12 is a Mantle template (yeah right). It's a template for AMD hardware and AMD hardware only. NV hardware and intel hardware is not designed similarly to AMD hardware, and direct metal optimizations are far different. And then non GCN AMD hardware cannot use Mantle, but here we are to believe Huddy telling us that competing non GCN hardware can use Mantle (YEAH RIGHT). I can just see how that works. Mantle "works" but it's a zero percent performance improvement because Mantle has literally nothing specific for NV or intel hardware. Mantle will never be the best API for anything except AMD hardware, as it is designed SOLELY for AMD hardware. And it doesn't even work on all AMD DX11 hardware.

Why can't AMD marketing just be up front and say Mantle is "ours" and is for "our hardware"? That would be completely fine. Mantle will never implement features for NV hardware or intel hardware. Period end of story. Therefore even if Mantle WORKED, which I highly doubt it would, it would not benefit other architectures. Hell AMD cannot benefit their own non GCN architectures with Mantle. LOL. Again, AMD marketing should just be up front instead of having clowns like Huddy lie through their teeth. Heck, AMD TOLD PCPer that they were closing Mantle due to DX12 being in development (they mentioned this on a podcast).

But NOW we have this clown Huddy who is still insisting it will work on everything. Yeah. Backtracking. Why can't this guy just be up front? Mantle is theirs. And that is okay. I don't care if Mantle doesn't work on competing hardware. The "good guy" image marketing is just nauseating at this point especially they're being especially hypocritical about gameworks. Just call Mantle what it is: a software value add for AMD users to help AMD sell cards. And that is completely fine. That's what AMD should be doing: creating value adds to make their cards more appealing, nothing wrong with that. But this nonsense they're pushing about Mantle being designed for everything and Mantle == DX12, please, excuse me while I roll my eyes. In the mythical world where Mantle COULD work on everything but not implementing ANYTHING feature wise for intel or NV hardware? And they demand source code to gameworks after pushing such hypocritical nonsense. Yeah........okay.
 
Last edited:

PPB

Golden Member
Jul 5, 2013
1,118
168
106
Right except the market we are talking about everybody is using that hardware base. Now we are discussing whether within that sphere is it better for us to be locked in at the software level(DX) or hardware level(AMD).

We are using the hardware base because we can't even choose to use another one. Luckily that soon will be stopping being the case, thanks to the software leaving behind the dreadful Wintel monopoly we are unfortunately so accustomed to.

You are trying to battle a suspected closed standard when your entire life was spent using another one a lot whole more harmful to the PC industry as a whole. Talk about hypocrisy.
 

Noctifer616

Senior member
Nov 5, 2013
380
0
76
MS demonstrated DX12 on NV hardware with NV beta drivers. How hilarious is that if people here claim DX12 is a Mantle template (yeah right). It's a template for AMD hardware and AMD hardware only. NV hardware and intel hardware is not designed similarly to AMD hardware, and direct metal optimizations are far different. And then non GCN AMD hardware cannot use Mantle, but here we are to believe Huddy telling us that competing non GCN hardware can use Mantle (YEAH RIGHT).

So, how does DirectX 12, which tries to solve the same issues as Mantle and also adds close to the metal control, manage to work on a wide range of hardware?

Is MS lying when they claim DX 12 is closer to the metal?
 

Genx87

Lifer
Apr 8, 2002
41,091
513
126
We are using the hardware base because we can't even choose to use another one. Luckily that soon will be stopping being the case, thanks to the software leaving behind the dreadful Wintel monopoly we are unfortunately so accustomed to.

This is where I take issue with your equivalency of what we are talking about. We all acknowledge DX is proprietary with the PC space. But we aren't discussing the wintel monopoly. We are talking about DX vs Mantle within PC gaming.

You are trying to battle a suspected closed standard when your entire life was spent using another one a lot whole more harmful to the PC industry as a whole. Talk about hypocrisy.

That is debateable. I remember when we had competing hardware and software API's in the 1990s. DX being a hardware agnostic proprietary standard unified PC gaming development, and made it easier for developers to bring out games. But that is a different thread.
 

Genx87

Lifer
Apr 8, 2002
41,091
513
126
So, how does DirectX 12, which tries to solve the same issues as Mantle and also adds close to the metal control, manage to work on a wide range of hardware?

Is MS lying when they claim DX 12 is closer to the metal?

Because it wont be as close as Mantle. While it should be closer to the metal. It wont be as close as Mantle. Also, hardware developers will have DX12 standards on the hardware and software layers that will make it compatible. If Mantle was open and not tied to AMDs GCN or if Intel and Nvidia adopted GCN the same could be accomplished using Mantle.
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
Claiming "I want the best API" is a cop out. Do you prefer proprietary standards on the software or hardware levels?

Neither. But I think you are overreacting. And wanting the best API is not a cop out. In the end DX is still there. Msft isn't abandoning it and leaving AMD to wield Mantle like a sword over nVidia's and Intel's heads. You mentioned PhysX. If nVidia decided to offer it up for the other IHV's to use, I'd be happy with that and I'm pretty sure the IHV's would be too.
 

Noctifer616

Senior member
Nov 5, 2013
380
0
76
Because it wont be as close as Mantle. While it should be closer to the metal. It wont be as close as Mantle. Also, hardware developers will have DX12 standards on the hardware and software layers that will make it compatible. If Mantle was open and not tied to AMDs GCN or if Intel and Nvidia adopted GCN the same could be accomplished using Mantle.

Sorry, but that is pure BS. According to an interview with one of the MS devs DX12 will be in the worst case scenario just as good as Mantle. He expects the API to be better than Mantle actually.

So, how can they accomplish better performance when it's not as close to the metal?

http://channel9.msdn.com/Events/Build/2014/9-004

This is the interview.
 

Genx87

Lifer
Apr 8, 2002
41,091
513
126
Sorry, but that is pure BS. According to an interview with one of the MS devs DX12 will be in the worst case scenario just as good as Mantle. He expects the API to be better than Mantle actually.

So, how can they accomplish better performance when it's not as close to the metal?

Shocking MS would claim their API better than the competition
 

sontin

Diamond Member
Sep 12, 2011
3,273
149
106
DX12 supports full Multithreading. Mantle is still missing this in the current release.
That's the reason why nVidia's DX11 driver is so much better than AMD's one and why they are able to win in some Star Swarm tests over Mantle.
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
DX12 supports full Multithreading. Mantle is still missing this in the current release.
That's the reason why nVidia's DX11 driver is so much better than AMD's one and why they are able to win in some Star Swarm tests over Mantle.

What you've said here doesn't make any sense.

DX12 doesn't support anything yet. It's not out. How can what DX12 is going to have influence nVidia's DX11 drivers vs AMD's DX11 drivers?
 

desprado

Golden Member
Jul 16, 2013
1,645
0
0
DX12 supports full Multithreading. Mantle is still missing this in the current release.
That's the reason why nVidia's DX11 driver is so much better than AMD's one and why they are able to win in some Star Swarm tests over Mantle.
This...........In even GTX 760 can beat R9 280X Dx11 in star swarm.
AMD stance has changed alot towards mantle first when it was announced than AMD was claiming that R9 290X with Mantle will be significantly faster than titan in BF4 but it did not happen like AMD planned and now Mantle is target customer are low and mid end user mostly.
 
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/    |