• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

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

Page 11 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
What is there to document? GCN has ACEs, older architectures don't. It doesn't work in DX 11.2 while it is supported in Mantle and DX 12.

I assume you cant document it? Why should I trust you then?

GCNs ACE is about equal to nVidias HyperQ. Something only found in Kepler and Maxwell.
 
Again, let's not make Mantle to be something it isn't.

imho,

When AMD releases the SDK, nVidia probably could create a driver to support it but is it worth spending precious resources on it ? Probably not!

One of my constructive nit-picks with ATI and now AMD is they may of felt that pro-active developer relations, investments in software were more marketing exercises and if we build the hardware the rest would follow!

Now as of late, AMD customers are seeing AMD invest into features like TRessFX and mantle and quite frankly wonderful to see!

Mantle is innovative, risky, welcomed, more efficient and brings competition and choice, brings awareness so industry standards may mature and evolve!

It's nice to see AMD risk instead of playing it safe! Standards are essential, paramount and make sense but in a very competitive landscape they can also restrain innovation and choices!

Proprietary is a nasty word to some but to move forward one has the ability to risk and bring more choice that the market may reward!
 
What is there to document? GCN has ACEs, older architectures don't. It doesn't work in DX 11.2 while it is supported in Mantle and DX 12.

Oh PLEASE. There is no AMD reference to this ANYWHERE, are you making this stuff up? Either find AMD documentation stating this specifically for Mantle or just stop saying it. AMD states that Mantle requires GCN, period, end of story.
 
Oh PLEASE. There is no AMD reference to this ANYWHERE, are you making this stuff up? Either find AMD documentation stating this specifically for Mantle or just stop saying it. AMD states that Mantle requires GCN, period, end of story.

http://developer.amd.com/wordpress/media/2013/06/2620_final.pdf

http://www.anandtech.com/show/4455/amds-graphics-core-next-preview-amd-architects-for-compute/5

http://community.amd.com/community/amd-blogs/amd-gaming/blog/2014/05/28/mantle-101
 
2 of your links only says that GCN got ACE.

And the last link says:
Asynchronous compute queue for overlapping of compute and graphics workloads.

So you have proven absolutely nothing of the claims you made.

Mantle supports AC, DX 11.2 doesn't. Show me where it says DX 11.2 has AC support. I can't find it.
 
imho,

When AMD releases the SDK, nVidia probably could create a driver to support it but is it worth spending precious resources on it ? Probably not!

One of my constructive nit-picks with ATI and now AMD is they may of felt that pro-active developer relations, investments in software were more marketing exercises and if we build the hardware the rest would follow!

Now as of late, AMD customers are seeing AMD invest into features like TRessFX and mantle and quite frankly wonderful to see!

Mantle is innovative, risky, welcomed, more efficient and brings competition and choice, brings awareness so industry standards may mature and evolve!

It's nice to see AMD risk instead of playing it safe! Standards are essential, paramount and make sense but in a very competitive landscape they can also restrain innovation and choices!

Proprietary is a nasty word to some but to move forward one has the ability to risk and bring more choice that the market may reward!
Why the constant exclamation marks?It is distracting and annoying to read.

If Nvidia doesn't care about mantle so what?
 
Mantle supports AC, DX 11.2 doesn't. Show me where it says DX 11.2 has AC support. I can't find it.

You havent delivered on your claim yet. And now you start to try kick the ball away to avoid it?

No. There is a specific feature which both DX 12 (the one that runs on current hardware) and Mantle support but DX 11.2 doesn't.

Common, you have to know this.

You havent even been able to show that Fermi and Haswell supports the feature you also claim is essential for DX12 And Mantle. Plus you havent even been able to show the feature is needed at all for the graphics portion of Mantle.

All you did was to make 3 links. 2 of them just saying that GCN got ACE. And the last that: Asynchronous compute queue for overlapping of compute and graphics workloads.

I even had to help you by telling you that Kepler and Maxwell supports it via the HyperQ.
 
I even had to help you by telling you that Kepler and Maxwell supports it via the HyperQ.

It's part of CUDA 2.0 so Fermi (at least the Tesla cards) support it. It is possible Fermi Geforce cards don't if they only had in their professional grade cards at that time.
 
Yea, since NV is not interested in Mantle why do we even bother if Mantle is NV hardware compatible ??

Because when Intel creates a driver for mantle......that will be nearly 70% of the market that uses mantle....

Its not if Intel does.....its when...😉 we have a couple more months before sdk will be released.....
 
Why the constant exclamation marks?It is distracting and annoying to read.

If Nvidia doesn't care about mantle so what?

Because they're subjectively stronger, emotional, opinionated wording!

I think nVidia does care! Based on there are improvements in efficiency and what person in their right mind is against efficiency?!

I believe nVidia likes or cares about the efficiency aspects but spending resources for a proprietary gaming API is not wise based on division and fragmentation; and is wiser to focus on current industry standards. Work and focus on getting the most efficiency out of DirectX 11 instead of offering it doesn't go far enough; and focus on DirectX 12 moving forward.
 
Because when Intel creates a driver for mantle......that will be nearly 70% of the market that uses mantle....

Its not if Intel does.....its when...😉 we have a couple more months before sdk will be released.....

I doubt it with the announcement of DirectX 12! Why would Intel spend resources on Mantle and DirectX 12?
 
Because they're subjectively stronger, emotional, opinionated wording!

I think nVidia does care! Based on there are improvements in efficiency and what person in their right mind is against efficiency?!

I believe nVidia likes or cares about the efficiency aspects but spending resources for a proprietary gaming API is not wise based on division and fragmentation; and is wiser to focus on current industry standards. Work and focus on getting the most efficiency out of DirectX 11 instead of offering it doesn't go far enough; and focus on DirectX 12 moving forward.
Yes I think that if Mantle cannot surpass DirectX in features and usability then it(Mantle) will die like so many other technologies that simply did not catch on for various reasons.
 
I believe nVidia likes or cares about the efficiency aspects but spending resources for a proprietary gaming API is not wise based on division and fragmentation; and is wiser to focus on current industry standards. Work and focus on getting the most efficiency out of DirectX 11 instead of offering it doesn't go far enough; and focus on DirectX 12 moving forward.
Wait, you know you are talking about Nvidia right? PhysX, Gsync, CUDA, GameWorks etc. these are all proprietary, closed systems. And DirectX is proprietary, I'm amazed how many people don't seem to realize this.

Also I've lost count how many times you've said that "division and fragmentation may bring innovation" now you're telling us all about how Nvidia should be pushing industry standards?
 
I doubt it with the announcement of DirectX 12! Why would Intel spend resources on Mantle and DirectX 12?

Interchangable - we already know that mantle will port very easy to dx 12; and dx 12 to mantle.....there is a reason for that......😉 DX12 first and formost isn't for PC but to save Xbox1.....MS still doesn't care about PC gamers; they would rather have everyone locked into Xbox platform......

Two - mantles made it into almost all the major engines. Once its there - any game made with those engines you won't need to do a lot for it. Your game will already support mantle. *as seen with PVZ Garden Ware*

three Intel isn't above using someone's else tech when it would help their igpus......

Four - PC gaming is still one area that is driving PC world; more and more people are wanting to game on their computers......doesn't matter if its a laptop or desktop....

Five - Intel's interested; I know how interested also. I can't say from who back in Oregon......but they are chomping at the bit to get ahold of it.

Six Intel has adopted AMD tech before....IE AMD 64; *of course that was a little bit of kicking and screaming to adopt; but non the less they did*.
 
Yes I think that if Mantle cannot surpass DirectX in features and usability then it(Mantle) will die like so many other technologies that simply did not catch on for various reasons.

It doesn't have to surpass though, as it is not linked to an OS. If it can be as good or nearly as good and keep running on Win7/8 and deliver performance and features... then it will continue to have a place in the market.
 
Yes I think that if Mantle cannot surpass DirectX in features and usability then it(Mantle) will die like so many other technologies that simply did not catch on for various reasons.

DirectX 12 will inevitably require windows 8. The fact that mantle can run on Windows 7 is a big deal. If AMD ports it to Linux, its an even bigger deal. Although I know a linux port is probably a year or two away. Mantle 1.0 will get it out there in a stable releasable form for everybody to use. Mantle 2.0 is going to add features and expand on what is already there.
 
It doesn't have to surpass though, as it is not linked to an OS. If it can be as good or nearly as good and keep running on Win7/8 and deliver performance and features... then it will continue to have a place in the market.

DirectX 12 will inevitably require windows 8. The fact that mantle can run on Windows 7 is a big deal. If AMD ports it to Linux, its an even bigger deal. Although I know a linux port is probably a year or two away. Mantle 1.0 will get it out there in a stable releasable form for everybody to use. Mantle 2.0 is going to add features and expand on what is already there.
I want Mantle to succeed.Nvidia needs to be kept in check in terms of their prices.I hope for 50/50% market share in discrete cards.
 
DX12 first and formost isn't for PC but to save Xbox1.....MS still doesn't care about PC gamers; they would rather have everyone locked into Xbox platform......

Sorry buts that's absolute nonsense. The Xbox One already has a low level API, and while D3D12 will bring refinements to that, the major gains will definitely be on PC..

DirectX has always been about PC...
 
To make it a Mantle/AMD shortcoming rather than an nVidia decision. nVidia said they aren't going to support it, not that they can't.

Is there really a distinction between a legal/technical prohibition and a business prohibition?

Nvidia can't take the risks involved in trying to support Mantle as long as AMD maintains full control. Is that really somehow more "open" than AMD saying outright that they can't use it?

Same result of Nvidia not supporting Mantle. Same entity, AMD, that's making the decisions that leads to the result.
 
Interchangable - we already know that mantle will port very easy to dx 12; and dx 12 to mantle.....there is a reason for that......

Wait...we know this how? Has either a Mantle SDK or ... well anything actually specific about DX12 been released? Sorry, I've been away for a while. And paging through this thread, I seem to remember why.
 
Back
Top