The AMD Mantle Thread

[smackababy]

You won't see anyone dropping DX support. M$ would buy AMD before they let that happen. Besides, There's too many non GCN systems out there that run DX for anyone to consider not using it.

They wouldn't have to buy AMD. Just block their hardware from Windows. "Mantle is now only available on Linux" = "MS just killed Mantle". >_>

I hope AMD actually announces a Mantle SDK release date. I'd be happy with access to even a beta. Just got a new 290 and wouldn't mind testing out Mantle myself.

 

[tg2708]

They wouldn't have to buy AMD. Just block their hardware from Windows. "Mantle is now only available on Linux" = "MS just killed Mantle". >_>

I hope AMD actually announces a Mantle SDK release date. I'd be happy with access to even a beta. Just got a new 290 and wouldn't mind testing out Mantle myself.

Unless they want to be sued for being anti-competitive or the like amd has nothing to worry about.

 

[Saylick]

Just to make you work harder (inserting new line) I'll go with 60%.

Thanks.

-----------------------
Assuming we will get results in the form of a bar chart across various games, we have the following estimates guesstimates for what Johan specifically will show us as the average improvement using whatever bench and settings he has:

0 - 4.99%: 0
5 - 9.99%: 0
10 - 14.99%: 1
15 - 19.99%: 0
20 - 24.99%: 3
25 - 29.99%: 0
30 - 34.99%: 3
35 - 39.99%: 1
40 - 44.99%: 0
45 - 49.99%: 0
50 - 54.99%: 2
55 - 59.00%: 0
60 - 64.99%: 1

 

[moonbogg]

I am really excited about mantle, but something tells me that its just too good to be true. Too many haters in the industry that don't want AMD to pull it off. I expect a few PC games to use it and that will be great, but I wouldn't expect actual wide spread adoption, because that would actually be really cool and we can't have that now can we? No, of course not.

 

[Erenhardt]

I am really excited about mantle, but something tells me that its just too good to be true. Too many haters in the industry that don't want AMD to pull it off. I expect a few PC games to use it and that will be great, but I wouldn't expect actual wide spread adoption, because that would actually be really cool and we can't have that now can we? No, of course not.

Not every game will support mantle. Don't expect Batman to be on the list, as they rejected patch from AMD to improve performance (can't find link now;/)

 

[bubbarock]

New here, but couldn't resist adding another line

0 - 4.99%: 0
5 - 9.99%: 0
10 - 14.99%: 1
15 - 19.99%: 0
20 - 24.99%: 3
25 - 29.99%: 0
30 - 34.99%: 3
35 - 39.99%: 1
40 - 44.99%: 0
45 - 49.99%: 0
50 - 54.99%: 2
55 - 59.00%: 0
60 - 64.99%: 1
>70% = Ridicule: 1

 

[tg2708]

Not every game will support mantle. Don't expect Batman to be on the list, as they rejected patch from AMD to improve performance (can't find link now;/)

What a pathetic move. I would love to see that post.

 

[AnandThenMan]

I am really excited about mantle, but something tells me that its just too good to be true. Too many haters in the industry that don't want AMD to pull it off. I expect a few PC games to use it and that will be great, but I wouldn't expect actual wide spread adoption, because that would actually be really cool and we can't have that now can we? No, of course not.

I've had a similar train of thought. Will be interesting to see if/how Intel and Nvidia will attempt to hamper Mantle adoption.

What a pathetic move. I would love to see that post.

I think it was a statement by AMD. But given what has happened in the past, it is very likely true.

 

[Erenhardt]

I've had a similar train of thought. Will be interesting to see if/how Intel and Nvidia will attempt to hamper Mantle adoption.

I think it was a statement by AMD. But given what has happened in the past, it is very likely true.

Fortunately nvidia users should be too concerned as amd and GE partners care about gamers. When nvidia required patch to Tomb Raider (2013) to fix the issues, it was swiftly patched. Going this route, games that opted for Mantle should still be running good on directX.

 

[SlowSpyder]

Mantle for BF4 is supposed to come in December? Does anyone know the date? Sorry, I haven't been following the whole thread.

 

[psoomah]

I've had a similar train of thought. Will be interesting to see if/how Intel and Nvidia will attempt to hamper Mantle adoption.

I think it was a statement by AMD. But given what has happened in the past, it is very likely true.

Not sure why Intel would feel a need to hamper Mantle, while Nvidia is no doubt desperate to hamper Mantle right about now, but what could either of them do if they wanted to?

With Mantle looking like it will become pervasive on next gen PC games, Nvidia is going to be THE big loser, at least short term, but Microsoft might be another big loser with Mantle potentially providing a ginormous boost to Steam OS.

I know Gabe keeps focusing on Intel and Nvidia, but a fully HSA/hUMA capable Kaveri + GCN AIB is going to be the killer cost/performance Steam Machine and AMD's 'Thracks' indicated AMD will be ready with optimised high performance drivers when AMD powered Steam Machines launch.

 

[psoomah]

Mantle for BF4 is supposed to come in December? Does anyone know the date? Sorry, I haven't been following the whole thread.

Date unspecified, though DICE's Johan Andersson teased a possible late November BF4 Mantle update.

 

[AnandThenMan]

I'm going on the assumption that Intel wants to be seen as viable for entry to entry/mid level gaming performance, that's not true? As for Nvidia, they will continue to push their differentiating features such as PhysX. Of course it all hinges on how potent Mantle really is, if it offers 30% better performance, that's as good as a product lineup refresh without any new silicon.

This is a scary proposition for Nvidia if it comes to fruition. Big if though IMO, sounds too good to be true. We'll see.

 

[psoomah]

I'm going on the assumption that Intel wants to be seen as viable for entry to entry/mid level gaming performance, that's not true? As for Nvidia, they will continue to push their differentiating features such as PhysX. Of course it all hinges on how potent Mantle really is, if it offers 30% better performance, that's as good as a product lineup refresh without any new silicon.

This is a scary proposition for Nvidia if it comes to fruition. Big if though IMO, sounds too good to be true. We'll see.

Intel can afford to watch what happens in what is for them a niche market as they grow their GPU capabilities.

Nvidia on the other hand is all but shut out of every mobile market but it's own Tegra chip wins and is about to have a chainsaw taken to it's profitability in the AIB space. So very scary indeed.

Sounding too good to be true doesn't mean it is too good to be true. Indications are positive on that count.

 

[SlowSpyder]

Date unspecified, though DICE's Johan Andersson teased a possible late November BF4 Mantle update.

Thanks.

 

[selni]

Mantle is immediately and specifically targeting gamers looking to upgrade their present AIBs to prepare for next gen gaming demands.

The PR effort sure might be, but we have no idea whether the API is or not yet so I'm not sure why you're phrasing that it's aimed at high end gaming as a fact.

But the thing is, everything from AMD from now on will be based on GCN. Every product from the lowest end APU and highest end GPU will be based on it.
If Mantle were to live up to the hype and give huge performance increases, enough to make modern demanding games to run ok even on APUs, it could be a big win for the studios.
Only one day to go now, hopefully most questions will be answered.

DirectX 10 and 11 didn't start off with a huge userbase either, and they both required the users to get Vista or Win7 in order to use it.

Yeah it could be very significant for APUs - that and reducing single thread CPU performance requirements (which is an issue with DX that can heavily favour intel) are what I suspect the real focus is. Looking at the performance/resolution the new consoles are rumoured to achive and I'd think we're still probably talking 1366x768 at low/medium settings, but that in itself wouldn't be a bad achievement for integrated graphics by any means.

I don't think DX adoption is the best comparison - Microsoft's market position means they can do pretty much whatever they want and bully everyone into following.

 

[redzo]

I am really excited about mantle, but something tells me that its just too good to be true. Too many haters in the industry that don't want AMD to pull it off. I expect a few PC games to use it and that will be great, but I wouldn't expect actual wide spread adoption, because that would actually be really cool and we can't have that now can we? No, of course not.

It's not about haters. It's just not healthy for the industry to support a vendor specific api. Same think thing applies for nvidia physx. Even if physx and mantle serve different purposes, they suffer from the same thing: they're both vendor specific. It isn't economically feasible to support an api that increases dev cost.

 

[Saylick]

It's not about haters. It's just not healthy for the industry to support a vendor specific api. Same think thing applies for nvidia physx. Even if physx and mantle serve different purposes, they suffer from the same thing: they're both vendor specific. It isn't economically feasible to support an api that increases dev cost.

Have you considered the case where the opposite was true? What if it turns out that the net effect of using Mantle was that developers are able to generate more than enough revenue to offset the time/cost of writing code in Mantle?

The way I see things is that if Mantle indeed provides a significant performance uplift for little to no additional hardware cost then that means more people are capable of meeting the hardware requirements which translates into more sales. More importantly, these performance improvements helps the mainstream consumer more than the hardcore enthusiast, which is important to note because there are far more mainstream consumers to begin with.

See attached image:

 

[Will Robinson]

It's not about haters. It's just not healthy for the industry to support a vendor specific api. Same think thing applies for nvidia physx. Even if physx and mantle serve different purposes, they suffer from the same thing: they're both vendor specific. It isn't economically feasible to support an api that increases dev cost.

Yes,except GCN hardware is the basis for the two most powerful/popular consoles on the market that will sell in the 10's of millions and also AMD graphics cards which have about 40% of the discreet graphics card markets.
That's a lot different to Physx which is limited to NVDA graphics cards.

 

[dacostafilipe]

Article about Mantle on hardware.fr with new information:

http://www.google.com/translate?hl=.../13445/apu13-amd-mantle-premiers-details.html

 

[SiliconWars]

Article about Mantle on hardware.fr with new information:

http://www.google.com/translate?hl=.../13445/apu13-amd-mantle-premiers-details.html

The application also acquires the ability to take control of the multi-GPU and to decide where to run each command issued. Why AMD has provided in Mantle access CrossFire compositing engine, data transfer between GPU etc.. Will allow multi-GPU modes that go beyond the AFR and adapt better example to use GPU computing in games or asymmetric multi-GPU systems as is the case for APU combined with a GPU. For example it is possible to imagine the GPU load based rendering and APU handle the post processing.

This is it, AMD's holy grail.

 

[Saylick]

Article about Mantle on hardware.fr with new information:

http://www.google.com/translate?hl=.../13445/apu13-amd-mantle-premiers-details.html

The link doesn't display images for me. Here it is reposted.

To quote the article [translated]:

In the end, according Mantle will allow AMD to help entry-level systems, to enable developers to better predict the behavior and performance of their engines as well as sharing optimizations between PC and next gen consoles, and more long term to allow the introduction of new rendering techniques.

To do this, however, AMD need effective tools and indicates that everything is on track. In addition, AMD has integrated a large part of its debugging features and validation directly in the API, which facilitates the creation of efficient and accurate tools on possible bottlenecks. Developers can also integrate directly access to various counters GPU in their own tools, a more comprehensive manner than at present.

AMD says its next development is progressing well and the timing for the December patch Mantle Battlefield 4 remains valid. Currently, Mantle is available in alpha release to a handful of developers, but next month a beta version should be available to all developers who request it. For public documentation, it will probably take the GDC in March 2014 while the availability of the final version of Mantle is scheduled for the second half of next year. In the meantime, the API support will of course included in all drivers.

AMD insists that Mantle was not intended to be limited to architecture. The base Mantle merely relatively generic functions that could be supported by other architectures while a extended Mantle level adds support for specific functions currently Radeon. Mantle and could potentially become a standard with extensions, but nothing says that interest Nvidia or AMD that management will not cause it difficult to accept conditions.

Note that the AMD forum being closed to anyone, the output of the presentation linked to Mantle, we have crossed several employees of NVIDIA, one of the main architects of the GeForce GPU. AMD's competitor appears visibly as curious as us about Mantle!

Looks like we won't get hard numbers after all... until December's BF4 patch at least.

 

[Abwx]

Article about Mantle on hardware.fr with new information:

http://www.google.com/translate?hl=.../13445/apu13-amd-mantle-premiers-details.html

Last lines are funny..

Note that the AMD forum being closed to anyone, the output of the presentation linked to Mantle, we have crossed several employees of NVIDIA, one of the main architects of the GeForce GPU. AMD's competitor appears visibly as curious as us about Mantle!

 

[frozentundra123456]

I just can't believe people keep claiming mantle will cut costs. The only way it can cut costs is if fewer resources are directed toward direct x. Or at least for the developers, if AMD subsidizes the development. The additional resources needed may be worth it, but it will require additional time (money) to add another code path and test/debug it.

I also an not convinced it will lead to a huge increase in game sales. Console sales are probably 90% over the market for new big name titles. Even if mantle were to increase PC sales of a game 50% (highly unlikely in my opinion) that would only be a 5% increase in total sales, assuming none of those sales were taken away from consoles.

 

[GaiaHunter]

I just can't believe people keep claiming mantle will cut costs. The only way it can cut costs is if fewer resources are directed toward direct x. Or at least for the developers, if AMD subsidizes the development. The additional resources needed may be worth it, but it will require additional time (money) to add another code path and test/debug it.

I also an not convinced it will lead to a huge increase in game sales. Console sales are probably 90% over the market for new big name titles. Even if mantle were to increase PC sales of a game 50% (highly unlikely in my opinion) that would only be a 5% increase in total sales, assuming none of those sales were taken away from consoles.

Apparently Mantle is also a set of developing tools.