DirectX and OpenGL to get low level access according to leaked statements (GDC 2014)

Page 8 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

el etro

Golden Member
Jul 21, 2013
1,581
14
81
At least PhysX runs on other hardware.

Like butter. :biggrin:


Lolwut. What are you? AMD marketing guy? Mantle will hurt nVidia user because AMD partners are not allowed to implement proper DX support.


Proper DX support can be implemented outside of a blackbox in the code, no?
Why restrict AMD to access to this code?



No difference to Tessellation (only PN triangles is allowed) which is nearly absent in every supported game - BF4, Crysis 3, Thief 3, Sleeping Dogs, Hitman Absolution (only PN)...


Problem in Batman is the performance disparity between cards, like in Splinter Cell when HBAO+ and all other Filters are enabled.


Four months after TombRaider Launch: http://tpucdn.com/reviews/Gigabyte/GeForce_GTX_760_WindForce_OC/images/tombraider_1920_1080.gif

Hitman problem is like AC3 problem: http://tpucdn.com/reviews/ASUS/ARES_II/images/ac3_1920_1200.gif then http://tpucdn.com/reviews/EVGA/GTX_780_Ti_SC_ACX_Cooler/images/ac3_1920_1080.gif Where's AC4 not.





And why didn't you post a link to AMD's forward+ implementation in Dirt:Showdown or the sabotage act in Tomb Raider? I mean you are right, Mantle will not hurt nVidia's userbase. Because like Tomb Raider: It wont run on their hardware. :biggrin:

Dirt Showdown benchmark nowdays: http://imagescdn.tweaktown.com/content/5/5/5566_28_nvidia_geforce_gtx_760_2gb_video_card_review.png



I honestly think that TressFX bogging down on Kepler is because Kepler just isn't good at compute like GCN or even Fermi.

Look at latest TPU Tomb Raider Benchmark: http://tpucdn.com/reviews/MSI/R9_290X_Gaming/images/tombraider_1920_1080.gif
 
Last edited:

Phynaz

Lifer
Mar 13, 2006
10,140
819
126
Mantle, TressFX, Physx.

My Spidy Sense tells me you guys want this thread locked.
 

BFG10K

Lifer
Aug 14, 2000
22,709
2,980
126
Physx investment goes back?
No, which is why hardware PhysX is the same failure as Mantle is. Nobody else will ever support it so it remains a niche that only works on certain hardware, and only in rare games.
 

desprado

Golden Member
Jul 16, 2013
1,645
0
0
But it still compatible with all of x86 based computers environment.



ShintaiDK constructive posts rate is 1/10.
Sir this is technical thread so dont make any BS statement and false claims that Console will support mantle and even Mantle is not support by all amd than let alone other platform.Some people here are crying about Mantle because deep down they now Mantle is only here for limited time, limited games and it has been a big fail for high end PC.

Some people here dont even now that Mantle is liability for developers because AMD has to pay for each game for applying Mantle just for 5% of user and 5% to 15% performance increase with additional bugs and lower image quality.
 
Last edited:

VulgarDisplay

Diamond Member
Apr 3, 2009
6,193
2
76
This thread cracks me up. Alpha software that is in a game already is now irrelevant because of vaporware that doesn't exist and existing software that no triple AAA devs use.

There are no guarantees that low level dx will be as good as mantle but we are already giving it a knock out victory. If Microsoft has proven anything to me it's that they are just as capable of over promising and under delivering ad amd with the added problem if being completely out of touch with gamers and developers.

I think the route they need to go is like mantle. Let the GPU vendors do the abstraction layers and Microsoft come up with feature sets and the shader language.
 

VulgarDisplay

Diamond Member
Apr 3, 2009
6,193
2
76
Sir this is technical thread so dont make any BS statement and false claims that Console will support mantle and even Mantle is not support by all amd than let alone other platform.Some people here are crying about Mantle because deep down they now Mantle is only here for limited time, limited games and it has been a big fail for high end PC.

Some people here dont even now that Mantle is liability for developers because AMD has to pay for each game for applying Mantle just for 5% of user and 5% to 15% performance increase with additional bugs and lower image quality.

When calling someone out for lying at least make sure you understand what he is saying. There is a huge difference between hardware being compatible with mantle and hardware using mantle.

As for your claims about mantle only giving a 5-15% performance increase I will tell you that you are wrong. Its not your fault that reviewers aren't giving extensive coverage of mantle though. I can only assume its because of pressure from corporate sponsors that wanted mantle reviews to be shown in a negative light. We saw 5 articles a week about horrible frame pacing on amd cards, but now that mantle is working well in bf4 we haven't seen squat from reviewers. These guys make their living off page views but we aren't seeing articles about mantle updates anywhere.

In my testing as a competitive player on a mid range quad core my minimums have increased by over 45fps in bf4. 62% increase in the worst case scenarios and a basically flat frame time graph the whole time I play.
 

el etro

Golden Member
Jul 21, 2013
1,581
14
81
No, which is why hardware PhysX is the same failure as Mantle is. Nobody else will ever support it so it remains a niche that only works on certain hardware, and only in rare games.

IMO the greatest benefit on Mantle is that the API is not tied to Windows.

And part of AMD efforts in the Close-to-Metal research are being used in the new API initiatives.... (the extensions to improve OpenGL for example).



dont make any BS statement and false claims

Some people here dont even now that Mantle is liability for developers because AMD has to pay for each game for applying Mantle just for 5% of user and 5% to 15% performance increase with additional bugs and lower image quality.
Some people here are crying about Mantle because deep down they now Mantle is only here for limited time, limited games and it has been a big fail for high end PC.
 

desprado

Golden Member
Jul 16, 2013
1,645
0
0
When calling someone out for lying at least make sure you understand what he is saying. There is a huge difference between hardware being compatible with mantle and hardware using mantle.

As for your claims about mantle only giving a 5-15% performance increase I will tell you that you are wrong. Its not your fault that reviewers aren't giving extensive coverage of mantle though. I can only assume its because of pressure from corporate sponsors that wanted mantle reviews to be shown in a negative light. We saw 5 articles a week about horrible frame pacing on amd cards, but now that mantle is working well in bf4 we haven't seen squat from reviewers. These guys make their living off page views but we aren't seeing articles about mantle updates anywhere.

In my testing as a competitive player on a mid range quad core my minimums have increased by over 45fps in bf4. 62% increase in the worst case scenarios and a basically flat frame time graph the whole time I play.

Yes it is fault of every one that they did not give good reviews about Mantle even good site like Pcper,Anandtech and guru3d did not even have 7% of advantage on 2560 by 1440 even AMD offical Benchmark is lie about mantle because there not much gain on high resolution that why Mantle is not here for so much time.

Do u now that Thief is badly flopped and on Nvidia offical site many fans has boycott that game even AMD guys has not bought that games because they were promised that AMD will provide Mantle as soon as game release.That is why Mantle is headache if Nvidia fan boys don't buy that game than only game will purchase between 30% of PC user meaning that Developers will be wasting time on PC.
 

VulgarDisplay

Diamond Member
Apr 3, 2009
6,193
2
76
Yes it is fault of every one that they did not give good reviews about Mantle even good site like Pcper,Anandtech and guru3d did not even have 7% of advantage on 2560 by 1440 even AMD offical Benchmark is lie about mantle because there not much gain on high resolution that why Mantle is not here for so much time.

Do u now that Thief is badly flopped and on Nvidia offical site many fans has boycott that game even AMD guys has not bought that games because they were promised that AMD will provide Mantle as soon as game release.That is why Mantle is headache if Nvidia fan boys don't buy that game than only game will purchase between 30% of PC user meaning that Developers will be wasting time on PC.

You prove my point for me. Review sites all did cookie cutter reviews with $1000 CPUs and only used GPU bound settings and called it a day. Many competitive and 120hz users do not play games like that. For them mantle provides the most benefit. No reviewers have looked into such scenarios at all, and it really shows that they are doing it on purpose. I can't even remember a review that studied minimum fps and frame pacing in depth. Sad really because most people now have a skewed perception of what mantle can provide them.
 

el etro

Golden Member
Jul 21, 2013
1,581
14
81
Yes it is fault of every one that they did not give good reviews about Mantle even good site like Pcper,Anandtech and guru3d


Is related to the possibilities that you can build your hardware. If tech is showing 5-7% gains even the worst case possibilities, it can't not be good.

This Mantle performance advantage(in FPS or frametime numbers) is what we can expect in from last DX/OGL initiatives from Nvidia/AMD/Intel/MSFT.

Consoles have closer to metal ways of coding, but due to hardware limitation developers need many times to do(especially in multiplataform games) tricks to attenuate the loss of quality of their games(like upscaling and improvise softer anti-aliasing forms) when the original(source) game has too many detail.

did not even have 7% of advantage on 2560 by 1440 even AMD offical Benchmark is lie about mantle because there not much gain on high resolution that why Mantle is not here for so much time.

Mantle is for CPU bound situations and for CPU functions on PC gaming. 1440p turns the things so much GPU-bound. Would be good to see what Two top-end cards(Like two 290x Lightning on the same CPU) can do in this same condition.
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
Lolwut. What are you? AMD marketing guy? Mantle will hurt nVidia user because AMD partners are not allowed to implement proper DX support. No difference to Tessellation (only PN triangles is allowed) which is nearly absent in every supported game - BF4, Crysis 3, Thief 3, Sleeping Dogs, Hitman Absolution (only PN)...
And why didn't you post a link to AMD's forward+ implementation in Dirt:Showdown or the sabotage act in Tomb Raider? I mean you are right, Mantle will not hurt nVidia's userbase. Because like Tomb Raider: It wont run on their hardware. :biggrin:

How can you say that? Seems to me DX multi thread rendering works fine in BF4 even though AMD doesn't support it. So, it's obvious that AMD doesn't axe feature support because it won't help them and helps nVidia customers. And while Dirt SD and TressFX doesn't run as fast on nVidia hardware, it does work. nVidia is just weaker in those compute functions.
 

desprado

Golden Member
Jul 16, 2013
1,645
0
0
Is related to the possibilities that you can build your hardware. If tech is showing 5-7% gains even the worst case possibilities, it can't not be good.

This Mantle performance advantage(in FPS or frametime numbers) is what we can expect in from last DX/OGL initiatives from Nvidia/AMD/Intel/MSFT.

Consoles have closer to metal ways of coding, but due to hardware limitation developers need many times to do(especially in multiplataform games) tricks to attenuate the loss of quality of their games(like upscaling and improvise softer anti-aliasing forms) when the original(source) game has too many detail.



Mantle is for CPU bound situations and for CPU functions on PC gaming. 1440p turns the things so much GPU-bound. Would be good to see what Two top-end cards(Like two 290x Lightning on the same CPU) can do in this same condition.

I think u are new in PC gaming.Do u now that more than 90% of the games are GPU bound all the best and high demanding games are gpu bound so which games are u talking about.
 

bystander36

Diamond Member
Apr 1, 2013
5,154
132
106
I think u are new in PC gaming.Do u now that more than 90% of the games are GPU bound all the best and high demanding games are gpu bound so which games are u talking about.

Being GPU or CPU bound depends a lot of the genre, your target FPS, whether you are online or not, and your system.

People who shoot for 80+ FPS find themselves CPU bound often. People playing RTS's and multiplayer games also find themselves CPU bound often.

I like high FPS, so the idea of Mantle, or an improved DX does have an appeal to me, but I would prefer it be on a more universal API than Mantle is currently.
 

VulgarDisplay

Diamond Member
Apr 3, 2009
6,193
2
76
I think u are new in PC gaming.Do u now that more than 90% of the games are GPU bound all the best and high demanding games are gpu bound so which games are u talking about.

The important thing to remember is that low level API's are going to become increasingly important as GPU's outstrip CPU's in performance. With 20nm GPU's Mantle/DX Low Level/Open GL is going to really push past older style DX.
 

bystander36

Diamond Member
Apr 1, 2013
5,154
132
106
The important thing to remember is that low level API's are going to become increasingly important as GPU's outstrip CPU's in performance. With 20nm GPU's Mantle/DX Low Level/Open GL is going to really push past older style DX.
If solid CPU multithreading that can expand to unlimited cores, they can make CPU's keep up with the GPU quite easily. If CPU multithreading for draw calls becomes the norm, CPU companies can join the GPU race by adding cores. Currently, adding more CPU cores has limited uses, so they've focused on speeding up the cores they have, that would change with multithreaded games.
 

SiliconWars

Platinum Member
Dec 29, 2012
2,346
0
0
I think u are new in PC gaming.Do u now that more than 90% of the games are GPU bound all the best and high demanding games are gpu bound so which games are u talking about.

So there's no point in Microsoft bothering with DX12 multi-threading?
 

PPB

Golden Member
Jul 5, 2013
1,118
168
106
I think u are new in PC gaming.Do u now that more than 90% of the games are GPU bound all the best and high demanding games are gpu bound so which games are u talking about.

So potentially 90% of the games could be held back in terms of adding new elements into their games because they could be draw call limited.


You only prove that the 1# "optimization" in today's game really involves in cutting stuff out of the game to reach certain CPU performance level. I sure know all the stunts the guys at DICE had to pull in order to stay in the 3-7k draw call range. We sure all have seen how many times levolution effects hasdbeen nerfed in order to reach better CPU performance.
 

Carfax83

Diamond Member
Nov 1, 2010
6,841
1,536
136
If solid CPU multithreading that can expand to unlimited cores, they can make CPU's keep up with the GPU quite easily. If CPU multithreading for draw calls becomes the norm, CPU companies can join the GPU race by adding cores. Currently, adding more CPU cores has limited uses, so they've focused on speeding up the cores they have, that would change with multithreaded games.

They're probably going to need a combination of both methods to make sure that future GPUs are well fed..

More and faster cores, plus much lower overhead. The latter in particular is probably going to give them the biggest benefit near term, as it's much harder for developers to exploit increased CPU parallelism.

One thing I've been wondering, is how the Mantle API allows the CPU to issue draw calls?

I know it's multithreaded, but are ALL threads issuing draw calls at the same time? That can be problematic, because 3D rendering is still an essentially serial process as many things need to be done in precise order..

DX11 multithreading got around this by using command lists, which used multiple threads to upload data at the same time, but all of that data still went through a single thread for final dispatch to the GPU as far as I know..

If it does turn out that Mantle is using multiple threads to issue draw call commands simultaneously without an immediate device thread like DX11, could that be responsible for the stuttering problem?

Having multiple threads issuing draw call commands at the same time with no check would create a whole lot of sync issues I'm sure..
 

jj109

Senior member
Dec 17, 2013
391
59
91
The first thing MS needs to focus on is reducing DX11 overhead. The mechanisms for multithreaded rendering are there but it's not as good as it can be.

Spreading out the API calls by using command lists is an incomplete solution: while it's better than manually threading or rendering in solely immediate context, the CPU resource demand from DX11 can eat up way too many cycles. It can make certain games worse if the game has very heavy simulation/AI threads and the DX11 render threads collide with those.
 

sontin

Diamond Member
Sep 12, 2011
3,273
149
106
The first thing MS needs to focus on is reducing DX11 overhead. The mechanisms for multithreaded rendering are there but it's not as good as it can be.

The Star Swarm alpha versions scale at least up to 4 cores on a nVidia plattform with deferred context (~90% utilization).
 
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/    |