The AMD Mantle Thread

Page 153 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Status
Not open for further replies.

monstercameron

Diamond Member
Feb 12, 2013
3,818
1
0
we need to separate mantle the api from mantle the promise.
the api
the api is more than a performance boost
the api will offer more flexibility at the cost of complexity

the promise
more performance
less bugs
less cpu load

let me know what I am missing.
 

bystander36

Diamond Member
Apr 1, 2013
5,154
132
106
If what we're hearing is true, then Mantle will be able to be used on both intel and nvidia gpu's . At their discretion, of course .
A dev mentioned it "could" be used for any brand, but AMD never said this. AMD said it works only for GCN.

The reality is that the dev could be right, in the sense that they could Jimmy-rig it to work for any brand, but that doesn't mean it would work well for any architecture.

I would be happy with a new API, as long as it would work well for all brands and architectures....of course that means it would be a high level API, which is not the concept of Mantle.
 

desprado

Golden Member
Jul 16, 2013
1,645
0
0
we need to separate mantle the api from mantle the promise.
the api
the api is more than a performance boost
the api will offer more flexibility at the cost of complexity

the promise
more performance
less bugs
less cpu load

let me know what I am missing.
U are missing API which was meant and promised to released in Dec so now where is it?
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
Where are you even getting 100K? The oxide demo only went up to ~60K and the frame time was up to 40ms.

A screen grab was posted in this thread showing 105K @ 15.4ms. Threads too long to find it, maybe the original poster can repost it.
 

jj109

Senior member
Dec 17, 2013
391
59
91
A screen grab was posted in this thread showing 105K @ 15.4ms. Threads too long to find it, maybe the original poster can repost it.

I see. I do wonder why the Oxide guys say they hope to hit 100K when they already hit 100K then.

Back on to the topic of deferred rendering, here it is with 8 threads on a DX11 demo:

Deferred 65536 draw calls with simulated workload: 22 fps
Immediate 65536 draw calls with simulated workload: 6.5 fps

Deferred 40K: 35 fps
Immediate 40K: 10.5 fps

Deferred 20K: 70 fps
Immediate 20K: 22 fps



At 65K cubes I can't even tell what it was originally trying to render.
 

Carfax83

Diamond Member
Nov 1, 2010
6,841
1,536
136
I suppose you can show us 100K draw calls @ 65fps using DX multi thread?

What game could possibly use that many draw calls? You'd overtax the GPU and have so much clutter on the screen you wouldn't even be able to see anything.

It's great that Mantle offers such high performance and reduced overhead, but quoting big numbers is a bit of a spurious argument in my opinion, because the majority of games are not really draw call limited.

DX11 multithreading can probably do a maximum of 20K draw calls on a well designed multithreaded engine like Firaxis's Lore Engine. I have no idea how many draw calls 3D engines like Frostbite 3 and CryEngine 3 as they use immediate context via manual threading.

And PCs also have something called instancing, which allows you to batch draw calls for objects with similar properties. This can be very effective at reducing the amount of draw calls, as the CPU doesn't care about the size, only the amount.

At any rate, IF we ever have games that issue that many draw calls, by the time they roll around, we'll have DX14 or something which will have a reduced operating overhead similar to Mantle, but not as thin as to retain a reasonable degree of abstraction and backward compatibility.
 

mindbomb

Senior member
May 30, 2013
363
0
0
my current theory on mantle is that it will technically be supported on a variety of architectures, however, the games themselves will generally require a specific architecture. So mantle api support won't necessarily mean you can play mantle games.
 

Lonbjerg

Diamond Member
Dec 6, 2009
4,419
0
0
And what exactly would be bad for consumers with Mantle making AMD's CPUs perform better?
It's not as if it won't work on Intel CPUs too, and while AMD needs it more than Intel, don't forget that there still are users with Core 2 Quads. The Nehalem CPUs could get even greater longevity for high-end games.

Being able to target weaker systems matters alot for game studios, and Mantle could also make a great difference for AMD's APUs



And I don't see why everyone would expect the consoles to be "maxed" at release, that has never happened. Optimization takes time, and even a knowledgeable guy like Timothy Lottes said he expected the launch titles would be little more than DX11 ports.
Heck, just compare the 360's release games to today's games. Or the original Xbox' release games to the last ones.

We already have 2 API's (DirectX and OpenGL)...we don't need more...I remember the API's wars back in the days...I don't want to back to that mess.
 

jj109

Senior member
Dec 17, 2013
391
59
91
What game could possibly use that many draw calls? You'd overtax the GPU and have so much clutter on the screen you wouldn't even be able to see anything.

With 65k draw calls, you can paint a 1000px by 1000px square with 4px wide rotating, colorful cubes :biggrin:
 

bystander36

Diamond Member
Apr 1, 2013
5,154
132
106
my current theory on mantle is that it will technically be supported on a variety of architectures, however, the games themselves will generally require a specific architecture. So mantle api support won't necessarily mean you can play mantle games.
At first, it definitely is only going to work on all GCN AMD cards. I suppose you may be thinking about future games when new AMD hardware comes that may have revised versions of GCN, or possibly new architectures.

I imagine it is possible in the future, that AMD might change architectures, which may fragment Mantle. I don't know if that is likely or not. I imagine Mantle has to be a hit first.
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
Can you go back for yourself and find the post where I linked to OpengL and 100.000 draw calls...above the 65.000 in the Mantle Tech demos...or do I need to make a repost?

Are you talking the post where it was draw calls per second and it wasn't in a game?
 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
Oh, you have a link to a mantle game I can try?

The game is still in development, as you are aware. Nobody ever claimed there was a game you could play available.

I also didn't ask for there to be a game that I could play using DX multi thread rendering that matched Mantle's performance. Desprado specifically claimed that with better drivers AMD could match Mantle's performance. I don't care what he uses to back that claim.
 

Lonbjerg

Diamond Member
Dec 6, 2009
4,419
0
0
The game is still in development, as you are aware. Nobody ever claimed there was a game you could play available.

I also didn't ask for there to be a game that I could play using DX multi thread rendering that matched Mantle's performance. Desprado specifically claimed that with better drivers AMD could match Mantle's performance. I don't care what he uses to back that claim.

So that would be a "no"...gotcha.
 

Adampa1006

Member
May 29, 2013
38
0
0
I guarantee we can all agree on something. The performance we get from direct X is awful. We know how much "horsepower" our graphics cards have (ESPECIALLY when compared to consiles) , but how they are squandered away and we end up with games that don't look that great, and don't run that great.

AMD is finally trying to solve this problem.
 

Lonbjerg

Diamond Member
Dec 6, 2009
4,419
0
0
I guarantee we can all agree on something. The performance we get from direct X is awful. We know how much "horsepower" our graphics cards have (ESPECIALLY when compared to consiles) , but how they are squandered away and we end up with games that don't look that great, and don't run that great.

AMD is finally trying to solve this problem.


I disagree....my performance under DirectX is just fine.
I max out games like ARMA3 and Crysis 3...could you give me some examples of the "awfull" performance?
 
Status
Not open for further replies.
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/    |