The AMD Mantle Thread

[SiliconWars]

http://forums.galciv2.com/?aid=451041

(note that the video demo of “Star Swarm” was from November, it looks even better now, we were able to throw that together in basically a month or so with Nitrous).

 

[Lonbjerg]

It is just a video for you, me and him.

Unless you have a download link to the techdemo?

 

[Noctifer616]

It is just a video for you, me and him.

Unless you have a download link to the techdemo?

And then what? Are you going to complain that you can't run it in Mantle cause you have no GCN card?

 

[SiliconWars]

That's some pretty rampant denial going on there Lonbjerg. You can't actually believe this is somehow faked? If you watched it you'd know that he says at the start that it would be impossible to fake it even with a full film crew, there's just too much going on screen.

http://www.youtube.com/watch?feature=player_detailpage&v=QIWyf8Hyjbg#t=1451

 

[Arkadrel]

It is just a video for you, me and him.

Unless you have a download link to the techdemo?

When you see a movie trailer, do you say.... thats just a movie trailer!
that ll NEEEEVER... be made into a REAL movie!

Thats akin to what your saying.

Why is it so hard for you to just take someones word for it? no trust in your fellow man?

If a game dev shows you a working demo of their game engine, and say its from a game their developing, why is it you automatically assume its fake?

You love to spread FUD, thats the most likely explaination I can come up with.
(and you really love nvidia, and reeeaally hate amd, we get it already, but enough with all the fud)

 

[desprado]

That's some pretty rampant denial going on there Lonbjerg. You can't actually believe this is somehow faked? If you watched it you'd know that he says at the start that it would be impossible to fake it even with a full film crew, there's just too much going on screen.

http://www.youtube.com/watch?feature=player_detailpage&v=QIWyf8Hyjbg#t=1451

Of course it can be.How because they did not showed any specs they just tell specs like marketing and there was no compression with Dx but just saying this or that.

 

[Noctifer616]

Of course it can be.How because they did not showed any specs they just tell specs like marketing and there was no compression with Dx but just saying this or that.

They said specs like it was marketing? What does that even mean. They tell you exactly what CPU and GPU it runs on. Considering the demo is GPU bottlenecked they CPU clock is not relevant, and since it's a demonstration from a month ago, you can assume it's a reference Hawaii board, so it's not going to clock that well.

They tell you that with DX they run into problems past 15k batches. While in Mantle you can see the demo reach 100k batches at decent FPS. The only problem with that number is that we don't know if it's a GPU or CPU bottleneck. But considering how limited DX core scaling is, even if it was a CPU bottleneck, they still couldn't reach the numbers Mantle is having.

For those that actually want to try the demo out, they said they will release it in Q1.

 

[krumme]

Regarding BF4 there must be more than enough work existing, otherwise Andersson wouldn't say perfect parallel rendering - Mantle utilizes all 8 cores.




Which is unrealistic with only 20%.




Considering that I see up to +20% in an unoptimized DX11MT over DX11 in Pcars I expect something better from a low level API with much more advantages. Unlike DX11MT Mantle is low level>no driver overhead, reduced latency, much easier CPU scaling, drawcall limits much higher and reduced drawcall overhead. All the talk about the bad Directx performance wouldn't make sense if Mantle can do just 20% better.

What you said was 50% more on a 6c cpu in bf4 or it would be waste. And i asume it was not the phenom x6 you were referring to. It was just a pethetic attemp to create a strawman before mantle patch hits.

The only way a i7 6c will be cpu limited in bf4 is if the drawdistance can be improved. A nice trick. But perhaps an idea for amd to get. Nv would have no problems using that kind of trick even if its not visible. Then all without a gcn card will be rediculed. I dont like fps to be turned into rts and all with an nv card to suffer. I like stellar performance for all.

So could your predictions be a proposal for amd to drastically higher draw distance for bf4 and for all future games?

 

[Silverforce11]

What you said was 50% more on a 6c cpu in bf4 or it would be waste. And i asume it was not the phenom x6 you were referring to. It was just a pethetic attemp to create a strawman before mantle patch hits.

The only way a i7 6c will be cpu limited in bf4 is if the drawdistance can be improved. A nice trick. But perhaps an idea for amd to get. Nv would have no problems using that kind of trick even if its not visible. Then all without a gcn card will be rediculed.

So could your predictions be a proposal for amd to do that and for all future games?

You mean we could perhaps finally enjoy Arma 3 at blazing fast performance?! Sign me up!

 

[jj109]

They tell you that with DX they run into problems past 15k batches. While in Mantle you can see the demo reach 100k batches at decent FPS. The only problem with that number is that we don't know if it's a GPU or CPU bottleneck. But considering how limited DX core scaling is, even if it was a CPU bottleneck, they still couldn't reach the numbers Mantle is having.

For those that actually want to try the demo out, they said they will release it in Q1.

Well, is that 15k number with AMD DX drivers or Nvidia DX drivers? :biggrin:

A correction: They said they want to reach 100k with optimization. The highest I saw in the busiest scenes was 60k or so.

As a side:
I'm impressed by their simulation and AI model, if it is doing what they claim it is doing. The problem with large scale strategy games like SupCom was the awful scaling as unit counts and AI complexity grew. I remember SupCom games slowing down to a crawl on the simulation side once unit counts hit 1500 or even less if AI is involved. When several hundred fighters started swirling around dog fighting, it got even slower. If Nitrous can deliver an engine that easily handles 3000 units with detailed simulation, that would be a boon to the strategy game genre.

You mean we could perhaps finally enjoy Arma 3 at blazing fast performance?! Sign me up!

Don't count on it.

 

[Lonbjerg]

I am in denial because I remember this?
http://www.youtube.com/watch?v=h1E6vLsX2UI
AMD GPU physics demo...

Never came to life...

Like so many other broken promises.

And my ARMA 3 runs just fine...but since renders a MUCH lager world than BF4...you might want to upgrade your hardware...like I said...I run s socket 1366 PC....max settings...I even have the performce overhead to record video's...at max settings
EDIT:
Again i7 990X + Titan running ARMa3 at max settings (view distance 7400 meters)...show me where I am draw call limited?
http://www.youtube.com/watch?v=SkHQlId8GfI&feature=c4-overview&list=UUb2qvg7CbWkPLE4wc9ULT4g

Why all the false "arguments"...in defence for PR?

 

[krumme]

I don't know about you, but features don't grow on trees where I work.

The 'hostility' you perceive comes from the fear that developers trying to code for both DX11 and Mantle and doing a poor job on one/both of them. There's a real chance that Mantle becomes an anti-feature for non-GCN users.

G-sync and Shadowplay, on the other hand, are at worst harmless gimmicks.

Its a valid and only consern about mantle imo.

But we cant continue with dx. The change will happen at some time. Mantle or not mantle. But the change comes during a period of several years so its not happening over night. Plenty of time for all to adapt.

And its also very simple. If mantle does not deliver for the devs and consumers it will just fade away. All the worrying looks more like protections of a brand and a company than striving for better games.

 

[Noctifer616]

I am in denial because I remember this?
http://www.youtube.com/watch?v=h1E6vLsX2UI
AMD GPU physics demo...

Never came to life...

What are you talking about? Never came to live? There are games actually using that physics engine.

 

[Lonbjerg]

What are you talking about? Never came to live? There are games actually using that physics engine.

Runnning on AMD's GPU's ? ^^
Or running on a x86 CPU?

 

[krumme]

You mean we could perhaps finally enjoy Arma 3 at blazing fast performance?! Sign me up!

Hopefully we will see a new generation of games. Looking what ceo of oxide says in prior link post 3673 devs are going to create new rts and roleplaying games using the new engine. I hope that happens.

 

[Noctifer616]

Runnning on AMD's GPU's ? ^^
Or running on a x86 CPU?

OpenCL, it can run on both.

 

[desprado]

I am in denial because I remember this?
http://www.youtube.com/watch?v=h1E6vLsX2UI
AMD GPU physics demo...

Never came to life...

Like so many other broken promises.

And my ARMA 3 runs just fine...but since renders a MUCH lager world than BF4...you might want to upgrade your hardware...like I said...I run s socket 1366 PC....max settings...I even have the performce overhead to record video's...at max settings
EDIT:
Again i7 990X + Titan running ARMa3 at max settings (view distance 7400 meters)...show me where I am draw call limited?
http://www.youtube.com/watch?v=SkHQlId8GfI&feature=c4-overview&list=UUb2qvg7CbWkPLE4wc9ULT4g

Why all the false "arguments"...in defence for PR?

That what i am saying AMD has a habit to hype alot but when it come to reality it being seen as a lie.If Mantle come of course it will increase the performance but not as much people hoped.As i said Mantle will totally be depended on the system AMD CPU+AMD GPU more performance increase around max 15% and if only AMD GPU and not more than 10%

As a fact if AMD improve there driver they do not need mantle to improve there GPU performance.

 

[Noctifer616]

As a fact if AMD improve there driver they do not need mantle to improve there GPU performance.

Then why don't they? Why would they spend so much time and effort to create an API if they could get the same benefit with better drivers?

 

[Lonbjerg]

Hopefully we will see a new generation of games. Looking what ceo of oxide says in prior link post 3673 devs are going to create new rts and roleplaying games using the new engine. I hope that happens.

So now "hope" is being used as argument?
Look at my video of ARMA3 and tell me where I am limited...I mean 8xMSAA just for starters(ALL settings at MAX, 7400 meter viewrange)....trying to break my rig...didn't work....DX didn't fail, sorry.

If you expect ARMA3 to run maxed out on a crappy APU...the problem isn't the Reality Engine...it's you.

 

[Lonbjerg]

OpenCL, it can run on both.

So you can name a game running physics on AMD's GPU's?
Still waiting for that one..

 

[ShintaiDK]

OpenCL, it can run on both.

There are no used physics engine running on AMD GPUs with OpenCL. ATI/AMDs several physics attempt all failed.

 

[jj109]

Its a valid and only consern about mantle imo.

But we cant continue with dx. The change will happen at some time. Mantle or not mantle. But the change comes during a period of several years so its not happening over night. Plenty of time for all to adapt.

And its also very simple. If mantle does not deliver for the devs and consumers it will just fade away. All the worrying looks more like protections of a brand and a company than striving for better games.

You say as though DX is a API set in stone, but the truth is that it isn't.

DX9 from DX10 was a HUGE change for the better technically. It was faster, had fancier graphics features, and supported a much better driver model.

I agree that any change will come over a period years. Remember DX10? Transitioning to DX10/11 was like pulling teeth.

For the first 3 years, we got a slew of games running on hacked together DX10/11 engines that ran slower on a API that was should have been faster. DX10/11 modes didn't even look any better! Out of all the big pre-2010 games, the only one I remember doing better in DX10 with the same IQ was Far Cry 2.

So, just because an API is technically faster doesn't mean the end result will be immediately better. If past history is any indication, the new API gets half-***ed for an eternity in tech years.

Then why don't they? Why would they spend so much time and effort to create an API if they could get the same benefit with better drivers?

From what I read, when they tried to implement what Nvidia did in the 270 release, they saw a performance regression.

 

[ShintaiDK]

Then why don't they? Why would they spend so much time and effort to create an API if they could get the same benefit with better drivers?

Why have now 4 years passed without AMD supporting DX Command list, D3D9 Overlay, DXVA-HD and so on? Or the complete lack of DX11.2 support on Hawaii chips?

Mantle at best, simply looks like a way to get out of the DX war. Deperate times calls for desperate actions. dGPU only accounts for 1/3rd of the marketshare. And the CPU division is in utter ruins.

And how many games are gonna support Mantle again? Because its certainly not all Frostbite games. And right now BF4 mantle is MIA. No mantle for NFS:Rivals, C&C got cancelled. So now the next bet is for Thief to use Mantle. But will it be at release? Or will it also be some patch..that may or may not come at a later date?

 

[jj109]

Or the complete lack of DX11.2 support on Hawaii chips?

Uh, please elaborate.

Edit: Nevermind, I found the thread. The hardware is willing, but the driver team is ... I'll just stop there.

 

[ShintaiDK]

Uh, please elaborate.

http://forums.anandtech.com/showpost.php?p=35865089&postcount=6