The AMD Mantle Thread

[Gloomy]

I have a 290 and BF4. I have no problem blasting Mantle if it sucks, don't worry about that.

Hopefully I'll have my PSU sorted out by the time Mantle is released. Thank God it wasn't the motherboard, if I had to deal with an RMA to Asus I'd have opted for suicide instead.

 

[richiegore]

Has there been any mention of triple buffering and mantle? More consistent frames per second due to fewer CPU hiccups with triple buffering could mean a really great gaming experience... If I was AMD I'd make triple buffering mandatory on all mantle games...

 

[chimaxi83]

I'm looking forward to trying BF4 on my 290CF, but to be honest, not really getting my hopes up anymore.

 

[jj109]

I have a 290 and BF4. I have no problem blasting Mantle if it sucks, don't worry about that.

I don't think Mantle is going to suck for AMD users. My benches against HardCoreWare's A10-7850K + 290X BF4 results tells me that it's going to fix AMD's DX11 problem. But that isn't justification for making a whole new graphics API.

http://www.hardcoreware.net/kaveri-review-a10-7850k/4/

A10-7850K averages 72.7 FPS on the run at high preset 1080p.

Here's my i5 at 2.8 GHz which should match the A10-7850K in BF4.

So there is a huge problem that the majority of reviewers have consistently missed. For CPU tests they only use one brand of GPU which isn't sufficient. More obscure sites like hardwarepal or pclab.pl do NV and AMD CPU bound tests, and you can see them verify my findings.

 

[monstercameron]

I don't think Mantle is going to suck for AMD users. My benches against HardCoreWare's A10-7850K + 290X BF4 results tells me that it's going to fix AMD's DX11 problem. But that isn't justification for making a whole new graphics API.

http://www.hardcoreware.net/kaveri-review-a10-7850k/4/

A10-7850K averages 72.7 FPS on the run at high preset 1080p.

Here's my i5 at 2.8 GHz which should match the A10-7850K in BF4.

So there is a huge problem that the majority of reviewers have consistently missed. For CPU tests they only use one brand of GPU which isn't sufficient. More obscure sites like hardwarepal or pclab.pl do NV and AMD CPU bound tests, and you can see them verify my findings.

maybe it is because youtube has stuck to 30fps for so long but the video seems super sped up...

 

[Gloomy]

Nvidia has always had better performance than AMD/ATi in CPU-limited scenarios. I can't remember a time when this was not a running theme, actually.

If you're implying that this is directly related (or due) to Direct3D's overhead, then that is an interesting thought and (in my opinion) it wouldn't be that out of the ordinary for Nvidia cards to have a higher draw call ceiling than AMD's.

 

[Gloomy]

maybe it is because youtube has stuck to 30fps for so long but the video seems super sped up...

I think you need to play more games at >60 FPS

 

[Grooveriding]

I have a 290 and BF4. I have no problem blasting Mantle if it sucks, don't worry about that.

Hopefully I'll have my PSU sorted out by the time Mantle is released. Thank God it wasn't the motherboard, if I had to deal with an RMA to Asus I'd have opted for suicide instead.

I just completed an advanced RMA with Asus for my motherboard, call it a miracle, but it went off without a hitch and I had the replacement in a little over a week.

All that can go wrong at this point is them claiming there is some sort of physical damage to my board that went bad I sent in. lol, still possible with Asus, but I made sure to take pictures of everything.

 

[jj109]

maybe it is because youtube has stuck to 30fps for so long but the video seems super sped up...

Did you see that there's BF4's built-in FPS meter in the top right corner?

 

[ASM-coder]

At least someone here has a 290/X and BF4 right?

What is it that you think those benchmarks will be able to tells us without the ability to compare with a Mantle version?

 

[jj109]

What is it that you think those benchmarks will be able to tells us without the ability to compare with a Mantle version?

If you missed it, AMD posted a BF4 Mantle video during CES. They also claimed a 45% improvement over DX11 using the same hardware.

 

[Final8ty]

Doom 3 AstoundSound/TrueAudio
http://www.youtube.com/watch?v=mb57qpc0RD0

Freaky that voice
http://youtu.be/mb57qpc0RD0?t=4m18s

Alien Planet AstoundSound 3D demo
http://www.youtube.com/watch?v=FVRUHFB0ATE


Last time I checked TrueAudio isn't the same as Mantle. Let's start posting on topic please.

-Rvenger

 

[ASM-coder]

If you missed it, AMD posted a BF4 Mantle video during CES. They also claimed a 45% improvement over DX11 using the same hardware.

Yes, yes, yes, we know all that. But what will BF4 on a random R9 290X machine prove until we have Mantle to run on the exact same machine? Don't we need a before and after comparison to tell us anything?

 

[jj109]

Yes, yes, yes, we know all that.

Perhaps you should say what you do know first.

But what will BF4 on a random R9 290X machine prove until we have Mantle to run on the exact same machine? Don't we need a before and after comparison to tell us anything?

So you don't know that we have a Mantle demo on BF4.

 

[ASM-coder]

Perhaps you should say what you do know first.



So you don't know that we have a Mantle demo on BF4.

Yes, this I know. But we don't know the exact setup of that machine to replicate a "without Mantle" test.

 

[DamnedLife]

Almost 5000 posts of baseless speculation and you're still willing to sit on your hands? If you aren't willing to put evidence on the table when all the tools are available, don't even bother coming at me.

At least someone here has a 290/X and BF4 right?

I am not into this whole speculation/guesstimate thing tbh. I have an r9 290x and a BF4 on windows 8.1 so DirectX but still I can't sample for Mantle even if I downclock, overclock, undervolt my cpu and/or my gpu.... etc. because I do not have Mantle period. Any benches I would do is just PURE speculation and guess and estimate.... So nope.. No need to spread FUD with these less than credible benches.

Developers words are just that, words. But still they are more credible than people here who got a thing going I'm just not that into. Words of the credible people can be taken seriously, like the phrase "take my word for it"... it's like they are swearing it will work but can't prove it just yet due to some NDA or I dunno something like a contract for a time period for release. I would take their words for it but not yours. And it's just the way it is and will be. Because we know their real names, their faces, their products we bought and played, people of internet with all their guessing are just nameless, faceless ppl who we can't trust. I take everything these internet ppl throw at me with grain of salt, but on the other hand the developers words are more thrustworthy, not completely but more than internet ppl, if the developers break their promises, can't reach their targets, then they lose their credibility, then we know exactly who we shouldn't trust next, internet ppl on the other hand will be forgotten as soon as their make their baseless accusations, guess works, spreading FUD... etc. so no I am not jumping on that wagon.

 

[DamnedLife]

Yes, this I know. But we don't know the exact setup of that machine to replicate a "without Mantle" test.

Also we lack mantle as of yet to replicate any "with mantle" bench.... some ppl do not get this.

 

[Enigmoid]

If you missed it, AMD posted a BF4 Mantle video during CES. They also claimed a 45% improvement over DX11 using the same hardware.

And they also said a 30% IPC increase for Steamroller. We all know how that turned out. How much faith do you place in this 45% claim?

 

[Spjut]

Perhaps this has been mentioned before, but regarding the claimed improvement being"up to 45% faster" on the A10 and R9 290x, couldn't some tasks be offloaded to the APU's IGP? One of AMD's slides did mention better potential for Crossfire and APU + discrete GPU systems

 

[ASM-coder]

And they also said a 30% IPC increase for Steamroller. We all know how that turned out. How much faith do you place in this 45% claim?

Didn't they say "up to 20%" on the CPU and "up to 30%" on the GPU?

 

[BrightCandle]

Actually it ended up like 0 on the CPU and quite a bit less than 30% on the GPU. The problem with marketing numbers is they are cherry picked. That up to 45% is both weasel word delivered ("up to") and doesn't come with enough information to reproduce or refute it. Its bad science from a company that does have a habit of being a bit overly optimistic with its performance improvement numbers.

 

[ShintaiDK]

Didn't they say "up to 20%" on the CPU and "up to 30%" on the GPU?

Nope. 30% for the CPU.

 

[McLovin42]

If you missed it, AMD posted a BF4 Mantle video during CES. They also claimed a 45% improvement over DX11 using the same hardware.

Yeah it was also proven that those improvement #'s were bunk too..........Mantle ='s broken vaporware for a broken Alpha game.

Warning issued for trolling.
-- stahlhart

 

[Hitman928]

I wasn't aware AMD ever claimed any specific IPC performance numbers for Kaveri. Does someone have a link?

 

[Enigmoid]

Nope. 30% for the CPU.

Yep

(This is not directly related to mantle but I am using it to show that 45% is overly optimistic).

Performance increase was 0%. Not 10-15%.

In their defense the improvement is simulated but it clearly shows that this is not an 'up to' statement. They also said these were the results of internal predictions/simulations on gaming, digital media, and productivity, none of which showed any improvement outside OpenCL.

This is why the 45% claim is very likely over estimated.

I mean, just looking at common logic would show that BF3 and BF4 are extremely CPU light in singleplayer, using only a couple cores and getting 200+ fps per second on an i5 (hitting the engine cap). Frostbite does not have a draw call problem. Multiplayer CPU dependancy is likely as a result of other things such as networking, AI, and other stuff. There are a lot more players in MP but the disparity is larger than that.


IMO mantle is a mistep. AMD should have spent the time developing a physics engine for use on their igps, making it easier on the CPU. RTS should use it for pathfinding and AI which would probably would increase performance just as much if not more. The oxide demo is not going to run nearly so smoothly once they add in proper AI and animations. There is no pitch and yaw on the ships. There may be 1000's of units but they really aren't doing anything.

Then you watch a video like this and you really are not that impressed anymore.

http://www.youtube.com/watch?v=NRTmVr4pU8Y