March 18th. - Thief 2014 getting Mantle and TrueAudio update

O

octoberasian

2[H]4U
Joined
Oct 13, 2007
Messages
4,082
http://www.pcgamer.com/2014/03/17/thiefs-mantle-update-incoming-on-march-18th/

Now though, AMD have teamed up with Eidos and are set to release a new update to the latest Thief game, wrestling it away from the Microsoft clutches of DirectX and giving it some Mantle lovin'.

...

Alongside the March 18th release of the Mantle update for Thief, AMD and Eidos are also releasing the first TrueAudio update for any game in the market. TrueAudio utilises the compute power in the graphics card to offload intensive audio effects from the CPU, which should give richer sounds to games.
Click to expand...
 
Early performance tests: http://gamegpu.ru/action-/-fps-/-tps/thief-mantle-vs-directx-11-test-gpu.html

http--www.gamegpu.ru-images-stories-Test_GPU-MMO-ArcheAge-mantle-tdh_1920.jpg
http--www.gamegpu.ru-images-stories-Test_GPU-MMO-ArcheAge-mantle-tmh_1920.jpg


CPU scaling is pretty damn good I must say. More impressive is the fact that Mantle support was bolted onto the game engine (THIEF uses Unreal engine 3). Games developed from the ground up using Mantle should see further performance gains.
http--www.gamegpu.ru-images-stories-Test_GPU-Action-Thief_-test-proz.jpg
http--www.gamegpu.ru-images-stories-Test_GPU-MMO-ArcheAge-mantle-td11_proz.jpg
http--www.gamegpu.ru-images-stories-Test_GPU-MMO-ArcheAge-mantle-tm_proz.jpg
 
Last edited:
So, on graphics performance alone, an extra 4 to 5 FPS on average.

But, on CPU performance, Mantle is adding anywhere from 5 to 10 FPS.
 
You have to look at all the graphs before making any generalizations. I purposely posted the GPU limited scenario with high Super Sampling AA (which is 4X) to show that Mantle still gives you a 10% performance increase.

Lower quality tiers give even greater gains:
http--www.gamegpu.ru-images-stories-Test_GPU-MMO-ArcheAge-mantle-td_1920.jpg
http--www.gamegpu.ru-images-stories-Test_GPU-MMO-ArcheAge-mantle-tm_1920.jpg
 
Last edited:
I've been waiting for this patch to try out the game. Even if Thief is only mediocre, I really am hopeful that TrueAudio brings some life back to the PC audio scene. It's been a long, long time since the Aureal A3D days and I still haven't heard anything that blows me away like those cards did compared to the competition back then. We shall see I suppose.
 
Rizen said:
I've been waiting for this patch to try out the game. Even if Thief is only mediocre, I really am hopeful that TrueAudio brings some life back to the PC audio scene. It's been a long, long time since the Aureal A3D days and I still haven't heard anything that blows me away like those cards did compared to the competition back then. We shall see I suppose.
Click to expand...

I miss the A3D stuff. EAX wasn't horrible either.
 
its a DSP, drivers already have the APIs.

It isnt something you can "turn on"

Programs have to be specifically coded to use the APIs.
 
Lorien said:
You have to look at all the graphs before making any generalizations. I purposely posted the GPU limited scenario with high Super Sampling AA (which is 4X) to show that Mantle still gives you a 10% performance increase.

Lower quality tiers give even greater gains:
http--www.gamegpu.ru-images-stories-Test_GPU-MMO-ArcheAge-mantle-td_1920.jpg
http--www.gamegpu.ru-images-stories-Test_GPU-MMO-ArcheAge-mantle-tm_1920.jpg
Click to expand...

Now these are impressive! It's nice to see even that even on the high end (in your first post) to see the latest Intel CPUs getting a noticeable boost, not just the low end. I can't get over how well the FX series benefit from Mantle..It is like AMD just released a new CPU that is ~20% faster then their current offerings. It is a game changer. Period.
 
Lol, good to see a 3 generation old stock clocked 2600k still beats the 5ghz 9590. Even with mantle. Nope AMD cpu don't bottleneck.:rolleyes: someone needs to move those cpu benchmarks into Mr. 8350 doesn't bottle eck quad 290x thread. Cuz this clearly shows it bottlenecks even 1.
 
HOODedDutchman said:
Lol, good to see a 3 generation old stock clocked 2600k still beats the 5ghz 9590. Even with mantle. Nope AMD cpu don't bottleneck.:rolleyes: someone needs to move those cpu benchmarks into Mr. 8350 doesn't bottle eck quad 290x thread. Cuz this clearly shows it bottlenecks even 1.
Click to expand...

Once again you demonstrate that you don't understand what the word bottleneck means.
 
Seriously. If the only variable you are changing is the cpu and you get significantly lower framerate. Than that means the cpu is bottlenecking. Wtf r u talking about I don't understand. You think bottleneck means unplayable. It doesn't. It means holding back the potential of another component.

so funny just notice in DX11 even the 4330 beats the 9590 lol. You fan boys are in lala land.
 
HOODedDutchman said:
so funny just notice in DX11 even the 4330 beats the 9590 lol. You fan boys are in lala land.
Click to expand...

You are in LALA land. It doesn't matter a damn what CPU beats what CPU. This is about getting the best performance from your video card. All that the graphs show is how much performance improves with each CPU.

Mantle doesn't care what CPU you have In your system.
 
Clearly mantle does. Since intel is still at top of the chart. At around 35-40% lower clock speed than the 9590 too.
 
HOODedDutchman said:
Clearly mantle does. Since intel is still at top of the chart. At around 35-40% lower clock speed than the 9590 too.
Click to expand...

Stop bringing the stupid CPU wars into this. Mantle helps all CPU's. As Mantle gets better it will offer more performance to even high powered CPU's.
 
HOODedDutchman said:
Seriously. If the only variable you are changing is the cpu and you get significantly lower framerate. Than that means the cpu is bottlenecking. Wtf r u talking about I don't understand. You think bottleneck means unplayable. It doesn't. It means holding back the potential of another component.

so funny just notice in DX11 even the 4330 beats the 9590 lol. You fan boys are in lala land.
Click to expand...

If the CPU was the bottleneck then going from 1 290X to say quad 290X would give you exactly 0FPS incress. but clearly crossfire 290X is faster then a single 290X.

So saying the CPU is a bottleneck even with a single 290X is wrong.
 
If I were looking at it at face value, is there a fracking problem going from 61 FPS in DirectX to 91 FPS in Mantle on a single R9 290X whilst using an AMD FX -9570?

That right there tells me that the game is still very playable.

What is being held back here?
 
Spazturtle said:
If the CPU was the bottleneck then going from 1 290X to say quad 290X would give you exactly 0FPS incress. but clearly crossfire 290X is faster then a single 290X.

So saying the CPU is a bottleneck even with a single 290X is wrong.
Click to expand...

Your average will be higher because cpu bottleneck is not a constant but the situations where the cpu is the bottleneck will effect the minimums. So 290x crossfire would have approximately the same minimum fps. Depending on the resolution. Triple screens obv would be less like to cpu bottleneck. But still there.
 
octoberasian said:
If I were looking at it at face value, is there a fracking problem going from 61 FPS in DirectX to 91 FPS in Mantle on a single R9 290X whilst using an AMD FX -9570?

That right there tells me that the game is still very playable.

What is being held back here?
Click to expand...

Your talking a 3.4ghz i7 vs a 5ghz fx. Clock the i7 close to 5ghz and run a 120hz monitor then compare the two. Massive difference. Even with mantle. I'd love to see a clock for clock review with FX vs i7 and dx vs mantle.
 
HOODedDutchman said:
Your talking a 3.4ghz i7 vs a 5ghz fx. Clock the i7 close to 5ghz and run a 120hz monitor then compare the two. Massive difference. Even with mantle. I'd love to see a clock for clock review with FX vs i7 and dx vs mantle.
Click to expand...
Again, is there something wrong with a 30 FPS increase on one processor?

I'm not looking at the other Intel processors, just the AMD processors. With Mantle using a single R9 290X, there is an increase in framerate. Is that an issue?

And, I'm not talking about Crossfire either. Yes, when we start adding more than one card, then obviously we want a better processor feeding two or more video cards. That's logical.

But, when it comes to one card which the majority of gamers have, an increase in framerate is an increase in framerate no matter what processor is in use.
 
My point is not to down what mantle does. It is good for people lacking in the cpu department. I'm just saying that even with mantle the AMD cpus cannot keep up with intel. It does make it a much better story for those already on AMD cpu and running 60hz tho. Those running 120hz are still going to want the best available which is any 2nd gen or higher i5 or i7.
 
People with AMD FX processors are going to be loving this update. Damn impressive gains.
 
Yeah Aureal A3D was cool , sadly Creative killed them, from what I understand back then the dev kit was pretty good and A3D was easy to implement in games.
 
HOODedDutchman said:
My point is not to down what mantle does. It is good for people lacking in the cpu department. I'm just saying that even with mantle the AMD cpus cannot keep up with intel. It does make it a much better story for those already on AMD cpu and running 60hz tho. Those running 120hz are still going to want the best available which is any 2nd gen or higher i5 or i7.
Click to expand...

You still don't get it. IT's not about what CPU you have. Mantle isn't there to make AMD CPUs match Intel CPUs, it's about getting more performance from your graphics card.

And you shouldn't really be comparing AMD CPUs versus Intel CPUs for mantle. You should be comparing a specific CPU with DX and the same CPU with mantle to see how much performance gain mantle will bring to that CPU.

Mantle allows you to get better performance no matter what CPU you have.
 
Marcdaddy said:
Those graphs just show that my 2600K cpu might last forever for gaming lol.
Click to expand...

Yes and no :) it all depends if the amount of data that is now just limited by DirectX is going to get a bump in games using Mantle. That means that a lot more can be done when the cpu is doing nothing that is what the downclocks are showing in these tests.

This is new to developers so it might take a good while for this to change, were talking years but not if developers decide to push more data to the cpu.

Check APU14 videos on youtube from AMD with Nitrious engine developers they have some pictures of how much of the cpu is actually being used when doing Mantle in their engine. It does a whole lot of nothing :)
 
You must log in or register to reply here.
Back
Top