Mantle API Vs. DirectX Thief Benchmarks

[warmmilk]

Dont feed the nVidia troll. In other news, wccftech isnt a source to be trusted. Other than masquerading factually inaccurate content as news, they dont really do any journalism themselves. They repost stuff from some shady corners of the internet. Fox News and The Onion have more newsworthiness than these guys. For some reason, the OP keeps using it. The original source, in german isnt a well known source either. I'll reserve judgement until someone with some real street cred benchmarks this. 

 

how does him not caring about a certain game make him an nvidia troll?  

 

However, that statement does make you sound like an AMD fanboi...

[nicehat]

how does him not caring about a certain game make him an nvidia troll?  

 

However, that statement does make you sound like an AMD fanboi...

 

The bench shows that Mantle performs better than DX right?. Im saying Id rather reserve judgement until a more credible benchmark. How am I a fanboy again? 

[warmmilk]

The bench shows that Mantle performs better than DX right?. Im saying Id rather reserve judgement until a more credible benchmark. How am I a fanboy again? 

you quoted the guy responding to the that this test is on a game no one cares about, saying not to feed the nvidia troll...  the fact that you're calling someone out as an "nvidia" troll over a game title makes me think you're an AMD troll

[nicehat]

you quoted the guy responding to the that this test is on a game no one cares about, saying not to feed the nvidia troll...  the fact that you're calling someone out as an "nvidia" troll over a game title makes me think you're an AMD troll

 

If you feel like wasting your time, go ahead and search the guys other posts in any thread that's AMD related. Take a look at what other people have to say as well. Get back to me when you know whats going on. Cool?

[LAwLz]

I would gladly take the ~20% performance increase you get with a 290X at 1080p if I was interested in Thief. I would have liked to know what CPU they used though. "I am pretty sure it was a i7-4770K (downlocked, no HTT)" is not good enough for me.

I am more interested in what kind of results you get with for example an overclocked 8350 or overclocked i5-4570K.

[mr moose]

To be Honest, I'm just not that impressed, in the grander scheme of things you either pay out and get good performance or you spend your life making tweaks to get reasonable performance.   In a years time (which is blip on the tech radar) who is going to be weighing up a few frames when the difference in GPU is $50 and there are other factors that play a larger role in performance? Plus I don't know about anyone else but I can never tell the difference between 50fps and 60fps or above anyway, In fact I rarely even have an FPS monitor running so I wouldn't know half the time what it is anyway.

[warmmilk]

If you feel like wasting your time, go ahead and search the guys other posts in any thread that's AMD related. Take a look at what other people have to say as well. Get back to me when you know whats going on. Cool?

 

if thats the case, I take back my comments on the subject, I was only going off what I saw in this thread.

[Mandresel]

No way in hell the 780ti only runs at 87FPS at 720p.....

[xAcid9]

No way in hell the 780ti only runs at 87FPS at 720p.....

 

Looks like few card hit CPU bottleneck wall at 720p.

[Rageodontist]

Well this is disappointing to me. My hope was Mantle helping more with higher resolutions. Well DX12 here I come.

[StormEye]

Reading the article, its most likely intentionally handicapped 4770K simulating low end CPU, which suggests that it will have probably near zero difference if CPU reaches certain level.

 

Basically, its a tech for budget gamer.

[Sky Daddy]

Reading the article, its most likely intentionally handicapped 4770K simulating low end CPU, which suggests that it will have probably near zero difference if CPU reaches certain level.

 

Basically, its a tech for budget gamer.

BF4 runs much smoother for me with mantle, I would never go back to dx on BF4.  (on my high end pc)

 

I wouldn't say the tech is just for budget gamers, mid/high range are going to need optimizations to keep high framerates for VR, preferably with lots of AA. (probably will end up around 90hz)

That means star citizen is going to need mantle to make oculus rift for a bunch of users comfortable.

 

Optimization with occlusion culling is going to be extremely important too.

[Misanthrope]

To be Honest, I'm just not that impressed,

 

Really? It's certainly better performance gains than going up a price tier or a generation, for free. So it might not be ideal for all situations and it's only available on 2 games both (arguably) mediocre sure, but who turns his head at free stuff? Other than resentful and overzealous Nvidia "ethusiasts" so to speak

[mr moose]

Really? It's certainly better performance gains than going up a price tier or a generation, for free. So it might not be ideal for all situations and it's only available on 2 games both (arguably) mediocre sure, but who turns his head at free stuff? Other than resentful and overzealous Nvidia "ethusiasts" so to speak

 

In a years time (which is blip on the tech radar) who is going to be weighing up a few frames when the difference in GPU is $50 and there are other factors that play a larger role in performance? Plus I don't know about anyone else but I can never tell the difference between 50fps and 60fps or above anyway, In fact I rarely even have an FPS monitor running so I wouldn't know half the time what it is anyway.

[Imabigmac]

It interesting to see the improvements. I just want to know what cpu was used, it's not a fair representation if they used an underclocked i7. Yes they have proven it makes a difference, but the real test is a stock i7 with 290x running mantle vs. directx.

[Rorgash]

and this is another game that doesnt take much advantage of mantle AT ALL, until a huge RTS comes out using mantle i wont bother with benchmarks, since unless the game ACTUALLY uses 10k+ drawcalls Mantle wont make a big difference to mid to high end users (DirectX can do about 6-7k, Mantle, last i heard months ago it did 100k)

[iTile]

My Results

 

My PC Basic Specs:
AMD FX8350 @ Stock 4Ghz
16Gb DDR3
​MSI R9 290 4GB OC Gaming Edition
Xonar Essence STX (Tru-Audio is processed by the GPU and passed through to your sound card, onboard or extra which ever is your primery, I have not tested it yet, but it is advised to disable dolby surround or any virtual surround sound and use pure stereo with your Headset)

All graphics Settings maxed out(texture limit on default), 64bit executable
Results:

Direct X 11.1

This image has been resized. Click this bar to view the full image.

AMD Mantle:

This image has been resized. Click this bar to view the full image.

 
 

[Vitalius]

-snip-

Dat Minimum FPS though...

Seriously. That is so Quality of Life improving that it's nuts. I assume this has to do with Multi-core optimization specifically.

[EmmaLong]

http://www.tomshardware.com/reviews/thief-mantle-benchmarks,3773.html

[Guest]

Well this is disappointing to me. My hope was Mantle helping more with higher resolutions. Well DX12 here I come.

Isn't dx12 rumored to pretty much be mantle?

[Xsilent(X)]

Gotta love some comments here

[SkilledRebuilds]

Well this is disappointing to me. My hope was Mantle helping more with higher resolutions. Well DX12 here I come.

It does, however mainly with lower end CPU's like APU's or dualcores(or older quads), Mantle/Any Low Level API (DX12) should be able to allow the GPU to power through higher resolutions that were otherwise non-playable with a dualcore powering a beast GPU with non-low level drivers.

 

DX12's low level access is most likely the exact same thing as Mantle (finished however ), just enabled for every GPU thats all... not just AMD.

Gotta love it, when its here!

[warmmilk]

and this is another game that doesnt take much advantage of mantle AT ALL, until a huge RTS comes out using mantle i wont bother with benchmarks, since unless the game ACTUALLY uses 10k+ drawcalls Mantle wont make a big difference to mid to high end users (DirectX can do about 6-7k, Mantle, last i heard months ago it did 100k)

 

if they made a game that actually did 100k draw calls then it'd only be playable with cards able to run mantle...  thats a very small number of PC gamers considering only the latest gen AMD cards run mantle (and I wanna say not all of them but I'm not sure)...  I don't see anyone making a 100k draw call game til all cards are able to run mantle or dx12 if it does something similar to mantle...

[Rageodontist]

I haven't heard really any rumors about DX12 but if that is true than I guess there is nothing to look forward to sadly.

[Rageodontist]

Isn't dx12 rumored to pretty much be mantle?

I haven't heard really any rumors about DX12 but if that is true than I guess there is nothing to look forward to sadly.