Mantle = 45% better performance

[Katness Everdeen]

Remember its says up to, im thinking the biggest difference is with lower end cards eg if a card go from 16fps to 20fps then thats a big difference.

I think it will depend on your CPU, if it's very fast the performance improvement will be less, however if it's a low end CPU taking the load off of it will improve performance immensely.

 

Someone already posted this.

where ?

 

[brownninja97]

I think it will depend on your CPU, if it's very fast the performance improvement will be less, however if it's a low end CPU taking the load off of it will improve performance immensely.

 

where ?

 

and i have an i3 MWHHAAHAHAAHAHAHHA.

[Oshino Shinobu]

I think it will depend on your CPU, if it's very fast the performance improvement will be less, however if it's a low end CPU taking the load off of it will improve performance immensely.

 

where ?

 

Already posted here http://linustechtips.com/main/topic/98505-mantle-45-better-performance/

[M-ursu]

Lies... all lies until i see benchmarks, also this is a repost.

http://linustechtips...er-performance/

[Katness Everdeen]

Already posted here http://linustechtips.com/main/topic/98505-mantle-45-better-performance/

It says 45% which is wrong.

[Dragonzeanse]

It says 45% which is wrong.

What are you talking about? By that logic, your thread is wrong, too. The thread wasn't just talking about Battlefield 4 performance.

[Katness Everdeen]

The thread wasn't just talking about Battlefield 4 performance.

That's why it's wrong 45% only in BF4 in Oxide it went from 8 FPS to 30 FPS which is more than 300%.

[Oshino Shinobu]

It says 45% which is wrong.

Both this thread and that thread are about the same thing and use the leaks from purepc.

[Morpheus]

Remember its says up to, im thinking the biggest difference is with lower end cards eg if a card go from 16fps to 20fps then thats a big difference.

delete the picture when your quote man

[Kobz360]

But it's not an apple to apples to comparison, APU's (i.e. Kaverii) are meant to be used as stand alone while it is expected that people who invest in an i5 4670K would get a dedicated solution (i.e. 290, 780, etc.) 

Well to be honest, what else would they compare it too? I mean the cheapest comparable graphics solution to the APU would be either an Intel CPU with Intel HD by itself or an Intel CPU with at the very least, a 7750 which once again like I said before, would cost considerably more money. 

 

I understand where you guys are coming from, but there's no better alternative way to compare. 

[Dragonzeanse]

That's why it's wrong 45% only in BF4 in Oxide it went from 8 FPS to 30 FPS which is more than 300%.

Irrelevant. You're discussing the same thing. You don't get to make another thread just because there was one minor inaccuracy that everyone could deduce by reading the article.

[brownninja97]

delete the picture when your quote man

I deleted the pictrue that was irrelevant to the topic and kept the one i cared about.

[Scionyde]

Even if I trusted the source of the leak, the 300% gain came from starting at a crappy framerate and improving to a slightly less crappy framerate on integrated graphics. I'm not even sure I'd trust numbers from some Oxide demo in an engine we haven't seen yet. 

[Morpheus]

I deleted the pictrue that was irrelevant to the topic and kept the one i cared about.

or you could just write "the bf4 picture" and be considerate of others lol

[Butterfish]

I'm kinda getting sick of all this hype. Until i actually see some real proof of this, i can't give two ducks about it..

[kuddlesworth9419]

That is just comparing their APU's though. We already know that Intel APU's are not as good as AMD's. So where exactly is the performance increase. 

[Guest GRRigger]

up to OVER 9000% IMPROVEMENT in carboxide new engine.

 

so far they didnt show much interesting, and very little or no proof. If they really did remove alot of cpu overhead then thats cool. Gaming with low end cpu and high end gpu, no more bottleneck in cpu intensive games.

 

I want to see comparison with nvidia cards and stuff. And benchmarks.

[Chevaishr]

I am very, very sceptical about it. Even if it performs as advertised, how far can proprietary technologies get you these days? I mean, Physx hasn't really taken the world by storm because it only benefits users of Nvidia. 

Considering that Mantle is not even compatible with previous generations, how many developers will ditch the HUGE DirectX market base for Mantle?

thing is its open source and anyone can implement the underlying code, eg nVidia can make the cpu optimization part of mantel work with there gpus 

[TopWargamer]

Man I really hope this is true. At this point, I just hope that Mantle will live up to everybody's expectations.

[Fetzie]

This is the closest to an apples to apples comparison IMO:

 

Kaveri A8-7600, Mantle performance versus DX11.

 

[TechFan@ic]

That is just comparing their APU's though. We already know that Intel APU's are not as good as AMD's. So where exactly is the performance increase. 

Ahh no, the HD 4600 graphics in the 4770K doesn't support Mantle.

That 4770K & 7600 comparison had to be done using a discrete AMD graphics card.

[Kuzma]

Man I haven't played tomb raider yet... Is it worth the time?

yes. SO MUCH YES!

[Oshino Shinobu]

yes. SO MUCH YES!

I agree with this! Can't wait for the next one. 

 

Heard they're making a film too. 

[fade]

I'm actually playing through Tomb Raider right now and I'm kicking myself for not playing it sooner.  Excellent game.  Only downside is that there still seem to be alot of hassle if you have an nvidia card.

[Fetzie]

I'm actually playing through Tomb Raider right now and I'm kicking myself for not playing it sooner.  Excellent game.  Only downside is that there still seem to be alot of hassle if you have an nvidia card.

 

If you have an nVidia card, you have to disable at least TressFX, because that murders your framerate.