How much does Mantle benefit low end CPU setups, lets run a few tests.

[LtMatt]

You've got to have a hole in the head to get an AMD CPU though

And Matt, every time I turn around your 290s are running a lower core speed! For Gods sake pick a number and stick with it

Lol do you mean a higher core speed? 4.8ghz was my 24/7 but ive now created a 5Ghz 24/7 profile and tweaked my secondary timings to perfection.

Derp

Dammit it lol.

Jedi i use 975/1250 or 1000/1250 for Mantle as even that is overkill. FOr other more demanding games like Crysis 3 i use 1050/1250 to get a few extra fps though.

 

[Gregster]

was that not the whole idea of mantle though, to give weaker AMD cpus a boost in games?????

Seems like it

 

[LtMatt]

It's kinda irrelevant than Mantle benefit AMD CPU's more, I have no problem with that, they're boosting everyone, the fact that AMD CPU's see a higher gain isn't a problem at all.

I think its quite impressive than even a 5ghz i7 can see a 33% boost.

 

[Martini1991]

I think its quite impressive than even a 5ghz i7 can see a 33% boost.

Like I say, I've been impressed with that Mantle did to Thief.

 

[Marine-RX179]

Good testing Kaap and comparing your results to Humbugs really does show how bad AMD CPU's are

Mantle making AMD CPU's look better... Who created Mantle? Nuff said

To be honest though, even with higherly overclocked Intel there are games which would graphic card would still get horribly bottlenecked. I mean og Guild Wars 2, even my old 5850 would get bottlenecked by my i5 2500K at 4.80GHz with frame rate dipping down to 28-30fps (on a 120Hz monitor), which I think is a joke since they game would use up to 4 threads. The game might not had been perfectly coded, but I think the lots of the performance has been lost to the inefficiency of dx and unnecessary CPU overheads.

 

[Gregster]

To be honest though, even with higherly overclocked Intel there are games which would graphic card would still get horribly bottlenecked. I mean og Guild Wars 2, even my old 5850 would get bottlenecked by my i5 2500K at 4.80GHz with frame rate dipping down to 28-30fps (on a 120Hz monitor), which I think is a joke since they game would use up to 4 threads. The game might not had been perfectly coded, but I think the lots of the performance has been lost to the inefficiency of dx and unnecessary CPU overheads.

I was mostly jesting and I do think Mantle is a good thing in truth. I would like to see it done in a game that needs it though. Something like Civ 6 could be handy and that game is very CPU heavy, so anything taken off of the CPU would be very useful.

 

[Rusty0611]

I think its quite impressive than even a 5ghz i7 can see a 33% boost.

In crossfire I think the gains are good but not surprising. GPU utilisation in DX with 2 or more GPUs can be painfully bad.

Unless I'm mistaken your screenshot is standing still so it would be better to see the gains on CF over a round. I'm sure they've been posted elsewhere but I've not been reading much in here lately

 

[LtMatt]

In crossfire I think the gains are good but not surprising. GPU utilisation in DX with 2 or more GPUs can be painfully bad.

Unless I'm mistaken your screenshot is standing still so it would be better to see the gains on CF over a round. I'm sure they've been posted elsewhere but I've not been reading much in here lately

I will post some results when i can be bothered. I'm taking a break from BF4 at the moment until some of the issues are fixed.

 

[Ian Evey]

I agree this standing around proves that mantle is a lot more efficient in certain hotspots but actually playing a round or 2 would not be 50 fps difference like it is stood on the spot.

i wonder where nvidia make the fps up on amd cards like for like ( 290 + 780 , 280 + 770 etc ) the more intensive battles, or just in certain areas of bf4 maps more max fps.

 

[GoogalyMoogaly]

I will post some results when i can be bothered. I'm taking a break from BF4 at the moment until some of the issues are fixed.

Me too, been about 4 months so far...
Mind you, the number 1 issues for me is "The game is a bit pants" and they're taking their time fixing that one!

Kaap, how about repeating the tests with 2, 3 and/or 4 cards in crossfire to see how Mantle helps with a CPU bottleneck then?

Also occurred to me that there was no way Humbug could really run a comparison with your CineBench score as you used HT (2C4T) which he can't replicate. Maybe try 2 cores or 4 cores with no HT. Maybe also clock them down to 2GHz to see if Humbug was right about 4GHz AMD = 2GHz Intel?

 

[shankly1985]

Good testing Kaap and comparing your results to Humbugs really does show how bad AMD CPU's are

Mantle making AMD CPU's look better... Who created Mantle? Nuff said

How does crossfire scale with Mantle chaps?

sorry, I couldn't resist

Crossfire working fine here on mantle and BF4
Intel cpu here and also gain

 

[robertchalmers1]

Crossfire works perfectly for me as well in bf4 with mantle

 

[Mtom]

If Star Citizen will ever release that will be a good measure for mantle.
It has so high draw calls, that even if you have a Titan, it can't run much faster than smaller cards because of CPU bottleneck.

 

[LtMatt]

Crossfire working fine here on mantle and BF4
Intel cpu here and also gain

Crossfire works perfectly for me as well in bf4 with mantle

290 crossfire in Mantle makes a mockery of multi gpu DX11.1 imo.

 

[humbug]

Good testing Kaap and comparing your results to Humbugs really does show how bad AMD CPU's are

Mantle making AMD CPU's look better... Who created Mantle? Nuff said

How does crossfire scale with Mantle chaps?

sorry, I couldn't resist

*Can't tell if serious*

Its got to be a good thing that my "Bad AMD CPU" out performs your £450 (3930K @ 4.6Ghz) if i use Mantle.

I mean, thats the real joke here is it not?

Spoiler

3930K @ 4.625 - Single Titan @ 1267/3767

Min - 45.4
Avg - 89.3
Max - 115.4

Thanks

Ok. Tommy, i have different settings to you?
See how i have 'Limit Texture' and 'Image Based Reflection'.

Phenom II x6 1090T @ 4Ghz
Radeon R9 290P @ 1075 / 1400

My Score are;

DirectX
Min: 31.4, Max: 73.5, Avr: 45.4

Mantle
Min: 57.1, Max: 89.6, Avr: 67.8

Mantle rocks, Again!





DirectX

Spoiler

Mantle

Spoiler

 

[Rusty0611]

Yawn

 

[LtMatt]

Yawn

No smoke without fire.

 

[triss]

Avr: 67.8 how does this beat a Avg - 89.3 btw or am i just bad at maths?

 

[humbug]

Avr: 67.8 how does this beat a Avg - 89.3 btw or am i just bad at maths?

Minimum FPS is the cause of a CPU Bottleneck. in that.

Gregster:
3930K @ 4.625 - Single Titan @ 1267/3767
Min - 45.4

Me
Phenom II x6 1090T @ 4Ghz
Radeon R9 290P @ 1075 / 1400
Mantle
Min: 57.1

 

[tommybhoy]

i wonder where nvidia make the fps up on amd cards like for like ( 290 + 780 , 280 + 770 etc ) the more intensive battles, or just in certain areas of bf4 maps more max fps.

Frame times/smooth fps.

Relaxing latency output increasing frametimes=higher fps with lower lows resulting in reduced smoothness.

Drop DX11 frametimes to=Mantle frametimes=comparable smoothness with a fps reduction.