Does Mantle... work enough to rock your world and change your next Graphics Card?

[directhex]

I suspect the GTX6600 is a typo, as there was no such card (only a 6600 GT).

Typos fixed.

[Revvvhead]

From my experience Mantle made a huge difference.
I have an i7 920 @4.0Ghz and a Sapphire tri-x R9 290.
When I play BF4, looking at my MSI afterburner charts I see that my GPU memory usage jumps from under 2GB to almost 3GB.
This means I can play at a smooth almost 120fps on ultra with mantle, as opposed to a stuttery 35-70fps without mantle.
Of course mantle is optimised only for certain games, as is Physx on Nvidia.
But for my money AMD cards are the best value, as long as you get a non-reference cooler.

[dfour]

@Keyboarddemon

Sorry its taken me so long I forgot about this thread.

Thief test

low dx 11

min 38.5
ave 54.4
high 85.3

low mantle

min 70.4
ave 85.7
high 110.0

normal dx11

min 36.9
ave 52.4
high 77.3

normal mantle

min 59.4
ave 73.6
high 100.0

high dx11

min 37.0
ave 49.5
high 64.8

high mantle

min 48.3
ave 58.0
high 78.8

v high dx11

min 35.7
ave 43.0
high 60.7

v high mantle

min 40.4
ave 46.3
high 64.0

My rig

fx 3850 @ 4.75
16 gig @1760 mhz
nb 2375 mhz
r9-290x at stock clocks.
1440p monitor.

Im using the 14.4 driver pack from may.

All tests run on same day and only run once as I have important things to do than run each test multiple times and create averages and then excell them.

If you take the average FPS you will see a 5-20% gain roughly in mantle over dx.

Hope this helps.

[KeyboardDemon]

Hope this helps.

It does help, the performance difference between a GTX 780 Ti and a 290X in Battlefield 4 alone is disproportionate to the difference in price between the 2 cards, and I think it is quite unlikely that I would buy a GTX 780 Ti now, see those difference.

That said, I have recently swapped out my two 60hz 23.5" 1080p monitors for a single 24" 1080p that runs at 120hz and have been messing around with 3d Vision which I have actually enjoyed, even though most of my games don't officially support it many of those games look amazing in 3d, such as Elite: Dangerous.

[AMDMatt]

@Keyboarddemon

Sorry its taken me so long I forgot about this thread.

Thief test

low dx 11

min 38.5
ave 54.4
high 85.3

low mantle

min 70.4
ave 85.7
high 110.0

normal dx11

min 36.9
ave 52.4
high 77.3

normal mantle

min 59.4
ave 73.6
high 100.0

high dx11

min 37.0
ave 49.5
high 64.8

high mantle

min 48.3
ave 58.0
high 78.8

v high dx11

min 35.7
ave 43.0
high 60.7

v high mantle

min 40.4
ave 46.3
high 64.0

My rig

fx 3850 @ 4.75
16 gig @1760 mhz
nb 2375 mhz
r9-290x at stock clocks.
1440p monitor.

Im using the 14.4 driver pack from may.

All tests run on same day and only run once as I have important things to do than run each test multiple times and create averages and then excell them.

If you take the average FPS you will see a 5-20% gain roughly in mantle over dx.

Hope this helps.

Thank you for taking the time dfour to share your results with us. Very nice.