• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

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

PhysicsMan94 said:
Lol, come on Matt. Input lag and VSync has basically nothing to do with Mantle, it definitely does not solve the input lag problem. Mantle is a graphical API built to low-level so that it can communicate directly to the GCN architecture more efficiently rather than a high level coded API which has lots of structure overheads. This allows the CPU to deal with more draw calls per unit time than DirectX can currently offer.

But it has nothing to do with the monitor-GPU sync rate problem which causes input lag when VSync is on and tearing when it's off. Mantle doesn't solve that problem at all.
Click to expand...

Lol. Do you think im just making this up for the fun of it? Try Mantle and you'll see. There is no input lag when using vsync. I can't tell the difference with it on or off when using Mantle, but i can with DX immediately. I think its quite noticeable on my screen as i have a 60 Hz 27" 1440P IPS screen. I can't explain why this is the case, so i will ask Johan Anderson.
 
LtMatt said:
Lol. Do you think im just making this up for the fun of it? Try Mantle and you'll see. There is no input lag when using vsync. I can't tell the difference with it on or off when using Mantle, but i can with DX immediately. I think its quite noticeable on my screen as i have a 60 Hz 27" 1440P IPS screen. I can't explain why this is the case, so i will ask Johan Anderson.
Click to expand...

Because your getting lower, more consistent Frame Latencies.

If the time between stitching one Frame to the next is shorter and more consistent, not only will it feel smoother, it will also reduce the time difference between objects calculated and objects seen on your screen, (IE the Latency between what is there and what you actually see on your screen)



 
Last edited:
You should still be able to feel a difference between vsync on and off though in my opinion despite the lower latency.
 
Here's how Mantle benefits my low end CPU:

UnoLWAH.png


Two whole matches. Same map, same settings, Mantle was 64 players, DX was around 34 players. Golmund Railway.

Here's the nitty-gritty:

Mantle FPS: 99.7 average
DX FPS: 92.3 average

Mantle was only 8% faster. However. DX had 686 frames slower than 16.7ms (my monitor's refresh rate). Mantle had only 3. This means that DX was dropping below my monitor's refresh rate regularly. Meanwhile Mantle was perfectly smooth. Like glass.

This is on a humble 4770K at 4.5GHz. I know it's not the best, but I think the best part about Mantle is that poor people like me with budget CPUs can get perfect gameplay without drops or stuttering. People with fast CPUs don't need Mantle but for me with my slow CPU it's really nice to have every single frame above 60 consistently. :o
 
Gloomy said:
Here's how Mantle benefits my low end CPU:

UnoLWAH.png
Two whole matches. Same map, same settings, Mantle was 64 players, DX was around 34 players. Golmund Railway.

Here's the nitty-gritty:

Mantle FPS: 99.7 average
DX FPS: 92.3 average

Mantle was only 8% faster. However. DX had 686 frames slower than 16.7ms (my monitor's refresh rate). Mantle had only 3. This means that DX was dropping below my monitor's refresh rate regularly. Meanwhile Mantle was perfectly smooth. Like glass.

This is on a humble 4770K at 4.5GHz. I know it's not the best, but I think the best part about Mantle is that poor people like me with budget CPUs can get perfect gameplay without drops or stuttering. People with fast CPUs don't need Mantle but for me with my slow CPU it's really nice to have every single frame above 60 consistently. :o
Click to expand...

Your CPU is not slow, think of me and my extinct poor persons AMD Phenom II x6 1090T :o
 
Good results. Golmud is not a very heavy map on the cpu compared to some. Nonetheless would've been interesting to see how DX coped with 64 players.
 
You must log in or register to reply here.
Back
Top Bottom