PLA (or Passion Leads Army) is Direct3D 11 based on the Unreal Engine 3 game engine. The benchmark features hardware tessellation and PhysX-based physics simulations (flags, particles/explosions, etc.).
Here is a quick test with three graphics cards: a GeForce GTX 680, a GeForce GTX 580 and one GTX 480. I didn’t included Radeon cards because PhysX seems to be completely disabled (no CPU PhysX) with AMD’s cards.
The specs of the Z77 testbed used can be found HERE. I used the R301.24 beta drivers.
TEST 1 – default settings: 1280×720, texture quality: normal, PhysX: medium
| EVGA GeForce GTX 680 (GPU@1097MHz, mem@3004MHz) – 135 FPS  |
| EVGA GeForce GTX 580 (GPU@797MHz, mem@2025MHz) – 124 FPS |
| EVGA GeForce GTX 480 (GPU@700MHz, mem@1848MHz) – 105 FPS |
TEST 2 – custom settings: 1920×1080, texture quality: high, PhysX: high
| EVGA GeForce GTX 680 (GPU@1097MHz, mem@3004MHz) – 75 FPS |
| EVGA GeForce GTX 580 (GPU@797MHz, mem@2025MHz) – 62 FPS |
| EVGA GeForce GTX 480 (GPU@700MHz, mem@1848MHz) – 50 FPS |
You can find a download link in this thread on Geeks3D’s forum.
i was expecting more from 680 …
me too! the difference is only 11-13 fps! poorly!
I am starting to think the 600 series improvements are coming from more refined and better coding practices in some benchmarks and games vs. the 580… I mean the 680 is fast, but I am not so sure the horsepower is quite as much as people think it is over the 500 series… Hopefully Maxwell will slaughter Kepler…
GTX680 is most likely bandwidth starved in many benchmarks. It has more than doubled both the compute capacity and texture fill-rate, but still relies on the same memory throughput, as in the reference GTX580 board (~192GB/s).
@fellix: indeed, and that is what is the most limitating with deferred renderers.
On my GTX 570 SLI setup (and i5 2500k @4Ghz), I got with the cards running @850Mhz
75.6 fps Single card no PhysX
73.9 fps Single card with Physx (Physx at auto in Nvcpl-second card was showing activity)
59.7 fps Single card with Physx (Physx forced at card no1 in Nvcpl)
86.6 SLI with Physx (Physx at auto in Nvcpl)
128.5 fps SLI no Physx
Nvidia has proven time and again that PhysX is a total waste of resources, relative to what it offers in visual rewards, so why do they still bother?
^Using the settings that JegX used in test 2.
I never liked Unreal Engine, and this demo confirms it. Terrible reflections, terrible particles, etc. (But maybe it’s because of the data and not the engine). Anyhow I made two runs with two GTX 580 @945/2280@1.163v.
Standard 1280×720 and my native 2560×1600 (and with higher settings).
http://i.minus.com/j5YHunlptrFqC.jpg
and
http://i.minus.com/jpUJheoJwR3cc.jpg