Metro last light benchmark with gtx 780 sli on 4k

Metro last light benchmark with gtx 780 sli on 4k full#
Metro last light benchmark with gtx 780 sli on 4k code#
Metro last light benchmark with gtx 780 sli on 4k Pc#

Oh, and even with a pair of Titan X cards running at full pelt they still drew less power from the wall than a single R9 295X2.ДirectX 11 4K synthetic performance - Heaven 4.0 Minimum FPS Average FPSДirectX 11 4K gaming performance - BioShock Infinite Minimum FPS Average FPS Heaven 4.0 and GRID 2 see the AMD cards still keeping their lead, but in Bioshock, BF4, and Metro Last Light the two Nvidia GPUs outshone the four AMD chips. Testing against a pair of the twin-GPU Radeons-to make a quad-CrossFireX setup-had the Nvidia cards coming out on top in a surprising number of my benchmarks. Interestingly, with a pair of GTX Titan X cards the roles are practically reversed with the R9 295X2. 49FPS is still not the 60FPS target weвЂ™re all hoping for, but itвЂ™s still pretty darned glorious. IвЂ™m a benchmarking nerd and I almost wept seeing the Heaven test move so lithe across my screen at 3840 x 2160. Running them at 4K, especially on a 28-inch G-Sync screen, was an absolute, non-headachey joy. Not pleasant, right? Imagine those green swatches strobing up and down the screen at irregular intervals all the way through my testing runs, bizarrely avoiding the specific light-sources in-game.Д«ut all that was happening at the sort of resolutions youвЂ™d be mad to run your twin Titan X cards at.

Something weird was happening with them when I wasnвЂ™t running at the 4K native resolution of my panel, but frustratingly not every time

Metro last light benchmark with gtx 780 sli on 4k Pc#

With a pair of GTX Titan X cards powering my PC I was getting ludicrous gaming performance, but I was also getting a bit of a migraine. If thatвЂ™s why a single GTX Titan X is better than a twin-GPU R9 295X2, why am I even bothering to SLI two of NvidiaвЂ™s latest cards? ItвЂ™s all about the chase for 4K gaming at the highest settings, isnвЂ™t it?Д«ut in that pursuit, admittedly with early drivers, the problems of flaky multi-GPUs came to the fore. Sometimes that happens via a day one patch, sometimes it can be a couple of weeks after the fact and sometimes it doesnвЂ™t happen at all.

Metro last light benchmark with gtx 780 sli on 4k code#

When a new game, new bit of software, or a new piece of hardware is released weвЂ™ll often need someone somewhere to code in a fix to get things working again. It was all to do with the fact that a quick single GPU is going to deliver far less gaming stress than a multi-GPU array.ДЄs much as multi-GPU computing has improved in the last few years, not just in the scaling of performance but in the overall reliability, you'll still hit times when your pair of cards or GPUs is cut in half by an uncooperative driver-something that very, very rarely happens with a single GPU.ДЄn esteemed colleague made the point that essentially the default state of multi-GPU is to be broken and in need of a fix. I got a bit of flack for suggesting that, despite the Radeon offering higher average frame rates, the GTX Titan X offered a much better gaming experience overall.

So into my test rig go a pair of GeForce GTX Titan X cards, and during our time together they ably demonstrate exactly why NvidiaвЂ™s magnum opus is a better graphics card than AMDвЂ™s slightly quicker Radeon R9 295X2. Nvidia gives us the fastest GPU thatвЂ™s ever been produced and whatвЂ™s my response? 'What happens if I strap two together?'