Gadgetory

welcome back to harbor unbox now recently we did check out how the Radeon RX Vega 64 and GeForce GTX 1080 stacked up in 2018 using the latest drivers and batch of games it was a pretty even fight across the 27 games tested but there were a few big wins going in videos way and this did allow the GTX tane to win by a 5% margin over all the games that offered support for DirectX 11 and DirectX 12 I did test both the api's I did notice some very mixed results which I didn't discuss in the previous benchmark video so that was something we could quickly do now so I don't think I need to explain too much more I think we'll just talk about the test system and then we'll get to the benchmark results so as usual I'm using our Corsair GPU test rig which uses the core i7 87 okay which is clocked at 5 gigahertz and it has 32 gigabytes of ddr4 3400 memory I know this will come up with a few of you and would have been nice if we had have included the Rison 720 700 X but due to time constraints because there is a bit going at the moment I wasn't able to update with all those results and believe it or not that kind of testing does take quite a large amount of time and as I said I just don't have that time available at the moment and this is testing I would like to do but I think at this point I will reserve it for the next-generation GPUs anyway we have half a dozen titles to check out so let's get into it first up we have the battlefield 1 results and here we see when using the DirectX 11 API the performance between Vegas 64 and the GTX 1080 is very similar we see the exact same 106 FPS on average with a slightly better 1% lower result for the radio and GPU however once we move to DirectX 12 the GTX 1080 falls away while Vegas 64 maintains the same performance scene when using DX 11 so providing you have a high-end CPU it doesn't really matter which API Vega owners use while GeForce owners will certainly want to stick with DirectX 11 this time when testing with day 6 mankind Nevada we see a similar average frame rate using either DirectX 11 or DirectX 12 the GTX 1080 was just 4 to 5 FPS slower than the value graphics card in both situations however we see a rather hefty 11% drop-off for the 1% result when using DirectX 12 which is obviously a very bad result in theory the low-level API should offer better frame time performance but in this implementation it's much worse this leads me to conclude that anyone with a high end CPU should avoid DirectX 12 entirely for this title here we see the complete opposite the MPC heavy hit man runs world's better using DirectX 12 for both the Radeon and GeForce GPUs the performance uplift really is massive but please note this hasn't always been the case about a year ago now there was very little difference between the two api's and we found for the most part that dx11 was more consistent today though the GTX 1080 is 18% faster than using DirectX 12 and Vega 64 was a whopping 27 percent faster truly massive gains there that means anyone playing hitman will certainly want to make sure they're using the DirectX 12 API moving on to some testing with Sniper Elite 4 here we see the GTX tonight you have a significant advantage when using DirectX 11 however once we test both GPUs using DX 12 we find a very similar performance performance that is comparable to the GTX 280s dx11 result warhammer vermintide 2 will likely play better using DirectX 12 with a Loren CPU but with a high-end CPU it seems DirectX 11 is the way to go at least with Vegas 64 even the GTX 1080 sees a massive 16 percent drop in performance for the 1% low result we're using DirectX 12 opposed to DirectX 11 and this isn't the first time we're seeing such a result so once again for high end hardware we recommend using DirectX 11 and this gives Vegas 64 a notable advantage Warhammer 2 throws up some mixed results here we see Vegas 64 actually providing better performance using DirectX 11 while the GTX 1080 is much the same using either API an odd result but it's another example of how baked in DirectX 12 performs okay so we're now deep into 2018 despite that we're still seeing mixed DirectX 12 performance and quite shockingly it's some of the oldest dx12 titles that are now showing the best results with the more modern API despite lacking full hardware support and video husband a little workaround Pascal shortcomings with driver trickery though we have seen this impact the performance of certain CPUs namely those from AMD I should just note the margins between DirectX 12 and DirectX 11 will vary depending on the CPU used Laurin CPUs may benefit more from DirectX 12 but for high-end rigs using well high-end CPUs and GPUs it's quite clear that you will need to do some research to work out which API are work best in a given title it was quite surprising to see the Vega 64 really only offered a performance boost over the GTX 1080 and DirectX 12 titles such as battlefield one and yeah it was just battlefield one wasn't it so hmm oh and for those of you donning the red warrior face paint this was a DirectX 11 versus DirectX 12 comparison so that's why the only true Vulcan games worth testing were included I know that's going to come up so much in the comments why don't you test volcán it always comes up DirectX 11 DirectX 12 is the reason here and on that note I'm running for cover as always subscribe for more content if you appreciate the way we do your hair unboxed and consider supporting us on patreon thanks for watching I'm your host Steve again next time