Gadgetory

hey guys welcome back to harbor unboxed today's video is an update of a video I did back in November if you recall I checked out if in videos new geforce gtx 1050 TI graphics card had the measure of 2012's flagship gtx 680 weeks prior to creating that video I was benchmarking with the new titanfall 2 game and I noticed well benchmarking that the 1052 I was actually able to match the performance of the gtx 680 in that title so out of interest i looked a bit further into some other games to see how the two compared and i found that actually overall the 1052 I was slightly faster and this is despite featuring 33% fewer cuda cores the 1050 Ti is of course based on a much more refined architecture Pascal is considerably more efficient than Kepler so the 1052 I might have fewer cores but the cause it does have are able to get more done and they're also clocked higher for good measure anyway in that video heaps of you asked me to retest with the Radeon HD 7970 included well actually okay so from the 500 plus comments like two of you asked for a retest but whatever I thought it was a good idea so here we are complementing the games tested previously I've thrown in some oldies as well the Tomb Raider reboot along with Sleeping Dogs has been added this takes the total games tested to eight all three graphics cards have been tested IR core i7 6700k test rig using the latest available drivers so then on to the benchmarks first up we have f1 2016 and this was one of the games tested previously whether 1052 I was a good bit faster than the GTX 680 here you can see the newer budget GP was 12% faster when comparing the average framerate although the minimum frame rate for the 7970 was slightly lower than the 680 it pulled ahead for the average this meant the 1052 I was just 4% faster than the 7970 when comparing the average frame rate that's what I should note that it was also 23% faster when comparing the minimum frame rate the 1052 is able to slightly improve upon the GTX 680 m Mirror's Edge catalyst however when compared to the 79 7 it was 10% slower and we see here that AMD's flagship GPU from 2012 runs away with it rendering 73 FPS on average for those of you keeping score at home this meant the 7970 was 14% faster than the 680 generally speaking overwatch plays slightly better on NVIDIA Hardware and here we're seeing that still with the ultra quality preset in place the 7970 turned out an average of 100 FPS this meant the 680 was 11% faster while the 1052 I enjoy a 13% performance advantage what about 2016 s most epic title battlefield 1 well here the 7970 was good for 50 FPS on average making it 6% slower than the 680 when compared to the 1050 CI the 7970 was 14% slower though it was just 4 percent slower comparing the minimum frame rate this is another title developed closely with Nvidia so it's not entirely surprising that 680 comes out ahead of the 7970 here that said the margins were extremely slim the 680 was just 5% faster when comparing the average frame rate the 1052 on the other hand was 18% faster you can probably thank driver optimization for that the GTX 680 and 1050 Ti provided very similar performance when testing with titanfall 2 both average 62 fps this made them 6% slower than the 7970 which average 66 fps that is a solid result for AMD's almost five-year-old graphics card like the GTX 680 and HD 7970 sleeping dogs was released back in 2012 comparing the three GPUs in this game shows fairly similar performance across the board the GTX 680 was 4% fast in the 79 70 for the average result but 7 percent slower for the minimum meanwhile the 10 52 I delivered a strong average result with a weaker than expected minimum this isn't this year's rise of the Tomb Raider but rather 2013 s reboot with tress-fx enabled you could argue that AMD has an advantage here but we also test The Witcher 3 Wild Hunt with hair works enabled for the most part so you'll just have to deal with it even if things do get a bit hairy anyway the 7970 was just a single frame faster on average than the 1050 Ti that was 3 FPS faster when comparing the minimum frame rate finally we have the total system power consumption figures and this is where the real gains have been made the core i7 test system consumed a hundred and sixteen what's with the 1050 Ti handling the rendering work that's almost half the power consumption of the 680 while it's less than half of that of these 7970 looking back it doesn't seem like the GTX 680 or the 1050 Ti got trashed by the 7970 like many viewers suggested it would have in the previous comparison video in fact once we tally up the numbers we find some very interesting results I'm sure some of you won't believe me but on average the 680 and 7970 delivered the exact same performance in the eight games tested they both came out with an average of 71 FPS while the 1050 TI node to head with 74 fps so what about the minimums though well again there wasn't much in it the 1050 TI and 7970 both averaged 56 FPS while the 680 was just 1 FPS slower honestly these results aren't at all surprising earlier this year we created a video that compared the 680 and 79 70 and nine new games as well as 9 old games AMD and NVIDIA traded blows in both the new and old titles to live a similar performance overall finally I'd just like to point out that I'm not trying to paint the GTX 1052 as an amazing graphics cart with this video it's simply a look at where we are now and what we've come from the GTX 1050 Ti isn't the best value sub $200 GPU we've already awarded that onto the RX 470 all the r9 to know if you're willing to go second hand and take a big hit on power consumption the reason we use the 1052 is the test subject is simple it's the most recently released GPU making it a bit more of a hot topic personally I really enjoy these new and old comparisons and I know although you guys do as well so be sure to let me know what you thought of the results in the comments below and if you have any ideas for comparisons I could make in the future be sure to let me know those as well anyway that's it for this one guys thanks for watching I'm your host Steve and I'll catch you next time you