WEBVTT 00:00:04.880 --> 00:00:14.879 welcome to my performance review at 1080p resolution for the GeForce GTX 00:00:12.599 --> 00:00:19.600 Titan this card is in a class of its own it has over 2500 Cuda course overclocked 00:00:18.240 --> 00:00:25.080 and we run all of our Graphics benchmarks overclocked we were able to 00:00:21.680 --> 00:00:28.800 run it consistently at well over 1.1 GHz 00:00:25.080 --> 00:00:32.200 around 1.13 to 1.4 GHz with GPU boost it 00:00:28.800 --> 00:00:34.200 features GPU boost 2.0 which basically 00:00:32.200 --> 00:00:39.480 amounts to more customizability in terms of the fan profiles as well as the Boost 00:00:37.079 --> 00:00:43.160 clocks of the card and even overvolting options that we've never seen before and 00:00:41.600 --> 00:00:51.760 I think you guys are going to be pretty impressed with the performance this is based on the same Kepler architecture as 00:00:47.199 --> 00:00:53.680 the GK 114 or GTX 680 which was the 00:00:51.760 --> 00:00:59.039 previous ruler of the roost in terms of the Invidia stack but this one comes in 00:00:56.520 --> 00:01:02.960 at a much higher performance level and a much higher price has 6 gigs of RAM 00:01:01.480 --> 00:01:09.479 which probably won't benefit it much at 1080p but you can check out my other 00:01:04.600 --> 00:01:12.080 videos at 2560 x600 as well as 5760 x 00:01:09.479 --> 00:01:16.200 1080 in three ways surround in order to get some idea of what you think about 00:01:14.240 --> 00:01:20.360 the 6 gig frame buffer now the contenders for this video this is meant 00:01:18.200 --> 00:01:26.960 to be a heavyweight Showdown there are no no mid-range Graphics Solutions 00:01:23.479 --> 00:01:29.560 included so we've gone with 660ti SLI to 00:01:26.960 --> 00:01:34.159 show sort of what two better value but lower performance cards in SLI might be 00:01:31.600 --> 00:01:39.280 able to do we've gone with a 7970 which is the best single GPU card from the AMD 00:01:36.880 --> 00:01:43.520 Camp to go up against NVIDIA's best single GPU card however they're in very 00:01:41.200 --> 00:01:48.759 different price ranges we've got a GTX 680 which is NVIDIA's previous top 00:01:46.520 --> 00:01:53.799 top-of-the-line single card and to represent dual 797s since we did take 00:01:51.840 --> 00:02:01.479 two lower-end gpus from NVIDIA and run them in SLI we have an ASUS Aries 2 so 00:01:57.920 --> 00:02:03.039 this is 27970 GHz Edition gpus this 00:02:01.479 --> 00:02:08.560 one's pretty expensive but the results should be pretty close to what you'd get 00:02:04.719 --> 00:02:10.080 with two normal cards as well so let's 00:02:08.560 --> 00:02:15.360 start with our first game which is Crisis the old joke can it run crisis 00:02:12.400 --> 00:02:20.319 the answer is yes yes indeed it can the GeForce GTX Titan destroys crisis it 00:02:17.680 --> 00:02:25.280 beats the Aries 2 which is priced about $500 higher for an actual Aries 2 or 00:02:23.400 --> 00:02:32.599 pretty similarly if you buy a couple 797s it beats 660 Ti SLI and leaves the 00:02:29.480 --> 00:02:36.640 7 970 and the 680 the other only single 00:02:32.599 --> 00:02:38.560 GPU Solutions in the dust Crisis 3 00:02:36.640 --> 00:02:44.120 obviously just launched so this is another one ha can it but but can it run 00:02:41.280 --> 00:02:50.040 Crisis 3 the answer is once again yes it can the only solution that beats out the 00:02:46.200 --> 00:02:53.360 GTX Titan in Crisis 3 is the GTX 660ti 00:02:50.040 --> 00:02:56.800 SLI solution so it beats the 680 by a 00:02:53.360 --> 00:02:59.360 handy 25 to 30% beats the Aries 2 00:02:56.800 --> 00:03:04.280 handily and the 7970 you can see here the Aries 2 didn't scale very well in 00:03:01.080 --> 00:03:06.280 Crisis 3 probably not optimized yet for 00:03:04.280 --> 00:03:10.280 dual GPU on AMD it wasn't much faster than a single 00:03:07.560 --> 00:03:14.080 7970 moving right along Far Cry 3 another popular title that's actually 00:03:12.319 --> 00:03:17.080 pretty new as well and something you might notice about all these graphs is 00:03:15.680 --> 00:03:22.560 we've labeled what settings we're running at so you can reproduce the results if you want and we've gone with 00:03:21.000 --> 00:03:27.680 higher settings than we normally do because this is a heavyweight battle so 00:03:24.400 --> 00:03:31.239 the Titan comes out on top again in Far 00:03:27.680 --> 00:03:33.720 Cry 3 Beating out the Aries 2 the 7970 00:03:31.239 --> 00:03:38.599 660 Ti SLI really didn't Fair very well in Far Cry 3 barely scaling at all and 00:03:36.480 --> 00:03:44.840 just edging out the GTX 680 which comes in at the bottom of the Heap for this 00:03:41.040 --> 00:03:46.239 particular uh game Skyrim now Skyrim 00:03:44.840 --> 00:03:50.720 we're doing things quite a bit differently in the past we've run a bone 00:03:48.519 --> 00:03:55.360 stock Skyrim with just the high-res texture pack that's gone completely out 00:03:52.959 --> 00:04:00.000 the window we are running 18 mods from the Steam Workshop and Slick's going to 00:03:57.280 --> 00:04:03.439 be creating a Skyrim benchmarking guide hopefully sometime in the next couple 00:04:01.519 --> 00:04:08.200 weeks so you guys can check out how we run Skyrim exactly because it gives it 00:04:06.120 --> 00:04:13.079 it makes it much more demanding and gives it more of a more of a PC like uh 00:04:11.560 --> 00:04:17.000 well makes it so there's a point having a high-end graphics card to run it gives 00:04:15.239 --> 00:04:20.959 you more of a PC gaming experience rather than dumbed down console Graphics 00:04:18.959 --> 00:04:26.040 lots of trees better textures on everything looks absolutely fantastic so 00:04:23.080 --> 00:04:30.039 the Titan comes out on top in Skyrim and uh the Aries 2 is only beat by the Titan 00:04:28.280 --> 00:04:35.280 then the single GPU Solutions fall a little bit down and GTX 660 Ti SLI 00:04:32.840 --> 00:04:38.919 really didn't scale that much from 680 well the thing that I do notice here is 00:04:37.120 --> 00:04:42.400 that we still don't see maybe as much performance difference in Skyrim as we 00:04:40.520 --> 00:04:46.520 could if we added even more mods so by the time we do our performance guide 00:04:43.720 --> 00:04:49.639 it's possible we'll have changed to this Battlefield 3 is getting to be a bit of 00:04:48.199 --> 00:04:54.840 an older title but it's still very graphically demanding AMD Crossfire X 00:04:52.720 --> 00:04:59.880 scales incredibly well in this game so you can see the Aries 2 actually 00:04:56.440 --> 00:05:02.080 performs about double as well as a 797 00:04:59.880 --> 00:05:06.880 it's so rare to see perfect Crossfire scaling then the Titan performs 00:05:03.919 --> 00:05:10.479 similarly to a 660 Ti SLI so this is where that discussion about do you want 00:05:08.400 --> 00:05:16.479 a single GPU or do you want a dual GPU solution really comes to comes to a head 00:05:13.960 --> 00:05:20.639 because dual GPU Solutions like this where the Aries 2 just destroys 00:05:18.199 --> 00:05:23.840 everything else look pretty compelling grab a couple of these you're going to 00:05:21.880 --> 00:05:30.400 get the best possible gaming experience in Battlefield 3 hands down however the 00:05:27.960 --> 00:05:33.880 Titan because it's a single GP is consistent across the board it 00:05:32.240 --> 00:05:38.680 consistently performs better than the 680 and the 7970 single gpus whereas 00:05:37.000 --> 00:05:42.680 sometimes those dual graphics card Solutions perform the same or just a 00:05:40.919 --> 00:05:48.520 little bit better and sometimes they leave the single gpus in the dust our 00:05:45.479 --> 00:05:50.560 last two games are Metro 2033 which is 00:05:48.520 --> 00:05:55.000 also an older title but does continue to scale well which did incredibly well in 00:05:52.720 --> 00:06:00.600 crossfire again with the Aries 2 leaving everything behind the Titan still beats 00:05:56.919 --> 00:06:02.919 out the 660ti and destroys the GTX 680 00:06:00.600 --> 00:06:07.400 this seems to be more of an AMD favoring title uh in spite of the the way it's 00:06:05.600 --> 00:06:11.400 meant to be played branding on the title but you see that happen from time to 00:06:09.080 --> 00:06:15.360 time Witcher 2 is our last one we're using the ultra preset including Uber 00:06:13.599 --> 00:06:18.960 sampling now one of the things that made Uber sampling unappealing in the past 00:06:17.400 --> 00:06:23.000 was the fact that it was extremely demanding on the graphics card memory at 00:06:21.400 --> 00:06:26.240 a time when most graphics cards were shipping with one to 1 and a half gigs 00:06:25.319 --> 00:06:32.120 of vram not a big deal anymore six gigs of 00:06:29.960 --> 00:06:35.280 RAM you can run Witcher two at Ultra presets at pretty much whatever 00:06:33.520 --> 00:06:39.280 resolution you want that's what makes the Titan ready for not only current 00:06:37.440 --> 00:06:44.120 generation games but future generation games and future resolutions even 00:06:41.319 --> 00:06:49.120 surround was no big deal in terms of the memory usage on the GTX Titan so that 00:06:47.759 --> 00:06:53.120 pretty much wraps it up in terms of 1080p performance the conclusion is 00:06:51.039 --> 00:06:57.479 pretty simple if you're looking for the best possible performance from a single 00:06:54.880 --> 00:07:01.599 GPU card then go for it however it's not cheap and unlike what you're going to 00:06:59.800 --> 00:07:06.879 see in our other performance review at 2560 x600 the Titan doesn't really 00:07:04.680 --> 00:07:11.319 stretch its legs until you get to those higher resolutions with that said we did 00:07:09.319 --> 00:07:16.160 run the games at more demanding settings than we usually do with lower-end cards 00:07:13.319 --> 00:07:20.479 in order to show you that even at 1080p the Titan does distance itself from 00:07:18.160 --> 00:07:24.520 other single GPU Solutions thank you for checking out this video review on Linus 00:07:22.840 --> 00:07:29.479 Tech tips don't forget to subscribe for more unboxings reviews and other 00:07:26.080 --> 00:07:29.479 computer videos