WEBVTT 00:00:00.280 --> 00:00:09.920 so this video is going to be about the GTX 590 compared to its closest 00:00:05.600 --> 00:00:13.280 competition in terms of power and noise 00:00:09.920 --> 00:00:16.720 so that would be the 6990 as well as two 00:00:13.280 --> 00:00:18.439 GTX 580s in SLI so the first thing I'm 00:00:16.720 --> 00:00:24.320 going to do is I'm going to take an idle temperature as well as an idle power 00:00:20.880 --> 00:00:28.760 reading on the GTX 590 so the idle 00:00:24.320 --> 00:00:30.240 temperature right now is about 35 or 36° 00:00:28.760 --> 00:00:36.120 depending on whether we can actually get the camera to focus for you guys or not 00:00:32.040 --> 00:00:37.360 there we go so 30 let's say 35° on the 00:00:36.120 --> 00:00:45.480 GPU cores and then we've also got our idle 00:00:40.440 --> 00:00:47.520 power at the wall being about 96 00:00:45.480 --> 00:00:50.840 Watts so what I'm going to do is I'm going to put audio Snippets at the end 00:00:49.160 --> 00:00:54.280 of the video so if you just want to see just if you just want to hear how these 00:00:52.680 --> 00:00:58.239 cards compare to each other in terms of noise then you can skip there otherwise 00:00:56.280 --> 00:01:02.079 I'll show you guys the temperatures and power consumption all along for all the 00:01:00.039 --> 00:01:07.040 these different solutions just so you guys know for the max power I using the 00:01:04.720 --> 00:01:11.600 batch size test as recommended by Sky MTL from Hardware conu I'm going to 00:01:09.040 --> 00:01:18.600 repeat that test so that'll give me a nice high GPU temperature as well as a 00:01:14.600 --> 00:01:22.000 maximum power load so here we go so our 00:01:18.600 --> 00:01:25.560 load power consumption for the GTX 590 00:01:22.000 --> 00:01:29.520 clocks in at around 340 watt so we're 00:01:25.560 --> 00:01:32.079 just going to record that as 340 and uh 00:01:29.520 --> 00:01:37.200 you can see there's a cat investigating it and our maximum temperatures after 00:01:34.600 --> 00:01:41.399 running this particular little Benchmark or rather our stable 00:01:39.240 --> 00:01:47.040 temperatures are 78 and 79° on each of the gpus so we'll call 00:01:44.200 --> 00:01:47.040 that 00:01:48.040 --> 00:01:57.360 79 and I really wish that window would go away I don't care I don't care 78 so 00:01:53.880 --> 00:02:00.880 there you go or 79 rather um so we'll go 00:01:57.360 --> 00:02:03.600 ahead and record that in our 00:02:00.880 --> 00:02:11.080 little uh Power W there really is quite a dramatic difference 00:02:06.680 --> 00:02:13.080 between the maximum power and the uh and 00:02:11.080 --> 00:02:19.519 the minimum power NVIDIA's really done a good job of uh clocking down the GPU 00:02:16.800 --> 00:02:23.519 when it's not being used you can see right now it's actually stepping down 00:02:21.120 --> 00:02:29.360 right now so it goes all the way down to 51 mahz on the 00:02:26.000 --> 00:02:30.480 GPU Focus there you go very nice very 00:02:29.360 --> 00:02:36.720 cool all right so there's my idle power consumption for the Radeon 6990 so we'll 00:02:34.599 --> 00:02:43.920 call that 132 and I should write that on my handy 00:02:39.720 --> 00:02:46.920 little chart at some point and then um 00:02:43.920 --> 00:02:49.200 oh hold on let's just do that and then 00:02:46.920 --> 00:02:54.159 here are my idle temperatures so it settles in around 00:02:50.800 --> 00:02:56.360 41° on the GPU that has less ventilation 00:02:54.159 --> 00:03:01.120 so the one at the back of the card and 39° on the GPU at the front of the card 00:02:59.440 --> 00:03:07.200 so there there's our idle values so here's the max power 00:03:03.319 --> 00:03:11.440 consumption of the radon HD 6990 for the 00:03:07.200 --> 00:03:11.440 sake of argument why don't we say around 00:03:12.159 --> 00:03:19.120 400 varies let's take the Peaks almost 00:03:16.120 --> 00:03:21.840 the peak so we'll say about 405 Watts 00:03:19.120 --> 00:03:27.640 cuz it sort of Rises and dips so we're running the 3D Mark test and uh let's 00:03:25.159 --> 00:03:31.439 have a look at what we achieved in terms of maximum temperatures 00:03:31.840 --> 00:03:39.040 bringing up after burner here so our Max 00:03:34.879 --> 00:03:42.280 temps Peak this is artificial um but 00:03:39.040 --> 00:03:45.879 they peaked at around 85 and 00:03:42.280 --> 00:03:48.720 82° on both of the gpus so here I'm just 00:03:45.879 --> 00:03:56.760 going to make it focus focus focus thank you all right so 85 and 82° were my Peak 00:03:53.360 --> 00:03:56.760 temperatures on the 00:03:57.239 --> 00:04:05.319 gpus so let's call that a 15 50 wat 00:04:01.239 --> 00:04:07.159 idle total system power draw and bearing 00:04:05.319 --> 00:04:11.599 in mind that all of these numbers for the power are coming through an 80 plus 00:04:09.560 --> 00:04:15.480 silver Power Supply so you got to kind of uh factor that in if you want to 00:04:13.599 --> 00:04:20.560 figure out how much power they're actually consuming so this is two GTX 00:04:17.880 --> 00:04:26.960 580s both of them are stock clocked reference designs and our idle 00:04:24.160 --> 00:04:31.960 temperatures are actually come in similar on one of them okay so similar 00:04:29.479 --> 00:04:35.840 on this one this is the bottom GPU at 34° and then quite a bit higher so the 00:04:34.360 --> 00:04:40.160 reason that second one is quite a bit higher is that there's only that much 00:04:37.720 --> 00:04:44.600 space for the fan to draw air in whereas the GTX 00:04:41.400 --> 00:04:47.600 590 is getting fresh air for both of the 00:04:44.600 --> 00:04:50.000 gpus going out either side so uh let's 00:04:47.600 --> 00:04:55.160 get some load temps and load power consumption going on here and then we 00:04:52.520 --> 00:05:02.520 will have some commentary at about 00:04:57.680 --> 00:05:05.440 465 Watts the GTX 580 solution consumes 00:05:02.520 --> 00:05:10.320 the most power out of all of the configurations that are here so that's 00:05:08.160 --> 00:05:16.880 uh going with the two dedicated dual gpus rather than a single dual GPU card 00:05:14.560 --> 00:05:21.720 uh in both cases consumes more power it is 00:05:18.759 --> 00:05:26.479 um the highest performing solution out of all of the ones that we've looked at 00:05:23.720 --> 00:05:30.800 so that's to be considered as well although sometimes the margin is 00:05:28.479 --> 00:05:34.400 substantial and other times it's less substantial uh no it's fairly 00:05:33.000 --> 00:05:42.680 substantial for the most part just looking at uh at these numbers here 00:05:37.479 --> 00:05:46.120 anyway so uh yeah I'll show you the load 00:05:42.680 --> 00:05:49.080 temps it's me in my pink shirt so for 00:05:46.120 --> 00:05:54.639 load temps we're looking at 82 and 77° so that GPU uh that is starving for air 00:05:52.039 --> 00:05:59.960 in the middle uh really well does starve for air a little bit and I think that's 00:05:58.000 --> 00:06:03.039 pretty much it guys so stay stay tuned and I'm going to show you all the audio 00:06:01.240 --> 00:06:08.400 samples from these cards thanks for checking out this video don't forget to 00:06:04.319 --> 00:06:08.400 subscribe to lus Tech tips