{"video_id":"t2QpCtsL5L0","title":"NVIDIA GeForce GTX 980 Video Card - Performance Overview","channel":"Linus Tech Tips","show":"Linus Tech Tips","published_at":"2015-05-07T14:53:29Z","duration_s":683,"segments":[{"start_s":0.0,"end_s":5.359,"text":"Another day, another new graphics card.","speaker":null,"is_sponsor":0},{"start_s":3.28,"end_s":13.92,"text":"Welcome to the Lionus Tech Tips performance overview of the GTX 980 from","speaker":null,"is_sponsor":0},{"start_s":8.559,"end_s":13.92,"text":"NVIDIA, their most advanced GPU","speaker":null,"is_sponsor":0},{"start_s":18.279,"end_s":23.279,"text":"yet. DKL's new Rip Jaws 4 series is","speaker":null,"is_sponsor":0},{"start_s":21.279,"end_s":27.439,"text":"their ultimate DDR4 memory with speeds of up to","speaker":null,"is_sponsor":0},{"start_s":24.359,"end_s":29.519,"text":"3,333 MHz available. Click now to learn","speaker":null,"is_sponsor":0},{"start_s":27.439,"end_s":34.0,"text":"more. As usual, we'll start with the unboxing and physical overview. The","speaker":null,"is_sponsor":0},{"start_s":31.519,"end_s":40.719,"text":"Media Sample GTX 980s come packaged in this sweet box that has the card upright","speaker":null,"is_sponsor":0},{"start_s":37.92,"end_s":47.2,"text":"looking all majestic with its magnesium alloy fan housing, aluminum frame, and","speaker":null,"is_sponsor":0},{"start_s":43.719,"end_s":48.6,"text":"trivalent chromium plating with H.","speaker":null,"is_sponsor":0},{"start_s":47.2,"end_s":53.199,"text":"What's that? Sorry, you've seen this user","speaker":null,"is_sponsor":0},{"start_s":50.8,"end_s":58.32,"text":"configurable green LED lit GeForce logo before. NVIDIA hasn't really changed","speaker":null,"is_sponsor":0},{"start_s":55.68,"end_s":63.359,"text":"their cooler design since the 690. Ah, yes. No, I can see how you might think","speaker":null,"is_sponsor":0},{"start_s":60.399,"end_s":68.479,"text":"that, but you haven't seen the back yet. Check it out. Now, we get a full metal","speaker":null,"is_sponsor":0},{"start_s":66.479,"end_s":72.799,"text":"back plate with a classy looking design and surface. Now, the issue with back","speaker":null,"is_sponsor":0},{"start_s":71.04,"end_s":77.439,"text":"plates, and presumably the reason NVIDIA has never done a reference card with one","speaker":null,"is_sponsor":0},{"start_s":74.479,"end_s":82.0,"text":"before, is that it restricts the intake of a card installed in the next slot","speaker":null,"is_sponsor":0},{"start_s":79.92,"end_s":86.159,"text":"over in tightly packed SLI configurations. So NVIDIA made this","speaker":null,"is_sponsor":0},{"start_s":84.4,"end_s":92.0,"text":"strategic chunk of the back plate removable for users with lots of cards","speaker":null,"is_sponsor":0},{"start_s":89.28,"end_s":97.28,"text":"right next to each other in SLI. Not that you need to run SLI just to hook up","speaker":null,"is_sponsor":0},{"start_s":95.04,"end_s":102.159,"text":"enough monitors anymore. As we move over to the IO, yeah, that's some pretty good","speaker":null,"is_sponsor":0},{"start_s":99.6,"end_s":107.439,"text":"stuff right there. Three DisplayPort ports, so that's up to three G-Sync","speaker":null,"is_sponsor":0},{"start_s":104.4,"end_s":110.479,"text":"displays off of a single card. An HDMI","speaker":null,"is_sponsor":0},{"start_s":107.439,"end_s":114.32,"text":"2.0 port. So yeah, that's 4K 60 Hz over","speaker":null,"is_sponsor":0},{"start_s":110.479,"end_s":116.72,"text":"HDMI. and a dual link DVI port can all","speaker":null,"is_sponsor":0},{"start_s":114.32,"end_s":121.28,"text":"be found here. Any four of which can be used at one time on this beast of a","speaker":null,"is_sponsor":0},{"start_s":119.28,"end_s":127.92,"text":"graphics card. Which leads us to the next notable thing physically. This is a","speaker":null,"is_sponsor":0},{"start_s":124.04,"end_s":132.319,"text":"flagship 80 class card, but like the GTX","speaker":null,"is_sponsor":0},{"start_s":127.92,"end_s":133.92,"text":"680, it only has two six pin connectors.","speaker":null,"is_sponsor":0},{"start_s":132.319,"end_s":139.04,"text":"A hint at the incredible power efficiency of the Maxwell architecture","speaker":null,"is_sponsor":0},{"start_s":136.319,"end_s":144.16,"text":"at the core of the GM 204 chip that powers this card. The GTX 980 not only","speaker":null,"is_sponsor":0},{"start_s":142.48,"end_s":150.08,"text":"hung around at the top of our performance charts, but did it while","speaker":null,"is_sponsor":0},{"start_s":146.48,"end_s":153.519,"text":"consuming less power than GK 110based","speaker":null,"is_sponsor":0},{"start_s":150.08,"end_s":155.519,"text":"cards like the 780 and 780 Ti and while","speaker":null,"is_sponsor":0},{"start_s":153.519,"end_s":159.84,"text":"running extremely quietly. So, let's talk about Maxwell for a bit here.","speaker":null,"is_sponsor":0},{"start_s":157.599,"end_s":163.519,"text":"Without shrinking the manufacturing process, so they're still on the same","speaker":null,"is_sponsor":0},{"start_s":161.519,"end_s":169.44,"text":"28nmter node that they used to build the Kepler based GTX 680, NVIDIA has","speaker":null,"is_sponsor":0},{"start_s":166.319,"end_s":171.239,"text":"effectively reduced TDP by 30 W,","speaker":null,"is_sponsor":0},{"start_s":169.44,"end_s":176.8,"text":"increased the number of CUDA cores by 33%, increased base clocks and memory","speaker":null,"is_sponsor":0},{"start_s":174.239,"end_s":181.76,"text":"speed by around 12% and doubled the number of render output units, as well","speaker":null,"is_sponsor":0},{"start_s":179.2,"end_s":186.64,"text":"as adding some other cool hardware bits to enable some really cool technology.","speaker":null,"is_sponsor":0},{"start_s":184.08,"end_s":191.12,"text":"Pretty amazing, actually. So, first up in the new technology list is their","speaker":null,"is_sponsor":0},{"start_s":189.04,"end_s":195.28,"text":"improved third generation delta color compression. A technique that allows","speaker":null,"is_sponsor":0},{"start_s":193.12,"end_s":200.4,"text":"data coming out of the core to be compressed by a factor of 2:1, 4:1, or","speaker":null,"is_sponsor":0},{"start_s":198.0,"end_s":205.12,"text":"even 8:1 before being committed to memory by storing the value of a change","speaker":null,"is_sponsor":0},{"start_s":203.12,"end_s":210.159,"text":"versus the previous frame rather than the value itself. This combined with","speaker":null,"is_sponsor":0},{"start_s":207.519,"end_s":216.959,"text":"improved caching allows the GTX 980, in spite of its mere 256-bit memory","speaker":null,"is_sponsor":0},{"start_s":213.28,"end_s":219.28,"text":"interface, to achieve up to around 30%","speaker":null,"is_sponsor":0},{"start_s":216.959,"end_s":224.319,"text":"better effective memory performance compared to the GTX 680. In this scene,","speaker":null,"is_sponsor":0},{"start_s":222.4,"end_s":228.64,"text":"the pink highlighted screen elements are taking advantage of this technology.","speaker":null,"is_sponsor":0},{"start_s":226.56,"end_s":232.319,"text":"Next up is a feature that's all about getting the most out of older games","speaker":null,"is_sponsor":0},{"start_s":230.239,"end_s":238.64,"text":"where you have lots of GPU horsepower to spare. DSR or dynamic super","speaker":null,"is_sponsor":0},{"start_s":235.799,"end_s":244.12,"text":"resolution is mostly just another name for a feature that has existed forever","speaker":null,"is_sponsor":0},{"start_s":240.879,"end_s":246.319,"text":"but is now easier to use. Super sampling","speaker":null,"is_sponsor":0},{"start_s":244.12,"end_s":250.239,"text":"anti-aliasing. The way it works is it renders the scene at a higher","speaker":null,"is_sponsor":0},{"start_s":248.0,"end_s":254.64,"text":"resolution. So you can choose 2x, 3x, or 4x in the control panel. Then","speaker":null,"is_sponsor":0},{"start_s":252.48,"end_s":259.12,"text":"intelligently downscales it with a filter whose sharpness or softness can","speaker":null,"is_sponsor":0},{"start_s":256.56,"end_s":264.88,"text":"be adjusted to let you use your extra GPU horsepower that otherwise was just","speaker":null,"is_sponsor":0},{"start_s":262.16,"end_s":269.68,"text":"sitting there to yield potentially a much more pleasing gaming experience.","speaker":null,"is_sponsor":0},{"start_s":267.04,"end_s":275.44,"text":"It's not the same as actually running on a 4K display, but it's much better than","speaker":null,"is_sponsor":0},{"start_s":272.96,"end_s":279.68,"text":"running native 1080p. The effect is especially noticeable on the grass in","speaker":null,"is_sponsor":0},{"start_s":277.84,"end_s":284.08,"text":"Dark Souls II here in this footage that I took at the NVIDIA event. The next","speaker":null,"is_sponsor":0},{"start_s":282.4,"end_s":289.88,"text":"feature is about getting more performance when we don't have enough,","speaker":null,"is_sponsor":0},{"start_s":286.0,"end_s":292.199,"text":"and it's called MFAA or frame","speaker":null,"is_sponsor":0},{"start_s":289.88,"end_s":297.04,"text":"anti-aliasing. That's short for multi-frame sampled anti-aliasing. Grow","speaker":null,"is_sponsor":0},{"start_s":294.72,"end_s":301.919,"text":"up. The way this works is instead of taking a large number of samples with a","speaker":null,"is_sponsor":0},{"start_s":299.84,"end_s":308.0,"text":"static pattern to smooth the edges of on-screen objects or textures, MFAA","speaker":null,"is_sponsor":0},{"start_s":305.04,"end_s":313.44,"text":"takes fewer samples per frame, half as many from what we've been told, but uses","speaker":null,"is_sponsor":0},{"start_s":310.16,"end_s":316.4,"text":"new hardware in the GTX 980 and 970 to","speaker":null,"is_sponsor":0},{"start_s":313.44,"end_s":321.199,"text":"alternate the positions with each frame. Then it applies a clever filter. So the","speaker":null,"is_sponsor":0},{"start_s":318.479,"end_s":326.479,"text":"performance impact is about equivalent to two times MSAA, but the resulting","speaker":null,"is_sponsor":0},{"start_s":324.32,"end_s":331.759,"text":"image looks pretty much identical to four times MSAA. This feature isn't","speaker":null,"is_sponsor":0},{"start_s":329.6,"end_s":334.639,"text":"enabled in the driver yet, so expect an update once we get a chance to play","speaker":null,"is_sponsor":0},{"start_s":333.199,"end_s":338.96,"text":"around with it some more, but the initial results are impressive. Check","speaker":null,"is_sponsor":0},{"start_s":337.199,"end_s":343.36,"text":"out the effect on the transparent textures like these wires in this","speaker":null,"is_sponsor":0},{"start_s":341.12,"end_s":346.88,"text":"Battlefield 4 demo that NVIDIA was showing off at the launch event. Another","speaker":null,"is_sponsor":0},{"start_s":345.199,"end_s":350.72,"text":"thing we won't be able to test for a while is the improvements that NVIDIA","speaker":null,"is_sponsor":0},{"start_s":348.8,"end_s":355.68,"text":"has been making to the VR experience on their cards. The most notable of which","speaker":null,"is_sponsor":0},{"start_s":352.4,"end_s":357.68,"text":"is called auto asynchronous warp. A","speaker":null,"is_sponsor":0},{"start_s":355.68,"end_s":363.84,"text":"cheat, I guess, that allows the head tracking of a VR headset like the Oculus","speaker":null,"is_sponsor":0},{"start_s":359.84,"end_s":366.56,"text":"Rift to sample its position again right","speaker":null,"is_sponsor":0},{"start_s":363.84,"end_s":371.36,"text":"before a frame is finished being drawn to reduce the player's perception of the","speaker":null,"is_sponsor":0},{"start_s":368.96,"end_s":373.919,"text":"delay between physical motion and on-screen motion. And this will be","speaker":null,"is_sponsor":0},{"start_s":372.72,"end_s":378.639,"text":"important for folks who have trouble with motion sickness in VR. Which leads","speaker":null,"is_sponsor":0},{"start_s":376.08,"end_s":382.639,"text":"us to the final feature, their VxGI global illumination technique that we","speaker":null,"is_sponsor":0},{"start_s":380.479,"end_s":386.08,"text":"also can't really test here because while Unreal Engine 4 is getting","speaker":null,"is_sponsor":0},{"start_s":384.319,"end_s":390.319,"text":"support, it'll be a while before we get real games with the tech, but it looks","speaker":null,"is_sponsor":0},{"start_s":388.0,"end_s":394.4,"text":"so freaking cool. NVIDIA has a couple of demos to show it off. One where they","speaker":null,"is_sponsor":0},{"start_s":391.759,"end_s":397.68,"text":"render a Cornell box, so you can see the way the direct light enters at the top,","speaker":null,"is_sponsor":0},{"start_s":396.4,"end_s":403.52,"text":"so the tops of the shapes are the brightest with realistic soft shadows and color bleeding from reflective","speaker":null,"is_sponsor":0},{"start_s":400.72,"end_s":408.24,"text":"surfaces. and one where they recreated as closely as possible the conditions of","speaker":null,"is_sponsor":0},{"start_s":405.759,"end_s":413.6,"text":"this iconic photo from the 1969 lunar landing to demonstrate that it's","speaker":null,"is_sponsor":0},{"start_s":410.56,"end_s":416.4,"text":"completely realistic and wasn't taken","speaker":null,"is_sponsor":0},{"start_s":413.6,"end_s":420.72,"text":"somewhere in Area 51 or whatever. Now, these two demos wouldn't be impressive","speaker":null,"is_sponsor":0},{"start_s":418.24,"end_s":426.08,"text":"at all except that they're being done in real time instead of using path tracing,","speaker":null,"is_sponsor":0},{"start_s":423.759,"end_s":429.759,"text":"a phenomenally computationally intensive global illumination technique that","speaker":null,"is_sponsor":0},{"start_s":427.52,"end_s":434.0,"text":"involves following light rays as they bounce around the scene. like all of","speaker":null,"is_sponsor":0},{"start_s":431.44,"end_s":439.36,"text":"them. NVIDIA is using dedicated hardware to split the scene into cubes called","speaker":null,"is_sponsor":0},{"start_s":436.319,"end_s":442.16,"text":"voxels that act as dynamic light sources","speaker":null,"is_sponsor":0},{"start_s":439.36,"end_s":448.88,"text":"that approximate the results of a more accurate rayraced rendering method but","speaker":null,"is_sponsor":0},{"start_s":445.28,"end_s":452.08,"text":"in real time. All right, so as cool as","speaker":null,"is_sponsor":0},{"start_s":448.88,"end_s":454.24,"text":"all that technology is, how does the","speaker":null,"is_sponsor":0},{"start_s":452.08,"end_s":458.24,"text":"card actually perform? That's what you want to know. Great question, Luke. Take","speaker":null,"is_sponsor":0},{"start_s":456.56,"end_s":461.44,"text":"it away. Before we get into things, I want to point out that overclocking all","speaker":null,"is_sponsor":0},{"start_s":459.68,"end_s":465.039,"text":"of our cards is yet again backed by popular demand. And because of that, you","speaker":null,"is_sponsor":0},{"start_s":463.52,"end_s":468.88,"text":"will be able to find the overclocking results and settings in the graphics","speaker":null,"is_sponsor":0},{"start_s":466.96,"end_s":471.84,"text":"card overclocking dock, which should be linked in the description below this","speaker":null,"is_sponsor":0},{"start_s":470.24,"end_s":476.56,"text":"video. Also, when reading the settings for the GTX 980 in the overclocking dock","speaker":null,"is_sponsor":0},{"start_s":474.319,"end_s":480.319,"text":"below, there's no reason to worry. There is a reason why I ignore the memory","speaker":null,"is_sponsor":0},{"start_s":478.08,"end_s":484.4,"text":"clock speed. The 980, much like the 780 Ti before it, features a fantastic","speaker":null,"is_sponsor":0},{"start_s":482.319,"end_s":488.319,"text":"technology called autobalancing power rails, which in its most basic","speaker":null,"is_sponsor":0},{"start_s":486.56,"end_s":492.08,"text":"description, allows the card to take extra power overhead from parts of the","speaker":null,"is_sponsor":0},{"start_s":490.319,"end_s":496.4,"text":"card that aren't using it and move it elsewhere, allowing you to focus on one","speaker":null,"is_sponsor":0},{"start_s":494.08,"end_s":500.639,"text":"specific thing, which in this case resulted in the GPU clock speed being","speaker":null,"is_sponsor":0},{"start_s":498.639,"end_s":504.16,"text":"cranked through the freaking roof, which is awesome because then you can access","speaker":null,"is_sponsor":0},{"start_s":502.479,"end_s":508.08,"text":"all of that FPS that you wouldn't otherwise be able to. Performance of the","speaker":null,"is_sponsor":0},{"start_s":505.919,"end_s":512.08,"text":"GTX 980 is honestly a little bit surprising in a good way. To give proper","speaker":null,"is_sponsor":0},{"start_s":510.72,"end_s":517.76,"text":"context for that statement, we have to look at the rumor mill. Personally, I like to avoid rumors, but sometimes I","speaker":null,"is_sponsor":0},{"start_s":515.919,"end_s":522.159,"text":"don't really have a choice. For instance, on the September 12th version","speaker":null,"is_sponsor":0},{"start_s":519.36,"end_s":527.04,"text":"of the WAN Show, my brother at Cold SE on Twitter, and I discussed how the 980","speaker":null,"is_sponsor":0},{"start_s":524.48,"end_s":531.2,"text":"was, according to rumors, not supposed to beat out the 780 Ti. So, when I","speaker":null,"is_sponsor":0},{"start_s":529.76,"end_s":538.399,"text":"finally got the card in hand, I was delighted to see that more often than not, it was able to beat the 780 Ti in","speaker":null,"is_sponsor":0},{"start_s":535.12,"end_s":540.56,"text":"both 1080p and 4K benchmarks. Now, my","speaker":null,"is_sponsor":0},{"start_s":538.399,"end_s":547.12,"text":"overclock for the 980 feels a little bit high, running at a range of 1503 to 1492","speaker":null,"is_sponsor":0},{"start_s":544.04,"end_s":548.88,"text":"MHz, but I really have no idea as I have","speaker":null,"is_sponsor":0},{"start_s":547.12,"end_s":554.0,"text":"no one else's overclocks to compare to yet. And to be fair, my 780 Ti overclock","speaker":null,"is_sponsor":0},{"start_s":551.68,"end_s":559.2,"text":"is very aggressive anyways. So, I haven't exactly done the 980 any favors","speaker":null,"is_sponsor":0},{"start_s":556.0,"end_s":560.8,"text":"in regards to the 780 Ti's clock speed.","speaker":null,"is_sponsor":0},{"start_s":559.2,"end_s":564.56,"text":"But don't forget, all of this performance awesomeness doesn't just","speaker":null,"is_sponsor":0},{"start_s":562.56,"end_s":568.88,"text":"stand on its own. Not only are you paying considerably less for this card","speaker":null,"is_sponsor":0},{"start_s":566.56,"end_s":573.12,"text":"than you would pay for a 780 Ti, but you also pay less in the future. Underload,","speaker":null,"is_sponsor":0},{"start_s":571.12,"end_s":577.92,"text":"the 980 has considerably less power draw than the 780 Ti, which is nice if you're","speaker":null,"is_sponsor":0},{"start_s":575.68,"end_s":580.72,"text":"actually paying for your power bill. And if you're not paying for your power","speaker":null,"is_sponsor":0},{"start_s":579.36,"end_s":585.36,"text":"bill, maybe you can use that little tidbit of information to convince your parents that it might be time for an","speaker":null,"is_sponsor":0},{"start_s":583.68,"end_s":588.88,"text":"upgrade. Anyways, thanks for watching my segment, guys. Back to you, Linus.","speaker":null,"is_sponsor":0},{"start_s":586.959,"end_s":592.32,"text":"Thanks, Luke. All right, guys. So, you've seen the results and you've seen","speaker":null,"is_sponsor":0},{"start_s":590.56,"end_s":596.36,"text":"the pricing. Links as always in the video description for pricing. So, let","speaker":null,"is_sponsor":0},{"start_s":594.24,"end_s":599.519,"text":"us know what you think. For non-overclockers, as I'm sure you'll","speaker":null,"is_sponsor":0},{"start_s":598.24,"end_s":608.48,"text":"have noticed from the other reviews around the web, GTX 980 is a total tank.","speaker":null,"is_sponsor":0},{"start_s":604.399,"end_s":611.519,"text":"But when overclocking the GTX 780 Ti, we","speaker":null,"is_sponsor":0},{"start_s":608.48,"end_s":614.079,"text":"observed that this older GK 1110-based","speaker":null,"is_sponsor":0},{"start_s":611.519,"end_s":618.32,"text":"card becomes much more competitive if you're willing to sacrifice features and","speaker":null,"is_sponsor":0},{"start_s":616.24,"end_s":623.44,"text":"power consumption. So, let us know what you think. If you had over 500 bucks to","speaker":null,"is_sponsor":0},{"start_s":620.959,"end_s":627.68,"text":"spend on a graphics card, which one would get your money? Also, let us know","speaker":null,"is_sponsor":0},{"start_s":625.92,"end_s":631.839,"text":"in the comments what other coverage you'd like to see of this card. An","speaker":null,"is_sponsor":0},{"start_s":629.36,"end_s":635.92,"text":"overclocking guide, um, SLI performance scaling. Any other ideas? let us know.","speaker":null,"is_sponsor":0},{"start_s":634.48,"end_s":639.04,"text":"Thanks for watching, guys. Like this video if you liked it. Dislike it if you","speaker":null,"is_sponsor":0},{"start_s":637.76,"end_s":643.44,"text":"disliked it. You can also leave a comment if you have feelings that are just more complicated than this. Also,","speaker":null,"is_sponsor":0},{"start_s":642.399,"end_s":648.32,"text":"check out the link in the video description to support us if you want to buy a cool t-shirt, give us a monthly","speaker":null,"is_sponsor":0},{"start_s":646.88,"end_s":651.68,"text":"contribution, or change your Amazon bookmark to one with our affiliate code.","speaker":null,"is_sponsor":0},{"start_s":649.92,"end_s":654.8,"text":"So, when you buy your new graphics card, we'll get a small kickback as a, you","speaker":null,"is_sponsor":0},{"start_s":653.36,"end_s":658.24,"text":"know, thank you for providing us with these videos, Lionus. We sure do think","speaker":null,"is_sponsor":0},{"start_s":656.56,"end_s":662.32,"text":"you're awesome. Here's your affiliate code being used in Amazon. Helps us out","speaker":null,"is_sponsor":0},{"start_s":660.16,"end_s":684.01,"text":"a lot. Thanks again for watching. And as always, don't forget to subscribe.","speaker":null,"is_sponsor":0}],"full_text":"Another day, another new graphics card. Welcome to the Lionus Tech Tips performance overview of the GTX 980 from NVIDIA, their most advanced GPU yet. DKL's new Rip Jaws 4 series is their ultimate DDR4 memory with speeds of up to 3,333 MHz available. Click now to learn more. As usual, we'll start with the unboxing and physical overview. The Media Sample GTX 980s come packaged in this sweet box that has the card upright looking all majestic with its magnesium alloy fan housing, aluminum frame, and trivalent chromium plating with H. What's that? Sorry, you've seen this user configurable green LED lit GeForce logo before. NVIDIA hasn't really changed their cooler design since the 690. Ah, yes. No, I can see how you might think that, but you haven't seen the back yet. Check it out. Now, we get a full metal back plate with a classy looking design and surface. Now, the issue with back plates, and presumably the reason NVIDIA has never done a reference card with one before, is that it restricts the intake of a card installed in the next slot over in tightly packed SLI configurations. So NVIDIA made this strategic chunk of the back plate removable for users with lots of cards right next to each other in SLI. Not that you need to run SLI just to hook up enough monitors anymore. As we move over to the IO, yeah, that's some pretty good stuff right there. Three DisplayPort ports, so that's up to three G-Sync displays off of a single card. An HDMI 2.0 port. So yeah, that's 4K 60 Hz over HDMI. and a dual link DVI port can all be found here. Any four of which can be used at one time on this beast of a graphics card. Which leads us to the next notable thing physically. This is a flagship 80 class card, but like the GTX 680, it only has two six pin connectors. A hint at the incredible power efficiency of the Maxwell architecture at the core of the GM 204 chip that powers this card. The GTX 980 not only hung around at the top of our performance charts, but did it while consuming less power than GK 110based cards like the 780 and 780 Ti and while running extremely quietly. So, let's talk about Maxwell for a bit here. Without shrinking the manufacturing process, so they're still on the same 28nmter node that they used to build the Kepler based GTX 680, NVIDIA has effectively reduced TDP by 30 W, increased the number of CUDA cores by 33%, increased base clocks and memory speed by around 12% and doubled the number of render output units, as well as adding some other cool hardware bits to enable some really cool technology. Pretty amazing, actually. So, first up in the new technology list is their improved third generation delta color compression. A technique that allows data coming out of the core to be compressed by a factor of 2:1, 4:1, or even 8:1 before being committed to memory by storing the value of a change versus the previous frame rather than the value itself. This combined with improved caching allows the GTX 980, in spite of its mere 256-bit memory interface, to achieve up to around 30% better effective memory performance compared to the GTX 680. In this scene, the pink highlighted screen elements are taking advantage of this technology. Next up is a feature that's all about getting the most out of older games where you have lots of GPU horsepower to spare. DSR or dynamic super resolution is mostly just another name for a feature that has existed forever but is now easier to use. Super sampling anti-aliasing. The way it works is it renders the scene at a higher resolution. So you can choose 2x, 3x, or 4x in the control panel. Then intelligently downscales it with a filter whose sharpness or softness can be adjusted to let you use your extra GPU horsepower that otherwise was just sitting there to yield potentially a much more pleasing gaming experience. It's not the same as actually running on a 4K display, but it's much better than running native 1080p. The effect is especially noticeable on the grass in Dark Souls II here in this footage that I took at the NVIDIA event. The next feature is about getting more performance when we don't have enough, and it's called MFAA or frame anti-aliasing. That's short for multi-frame sampled anti-aliasing. Grow up. The way this works is instead of taking a large number of samples with a static pattern to smooth the edges of on-screen objects or textures, MFAA takes fewer samples per frame, half as many from what we've been told, but uses new hardware in the GTX 980 and 970 to alternate the positions with each frame. Then it applies a clever filter. So the performance impact is about equivalent to two times MSAA, but the resulting image looks pretty much identical to four times MSAA. This feature isn't enabled in the driver yet, so expect an update once we get a chance to play around with it some more, but the initial results are impressive. Check out the effect on the transparent textures like these wires in this Battlefield 4 demo that NVIDIA was showing off at the launch event. Another thing we won't be able to test for a while is the improvements that NVIDIA has been making to the VR experience on their cards. The most notable of which is called auto asynchronous warp. A cheat, I guess, that allows the head tracking of a VR headset like the Oculus Rift to sample its position again right before a frame is finished being drawn to reduce the player's perception of the delay between physical motion and on-screen motion. And this will be important for folks who have trouble with motion sickness in VR. Which leads us to the final feature, their VxGI global illumination technique that we also can't really test here because while Unreal Engine 4 is getting support, it'll be a while before we get real games with the tech, but it looks so freaking cool. NVIDIA has a couple of demos to show it off. One where they render a Cornell box, so you can see the way the direct light enters at the top, so the tops of the shapes are the brightest with realistic soft shadows and color bleeding from reflective surfaces. and one where they recreated as closely as possible the conditions of this iconic photo from the 1969 lunar landing to demonstrate that it's completely realistic and wasn't taken somewhere in Area 51 or whatever. Now, these two demos wouldn't be impressive at all except that they're being done in real time instead of using path tracing, a phenomenally computationally intensive global illumination technique that involves following light rays as they bounce around the scene. like all of them. NVIDIA is using dedicated hardware to split the scene into cubes called voxels that act as dynamic light sources that approximate the results of a more accurate rayraced rendering method but in real time. All right, so as cool as all that technology is, how does the card actually perform? That's what you want to know. Great question, Luke. Take it away. Before we get into things, I want to point out that overclocking all of our cards is yet again backed by popular demand. And because of that, you will be able to find the overclocking results and settings in the graphics card overclocking dock, which should be linked in the description below this video. Also, when reading the settings for the GTX 980 in the overclocking dock below, there's no reason to worry. There is a reason why I ignore the memory clock speed. The 980, much like the 780 Ti before it, features a fantastic technology called autobalancing power rails, which in its most basic description, allows the card to take extra power overhead from parts of the card that aren't using it and move it elsewhere, allowing you to focus on one specific thing, which in this case resulted in the GPU clock speed being cranked through the freaking roof, which is awesome because then you can access all of that FPS that you wouldn't otherwise be able to. Performance of the GTX 980 is honestly a little bit surprising in a good way. To give proper context for that statement, we have to look at the rumor mill. Personally, I like to avoid rumors, but sometimes I don't really have a choice. For instance, on the September 12th version of the WAN Show, my brother at Cold SE on Twitter, and I discussed how the 980 was, according to rumors, not supposed to beat out the 780 Ti. So, when I finally got the card in hand, I was delighted to see that more often than not, it was able to beat the 780 Ti in both 1080p and 4K benchmarks. Now, my overclock for the 980 feels a little bit high, running at a range of 1503 to 1492 MHz, but I really have no idea as I have no one else's overclocks to compare to yet. And to be fair, my 780 Ti overclock is very aggressive anyways. So, I haven't exactly done the 980 any favors in regards to the 780 Ti's clock speed. But don't forget, all of this performance awesomeness doesn't just stand on its own. Not only are you paying considerably less for this card than you would pay for a 780 Ti, but you also pay less in the future. Underload, the 980 has considerably less power draw than the 780 Ti, which is nice if you're actually paying for your power bill. And if you're not paying for your power bill, maybe you can use that little tidbit of information to convince your parents that it might be time for an upgrade. Anyways, thanks for watching my segment, guys. Back to you, Linus. Thanks, Luke. All right, guys. So, you've seen the results and you've seen the pricing. Links as always in the video description for pricing. So, let us know what you think. For non-overclockers, as I'm sure you'll have noticed from the other reviews around the web, GTX 980 is a total tank. But when overclocking the GTX 780 Ti, we observed that this older GK 1110-based card becomes much more competitive if you're willing to sacrifice features and power consumption. So, let us know what you think. If you had over 500 bucks to spend on a graphics card, which one would get your money? Also, let us know in the comments what other coverage you'd like to see of this card. An overclocking guide, um, SLI performance scaling. Any other ideas? let us know. Thanks for watching, guys. Like this video if you liked it. Dislike it if you disliked it. You can also leave a comment if you have feelings that are just more complicated than this. Also, check out the link in the video description to support us if you want to buy a cool t-shirt, give us a monthly contribution, or change your Amazon bookmark to one with our affiliate code. So, when you buy your new graphics card, we'll get a small kickback as a, you know, thank you for providing us with these videos, Lionus. We sure do think you're awesome. Here's your affiliate code being used in Amazon. Helps us out a lot. Thanks again for watching. And as always, don't forget to subscribe."}