WEBVTT 00:00:00.000 --> 00:00:04.720 When you're playing a video game you want to feel immersed. As much as possible, you want your 00:00:04.720 --> 00:00:09.440 character to feel like an extension of yourself. You want to see what they see, hear what they hear, 00:00:09.440 --> 00:00:13.440 feel what they feel. And when an enemy suddenly lunges out of nowhere, you want to let out just 00:00:13.440 --> 00:00:18.560 a little bit of pee. But isn't there a sense we're leaving out? Is it really enough that I can see 00:00:18.560 --> 00:00:26.880 here and feel Kratos? What if I could also smell Kratos? Taste him? Maybe not. That is exactly the 00:00:26.960 --> 00:00:32.240 kind of question that startup GameSend, makers of GameSend, is hoping to answer with a recently 00:00:32.240 --> 00:00:37.920 released GameSendSend system, which they claim automatically translates on-screen action into a 00:00:37.920 --> 00:00:44.320 number of scents. That number being six, and those scents being forest, gunfire, storm, explosion, 00:00:44.320 --> 00:00:49.760 racing cars, and a neutral pallet cleanser. They've also got a few upcoming scents like ocean, 00:00:49.760 --> 00:00:55.040 fresh-cut grass, sports arena, and blood. The perfect combination of smells to simulate Mario 00:00:56.000 --> 00:01:00.960 tennis. But is this the future of hyper-immersive video games, or just another flash in the pandemic? 00:01:00.960 --> 00:01:05.920 Well, let's look at the history of smellable media and see if past is prologue. 00:01:05.920 --> 00:01:10.320 A few theaters and cinemas experimented with adding scents to performances in the late 19th 00:01:10.320 --> 00:01:15.280 and early 20th century. It's hard to know exactly what audiences thought of this. Back then, 00:01:15.280 --> 00:01:20.800 they didn't have rotten tomatoes, so they just threw literal rotten tomatoes. In smaller venues, 00:01:20.800 --> 00:01:25.920 they might use a cloth soaked with scented oil and a fan to waft the desired smell across the 00:01:25.920 --> 00:01:30.160 audience. But in larger venues, they typically just pumped perfume in through the ventilation 00:01:30.160 --> 00:01:35.680 system or sprayed it down from the ceiling. Of course, this required a lot of perfume to pull 00:01:35.680 --> 00:01:41.040 off, and the smell tended to linger, sometimes for days afterward. That's a problem both because 00:01:41.040 --> 00:01:45.680 it's potentially expensive and because it means that it's hard to switch between scents without 00:01:45.760 --> 00:01:50.560 them winding up mixed and muddled, piling up on each other. Just imagine if you went to see 00:01:50.560 --> 00:01:54.320 Mamma Mia in the theater and it still smelled like the Fast and the Furious in there. 00:01:56.400 --> 00:02:00.960 Or you get home and your wife demands to know why you smell like Mulan Rouge. It was also easy for 00:02:00.960 --> 00:02:05.760 the film or play to have already moved on by the time the smell really hit you. It's also notable 00:02:05.760 --> 00:02:11.040 that these experiments were mostly done by venue owners rather than filmmakers, meaning that the 00:02:11.040 --> 00:02:16.720 audience, getting misted with perfume like a fancy vegetable, wasn't necessarily the experience 00:02:16.720 --> 00:02:21.040 intended by the creator of the content. One of the first filmmakers known to have considered 00:02:21.040 --> 00:02:27.760 including scents was Walt Disney as part of 1940's Fantasia, though this was dropped due to expense. 00:02:27.760 --> 00:02:33.040 Perhaps the most famous of these early experiments was Smelovision, originally called Centovision, 00:02:33.040 --> 00:02:39.040 which was demonstrated at the 1939 World Fair. Smelovision used a system of connected pipes, 00:02:39.040 --> 00:02:43.840 which all fed to individual theater seats, thereby allowing the amount and timing of the scent 00:02:43.840 --> 00:02:48.480 to be tightly controlled by the projectionist. Smelovision got a lot of excited press coverage, 00:02:48.480 --> 00:02:53.040 but the film industry, they just weren't interested, in part because it wasn't cheap for 00:02:53.040 --> 00:03:02.240 a theater to install, an estimated $25 to $30 per seat as of 1959, around $250 to $300 in real terms. 00:03:02.240 --> 00:03:08.720 The original Smelovision only wound up in one movie, 1960's Scent of Mystery, where the smells 00:03:08.720 --> 00:03:13.520 were designed to provide important hints about the film's plot. Unfortunately, the film was 00:03:13.520 --> 00:03:18.800 a massive stinker, making back only 15% of its budget. Also, while some of the smells were pretty 00:03:18.800 --> 00:03:24.800 accurate, others weren't, and still others were accurate, but too revolting to really be desirable. 00:03:24.800 --> 00:03:29.680 Modern day showings of Scent of Mystery typically use a system of scratch and sniff cards instead, 00:03:29.680 --> 00:03:34.000 which is far cheaper, but more of a distraction than an integrated part of the experience. 00:03:34.000 --> 00:03:37.920 You can also still experience something like Smelovision at select theaters that offer 00:03:37.920 --> 00:03:43.680 4DX experiences, which also involve haptic vibrations, wind, water, fog, and occasionally 00:03:43.680 --> 00:03:48.560 bubbles. Various researchers and entrepreneurs have played with the idea of a home version 00:03:48.560 --> 00:03:53.440 of Smelable Media, but outside of scratch and sniff concepts, it largely hasn't been economically 00:03:53.440 --> 00:03:59.440 viable enough to even go to market. There was that one VRSense system developer kit that was 00:03:59.440 --> 00:04:03.360 covered by our sister channel Linus Tech Tips, but it's difficult to tell how much progress 00:04:03.360 --> 00:04:07.680 they've made towards an actual consumer product in the time since we covered it. 00:04:07.680 --> 00:04:12.960 But if Smelovision, the concept, not the 20th century product, has never ever worked on a broad 00:04:12.960 --> 00:04:19.280 scale, why does anyone think it could work now? How are modern video games a better choice for 00:04:19.280 --> 00:04:24.880 Smelable Media than mid-century cinema? Let's take that basic smelly idea and apply it to video 00:04:24.880 --> 00:04:30.320 games. Can products like Gamecent actually overcome the pitfalls we've discussed? Can Smelovision 00:04:30.320 --> 00:04:35.680 perhaps gain a toehold NVIDIA games where it's been unsuccessful in television? The baseline bar 00:04:35.680 --> 00:04:41.280 that a modern Smelable Media needs to surpass is technical confidence and ease of use. First, 00:04:41.280 --> 00:04:46.000 it needs to provide the player with a clear, recognizable scent in sync with the action on 00:04:46.000 --> 00:04:50.640 the screen. This is pretty easily accomplished through a small module placed on your desk or 00:04:50.640 --> 00:04:55.200 nearby tripod. You could also probably use a mask of some kind to pump the scent directly into your 00:04:55.200 --> 00:04:59.680 face holes, but that's probably where we begin to see a clear trade-off in terms of comfort, 00:04:59.680 --> 00:05:04.560 which might negate the immersion benefit. It's very likely that, similar to early VR, 00:05:04.640 --> 00:05:10.400 smelly video games will only see limited consumer uptake and consequently limited developer support, 00:05:10.400 --> 00:05:15.200 which can be a self-reinforcing cycle. Gamecent, though, proposes getting around the need to have 00:05:15.200 --> 00:05:20.720 individual developers add smell support by using advanced AI to simply process what's 00:05:20.720 --> 00:05:26.400 happening in the game and generate appropriate scents accordingly. Finally, AI to the rescue. 00:05:26.400 --> 00:05:31.040 Due to the tendency for scents to linger, because there are literally molecules hanging in the air 00:05:31.120 --> 00:05:35.440 and going up your nose, video game scents would probably work best as a subtle, 00:05:35.440 --> 00:05:39.760 ambient background smell like grass when you're standing in a virtual meadow. 00:05:39.760 --> 00:05:43.920 There might even be some value in playing a VR racing game and actually being able to smell 00:05:43.920 --> 00:05:48.800 fresh asphalt with a hint of burnt rubber so long as it's relatively subtle. In a tight 00:05:48.800 --> 00:05:53.680 narrative experience, like a good film like we talked about before, it's all cut relatively 00:05:53.680 --> 00:05:57.840 fast with scenes typically lasting at most a few minutes and jumping quickly between 00:05:57.840 --> 00:06:03.520 times and locations. NVIDIA games, however, you're usually following the action in real time, 00:06:03.520 --> 00:06:07.840 allowing for a gradual transition between smells that feels more natural. 00:06:07.840 --> 00:06:12.160 Another issue to overcome is simply the range of smells available in current video game systems, 00:06:12.160 --> 00:06:18.640 which includes both Gamecent and predecessor Asylia. Unlike with color, there isn't a small set of 00:06:18.640 --> 00:06:24.160 primary smells that can be combined to create every other smell. It doesn't work like that. 00:06:24.160 --> 00:06:29.120 You're going to need more than six cents to capture the range of most gaming experiences. 00:06:29.120 --> 00:06:33.440 Plus, while our scent imitating abilities have definitely gotten more sophisticated since the 00:06:33.440 --> 00:06:39.360 1960s, there's still a lot of stuff we don't know how to fake well. Smellable media done well 00:06:39.360 --> 00:06:45.280 also requires easy customizability because while many players probably don't want to smell blood 00:06:45.280 --> 00:06:51.200 or gasoline, others would probably appreciate the viscerality of these arguably unpleasant smells. 00:06:51.200 --> 00:06:55.680 Still, it's pretty hard to imagine a large constituency of Resident Evil players who want 00:06:55.680 --> 00:07:00.800 to smell rotten flesh and sewage when they're fighting a zombie in an alleyway. Turn that crap down. 00:07:01.920 --> 00:07:07.600 But probably the biggest hurdle is simply economic viability. It's worth mentioning that 00:07:07.600 --> 00:07:12.720 several large VR experiences at places like theme parks have had olfactory effects as far 00:07:12.720 --> 00:07:18.000 back as the 1990s. It's just that those effects haven't been able to make the same jump to viable 00:07:18.000 --> 00:07:23.520 at-home use as VR headsets. Now, we all know that many gamers are very willing to spend a lot of 00:07:23.520 --> 00:07:28.640 dosh on fancy niche peripherals, but those peripherals still typically have to provide decent value 00:07:28.640 --> 00:07:33.360 for the money. Likewise, scent cartridges will need to be relatively long-lasting and relatively 00:07:33.360 --> 00:07:38.000 cheap to replace, or once this thing loses its novelty, it's just going to be a stinky paper 00:07:38.000 --> 00:07:43.600 weed. It's also hard to see the ability to smell a game being a better immersion game than the addition 00:07:43.600 --> 00:07:49.920 of a VR helmet, so it's very likely to lag far behind VR in adoption even after if it starts to 00:07:49.920 --> 00:07:54.720 gain traction. Still, there's an argument to be made that smell-o-vision, for all that it's been a 00:07:54.720 --> 00:08:00.160 punchline for the last 80 years, was just a good idea before its time. Even the latest iteration, 00:08:00.160 --> 00:08:05.840 Game Sense, is limited in that it's a third party trying its best to procedurally match sense to a 00:08:05.840 --> 00:08:10.720 scene made by other creators, just like those theater owners we mentioned before. In an ideal 00:08:11.200 --> 00:08:14.960 world, there would be a standard toolbox of sense that creators could work with to craft 00:08:14.960 --> 00:08:19.520 a scented experience that matches their creative vision for the content, regardless of which 00:08:19.520 --> 00:08:25.680 smelly box does the stinking. Though, if YouTube supported it, I'd probably turn off autoplay. 00:08:25.680 --> 00:08:32.320 So thanks for watching guys, if you liked this video, hit like, hit subscribe and hit up our other video where we discuss pneumatic tubes, the history of those, 00:08:32.320 --> 00:08:35.360 and why we don't really use them that widely, but they're cool.