# Generations of Wi-Fi Have Led to This ## Метаданные - **Канал:** Data Slayer - **YouTube:** https://www.youtube.com/watch?v=-soMNhNqEVc ## Содержание ### [0:00](https://www.youtube.com/watch?v=-soMNhNqEVc) Segment 1 (00:00 - 05:00) This is a new Wi-Fi chip that anyone can buy. It lets you take an off-the-shelf open-source router, plug in a simple thumb drive, and suddenly extend the range by about 10 times miles instead of feet. No special licenses, no exotic antennas, and it's not slow. This new version with improved data rates can push about 40 megabits per second, enough for voice, video, maps, and messaging. But speed isn't the point. The point is what this unlocks. Offgrid mesh networks that actually work. Routers that don't need the internet to be useful, and an entire category of tools that just wasn't feasible until now. Every new generation of Wi-Fi promises the same thing, more throughput. What they don't advertise is what you lose, namely range and how well the signal survives walls and distance. Notice how no one puts range on the box anymore. Because if I have to sit next to my router to get great Wi-Fi, shouldn't I just be using a wired connection anyway? Wi-Fi broke the gigabit barrier more than 10 years ago. So for most people, Wi-Fi hasn't been the limiting factor for internet speed in a long time. They continue to innovate in that direction, but it's not going to show up on your performance benchmarks since your connection is actually limited by your ISP. But this new Wi-Fi chip doesn't chase speed, which is what makes it so interesting. Instead, it chases reach. And once Wi-Fi starts going further again, an entirely new category of technology becomes possible. So, here's what we're actually trying to do. There are three stories running at the same time. The obvious one is the technical question. Can a normal consumer router using a tiny new Wi-Fi chip actually communicate at real distance? just off-the-shelf routers pushed way beyond what Wi-Fi is known for. But distance by itself isn't the point. The point is what becomes possible once distance stops being the limiting factor. Mesh networks that don't need the internet. Communication that still works when the grid doesn't. Tools that don't disappear the moment something goes wrong. That's the unlock. And here's the part most people never see. While big companies chase faster streaming and better ads, there's a growing group of builders doing something very different. They're experimenting with off-grid wireless solar nodes, long range mesh, local networks that don't touch the cloud at all. Not because it's trendy, but because they see the value of unstoppable networks managed by the people using decentralized, inexpensive tech that just about anyone can own. So, the real test here isn't speed. The real test is whether this thing can do something never before possible. And if so, it's not just another Wi-Fi upgrade. It's proof that this quiet off-grid approach is finally becoming accessible. And if it doesn't, then I'll happily categorize this as indie tech and just let you be. So, let's see if it's ready. Now, it's easy to get lost in the specs, throughput, multiband, IPX ratings, and in the process, it's easy to forget something very important, which is that connectivity isn't guaranteed for everyone. And I was reminded of that stark fact this week when I saw a post from a community member in our Discord. They're working on a real deployment funded by E, the same organization that literally writes the Wi-Fi standards. And the goal is a long-term mesh network for the Navajo farming community. This is a place that doesn't get fiber, doesn't get dense cell coverage, and frankly doesn't get treated like a priority. But communication still matters. Weather, equipment, safety, coordination across land. And helping these underconnected communities is something I care about. But that raises a lot of questions. Is this actually fast enough to be useful? Isn't it just slower Wi-Fi? Is it legal? Why haven't companies like Ubiquiti touched it? And if this is so promising, why hasn't it already changed everything? Because if this doesn't hold up, then none of the bigger ideas matter. So instead of speculating, I decided to test it with off-the-shelf gear that anyone can buy. Now, when I built that improvised militarystyle router a while back, what surprised me wasn't just the fact that you could beam internet connection over two meshtastic antennas. It was actually how many people were interested in repurposing that tech for their situation. This video is really a continuation of that project. And if this chip is going to enable a new category of tools, it has to work in the real world through walls, over distance without internet. So, we're going to build it, break it, and push it. Let's see what this actually unlocks. All right. And just for a tiny bit of context here, because I think it's useful, there has been a lot of development around Wi-Fi Halo. And most ### [5:00](https://www.youtube.com/watch?v=-soMNhNqEVc&t=300s) Segment 2 (05:00 - 10:00) of the boards and uh carrier boards, the hats that use those Halo cards are using the predecessor to this new chip, which is the MM6108. So things like the seed studio hat that we use, this sillex board, this alpha chip, this Heltech uh PCIe mini PCIe card, and even this Asia RF, they're all using that older Morris micro card, which is fine, but we're not talking about the MM6108. We're talking about the MM8108. And there's a number of enhancements that come along with this. The first one is just an increase in the data rate, which is pretty exciting. So, the new chip bumps up the maximum data rate from 32 megabits per second to 43 megabits per second, which is actually pretty substantial. This card is not about bandwidth, but it's still exciting that they are innovating along those lines. And I want to do a quick test in the real world to compare the old one with the new one as it relates to data rates. Okay. And then just as kind of a quick and dirty example of the bandwidth here, I have two Haven nodes and one is connected to my upstream router. So technically they have internet. Um, this one is only connected to the Meshgate router over Wi-Fi Halo. So what I'm going to do is I'm going to connect to this Haven's 2. 4 Wi-Fi on my phone, 2. 4 GHz Wi-Fi, and just run a Wi-Fi man speed test and see what kind of real world download and upload speeds that we actually get. And then we'll run the same test with the new chip and just see if it varies. So we're running on the 8 MHz channel which is kind of the widest channel in terms of meshtastic. It would be spreading factors kind of the analogous um fe uh configuration. And so we're trying to get as much data as we can. Not quite maximizing range uh but more maximizing data. So, I'm already connected to this guy's hotspot, and all I'm going to do here is go over to Wifi Man and uh run a speed test. Okay, so we got 6. 5 megabits per second down and 21 megabits per second up. Usually what I see um on a good day is 15 up, 15 down, but I've really never seen more than that with this chipset. So now we'll try uh the new MM8108. Okay, so now we have basically the same configuration, but instead of the Haven nodes, we have the Gleet routers running the new Morse Micro MM8108 chips. same antenna configuration, uh, same channel width, 8 MHz, and I'm going to just run the same Wi-Fi man test. I'm connected to this guy, the Glee point, not the Glee gate. So, their connection, the internet connection is only going through Halo. And so, I'm going to go ahead and run that test and see what we get. So, I've seen this pretty consistently. Um, these new chips seem to be getting about 19 megabits per second down. It says 47 megabits per second up. These tests aren't perfect, although I'll take that. So, really, uh, when it comes to internet speeds, um, over Halo, uh, this new chip, uh, really outperforms the old one. So, um I'm pretty impressed with that. It's just all the more reason that we're going to want to use this new chip uh in the next iteration of Haven. Okay, so we've shown that the new chip uh is more powerful in terms of bandwidth. But the next upgrade, which I'm super excited about, has to do with regionality. So to date, all the stuff that I've been talking about with Haven and open source mayonnaise has been relegated to uh really the United States and Australia because they use the same band for this sort of activity. Um the old chip, the MM6108 can technically support other regions like the EU, UK, Japan, etc. But it required hardware changes. So, the carrier boards that house the chip would have to implement things like uh PAS and uh saw filters and things like that. And I've I ordered all of them, frankly, and I tried to get them working with OpenWRT, and I just I really couldn't. The closest I got was with this Asia RF uh EU variant. I was able to get the radios to show up on um OpenWRT and look uh as if they were receiving a signal, but I was never able to get them to connect to each other. So, um, I didn't feel comfortable kind of endorsing that, but that all changes now. Um, with this new chip, it has a 26 dBm PA that allows regionality to be informed by the software. So, basically, it can support all the major regions. And when you first turn on the uh the router and run through the setup, it's super simple. ### [10:00](https://www.youtube.com/watch?v=-soMNhNqEVc&t=600s) Segment 3 (10:00 - 15:00) You're able to just select the region and uh start configuring. So, I'm going to run you through that real quick because the real question is uh with the EU, they are more rigid about um things like airtime and duty cycles and stuff like that. So, the question is like is Halo even worth it in some of these other regions and I did some tests to try to answer that question. But before we answer that question, let me run you through how easy it is to set up. So, we're going to take our fresh Gleet router with our Halo chip plugged in. it is uh connected to this antenna and we will run you through the setup of that. So when you're first setting this up, you always connect the uh the LAN Ethernet port to your computer so that you can directly log into the Lucy terminal. So that is what I'm going to do here. All right, let me connect that. Okay, there it is. And then we're going to grab the router IP address. Okay. And then there's not going to be any password on the initial setup. And so all you have to do here for the initial setup is select your region. And you can see all the regions that are supported. So we can see how easy this is. Um let's do EU. So I'm going to do EU. I'm going to call this the GLE point. And then I'll just give it a password. Go ahead and get this configured. All right. We're going to set this as the client. So, one of the kind of differences is um some of these other regions don't have all the bells and whistles. I don't think this has 802. 11s, which is mesh networking. Um but you can still do clients and access points and all that. And I think probably mesh networking will come at some point. All right. And then we should be good for that node. Now, I'm going to just do the same exact thing for our other um node here. And now this is going to be the node that has an internet connection. So just keep that in mind. Okay, we're connected again. Okay. And now we're going to do access point instead of client. Same SSID for Halo. And you can see for the EU um you only have the one and two MHz channel widths, which is fine cuz that's those give you the maximum range. And then let's just set this up. Ethernet zero will be WAN. Set it up as a router. That's what we want. Okay. 2. 4. So, we're going to do gleate Good to keep these um legible. And then gleate 5. And then so what we should now be able to do is we're going to plug WAN into the router because this gu is going to have an uplink because it's the gate mesh point will stand alone. That's fine. And then let's see if we can get these talking over um the EU bands here. So one thing is these are both 915 antennas. So it'll still work but it's uh the EU doesn't use 915. So I'll show you guys what it uses. So, let's go to the Glegate interface here and we'll see we'll be able to see exactly what band it's using. So, let's go over to network. Let's go to wireless. Again, the these GleNet routers have 2. 4 5 GHz in Morse Micro, right? So, we're going to do edit on Morse Micro so we can look at the stats. So, transmit power is 15 dBm. We're on channel six. Looks like the channel is automatically selected. Um, but you can see the frequency is 866. So, we know we're doing EU stuff. Not to mention, you can see the country code here and all that. Let's run the exact same speed test using this. Now, keep in mind, we're on 2 MHz, right? So, 2 MHz is not going to give us um the same kind of speed, but I do want to show that it does in fact work and just how easy it is. So, let me go to So, I'm going to connect to this guy. All right. And let's just do a quick internet test, internet speed test at that 2 MHz channel. So, we can see we're actually getting a very healthy bandwidth. Um, several megabits per second, which was actually what we saw, the same thing we saw for the old chip on its biggest channel, the 8 MHz channel. So, for the EU region on 2 MHz channel width, we got four down and 12 up. I'm pretty happy with that. Okay, so we proved that the bandwidth's pretty solid at those lower channel widths for the EU setup, but like what kind of range can we get, right? Because um we had overclocked on the Haven the transmit power to 27 dBm, which is right under the legal limit. For these EU um uh configurations on this chip, we're only doing 15 dBm. Now, for US it would be higher, but for EU it's set at 15 ### [15:00](https://www.youtube.com/watch?v=-soMNhNqEVc&t=900s) Segment 4 (15:00 - 20:00) uh, which is not as powerful, but I was actually surprised when I ran some other tests. And you can kind of always make up for transmit power with other kind of things like, um, antenna placement and antenna gain. So, let's take a look at the signal real quick. So, the signal between these two antennas, even though they're not 8, these aren't EU antennas technically, um, the signal is solid right now. The noise floor is at93, which is low. and the signal is at -9. So the sort of um margin there is is really solid. So but let's do some range stuff now. Let's see how far these can actually go or if they're just not powerful enough to be practical. All right. So in order to really test the signal, we're going to have to move one of these nodes. So I'm going to power it with a battery pack here. And what else am I going to do? I'm going to just um see how far I can go and still get a signal. Okay. So, my goal right now is to figure out uh what kind of signal we get on this guy as I move away from the bass node uh which is in my house. Right now, our signal is a negative 11 dB which is incredibly strong, more than enough to do anything that we are going to want to do. But I want to see what happens when we move away from my house. Now, it is raining out. So, I think I'm going to go in my car. All right. So, I have this situated in my car here. Let's go drive away and see what happens. So, I lost a little bit of the footage, but the long and short of it was with the EU frequency when I walked down the street in my neighborhood, I continued to get a signal between 500 and 1,000 ft, which was very similar to what I was able to achieve with the US band. So, this surprised me because the EU uh setup only operates with the 15 dBm transmit power whereas the US version has a much higher transmit power. So, um I was very impressed with that. So, I did take the router outside and did a little range test and I was getting a signal at about 500 ft. My setup here is pretty bad. I'm inside a concrete house, which is really not how you want to do it. But the point is, I was getting like comparable range to what I've seen with the Haven nodes, which is surprising and interesting. So, I'm going to give EU a kind of green light that it works. Um, and it's not all that dissimilar from the fact that like other bands like 2. 4 and 5 GHz have different sort of um you know uh rules and regulations in other regions as well. So, it follows suit that Halo would have those same sort of um deviations. But I do think that um although it's definitely not as um capable um and not as supported as it is in the US and some other regions, uh it is definitely still usable um in the EU as far as I can tell. So, one of the other things that excites me about these radios is their low power draw. And I remember when I was talking to Simon from Muzy Works, he really emphasized that Laura in particular um has an order of magnitude lower power draw than some of its peers like 2. 4 or 5 GHz Wi-Fi, which again opens up a whole new world of possibilities when it comes to remote solar, wind, turbine, or even just batteries that last a long time and just like other modalities of deploying nodes. So Laura has a power drop of about 0. 1 watts. Halo has about one watt. 2. 4 has about four watts and 5 GHz 10 watts. So those wider bands 2. 4 and five, you're probably not even going to try any sort of like solar or remote node. Uh but it's totally possible uh with Halo and Laura. And there's a lot of cool ways that you could deploy nodes um because uh there's such a lower power requirement to get them to go. And one of the things I'm going to do here, I have this hooked up, is I'm going to turn off the radios and show how it affects the wattage. Now, you have sleep and idle and active. So, you have different sort of modes for each radio, but I do kind of want to show how um you see the needle move when I turn off the wider spectrum radios, but not so much with Halo because it's just it's uh hardly noticeable. So, I'm going to go over to the UI um on this Gleegate router and I am going to uh toggle off the radio. So, I'll turn them off using the software and we'll see what comes out the other side. So, I'm just going to go over to network. Going to go to wireless. Again, we have three radios. Uh, Halo 2. 4 and five. So, let's start by turning off the 5 GHz radio. So, right now, the average power draw I'm seeing about 28. It's going between 2. 7 and 3. 2. 7 and ### [20:00](https://www.youtube.com/watch?v=-soMNhNqEVc&t=1200s) Segment 5 (20:00 - 25:00) three. So, now I'm going to disable the 5 GHz radio. Okay. And now we can see the wattage is down to 2. 5 2. 6. So you can see it has gone down a bit by disabling the 5 GHz radio. So now we're going to do the same thing for the 2. 4 GHz radio. Actually I'm using the 2. 4 to connect to that. So let's actually just do So I'm actually going to disable the Morse micro radio and we'll see if that wattage drops down materially in any way. So let's go over to Morse Micro. Again, it's about 2. 5 2. 6. Now, I'm disa disabling the Morse micro radio. Okay, the Morse micro radio is disabled. And you see no difference still at 2. 5 2. 6 because it's just it's so less demanding on um in terms of power that it's barely noticeable. So there's just so many cool things you could do in terms of getting these uh configured on like MCUs and just other low power draw units. Um that makes it pretty exciting. And so that low power draw is making a whole new set of uh carrier boards possible. For instance, this board right here. So this probably looks like an MCU. And if you think that, you would be wrong. This actually runs Linux. This has OpenWRT on it. It has a Halo chip, has a processor, a little bit of RAM, and it has an SMA connection for your antenna. And so, if I were to take this and pair it with like a lipo battery like this, I mean, you could put this on a buoy, potentially like a kite, a tree. I mean, there's really all different sorts of interesting places that you can mount these nodes. Um, and so that's super exciting to me that you don't need um kind of high energy, high power to get these things to go. Quick note, if you want to set up a node yourself, I have a guide on how to source, configure, and set up your very own Haven node. It's in the description below and it walks you through all the pieces of doing that. And real quick, if you're wondering how Halo dovetales with Laura, they're pretty complimentary, but they're very different. So Laura is its own modulation technique. In my opinion, they're good to run in parallel um as a form of redundancy. Based on what I've seen, Halo is great because it's real internet. It's full IP networking under the hood and there's some really cool stuff you can do with OpenWRT, layer 2 meshing, stuff like that. But Laura is still the king of range. It's just that modulation technique that they use. Yes, it's very low data, but if Halo stopped working, for instance, I mean, Laura is still there to kind of give you GPS coordinates, maybe text messages and stuff like that. So, in my opinion, if you can get both on your node, that's the sweet spot. But again, they run entirely in parallel. Laura does not run on OpenWRT. I guess it could, but no one really does that. It runs on it typically runs on its own MCU. Um, technically, you could get a Laura hat to go on Linux, but uh I'm not sure about the feasibility of that. So, usually the MCU just hooks into the device, and since it's so low power draw, um it doesn't really have any negative repercussions. And actually I have devices that do that. Like for instance this right here, right? So the blue part is an OpenWRT uh Linux board, but mounted on top is an MCU uh from Rack Wireless with a Laura chip. And so they literally run right next to each other. And it's just great for redundancy. If something happens with Wi-Fi Halo, you can still fall back to Laura. So, I think they work well together and ultimately are complimentary to um the entire rig. Okay. And so, for the final test, what we're going to do is run the new MM8108 chip on the Gleet routers using the US frequency and we're going to do FaceTime entirely offline and see how far we can go while still retaining a solid connection. So, we can see cellular is off on both of the devices. Okay, took a second there, but now we are facetiming and we're doing it just through uh Wi-Fi direct over these Wi-Fi Halo nodes. But now the question is how far can I go? So, I'm going to determine that right now. So, right now I have FaceTime on my iPhone. My iPhone is connected to this router right here over its 2. 4 for Wi-Fi radio, but Amelia is connected to her router, same router, and those two routers are only connected over this Wi-Fi Halo chip right here. Um, and this is a complete offline uh isolated network. So, we're doing FaceTime over Wi-Fi Direct offline. ### [25:00](https://www.youtube.com/watch?v=-soMNhNqEVc&t=1500s) Segment 6 (25:00 - 26:00) And you can still hear me, right? — Yes. — Nice. I'm pretty far out right now. You can still hear me? — I'm probably I don't know at least like a third of a mile away. Um and we're using we're just using FaceTime on my phone. No special settings. But what FaceTime doesn't know is that the connection is being driven by this Halo chip right there. — Yes. — That's crazy. All right, I'm going to come back. I think I proved the point here. So, — okay. — And you might be wondering, why even bother building stuff like this? And honestly, it's because communication shouldn't come with a monthly bill. You should be able to tinker with your gear, reuse it, fix it, not throw it away just because a company moved on. And the code running inside your house should be open enough that you actually know what it's doing, where your data is going, and why. I just don't think the future should depend on someone else's platform or a policy change that makes what you're doing today suddenly not okay tomorrow. This stuff is about building tech that actually serves communities, not optimizing for another quarterly earnings call. And ultimately, at the end of the day, like most things in life, we do it because we can. For more killer tech, click --- *Источник: https://ekstraktznaniy.ru/video/29814*