Brad & Will Made a Tech Pod.

326: Quantumly Entangled Keyboard Switches

69 min
Feb 15, 20262 months ago
Listen to Episode
Summary

Brad and Will discuss the evolution of magnetic switch technology in keyboards and mice, exploring how Hall Effect and TMR sensors replace traditional mechanical contacts. They examine real-world implementations like the ASUS ROG Falchata keyboard and Logitech G Pro X2 mouse, discussing benefits like configurable actuation points and drawbacks like software dependency.

Insights
  • Magnetic switches eliminate physical contact wear in input devices by using software-configurable actuation points instead of mechanical contact closure
  • Hall Effect and TMR sensors enable analog input detection, allowing per-key customization of actuation height and release point that was impossible with mechanical switches
  • Esports-focused features like Snap Tap/Speed Tap on magnetic keyboards provide competitive advantages but are being actively detected and banned by game publishers
  • Console manufacturers have not adopted Hall Effect sticks despite proven reliability benefits, likely due to cost-benefit calculations favoring warranty fulfillment over hardware improvements
  • The shift from hardware to software control creates long-term dependency risks if manufacturers discontinue support or web-based configuration tools become unavailable
Trends
Magnetic switch adoption expanding from gaming peripherals to mainstream keyboards and mice as manufacturing costs decreaseEsports competitive scene driving rapid innovation in input device technology, with professional players demanding sub-millisecond response timesThird-party controller manufacturers (8BitDo) outcompeting first-party console makers on both price and reliability through Hall Effect adoptionWeb-based configuration interfaces replacing proprietary software for input device customization, improving cross-platform compatibilitySoftware-configurable hardware enabling feature detection and bans in competitive gaming, creating new enforcement challenges for game publishersHaptic feedback replacing mechanical actuation in premium input devices, though current implementations still lack the tactile precision of mechanical switchesIncreasing complexity in input device firmware requiring ARM processors and sophisticated signal processing algorithmsGrowing market for aftermarket switch replacements and customization in gaming mice, creating new revenue opportunities beyond initial hardware sales
Topics
Companies
ASUS
Manufactures the ROG Falchata split keyboard with magnetic switches and web-based configuration interface
Logitech
Released the G Pro X2 mouse with magnetic switches and haptic feedback technology
Razer
Produces magnetic keyboards with Snap Tap feature that enables competitive advantages in esports games
Wooting
Manufactures magnetic keyboards with Speed Tap feature, competing in esports peripheral market
Nintendo
Switch Joy-Con controllers suffer from joystick drift, driving adoption of Hall Effect sensors in third-party alterna...
Sony
DualSense and DualShock 4 controllers experience stick drift issues, creating market for Hall Effect replacements
Keychron
Produces mechanical keyboards with QMK firmware and web-based configuration tools
8BitDo
Third-party controller manufacturer using Hall Effect and TMR sticks, outcompeting first-party console controllers
Cherry
Historical mechanical switch manufacturer whose patent expiration enabled proliferation of compatible switch designs
Corsair
Mentioned in context of mechanical keyboard switch market and customization options
DJI
Manufactures camera drones used in Olympic broadcast coverage with advanced 4K tracking capabilities
GoPro
Produces compact cameras used in drone applications for sports broadcasting
Ring
Mentioned for controversial data sharing practices with law enforcement
Intel
Manufactures ME (Management Engine) firmware requiring Windows-based updates on some systems
Mellanox
Produces 40 gigabit network cards that experienced driver issues during Windows updates
People
Nathan Edwards
Friend of the show who helped select silent U4 Boba White switches for Brad's Control Drop keyboard
Quotes
"I love it when technology's fun. Yeah, that's nice as opposed to the alternatives, which are many."
Brad and WillOpening
"The future is here. Unevenly distributed."
WillDrone discussion
"I can't believe I'm gonna say this but one of the favorite ones is actually on LinkedIn it's like a sudoku chess puzzle basically"
WillGaming discussion
"Has Windows gotten so shitty that it is now capable of eating another Windows? Is Windows now practicing auto cannibalism?"
BradTechnical issues segment
"I turned it on on this keyboard I fired up counter strike I immediately was able to stop and get like something I've never been able to do before I was able to stop and get the perfect headshots with an ak like tap tap tap no problem it was wild"
WillSnap Tap feature discussion
Full Transcript
Brad, I love it when technology's fun. Yeah, that's nice as opposed to the alternatives, which are many. I mean, look, there's a lot of alternatives. You know, when ring camera is like, hey, man, we can turn over your footage to help people find lost dogs and also help ice. That's bad. But, you know, sometimes technology is good. I was watching the Olympics. You know, I like to see how things change every four years. And, you know, the Winter Olympics are always good because it's, you know, it's the death sport Olympics, not the like, you know, who can go faster and jump higher Olympics. Going faster and jump high and jumping higher can also be a death sport. I mean, for what it's worth, but fair. But one of the things that's happened over the last like two or three Olympics is is that drones have gotten real good. So, you know, we used to cover drones over tested and I spent a fair amount of time with them over the years. and they started out as a thing that people would use instead of like boom shots so they use them like instead of crane shots and stuff like that that were relatively static i want you to watch this video i just sent over it is footage of people barreling down the mountain at 100 kilometers an hour in the alpine downhill the the death sportiest of snow skiing races i think and uh they have little micro drones that have really good 4k cameras the video quality on this particular one is bad when you watch it on the broadcast it looks fabulous of the drones just tracking the racers as they're going 100 kilometers an hour down the hill and it's kind of incredible they're using them all over the place i was watching the luge yesterday and the drone operators are flying the drones down the luge track just behind the racers so you get like an over the shoulder third person video game view of people going the luge track at 100 miles an hour do you think this is definitely human operated it's not it's not like some kind of computer vision motion tracking image tracking thing so i don't know for sure like that's definitely a technology that exists you can buy drones that will be like throw up in the air when you're going to ski and they'll track you on the way down the mountain um i i kind of assume that's the case for the lose track watching the mistakes that happen on the downhill i think it's probably not because i think they pick up the they they catch them on the way over the ridge when the drone would take a minute to acquire them uh and i think they're at least human operated for part of it sure i believe but it's wild and it's very cool and it made me happy and it's been fun because like last time they they did they weren't doing this and the footage from the drones looked crappy and like the drones advancing the camera's advancing it's all good man i want to see a drone for the drone i want to what you want a drone operator yeah i mean i'm kind of like yeah i mean sort of take or leave the winter sports stuff but i would absolutely love to see the drone in action i want to say like like how micro is micro here when we're talking about a micro drone. So I'm from what I've seen in the other shots because when they're doing the head on shots you sometimes catch the drone. It looks like they're 250 millimeter drones. That's a wow. Well no no it's like they're like the size of a shoebox basically. Yeah but that's still way I thought we might be talking like the size of a dog. I don't know why a dog is my size reference here but you know like something the size of like a coffee table or something like that's way smaller than I expected. They're definitely using like micro cameras like deep like gopro style size cameras or dgi style cameras um they're definitely not big giant octo quads you can hear them in their they sound like little mosquitoes which is usually mean smaller propellers going faster but yeah the the these are the ones the my hunch and i i haven't found a tech article explaining how they're doing it yet because drones are boring now and people don't write about that anymore um i haven't found an article talking about the drone pilots and the drones. But my hunch is that they're the like the racing class that we used to go to like industrial sites in South City to watch people zoop through like pipes and broken windows and shit like that in abandoned warehouses. Yeah, I didn't watch the Super Bowl this year, but we talked about it on the Ramblecast this week. And I meant to ask Vinny and Alex and did not think to if there were any kind of obvious drone shots at that. They had drone shots in the halftime show. They definitely do some drone shots of like the outside of the stadium and stuff like that. I didn't notice they use that spider cam in all the major stadiums. I mean, that's the thing that's on like a tether, though, right? Like a zip line. Well, it's on like four. It's tethered to corners to the cart to the like the corners of the of the field, basically at the top of the stadium. Still physically connected to something, though. Yeah. Yeah. As opposed to flying, because I mean, flying independently, flying cameras, buzzing around at a sports game is like that's not even some future shit. That's like some Star Wars shit. Yeah, kind of. Right. Like, and yet it's just here. The only difference is that we are using typical rotor. What's the term? No repulsor lifts, not whatever the hell a repulsor lift is. So we're not quite there yet, but it's like effectively the same thing. But yeah, it's funny because they're not filtering out the noise of the drones from the broadcast, which I was surprised by. Because that would be pretty easy to pull out, I think. Yeah, sort of. But no, you just hear the drones buzzing, which is kind of how you know that a drone shot's coming. It's nice. That's sad. The future is here. Unevenly distributed. Sometimes it's cool. Welcome to Brad and Will Made a Tech Pod. I'm Will. I am Brad. Brad, it's early in the morning. I am a sleepy boy. As it often is around these parts. By these parts, I mean this podcast. It's true. This is a morning podcast. I've always wanted to have an evening podcast. I've never had an evening podcast before. No, thank you. I am out of words by the evening. I've often thought it would be fun. I used to love doing the late night PUBG streams. and watching how weird everybody gets at about 11 o'clock at night. Oh, sure. Like a stream, a video game stream, totally fine. You can get sloppy. I don't mean sloppy in an inebriated sense, but you can just kind of be a mess on a video game stream and it's fine. I think doing like an Art Bell style call-in show, like take Art Bell and Car Talk and match them up into computer help. Yeah. And I would like to do that show late at night. That might be fine. I don't know. When people are having their computer problems. Because computer problems, I find, almost always come at night. Yeah. this is yes yes you're not wrong this this is a pretty technical show though i kind of i like i like the mental sharpness and clarity of the early morning recording as opposed to the end of very end of day or evening reporting that's true i so it's funny you say that though because i do there's a couple of like daily games that i play every day um one of i can't believe i'm gonna say this but one of the favorite ones is actually on linkedin it's like a sudoku chess puzzle basically and they time you and tell you how you do versus your average and versus people on your leaderboard i always do better if i do it at like 11 o'clock at night than at 7 o'clock in the morning weird always weird yeah i'm in bed before that at this point yeah you see we're both insomniacs at this point but we're on opposite ends because i'm going to i can't go to sleep and you wake up early i think my brand of insomnia is i fall asleep in five minutes or less but then can't sleep more than like six hours at the outside is a staying asleep problem i've been reading so many books in a way that i haven't really since the start of the pandemic so that's nice at least yes okay real quick before we have a topic this week but before we get into it i have a question for you speaking of computer problems on late night yeah and this is also a question that ultimately i'm going to put to our audience because i could use some advice potentially do you think it's possible Has Windows gotten so shitty that it is now capable of eating another Windows? I don't know what that means, Brad. Okay, let me expand. Is Windows now practicing auto cannibalism? Did you put a Windows inside another Windows? Let me just set the stage. I've talked about this before. I have a second Windows install on a USB drive that is just kind of a junk install that I use for things I don't want to touch my real install. Okay. Like, for example, the Logitech G-Hub. Yeah. Like shit that you know is going to hook deep into windows that you don't want to like gum everything up. I feel like that's less of a problem in the 21st century than it was 20 years ago. Yeah, it's maybe not as bad as I don't know. You know what? A lot of that type of vendor software is still pretty crappy. So remember, I did the testing on this last year. We did a video about a PC world where I took a year old Windows install and then wiped it and then benchmarked it versus the busted window and the busted old Windows install. And the performance was exactly the same. Yeah. No performance degradation. Yeah. it's just jank in crustacean and i also use that install for like you know sketchy game mods that i might not trust or like installing installing 25 year old windows games that may not be well behaved fair okay in terms of overwriting system files or whatever so you're kind of using it like a vm but it's bare metal because you're doing stuff that you require bare metal access to because like i would use g hub to update the firmware on my mouse for example even though i have absolutely no need of any of the other g hub functionality anyway i fired up that windows install last night to do some stuff and it's way out of date and I was like I should get it on a modern Windows I also use it for I have to update my NAS machine like there are some things like you can only update the Intel ME in Windows you may know which is a huge pain in the ass although actually there might be a Linux tool for that now I'm not sure actually yeah the Linux firmware app will do that but also you can do that if you on ASUS boards and certain other boards you can just upload the zip file that has the ME and the BIOS now and it'll do them both at the same time. So some boards do that. Some do not. Some do not. That workstation board I've got in there. Unfortunately, the BIOS update instructions are like, please update to Intel ME version blah, blah, blah before updating this. But yeah, it sucks. Anyway, if I could get that, if I could get the ME update working in Linux, I might actually just toss this drive and never use it again. But anyway, I was in that dummy install booted off this USB drive. I was like, I'll just let it Windows update to the 24 H2 or whatever it wants to do right now. Mm hmm. So it's, you know, on a patched secure supported version. Windows walked away for a while, came back. The update had failed. I let it reboot. It kind of unapplied the update for 20 minutes and then it was back to 23 H2, whatever. I went to bed. I got up this morning. The Mellanox card in this machine does not work anymore. Have you completely powered it all the way down? This machine? Like drained the power and everything? Like unplugged the power supply and hit the power button? Yeah, not turning off the power supply. I bet it got the firmware into a weird state. So it was on the card? Yeah. Those cards have a lot of logic on them. I don't know about that. I had not considered that, but I don't know if that sounds likely to me anyway. I can't believe I just gave you, have you considered turning it off as advice? That might be worth a shot. Basically, I looked in event, like none of my network mounts over that. That's the 40 gigabit card I use for that direct link to my NAS. None of the network mounts worked this morning, and I looked in event viewer, and that card was just Windows was trying to initialize that card and failing over and over in a five second loop infinitely so my guess is you have to rebuild the external windows drive as a um using the windows to go thing in order to get it to work and i haven't been able to get it to work i haven't been able to get windows to go to work with anything past 24 h2 interesting okay i had to use rufus the drive imaging tool that has an option for creating a portable like an external windows install uh here's a rub that i didn't mention, I believe this may be possible that at some point in that update process, you know, a big Windows update updates or reboots like three, four times, right? Yeah. My refind is set up to give me the boot menu. And if I don't hit any keys, boot into my regular windows after 10 seconds. At one point in the middle of all those reboots of the external install, I wasn't around and it booted into regular windows instead. So then, yeah, did it hose the regular windows too? Could that update process have touched the regular Windows install in any way? I don't think so. I don't think so either. I also don't think so. The way it hands stuff off during the upgrades and installs from Sleep and Fastboot and all that. I assume you have Fastboot and all that stuff turned off in both Windowses and in the BIOS. Oh, I don't know if Fastboot's turned off on the portable Windows, actually. If it's not, that'll do it every time. Ooh, I didn't think about that. Huh. Well, rebooting is supposed to always be a reboot. It's just shutdown that's not a reboot. That's not a shutdown. Right. But yeah, that's... I guess report back next week. Let's let us know how it goes. Yes. Broadly, the question is kind of updating a second Windows on a machine with an existing first Windows fuck up the first Windows. Maybe I bet that the Windows upgrade got your machine, got your card in a weird state and it hasn't been able to recover and power cycling will do it. Well, I'd have to remove the board and take it back out. This is like this has been multiple instances of me uninstalling it in device manager and letting it know I mean physically remove it from the machine. I've had problems before that solved when I pulled the CPU and we put the CPU in the socket. Like it's weird. Power lingers. Yeah. Yeah. But you know what doesn't linger? What's that? Magnetic switches. Wait. No, they don't. Yeah, that actually that works. Yeah, no, I'm gonna stick with that one. That's what we're here to talk about today. Yeah. There's been a little bit of a, it's been a magnetic switches have had a moment. They've been around for a really long time, but we're starting to see them in more places and keyboards and mice and stuff like that. And we thought it was a good time to talk about it. How they work, what they are, what kind of technology happens inside them, and then what it's like to actually use them a little bit on a couple of the newer devices that use them. So you've been putting hands on a fair number of magnetic switches from the sound of things. I have pause on keyboards and mice in the last few weeks. So, yeah, lots of stuff to talk about. Yeah. And these have been out there in, I feel like, joysticks, thumbsticks, and controllers were the first place I really started hearing about these. Hall effect and so forth. You may have heard the phrase, like, that's been around for several years now. Well, Hall effect has been around. Well, Hall effect is an ancient, like, ancient theory. I don't mean in concept. I mean, like you've been seeing them show up in actual consumer products in a way that is being promoted and is visible for a few years now. Yeah, certainly. Honestly, a lot of this for most people is downstream of joystick drift on things like the switch Joy-Cons. That's it. So the main difference, first we should talk about how regular switches work. So like a keyboard switch or a mouse switch typically is an enclosure that contains a couple pieces of metal. and when you push down on the switch part of the of the device it buckles one of the pieces of the metal until it touches the other one and depending on how that mechanism is constructed they can have just a bump they can make a noise they can be silent and just kind of push all the way down um there's a bazillion ways to make those and the the proliferation of keyboard switches after the cherry patent were ran out has has you know there are you can literally get hundreds and hundreds of different times of keyboard switches these days yeah that's just just important here it is a it is a mechanical action that is producing a physical electrical circuit yeah and a physical physical contact between the two places and that buckling that happens when you press the switch down tends to cause um uh mechanical stress on the on the on a piece on a physical item which will cause it to break over time wait when you say buckling wait yeah the metal's not bending yeah is it every actuation? Every activation. Wow. Yeah, because you're pushing a thing down that makes the metal bend until it touches the other piece of metal. Interesting. Wait, as a design? As an intentional design choice? Not as just a side effect of metal touching metal with stress on it? No, yeah. Interesting. I didn't realize that. Anytime the metal bends, it puts stress on it. Interesting. So magnetic switches use a variety of different techniques or effects. to do the same thing, but with no physical electrical contacts. So like a keyboard switch that's a magnetic keyboard switch will have the same basic shape mechanism, mainly so that you can use the same kind of keycaps and they fit in the same places and they're the right stability for a keyboard. But instead of having two pieces of metal that touch each other and close a circuit, a magnet just moves up and down on the post that moves up and down through the switch. Now, this is great. There's no physical contacts to wear out the mechanism, either in the mechanism of the switch or in the space between the switch and the keyboard. Right. You know, when you have those those those hot swappable switches, you jam the little copper wires in and they can bend or wear out or get corroded or whatever. The other benefit of this is that a lot of the stuff that was previously tied to hardware, like the actuation point for the switch, is software configurable. So you can control stuff like bounce back and actuation height and when the switch closes like when the mechanism stops reporting a signal when you lifting your finger back up off of a keyboard press for example Sorry I may have lost the thread here a little bit How does it gauge height? It's not just a binary on, off, yes, no. Well, we'll talk about that. Yes, no, it has crossed the boundary. No, it has not. Yeah, no, no. So the boundary is a software configurable thing. The way the sensors work, both Hall Effect and TMR, we'll talk about what both of those are and how they work more in a minute. but yeah they have analog signals so like in a magnetic keyboard if the software is set up right is essentially an analog device that you're putting thresholds on for when the press is trigger so and they do that so I mean let's just jump to it with hall effect sensors there's a little transducer in the in the sensor the thing that's mounted on the keyboard so the sensors typically are on the keyboard the magnets are typically on the switches because the magnets are moving and a hall effect sensor has a transducer that's in the integrated circuit and the changes in magnetic flux around it change the amount of current that flows through that transducer can you define what a transducer is it's like a little antenna basically for all intents and purposes it's just a conductor that's very very thin so it it reacts like the the current that flows through it changes the hall effect it says that in magnetic fields changing flux will change the amount of current that moves through a conductor. That's how it's measuring variable gradation then and how hard you're pressing. Yeah. So the amount of voltage that goes through it changes based on where the magnet is in the field. And so inherently it doesn't know that. Like often people use Hall effect sensors and things like RPM meters for industrial equipment or like bicycle rotation counters that give you a speedometer for your bicycle. Sometimes we'll put a hall effect sensor up on like your back fender and a magnet on the rim of the one of the rims of the bike and then you tell it how big your tires are and then it knows how far you're going and how fast you're going based on how many times that magnet hits the hits the hall effect sensor i want to say we use them on the early maker bots when we were building those to determine that the top and the bottom of the of the print uh of the print area right um so anyway uh yeah but the important thing is analog signals and you get a range and you can calibrate them based on positioning uh there's a lot of math there especially in noisy environments like a keyboard where there's a million of these switches and a million little magnets right next to each other this was my biggest question about this thing and i typed it into this doc until i saw you had addressed it elsewhere and then remove that question but yes it's a situation where there are upwards of 104 or 110 depending on your keyboard layout several dozen to a hundred different tiny magnets all very tightly packed and adjacent to each other so is there not an issue with like bleeding magnetic fields from one to the next like signal kind of crosstalk how does how does it correct for all of that well so there is but but because of the way um magnets work the field the field when the keys are all up is always the same right and then when you press down on the on a switch it's the same when you press down on the switch and the neighboring switch, that's a measurable change in the flux for both of the switches around it. And the magnets in the switches are really, really small. So the fields, their impact on the fields around them is relatively limited, as I understand. That was my guess was either either either or both of the fields are tiny and it's not a huge problem or they correct for it in terms of signal strength. Well, but also, like if you change switches in your keyboard or even I assume change switches from one spot to another, then you have to actually recalibrate the keyboard. which I haven't done yet. I assume that's just pressing a bunch of keys. Interesting. Yeah. It's just a lot of quick brown fox type stuff for a while until it's happy. I don't know. You want to try it? Sure. What could go wrong? I didn't think to do that. So anyway, but the point is because it's analog, you can control things like actuation height and like when the key press lets off, like when you're lifting your finger back up. So the activation height and the release height don't have to be the same in one of these mechanical keyboards. That makes sense. Yes. It's on at a higher height than it lets off of. That's what you're saying. Exactly. Or vice versa. I guess you could set it either way. So when I do a calibration, I'm calibrating my E switch right now. I press the selected key on the physical keyboard all the way down. And then I press the start button and it just says, oh, it's calibrated. so there's nothing it's just looking for what which sensors detect flux change or what the sensor underneath it detects when it changes all right well we'll get into more of the hands-on toward the end yeah for the specific keyboard you're testing so okay so there's there's also this other thing that you hear about in the in the game in the joystick joycon drift uh world uh which are tmr sensors that's short for tunnel magneto resistance yeah the first place i saw this was um i think it's like eight bit dues last couple of controllers yeah their their ultimate bluetooth twos i think are both tmr sticks they're starting to move to tmr instead of hall and they they definitely but of course they do kind of talk it up as these are better than hall this is the next generation of this thing i don't know how accurate that is but so that's that's accurate that's fair um this is a newer effect that's been just that was discovered in the 70s versus the 18th century 19th century i guess 1800s um and uh they use less power they measure the change in resistance of a thin insulator between two ferro magnets on an integrated circuit so imagine two magnets with something that insulates between them and the tunneling tmr effect the tunneling magneto resistance effect says that in a magnetic flux field electrons will tunnel across that insulator and into the other from one magnet to another. And that's a measurable effect because you can measure the resistance of that, um, of that, of that insulator. Yeah. Speaking of measurable effects, I'm sure you don't have numbers in front of you, but I wonder if the, the savings and electricity required to drive these is meaningful across, again, somewhere between 60 and a hundred switches on, on a, especially if it's a wired keyboard, it doesn't matter, but like I use a wireless keyboard and I'm wondering if, I wonder if that many switches makes a meaningful difference to battery life to have TMR overhaul. From what I can tell, it matters much more on small integrated circuits that are being used on smaller batteries. The batteries in mice and keyboards are relatively large. Yeah, fair. So the bigger thing is that they're faster and more precise than the Hall effect sensors generally. faster more precise and less power is a pretty uh compelling combo although and they're quantum so they use a quantum effect the tunneling can't be described in classical physics it's only a quantum effect so usually quantum mechanics wait so this would use tmr sticks so these are quantumly entangled keyboard switches i don't know how entangled they are but the it was very clear the articles i read about this all described this as quantum effects and i went and looked at it and i dug in further i was like oh right electrons jumping yeah that is quantum oh Electro okay that's not entanglement if it's One if it's one particle moving as opposed To two different particles and influencing each Other at a bizarre distance then that's not Yeah ask me anything about Quantum mechanics um But but so the tmr Effect is the same thing that's been in modern Uh hard drive heads for like 15 Years at this point uh the ones that Float above the surface of the disc on a cushion Of air um so Yeah the switches measure the chain the physical Setup for these switches is the same the sensors on the Board the magnets on the switch uh the switch measures the change in resistance in that insulator which is faster and like i said uses less power and more precise when you say faster are we talking on the order of nanoseconds i don't know microseconds i couldn't find that answer i looked for that and i think it's just i think it's i think it's not on a human scale faster right was my read probably milliseconds versus microseconds or milliseconds versus nanoseconds something like that i mean at the same time you're seeing esports pros now talk about wanting 600 hertz monitors and things like that or you straight up see people talking like they can't be competitive at lower than 480 hertz or some insane yeah i want to say there was a counter-strike tournament that was sponsored by intel and everybody's really raw because they put mid-range cpus into it yeah so the machines were only getting 180 200 frames a second yeah um now the the problem with the tmr sticks is the sensors are relatively new the ic sensors are relatively new that means they're more expensive So when I was looking at like replacement sticks for a DualSense, for example, which you can buy on Amazon, four pack of four Hall Effect sensors was $12. Sorry, I have a four pack right here, but it's a two pack of TMR sticks was 20 bucks. So they're significantly more expensive. That's what, $3 versus 10 a stick? Yeah, basically. I was going to ask something else. So, oh, this is mostly relevant in the context of a dual sense. Are those sticks and the circuit board integrated together? Or is that just the stick? Because on a dual sense, you have to solder if you don't have that whole contraption, I believe. So this was to desolder. These were units that would desolder the old sticks and re-solder on the new sticks. But you'd replace the stems and the caps from your board, I think. That's pretty substantial then. as opposed to joy cons that just use easily inserted ribbon cables like the stick replacement on joy cons is pretty dead simple as these things go but yeah so so we should talk about why um like in the old days thumbsticks and joysticks used potentiometers which are like sensors that measured the bend of a material inside the change the resistance based on the bend of a physical material usually there were a bunch of other ways that they could do that but that was the more common one um modern thumbsticks in like the switch and the dual sense have two touch surfaces they have a mechanism in there that has uh that that pivots on both axes and it moves these little they're like little virtual thumbs basically that move on a little touch surface back and forth on a linear way on the x and y axes and those sent those those pushers push down on these touch surfaces and that's what detects the movement those those touch surfaces because hey this is weird but touch surfaces are very inexpensive to make now yeah weird guess what else they also wear out as they grind against each other over time well so they do wear out with this grind against each other time in the switches case it was more often that dust in particular got into there and got stuck on that touch surface and would interfere with the with the touching part of it and then the software couldn't cope with that once it got beyond a certain level of dirtiness yeah that's speaking of i don't know if this is the time to say this speaking of software coping with physical effects physical anomalies yeah i mean i think let's let's um the fact that the shorting out was a problem was what started this whole conversation about magnetic switches i think in a lot of things but yeah so let's talk about software fixing other problems yeah like my first question about these when we were putting these notes together is do magnetic switches fix debouncing which is a phenomenon I have become acquainted with on this podcast over the last couple months as my as my as my relatively inexpensive but decent key cron started registering a whole bunch of extra inputs yeah and then I as I talked about recently I updated the firmware on it thinking that would not do anything and then in fact it did everything in the sense that they tweaked the debouncing algorithm in this thing so the debouncing the bouncing happens because the switches aren't the connections aren't perfect right yeah the the short version so i looked it up and from what i read the short version is that yes these do basically eliminate the bouncing effect but yes when you when you are slamming two pieces of metal together with a decent amount of force i assume it's everything from the amount of force you're applying to like tiny variations in the shape of the metal causes and and it's just the impact it's just the impact yeah i was gonna say on the hard impacts especially especially if you're a hard typer the switches can vibrate on and off of each other a little bit even the contacts and enough for you know current to flow and not flow multiple times per key press and so that's what bouncing is and that's what software and more often firmware is correcting for yeah typically i looked on the firmware for my like control drop and when you compile it yourself rather than using one of the pre-compiled binaries that they ship then you can actually change the debounce settings in the firmware the the newer firmware i put on this key chron also exposed a debounce setting and i think it's at like five milliseconds from by default right now, which has basically fixed everything. Although it did lead me to ask, since we're talking about this situation, it did make me wonder where is the debouncing happening? Obviously it's happening in firmware because we're talking about that and we can see the setting exposed for it, but it also made me wonder if the software receiving the input is doing any debouncing. Probably not. Whether OS. Also, what I looked up is that yes, in most cases on operating systems, they have to account for a wide range of input devices, not just highly precise gamer keyboards. So probably most generic USB keyboard drivers and things like Windows and Linux are doing some amount of smoothing for that stuff. And the reason this came to mind is that I've been in my UEFI a lot lately and I was seeing a lot more like double arrow key presses in the UEFI than I would ever see in Windows. Is this after the update? Yeah, this is in the last like week that I've been in there. I would be arrowing up and down through the menus and it would jump a little bit here and there. Can you increase that debounce more or is that a configurable option? Yeah, I can I can tweak it more. But the point is, it seems like maybe the operating system does a little bit of its own smoothing on top of whatever's happening in hardware or firmware. I mean, it's entirely possible. I mean, they remember like Windows has to account for everything from like terrible scissor keys on laptops to like dome keys to all sorts of other stuff. So, you know, there's going to be a generic keyboard driver that you're going through that's going to handle some of that stuff. Yeah. So anyway, there's other stuff that's a side effect of these switches too, right? Like Razer and Wooting both have magnetic keyboards that use features that are banned from use in Counter-Strike and Overwatch and games like that. Razer has a thing called Snap Tap. I think Wooting calls it something Tappy or Tappy Tappy Tappy or something like that. And the idea is that when you flip that on, it only prioritizes your last hit input. So it releases the previous hits, which doesn't matter in a lot of games. But in a game like Counter-Strike, where you have movement inertia and your gun accuracy is directly goes way, way bad the moment you are gets better the moment you stop moving. Counter-Strike pros do a thing where they do cross inputs or opposite inputs called snapping. It's some simultaneous natural opposing or something. The idea is if you're pressing D to strafe right and you want to stop, then you don't just let off D. You also press A for a split second as you're letting off D. And it's something that getting the timing on is one of the things that separates Counter-Strike mortals from Counter-Strike gods, right? Yes. Also, as with so much of what we talk about here, it all goes back to Quake. It does. Because remember the movement in Quake 1 and the amount of momentum you had when you let off the movement? Well, and when you did diagonals, yeah, of course. So, anyway, I can see why this might be a problem in the professional scene. Yeah, so the feature that they added basically prioritizes your last used input, which means it removes any overlap between the key presses, which is one of the things you have to learn how to deal with on any new keyboard when you're playing Counter-Strike at a high level. and it only replaces replace reports rather the last key that you press even while another key is still pressed so if you're strapping right yeah you can effectively just interrupt your movement by hitting the opposing direction exactly um and i'm gonna say i turn it on on this keyboard i fired up counter strike i immediately was able to stop and get like something i've never been able to do before i was able to stop and get the perfect headshots with an ak like tap tap tap no problem it was wild um and that was after about five minutes of practice so i understand why they've said hey don't use these uh it seems like they're able to detect it in software so they are issuing bans on on using this kind of stuff so don't do it seems like like they're kicking you from games it's not like a back ban okay basically you get a message that says hey don't do that yeah that's fine yeah like just just preventing you from playing the game with the device seems fine if it hit at the account level that would maybe be a little much yeah and just to be clear this is the thing you have to turn on in the firmware in the software for the for the keyboard like i had to log into the software interface and flip it on they call it speed tap in the asus software i wonder if they're using like kernel level detection features because this that seems like the kind of thing that would not be hard to spoof otherwise you could like my guess is they're tracking timing they're all kind of yeah maybe maybe that's what it is also wait wait but then but then you would start false positive catching people who are too good at this well Well my guess is that you don do it in a precise like my guess is that the timing is always the same every time when you use the feature And when a human does it it varies by a couple milliseconds I don know All I know is that, yeah, don't use these features or you're going to get yelled at. It really, the games that I've seen that it referenced are for different reasons. Overwatch, you can do it to completely change your movement and make you move like you have some sort of weird powers or cheating, actually, when you watch the videos of people using it in overwatch they literally look like they're cheating um it apparently doesn't matter in valorant because the movement inertia isn't the same in valorant you stop almost immediately when you let off um and yeah my understanding is that it's detected by heuristics not by some sort of kernel level anti-cheat or something like that um so and and just to be clear you can use the keyboard you just have to turn the feature off that's that's the oh yeah and per their rules you can still there's as long as you've got that off yeah it must be a detection thing then if they're if yeah if you can still play with that feature disabled but on that keyboard there's nothing fundamentally illegal about magnetic switches it's the feature that lets you turn off the other thing they do is they change the input at which the the key stops reporting input but that's a little bit less of an issue um so okay i think that's everything. People use these sensors in like all sorts of Hall Effect sensors are everywhere. Your fridge door likely has a Hall Effect sensor. If it doesn't have a physical switch that turns the light on and off that you can see, if the light just turns on and off when it opens and turns off when it closes, that's almost certainly a magnet embedded in the door someplace. You can test it by getting a big magnet and running it around your fridge and seeing what happens. But the place that we're mainly talking about are keyboards and mice. I haven't actually used, I've used Hall Effect thumbsticks before but not for any appreciable length of time all of my game pads that i use are still more or less stock but i'm kind of curious about swapping out the tmr sticks i mean the for some tmr sticks and hall effect sticks i did read a couple of comparisons uh like there was a person who did the the tmr switch and uh did basically put tmr sticks in one of his joysticks in one of his dual senses and hall effects sticks in one of his dual senses and then reported the differences and the tldr was they both seemed fine yeah yeah i have to it sounds like the differences would be imperceptible to human but like the the less electricity and stuff is meaningful yeah and i think like this is one of those places where human perception is a spectrum and like there's definitely going to be people out there that can tell the difference but for the most part probably they're functionally identical. It is so fucking frustrating to me that none of the console makers have embraced at least Hall Effect. Yeah. It sounds if it's really that ubiquitous and commodified like yes it probably would. I mean of course it would add a little bit of manufacturing costs but as bad as stick drift particularly was on the Switch 1 and Sony's last couple of generations of controllers have also I don't know how Xbox has been doing I had drift on some of my 360 controllers but. So my 360 controllers the problem i had with them wasn't drift it was the dead zone got wobbly so that so it would kind of jitter in the even if no you weren't touching the sticks i actually haven't any problems with dual senses um i haven't either but it's something you see a fair amount i've never had a dualshock 4 or a dual sense i don't think developed drift but but i do that's why the stick replacements are fairly common yeah i assumed but it's not as bad as joy cons those are absolutely the worst. And like, I assume Nintendo just made the old fight club car crash recall versus settlement calculation. I think because, you know, because you remember they started like a life. I think it was the policy was lifetime on the switch one, right? They would replace joy cons with drift lifetime of the unit, not like one year warranty style. And I assume it's just the standard thing of, Hey, it's still cheaper for us to just fulfill our maze on the people who will bother to RMA than it is than it is to take a i don't know what it is somewhere between like 50 cent to three dollar extra cost per unit produced kind of thing it's uh yeah i mean it seems like on the switch two they sealed them better at least so that you don't get as much dust and crap in there but my guess is that the hall effect sensors are significantly more expensive than the touchpad sensors it's a bummer though it look yeah it's i mean it's nice that the aftermarket controller like this is why you should buy an 8-bit dough controller instead of that 80 pro controller um so okay hands-on time uh i've been testing the rog falcata which is a split 75 percent uh gaming keyboard it uses uh the specific kinds of switches like this the so one of the things that's important just like in mechanical keyboard land you have to match your switch with your keyboard most everybody uses the same standards for switches at this point there are a handful of mechanical switches that are outside of like the cherry mx three or five supporter snap-in plastic pieces and two connectors when you say standard are you talking about physical form factor that would make them interchangeable in the same keyboard yeah okay so you can just kind of buy any brand of switch more or less there's like cherry amex compatible switches and then there's some other stuff like um but but but we don't the other stuff is relatively small um with this you need to make sure that the keyboards match your switch because the magnet placement matters like the asus the asus keyboard and a couple others use magnets on the side of the of the capsule and uh some of them use magnets on the center of the stem like they're they're all it's all over the place so like if you have a wounding keyboard you make sure you buy the wounding magnetic switches if you have an asus keyboard you make sure the asus one if you have the razor one you buy the razor one um switch selection is dramatically lower than um with mechanical keyboards because for obvious reasons yeah uh the most everybody is shipping what are effectively linear switches so no bump no click for kind of obvious reasons you know because the click is a mechanical process and the switch actuates at at a configurable height so like if you while you can move the actuation height you'll never be able to move where that click happens so you don't probably want to buy clicky switches for this until you figure out what height you want and then maybe you find a clicky switch that aligns with the height that you like i don't know um you can get clicky magnetic switches for other keyboards i haven't actually tried them. I don't think I would want that, but I'm also a linear keyboard switch guy. I like linears. I've, so I've, I've, I've actually, do people mix and match? Sometimes people put different switches on their, like the WASD and game keys. I got this keyboard as a tactile because that's all they have in stock. I thought I'd never, I'd never used one. So I had to guess at what I thought I would want, but I guessed that I would want, um, linear, but they only attacked all at the time. So I bought the tactile keyboard and later bought some linear switches. But by then I have kind of decided that I liked tactile enough that I sort of mixed and matched. I think all my letter keys are linear currently and everything else is still a tactile. Yeah, I like linear just because of noise. I like people to type a little bit while I'm on the microphone here. Yeah. But yeah, look, if I didn't do podcasts and didn't use my computer with the microphone on, I would have the clickiest, loudest, most obnoxious fucking switches you've ever heard. Yeah, yeah. You know, there's a lot of things about one's computing habits that flow downstream of having to be on a microphone a lot. yeah give me give me a big thonky like i want something to sound like a model 80 you know just thonk thonk thonk um so yeah uh the switches the the there's right now two switches on the market for this keyboard so i haven't tried the other ones the they're they're they weren't in stock when i tried to get them what kind of pricing are we talking about on these basically the same yeah i looked up i i look at keyron just because that's what i have not that they're like the best or anything but they a full set of like 110 or whatever was like 30 bucks for their all of line. So these are more in line with like a premium switch. When I looked at their let's see what is it called? The TTC golden magnetic switches. They were like $10 for $10 I think. 10 switches. For 10 switches which is like basically the same price I paid 10 years ago for my silent bobas on my control. Now here's the bigger question. What is wearing out in the switches that necessitates more switches? what do you mean like i mean on you know on the the standard mechanical switches it's pretty obvious why they would wear out over time yeah like what is what is wearing down in the in the magnetic switches that nothing well that's what i mean then why what are the scenarios in which you would need replacements is what i'm asking oh if you want like a stronger spring or a weaker spring you want less travel distance or more travel distance you want um oh so we're talking People buying switches that behave differently and not just switches to replace the old worn out switches. Yeah, 100%. Yeah, that's what I meant. It's a personal preference, not a – like the things to wear out are the physical, mechanical interface of the plastic rubbing on the other plastic. And since they're usually pre-lubed, like they should last a really, really long time. Right. I mean, I guess the magnet can fall out. I don't know. It seems like the magnet's pretty embedded in there, though. Yeah. So the weird thing about the switches, it's like the switch industry in shambles here. Like this is the, this is like the, the razor blade seller suddenly not being able to sell more razor blades. Right. I mean, kind of the keyboards are a little bit more expensive generally than the, than a comparative comparable magnetic or mechanical keyboard. Um, I think my, my assumption is that we'll see more preference come up and people will start making more like tactile and, and clicky switches. That makes sense. Um, the switches are really weird though, man because you open you when you look at the bottom of them there's no there's no conductors there's no like contacts that go into the into the keyboard it's just like three or five plastic pins which is bizarre to me it's all invisible it's all invisible you can't see electrons uh the the keycap interface is the standard cross though so you can use more or less whatever key caps you want on it um it's louder than my control drop but that that is with the help of friend of the show nathan edwards i picked out some silent u4 boba whites for that years ago and they have been consistently the quietest key mechanical switches i've ever used there there are i believe there are quieter ones now i don't think they make the the silent u4 bobas anymore um but uh like it's a little funkier it's a little clickier i mainly when you bottom out the switches now the weird thing about this keyboard is you don't have to bottom out the switches at all um like like you literally you can just kind of graze your fingers across the top as if it's almost like playing a harp is what it feels like well that's configurable right you could it's configurable yeah to make you have to do that or close to it if you wanted yeah you can so there's both a this one has a knob on the side that lets you manually adjust the activation height in software that's nice oh Wait, you still have to go into software? The knob just doesn't just do it on the fly? No, the knob just does it. There's like a series of leads on the side, and you twist the knob, and it goes up and down. So you can make it hair trigger or bottom out, depending on which one you want. It also has a web interface. So you go to the Asus GearLink site, and it connects to the USB HID. I looked to see what this spec is called. This is a thing that started, the first time I saw it was with the Via Firmware, the Via QMK Firmware programmer years ago. yeah um but it's like i've used some fractal stuff there's a fractal headset that uses it now there's a fractal um uh like fan hub and uh rgb controller that uses it now a couple of of uh of asus devices use it like it's really nice to not have to have a bunch of uh software running on them like it's it's nice to not have to use g-hub yeah i think i think it's probably still a net good both for not making you install a shitty software stack and also for being cross-platform presuming the driver interface exists on other operating systems. You can do it on any of them. It is mildly concerning if, for example, that company were to go away one day or even if the servers were down, like not being able to update your keyboard because their web-based configuration tool is suddenly unavailable is you know. That sucks. There are downsides here. Yeah. You would hope if they were going out of business, they would like release some local option for continuing to be able to support your device, but who knows? Well, so you can also, like on this one at least, you can always use Armory Crate on Windows if you want. But once I added the UDEV rule that gave my browser access or my user access to the web, the USB device, I was able to use the web tool on Linux with Chrome, no problem. Yeah, so my Keychron has the same stuff. It uses QMK and they have a web-based tool as well. I had to install some kind of open is either keyboard or microcontroller driver stack. I forget what it's called. It's a microcontroller driver stack, probably for teensy or whatever's in those, whatever's in those keyboards. I just cannot remember the name of that open source project, but it supports like a bajillion keyboards. But I did have to install that to make the web interface talk to the keyboard. But yeah, the thing that I think, I don't know, I assume people know this. We talked about on the mechanical keyboard episode, but like a modern mechanical keyboard can have more compute than like a computer that you had in the 90s. Probably. Totally. there's like i'd have to look at the specs but i know what arm cpu is in this keyboard yeah there's a bunch there's a bunch of like homebrew projects that just use teensies which are like like relatively cheap microcontrollers that also have an insane amount of compute on them for what they are for something that costs 10 bucks that's that whole thing of compute getting small and and cheap and relatively powerful and that good yeah like just to be clear i don't think the math to do like it's not like it's not like a an old dome uh plastic dome cap keyboard where the math on that is relatively simple it's just closing circuits and measuring which circuits are closed and then sending the right signal to the motherboard like there's there's real math happening on interpreting the analog signals from from the magnets in the halls hall of x sensors and stuff and that enables things like extra profiles you know different yeah different layout configurations for different os's and stuff like that yeah so this keyboard has multiple profiles The interesting thing about actually using it is that depending on where you set the actuation and the dead zone. So this lets you set activation point at a millimeter distance from the bottom, from the top of the key, the unpressed point. And you can set a dead zone where nothing happens. And you can set it on both the top and the bottom. And you can do it per key. So like you can have different settings on, say, your ASDF keys or ASDW keys than you do on your other typing keys. Or you mentioned profiles. It has on-device profiles and you can switch between them. So you have different profiles for like hair trigger games and just using your computer for typing. I tried a bunch of different actuation points. And my initial thought was, oh, I want this to be very near the top. Because it turns out the input latency on a keyboard is mostly in the travel distance of the key, right? Sure. So if you're talking about, I want to have microsecond or millisecond accurate key presses, having to bottom out that key is a travel of a couple of millimeters, and that takes time. Now, this instantly brings me to the question of when are we going to start seeing switches with next to no travel on them? Well, so they do that. they do low profile switches right now yeah yeah well yes i mean this this is a low profile keyboard but they still have a fair amount of travel i'm talking like i'm talking like millimeters like when when are counter-strike pros going to get to the point where they want like millimeters of travel on their keys sub millimeter so so the key travel on these keys is three point uh it's about four millimeters i think 3.5 millimeters oh okay yeah that's that's like how long the stem travel is on most cherry style key switches oh okay um my guess is that the what you'll do is is set the keys to hit very near the top i found that almost impossible because i was kept triggering when i was just resting my fingers on the keyboard right um well the other thing there though is that you still get very and this is for again this is the counter-strike pro problem i would not care about this but then you still have a variable amount of you'd have to train your fingers to only press so far because you would be wasting time on travel you don't need to activate the switch well that is a hundred percent like that's that's a big part of using these when you send them to the hair trigger that's that's where that's where like just having a physical blocker in there to only have to depress the thing half a millimeter or whatever would come in i mean i guess that so it would be easy to take something and make a like a ring that goes around the stem that prevents it from going further than x amount um i don't know that anybody's done that yet but that would be wild to see i be curious about that i bet it coming i sure you not wrong um the did this make me better at playing games i don probably not i don feel like it yeah uh is it nice to be able to control the actuation height for the switches yes absolutely it's i find myself bottoming out bottoming out a lot less when i'm typing fast and my typing i do have an occasional false positive where i where i hit a key just a little tiny bit too hard and it goes um but since that that activation actuation death depth is configurable what i did was i started high and just kept moving it down until i stopped having a bunch of false positives like a tenth of a millimeter at a time and i ended up at like 1.5 1.8 um for the for for typing keys uh space bar i let bottom out yeah even if you never touch any of the configuration like the actuation or activation height i should say or whatever like this still seems like a net gain yeah when you factor in that you never probably never have to replace the switches like they don't give out the way the mechanicals do and so forth i think probably that the keyboard will last like by the time the keyboard is gross from being in close proximity to human i'll i'll be ready to do a new keyboard i just check it in the dishwasher i don't yeah yeah yeah don't don't don't do that people used to do that i know i know but i'm joking modern modern not serious this one has batteries in both sides yes yeah that would be a bad idea um a couple other weird features on this that are kind of neat uh you can you can because it's a split you can it has screw mounts on the bottom uh and you can tent it if you want you can tilt you can do a reverse tilt like the old comfort curve um i i quite like it because i have it in like the normal traditional keyboard format when i'm typing and then i'm sorry when i'm playing games and when i'm typing i can angle it so i get a natural angle on the on the keys to type um uh so generally pretty good that that um that switch on the side you can use to alternate between a bunch of different things it has like media controls control the the rapid trigger uh sensitivity uh which which that that's a whole other feature we didn't talk about but basically if you flip a switch on the front of the keyboard It gives you the ability to make the typematic key press much, much faster, if that makes sense. So it just keeps typing really, really fast. So anyway, the keyboard's nice. I was kind of blown away by the keyboard. I was not expecting to be super impressed by it. I also want to talk about the new Logitech mouse. It's the Logitech SuperStrike X2. It's a G Pro mouse. It's basically the same shape as my old G Pro Superlight. It's Superlight 2. It's an ambidextrous shape, but only with thumb buttons for right-handers. Five buttons, I guess. Two regular buttons, the wheel that's clickable, and then two side buttons. And it uses their light speed stuff for up to 8,000 refresh rates per second in Windows, unfortunately only. um but what they've done is they've taken they put magnetic switches on the two main switch presses um so your two main mouse buttons are magnetic switches now and they've added a haptic uh bouncer that's kind of like uh the thing in modern track pads that don't that aren't mechanical actually mechanically actuated now uh and it's a little bit it gives you a lot of the same benefits as the keyboard right you can control the actuation height for the switches so like I can set it so I just barely graze my finger on the mouse and it clicks. Yeah. Mouse buttons are not something I think of as having much or any travel on them. They're kind of on off, but I guess it depends on the mouse. There's travel, man. There's a little bit of travel. It's not like a keyboard where you can really control. Like it clicks or it doesn't kind of like you can. There might be a little bit of variation in there, but it's not quite the same thing. Yeah. It's like if the keyboard is a three and a half millimeter travel, this is a half a millimeter travel on a mouse switch i have one right here i'm gonna pull out and in fact the feel of the mouse button is kind of what i was getting at with the the super shallow esports pro keyboard switch right you see that i think it may be blurry but you get what i'm saying yeah oh yeah yeah like like like basically basically keyboard keys that feel like mouse buttons and how hard they need to be depressed that kind of thing so you could make a keyboard with a bunch of these little tiny mouse buttons i don't know how to get the keycaps on there um but the uh the the there's a whole scene in the game in the in the high-end mouse community we've talked about the mouse review subreddit before where people are swapping out their their stock mouse switches with increasingly specialized switches that are either like you said really really fast throw um really quiet really uh like more loud different actuation points the whole thing um and this gives you again a software configurable version of that experience um unfortunately with this you have to do it all in the g pro software um the big thing is that you need it turns out you need the haptic input on the mouse to know when you've clicked because otherwise the the switches are so small and the travel distance is so short that without that click you don't know that you've clicked you like a mechanical click mechanism well so that's what the haptic thing is well no that's what i mean they like but i ask because on this g502x i have it uses i think what they refer to as mechanical optical like they're optical in terms of signal sensing yeah but i believe they just built a mechanical clicking mechanism in just to make them click that would make sense but they didn't do that here for whatever reason well this is they have the haptic thing in in place of that So you like, if you have the mechanical switch, you couldn't adjust the haptic. Like you'd still get the mechanical even with the haptic. So the haptics, I'm still, I've had it for four, three or four days at this point. I'm not sure how I feel about the haptics. The haptics feel really mushy compared to, um, if you think about how a mouse click feels like, like I'm going to hold up to the microphone and you can kind of hear it. Like there's an immediate, you know, exactly where it actuates or, you know, exactly where the click happens because you feel it and hear it. And with the haptics, you can't hear it at all. It's very, very quiet, which I do like, but the actuation point feels less precise than with a traditional mouse switch. I think my hope is the haptics should be software configurable. So you can have like different, like, like if you think about how a joy con works or the trigger rumble on a DualSense joystick gamepad works. There's a lot of variability in how you can present the sound waves that turn into those haptic bumps on those devices. And my hope is that they'll get to a point where they let you configure what those curves look like. So it's less of a hump and more of a square wave type look. So it's on, off, on, off, and is very precise and very, very sharp feeling. But the software is not there for that yet. um you can configure like i said you can configure the actuation depth so you can make it so you bottom out the the press or you just barely graze um and and i actually quite like that i went from i play a lot of i've been playing a lot of arc raiders lately and i've been playing a lot of guns that are semi-auto guns and even in like fortnight where a lot of the guns are full auto but you're but you're meant to if you're good at the game be tappa tapping the shots it's remarkable how much more accurate I am with the really light touch on the switch to do the triggering. You can make it so you just barely, barely graze, almost like it's a touch surface, which is kind of wild to me. And if you're in Windows and you're running G-Hub, you can change the profiles on the fly based on the game you're playing. I generally, on mice especially, advocate setting and forgetting your settings so that you're locked in because you want your brain to inherently know that moving the mouse is much is going to make you turn this much no matter what game you're playing and it's better to kind of normalize across a normalize like one dpi one one set of settings but yeah have you have you used the onboard memory manager with this one i have yeah it's the same as it's the same as it always is yeah because i'm using it right now in linux yeah that's oh do they make that for linux no i'm using the onboard i'm oh the old onboard memory manager no that's what i use instead of g hub to be clear because that that's just a lighter weight way to manage the mouse profiles and sensitivity and stuff so they only gave me this is the this the mouse is just out as we're recording this now okay uh so i'm using pre-release software and they only gave me g hub i don't know if the onboard memory manager does but i am using the onboard memory profiles yeah yeah to set up how i want it to work in linux i do still wish that this had a profile button on the bottom so that i didn't have to bind a key on the mouse to changing profiles um the the next thing to do is I have like three different kinds of mouse switches here that I want to do some testing with. I have some old KLH reds, which were the kind of default you want to change your mouse switches, which is the last time I went through this process. I also got these brown, I can't remember what they're called. They're brown with yellow bumps and they're supposed to be the standard for quiet switches these days, relatively hair trigger. And I'm going to try both of those and some other mouses that I have and compare them to this for upcoming PC world stuff. But yeah, like it's interesting. The whole thing about this is interesting because we talked about the optical mouse sensors last year or the year before as being a really impactful piece of technology because they turned something that was a physical sensor. You know, in the case of mouse, it was wheels with slots cut into them that they could shine light through that determined that were attached to rollers that rolled when you moved the mouse ball. And that was how we moved cursors up until the late 90s. Then the optical sensor came on and turned that into a silicon-based process where we got all the benefits of camera technology improving and silicon getting faster. And that gave us things like mice that are 32,000 DPI or something ridiculous, which just to be clear, unless you have an insanely high resolution screen, a 32,000 DPI mouse is kind of crazy. But an 8,000 DPI mouse makes a ton of sense. I'm using like 4000 DPI pretty much all the time on my mouse to get really smooth movement across my 4K displays. Right. Yeah. I'm at 3800. It's nice. Also, crucially, you don't have to clean them. Yeah. Which kind of ties into the thing with mechanical keyboard switches versus Hall Effect or TMR or whatever. Remember how often you had to turn over your ball mouse and take the ball out and clean all the lint out? And clean the ball and get the gunk that was stuck on the rollers. you um but but it also because moving stuff to software has traditionally been really good for giving you more control over it now the tactile stuff i'm really curious how this ends up working like i don't think this tactile solution that they have in this logitech mouse is ever going to replace like the feel of a mechanical interaction in a clicky switch or a tactile switch but it's also possible that their haptic implementation is just not up to snuff yet because like I think about the trackpad on current MacBook Pros. It's true. And that is basically good enough to fool your brain into thinking that like, like I still to this day am shocked there's no mechanical actuation going on when I press on that touchpad because it feels indistinguishable from back when they did use mechanical touchpads 15 plus years ago. It's been, it's 12 years since they switched to the haptic one, as I recall. And yeah, you're absolutely right. Like that, that is indistinguishable from the old way of doing it. Now, the weird thing about that is they haven't really done it for anything except for mimicking that old, old physical behavior, right? Maybe we'll get into this soon, but I don't know about the previous AirPods, but the new AirPod Pros also feel basically like I'm pressing a real switch. But they have a little scooped and dented. Oh, yeah, yeah. The twos do that, too. Touch area that actually feels like a switch you're pressing, but they also kind of convincingly feel like you're pushing a button. Yeah. um well yeah so anyway that's that's kind of it i'm curious to see what like i don't think these are going to replace mechanical switches and everything because there's more cost frankly yeah this is going to be an upgrade forever i think probably yeah i guess i guess in some contexts it makes sense for these to be a premium option but but again with like the default controllers on consoles it would be a real bummer if they never adopt this just because of the parts wearing out i mean those controllers have gotten very expensive for like the the default first party ones to be clear like the new joy cons are a hundred dollars it's wild to me even before tariffs the the the switch to uh pro controller was 80 bucks yeah and like i feel sense also about 70 80 bucks depending on color and stuff i think yeah like that like that's pretty expensive for something that wears out that aggressively i i also i mean especially when they're selling premium controllers and those premium controllers aren't using the the fancy sticks either i i'm i'm curious about why why that doesn't happen yeah and and on the flip side a lot of the 8-bit do controllers that do use hall and ntmr are cheaper than those yeah well who knows um i mean look it's this is the first time in my life that the third-party controllers have consistently been better than the first-party controllers yeah because like if you think about i think about going over to a friend's house in the n64 playstation era and i saw that they had a bunch of weird color controllers that they bought at walmart and i was like oh man this is gonna be a bad time and these days i if i go someplace and i and i see a bit do controllers i'm stoked so yeah um i i'm i'm curious to see how people continue to use these is i guess the kind of tldr um and i don't think like this falcata keyboard is nice there's they have a regular non-split version as well that also uses the magnetic switches i think if i were buying one of these i would buy one of the ones that's a little bit more user configurable because i wouldn't mind having more access to the firmware, even at the expense of the keyboard configuration interface being pretty nice on this and using the web thing. To your point earlier, I think I started to talk about this. In addition to using the web interface, on Windows, they have a mouse companion that lets you do things like change the profile based on the application that's running, because obviously the web-based interface can't do that. It also notifies you if there's a software update, or you can just configure it using uh their their um their main asus gamer uh software what is it called um uh god i'm completely blanking it's the thing that you use to control the lights on your motherboard too if you don't use open rgb or something like that um armory crate oh i started to say armory crate i mean that's that's the thing that pulls like motherboard drivers and stuff i didn't realize that was also their gamer utility so that controls uh fan speeds lights like if you have a ASUS AIO it controls that if you have you can do firmware updates and everything it's basically like they're all in one I think it's not great software I don't generally run it on my computers if I can avoid it but that is your option if the web interface goes away right is to get an old version of Armour Crate that still supports your device and yeah I'd love to know what you all think about this stuff I'm curious like I was curious just to try it because I hadn't used any of these switches before and was curious to see if you know software is better than hardware yeah it's cool to see stuff advance yeah good technology not bad technology um that's it for us this week i guess uh as always this is a listener funded podcast we wouldn't be here without you guys the listeners so if you would like to support the show and make make sure that we keep doing this uh you can go to patreon.com slash techpod again it's patreon.com slash techpod where for five dollars a month you get access to the fabulous tech pod discord which is full of beautiful nerds just like you uh as well as uh the monthly patron exclusive episodes where brad and i talk about ongoing projects and um sometimes we take extra questions if we have a bunch of good questions one month uh small topics that don't fit into a normal episode that may be too small to talk about in a normal episode i think we initially thought that this might be one of those this this mechanical keyboards uh magnetic keyboards rather No, there's plenty of meat here. And then, yeah, it ended up being really interesting when we kind of dug in a little bit. So you can go again to patreon.com slash techpod to sign up. It's $5. We appreciate each and every person who supports the show. But we especially appreciate our executive producer to your patrons, including Jason Lee, Enwee Felicitas Rips, Andrew Slosky, Jordan Lippet, Fiends 1 through 4, David Allen, James Scamic and Pantheon, makers of the HS3 high-speed 3D printer. Thank you also so much. Thank you. That'll do it for us this week. We'll be back next week with another edition of the TechPod. Until then, please consider the environment before printing this podcast.