It mostly depends on the actual network hardware, drivers and the access point, not the OS.
When a WiFi device does a 'scan' (which it needs to do to know if there might be a better AP to connect to, amongst other things), it has to retune the radio to another frequency. That means it can't listen on the original frequency, and packets get lost. Depending on the network design, they might be retransmitted from the access point (layer 2 retransmit, a few milliseconds), or by TCP after a timeout (1 network round trip, hundreds of milliseconds often).
A third option is your laptop tells the AP that it is going into a power saving mode and won't be listening. Packets are then queued for it while it retunes and does a scan.
Best would be to have multiple radios, one whose only job is scanning. Then there would be no latency impact at all.
as a wireless hater (and especially a hater of designed-to-fail rechargeables), I would love it if the trend reversed. I don't want to have to manage charging devices. I want a keyboard that works, always, if it is plugged in.
I've got a couple of wireless devices, and all of them charge using either micro USB or USB C, and my Logitech G mouse is the only one with a custom receiver, all the rest are Bluetooth. It's great. I no longer snag cables to the point of breaking with my chair, and if I do, they're just USB cables and are replaceable.
> Dongles. Even though all computers now have built-in Bluetooth, many Bluetooth accessories today still ship with proprietary dongles. I assume this is because the manufacturer was worried about inconsistencies or incompatibilities between their own Bluetooth implementation and your computer’s built-in Bluetooth hardware/drivers.
I've never seen this before. Is the author mistaking proprietary 2.4 GHz receivers (that do not use Bluetooth) for OEMs' "own Bluetooth implementation"?
They refer to swapping between built-in Bluetooth and the dongle, which sounds like Logitech's line of dual-mode keyboards that can use either Bluetooth or Logitech's proprietary "unifying receiver," which is based on Nordic Semiconductor's nRF24L, not Bluetooth.
What the author means: if you ever worked with actual Bluetooth you'll see that it's full of insane quirks and vendor bugs. This opens up the risk that your product will be seen as bad because the user has a crappy BT chipset in their computer. Just see the debate about Bluetooth headphone latency here on HN: even engineers don't realise that experience is very much dependant on what kind of chipset they have on BOTH sides.
This is why manufacturers build their own 2.4GHz proprietary dongles using proprietary protocols - it ensures consistency for the users.
Supposedly the drivers are also full of hacks. Blacklists and Whitelists for which devices actually support certain features. A device might say it supports a specific codec, but then when the phone starts streaming audio it causes the headset to lock up.
It's a bunch of hacks all hacking around the other hacks.
As someone not familiar at the low level, can somebody explain how this can still be happening today? Are there not libraries that help ensure proper implementation? I really don't understand.
Not an expert in this kind of low-level stuff, but bluetooth is mostly implemented in hardware. I don't imagine you could get a compact lightweight headset that can last 10 hours on a charge if it needed to run even something as lightweight as a C library. And obviously, you can forget something like Airpods.
No, bluetooth is pretty much entirely in software. I say that as someone who has worked on Bluetooth devices at CSR/Qualcomm.
Embedded software quite routinely runs hours on batteries. Indeed, some (entirely software) devices, last many years.
A Bluetooth device is much more complex than the proprietary wireless devices. The complexity is largely because Bluetooth hardware must support many different types of devices all with differing requirements. A proprietary keyboard dongle doesn't have to be able to deal with the audio traffic coming from a microphone or going to a headset. Furthermore, it doesn't have to support proper identification and authorization of each individual device; many of them will accept input from any compatible keyboard that happens to be near by.
There's multiple effects going on. Firstly the bluetooth spec itself is quite complex: the 'core specification' document is near 3000 pages of very dense technical details, because it's aiming to allow anything to connect to anything. Secondly, many bluetooth devices are both heavily power and cost optimised: you don't have a lot of silicon and you don't get to use it very often if you want your product to have a reasonable battery life. This is especially true of the raio in a bluetooth device, and adds a whole other level of complexity to your implementation, and is one area where bugs are very easy to introduce. It also requires a fairly complex dance between hardware and software (most of the bluetooth spec is implemented in software, usually C. The hardware mostly consists of a radio capable of transmitting on the right 2.4Ghz signals with the right modulation scheme, but when you add power management to that it gets gnarly).
Thirdly, hardware companies are generally not great at software. A lot of effort goes into physical testing and validation but the base level of quality of the code isn't great. Also the quality of a given bluetooth device often depends on the quality of the custom code for it: there's a generaly trend of running application code on the same CPU as the bluetooth code, and in an embedded context there isn't generally great seperation between the two (nordic's approach is probably the best here). A third-party making a cheap pair of bluetooth headphones is not likely to make a quality result, even if using a library which is relatively bug-free.
Finally, there's a vicous sprial effect: because you need to interact with a wide variety of other implementations, which may all be buggy or quirky in their own ways, you wind up needing to test with a wide array of other devices (expensive and may not always happen) and play whack-a-mole with the remaining issues which appear, which in turn creates more quirks and more opportunity for bugs.
The BT protocols are quite complicated and therefore the different implementations will be full of quirks IMHO. Try debugging an audio BT issue on a phone and you will see the amount of complexity.
Interoperability is tested, but not in an organized way and not repeatably - AFAIK 3 different teams are supposed to check that a new protocol works against each other. The different vendors verify their BT implementations but not against each other. That would be expensive, I guess.
The firmware in BT devices can also use a number of hacks which are not breaking the spirit of the specs as such .. and this is not being tested. On top of this often BT and WiFi has to work together (shared antenna) and this means that BT will not get all the time it needs but a percentage based on what Wi-Fi is doing at the time.
I try to use Ethernet every time I get the chance.
Bluetooth implementations are so stupidly quirky I can totally see proprietary dongles as being required to guarantee connectivity.
Fitbit won't pair with my phone, but works OK with my wife's and my tablet. Garmin fitness tracker syncs with all 3 devices, but will stop syncing with my wife's phone within 2-3 days and will need to be paired again..
My bluetooth headphones work with my phone and my wife's phone, but not my tablet. Every 6 months or so I still have to use a paperclip to reset the headphones to be able to re-pair them with my phone after they stop working for no reason.
I've seen similar problems with phones I've owned before.
Another example of this is with 'Windows Mixed Reality' VR headsets. The controllers connect with the computer via Bluetooth and will often be noticeably laggy or unreliable when using onboard Bluetooth due to interference from other devices on the motherboard. Switching to an external Bluetooth dongle on a USB extension cable results in significantly improved controller tracking, and this issue is partly to blame for the poor reputation of WMR controller tracking.
They also didn't use it to connect to the controllers. They used two nRF24L01 transceivers which use Nordic's proprietary shockburst protocol. Easier to set up and better latency.
Yeah, the second-gen[1] Samsung Odyssey+ headset likewise has a built-in dedicated receiver in the headset and it's a significant improvement over the first gen. (I like the first gen better in practically every other way though.)
[1] There's much debate about what 'first gen', 'second gen' actually mean in terms of VR, I just mean it's Samsung's second WMR model after the Odyssey HMD.
If apple made a new protocol called APAN (apple personal area network) where they get rid of all the BS in bluetooth and did it properly, kept it proprietary for 5 years and then released it to the general market, I bet they will sell like hotcakes.
Make AirPods 3 APAN & BT compatible, with the same latency characteristics of theatre wireless mics (really low). Same with the mice & keyboards.
I remember a friend had a PDA with Bluetooth over 15 years ago. It's dumbfounding that after being around for so long it's still such a train wreck even on mainstream products.
I think the author might not know that the dongles that Logitech ships are not Bluetooth. They use Nordic's nrf24L01 chips, which run a much simpler protocol, very fast and robust for simple communications.
I use a Logitech MX Master, and while I do enjoy its Bluetooth capability (for laptops), I use a dongle with my desktop, which results in zero latency, no stuttering, and perfect synchronization.
Right. Those old diNovos are a nightmare as well and not only did they ship with their own Bluetooth adapter but required you to install their driver which would install its own Bluetooth stack (the Broadcom one and they used to use I think BlueSoleil before that) over the Microsoft one.
I still have nightmares about those things failing to pair constantly.
It's many years ago so may be inaccurate, but I remember the dongle being the one stable solution. With the onboard BT of a raspberry pi on Ubuntu (or was it Raspbian?), recurring issues. With the dongle, perfectly stable, and IIRC no drivers not in the mainline repos.
Then there's also Bluetooth Low Energy, which is Bluetooth only in name and mandated in the 5.0 version. Apparently most news products use BLE instead of BT, unless they use a well-known profile for say, audio or game controller accessories.
I think the primary culprit is the power management on both the host and client
I have an MX Master and a K380. The MX Master has a noticeable lag in Bluetooth and both will have a noticable delay when first used after a period of inactivity before springing to life. The K380 seems to have a key buffer to mitigate the delay.
If however I have my Bluetooth headphones powered on and active to the host, keeping the bt radio active, most of the issues go away.
The 'Logi Options' software has been buggy for me on Mac Catalina and constantly stops remembering my custom MX Keys or MX ergo/MX Master shortcuts until I relaunch the app. Anyone else dealing with this?
I simply stopped installing the terrible Logi software. I'm fed up with the crap, the resident CPU-eating daemons, the insistence on reporting what I do to Logitech or registering online for a cloud account — enough of this garbage.
Fortunately, the mouse works just as well without the crappy software, once you get it configured.
The reality is that if you poke yourself out of the Mac & laptop world, there's plenty of desktop computers without integrated Bluetooth support, and Bluetooth <-> USB adapters are pretty cheap, so vendors usually ship with one bundled to ensure they don't have to deal with tech support issues from people thinking they have Bluetooth support in their computer when they don't.
You may have local issues. I ran a series of iperf3 tests in a small but non-trivial network and was surprised. There was a 100Mbps switch that had been forgotten about. Also a wifi network running off a Power over Ethernet link which was 3Mbps compared with 50Mbps over fairly poor Wifi it was supposed to be superior to, which looks pretty darn good to the POE link. A cheap switch and two LAN cables later things are so much better.
Hotels have universally switched to wifi, but I miss having an Ethernet jack. I think that at least some hotel Internet problems are actually wifi problems and not due to the crappy service provider they use.
A lot of hotels still have ethernet, but while the wifi might be free, most often than not, ethernet is not. In many cases some form of tunnelling allows to connect anyway (e.g. IP over DNS)
Yah, the main thing he doesn't mention--and in most households, it's the primary thing--is retry hogging by devices that have a poor connection to the access point.
High end access points use a time slicing algorithm (tdma-ish) and 802.11ax (Wi-Fi 6) mandates something along those lines. So Wi-Fi will eventually improve a little.
But for all the household all-in-one modem-router-switch-accesspoint boxes, one device with a poor connection can hog most of the access point's time, doing re-transmits of corrupted data.
If you're video calling over wi-fi, turn off every other device that uses the wi-fi.
If you still have problems after switching to wired, it's almost certainly bufferbloat. (Avery Penwarr has a nice writeup about this, getting usable video calling with his parent's low-end ADSL connection.)
Same here. My first one was an early Naga (around 2011) which could do wired and wireless. I loved having the battery in it, so it was a bit heavier. Unfortunately it wore out. The feet got a lot of wear and tear. Later Naga were only wired, and I missed it.
Now I have a Logitech G900 with a Logitech PowerPlay. So its always charged. Obviously, it has less buttons/keybinds than a Naga.
On Macs, I like the Apple Magic Touchpad. It can do wired and wireless, and the battery works for ages.
Any of these would also work with a magnetic cable. You'd plug it only when you'd have to charge it.
The wireless mouse has been the exception to this rule for me so far: the G Pro wireless (and some other mice) don’t have the latency issues, and have egregious battery life (and the charging cable fails over so it’s basically just a wired mouse).
The lack of a cable saves a bunch of space and avoids the floppy cable. Probably irrelevant if you don’t game, but definitely quite convenient for FPSes.
I don’t think I’ve ever heard egregious used with a positive connotation. It looked out of place at first, but makes perfect sense etymologically. Thanks!
I actually can't decipher strstr's post. Is it in favor or against wireless mice? Is the battery life good or bad? Are they using the mouse plugged in to the charging cable constantly or not? I'm legitimately baffled and not trying to be a grammar police.
They use egregious in the "extraordinary" sense, meaning the battery life's great. Once it does need charging, it can still be used via the cable during charging.
I have to agree, specifically mice have been the least painful Bluetooth experience for me, with a Microsoft. Annoying part is that it changes MAC address when pairing, so regularly using it for more than one computer is a frustration but that's nothing compared to the constant interruptions I've had with keyboards and codec issues with headsets.
I actually quite enjoy having a wireless mouse even though I don't game (nothing more than OpenTTD or Heroes of Might and Magic).
Like you, I have a "gaming" mouse (G700s), in my case I bought it for the high resolution. I used to have g9x, but it developed an unbearable whine. Battery life is downright atrocious, I basically have to recharge it everyday. It's also comically heavy.
I often have to move things around my desk, and having papers getting caught up in the wires is aggravating. Also, apart from gaming mice with very supple cables, I always found wired mice to be a pain because of the wire that's never quite in the right position and pulling and / or getting caught on things.
The same happens with keyboards, but to a lesser degree. I often use a trackball, which is like a keyboard in that it doesn't move around and I still prefer having those wireless.
Really depends on the on the mouse implementation: g pro is only 80g. It’s hard to get a wired mouse much lighter without a honeycomb shell (mice in that style can push into the 50s).
I run my G603 on a single AAA battery with an AAA to AA adapter. So far I've been exclusively using "dead" batteries from other devices – each lasts a couple weeks! (In low Hz mode at least.. I don't feel any benefit from 1000Hz mode, even in Counter-Strike.)
I've been using G602 for more than 5 years now and performs like a wired mouse and it doesn't eat battery.
I will never use a bluetooth headset, the technology is not there yet. Latency and battery usage of bluetooth audio devices are inconvenient and I don't understand people can bear with it.
On the topic of wired vs wireless... Has anyone had experience doing anything ridiculous such as getting a 40Gbit Ethernet adapter for say, their macbook? https://eshop.macsales.com/item/ATTO/TLNQ3402D00/
Kind of amazing that a macbook pro has four 40 Gb/s ports... and none of them are ethernet. Compare to 1 Gb/s ethernet which was rapidly adopted by Apple laptops as well as desktops. (Though you can currently get a Mac mini (!) with 10 Gb/s ethernet built in - if you can find something to connect it to.)
I've used thunderbolt cables to connect laptops together for fast data transfer. I think in order to get decent performance I had to get disable the software ethernet bridge.
Those are Ethernet ports, just SFP+ ports rather than the RJ45 copper you're probably used to. I've never seen RJ45 ports > 10Gbps. Even 10Gbps is more often than not SFP+. Even if Cat8 can theoretically can 40Gbps, it wouldn't really make sense to put copper for those speeds today IMO.
This device looks to be target for network engineers, who can use it for testing and benchmarking infrastructure, or for installations that will most likely be in or next to racks rather than on desktops.
You could just pop a transceiver module that will convert the port to RJ45. The price of those is peanuts compared to the 2k list price of this device, and the reverse isn't viable.
IIRC each side of the macbook pro only has a single Thunderbolt controller. So it's actually 2x 2 ports sharing 40Gb/s. If you need more than 40Gb/s, for say dual external gpus, you'd need to use cables on opposite sides.
Related: anybody have experience with powerline ethernet? Are there good adapters that can make high-quality/high-speed ethernet connections over power wiring inside a home? I know these are a thing but I have no idea how good their quality is or how strongly they depend on the wire itself.
In my first flat I had around 400mbit/s via some cheap powerline adapter, in my second flat I couldn't connect them at all...
You'll just have to try, it depends a lot on your home wiring and can be interfered by so many things. But if it works it's definitely nicer and more stable than WiFi.
I have the Extollo LANPlug 2000 and it has by far the best performance (rock solid pings of about 3ms in my house), but:
1. It works terribly if you use more than 2. There's just some bug in the L2 switching where it will get confused about the destination; broadcast packets get through, but some packets to specific hosts don't go through. Rebooting the adapter fixes it temporarily. I ended up moving one device to WiFi and as a pair it works reliably
2. The firmware it ships with is a bit limited, but they provided me with something more configurable when I contacted support about issue #1
The more advanced firmware they sent me allows tuning what broadcast traffic is allowed through. Note that with either firmware, broadcast traffic worked fine, but it would just seem to get confused about where a specific MAC was.
And also stomp all over various HF radio bands. They're a nuisance.
Anything that converts long runs of unshielded cable into an incoherent broadband transmitting antenna should fail FCC certification, yet they're still on sale.
PowerLine uses very similar technology as WiFi does, except that power cables are significantly more noisy than air. As such, they usually perform significantly worse than wifi. With exception of certain cases like buildings with 19th century 2m thick walls.
In most other cases, quality WiFi equipment performed better and was more stable.
This has been my experience. I bought some Devolo powerline adapters a few years ago to try in my apartment, and got quite poor speed (<100Mbps) and network was rather unstable. Switched to WiFi and got better speed and stability.
Last year I moved to a new place and faced the same issue of wanting internet in a different room and tried the Devolos again. Same experience except worse speed. Again I had to fall back to WiFi, which at least gets me ~150Mbps here.
That depends on where you are. In a drywall or wooden house, this would be true, but in concrete apartment building with thick walls and many other networks, it might not be the case.
I have the TP-Link AV2000 and found them good. My office has a bunch of metal laundry appliances and concrete walls between my desk and the wifi router, so wifi signal is terrible there. Solved it with the AV2000.
I also tried using a mesh wifi network, and was able to bounce a signal around the dead zone and get a WiFi signal with a higher bandwidth than the powerline ethernet, but I still found I got a lot of glitches in video calls, so went back to the powerline ethernet.
I've not noticed the problems other commenters here had with their adapters with the AV2000.
I initially tried a cheaper powerline ethernet, and it was worse than WiFi, so I'd recommend you buy the most expensive/best one that you can afford.
A run of ethernet cable costs money to install, but once done it uses no power, zero maintenance and gives you 1000Mbs or more full duplex reliably. (I'd discard Ethernet over Power before Wifi, cable is so much nicer).
I've been using ethernet cables at home recently, on account of having a bunch of work stuff here. Ethernet = better, more consistent throughput.
So I've settled on using rubber cable floor covers. They don't look very good, but they're perfectly practical: cheap, much reduced trip risk, no need to open up the walls or tape/nail anything to them to get everything arranged.
(Not sure how I'd feel about this on a permanent basis, though. Once it's back to just me, my laptop, my iPad, and my phone, the internet connection will be the bottleneck again, and I'll be switching back to nothing but wifi...)
All my walls are solid brick, so it's a choice of cabling tacked to skirting boards, cabling under the floors or excavations. I'd pick powerline over those options.
All my walls are solid concrete. I recently had the house cabled up because the powerline stuff dropped multiple times a day for 5 minutes at a time. It was infuriating.
The cabling runs through the ceiling and dropped down to port height via electrical conduit on the external side of the wall. It was expensive but I would have paid more considering the quality of life improvement.
I spent a couple of hundred on a mesh wifi system and while it has two downsides (double nat, and a higher base latency of around 6ms instead of <1ms to the router) it's otherwise flawless. I have a small flat (500 sq ft) and it took 3 carefully positioned mesh points (within LOS of each other) but it works great.
Powerline depends a lot on the state of your power wiring.
I have Devolo Magic2 G.hn powerline adapters. In my rental UK apartment (1990 or so vintage), I can barely get 100 Mbps for the 10 meters or so from my office to the living room. The same units with mesh-wifi sync at around 300 Mbps in my parents' much larger 1980 vintage house in France. Perhaps the difference is due to the UK's crackpot ring electrical standard.
I wish someone would make laser free-space optics for in-home use (they are available for company campus networks).
WiFi 6E adds a full GHz of bandwidth in the 6GHz band and should dramatically improve congestion, but you will still need multiple access points with some form of wired backhaul to get reliable WiFi.
I had a pair of tp link pa8010 gigabit adapters for a wire running about 10m. I had pairing issues every day, regular "random packet loss", large latency spikes to my router, none of which I was seeing on a wifi device. It was better than nothing, but not something I'd recommend.
I have Devolo dLAN powerline adapters and they deliver around 300mbps even though they are marketed as “AV1200” and have Gigabit Ethernet ports.
So yes there are adapters that are decent, but you need to aim for the fastest ones because real-world performance seems to be 1/10th of the advertised performance.
My desktop is connected through DLink AV1200 PLCs.
Every once in a while they lose connection (maybe once a month).
The speed is not great (~300Mbps) but I don't care because my inet connection is only 100Mbps. The latency is better than wifi, and really consistent (unlike wifi):
--- 192.168.1.1 ping statistics ---
18 packets transmitted, 18 received, 0% packet loss, time 48ms
I agree, wireless is overrated in “desktop” situations.
One solution that I hope more monitor manufacturers would follow is the recent Dell “hub” monitors with built-in Ethernet (RJ-45), USB-C, downstream power, USB-A, etc.
Apple has been doing this since the advent of Thunderbolt, and I have to believe it was one of the reasons they wanted it and then pushed toward USB-C.
With MiniDP and MagSafe it was nice but clunky; with USB-C and USB PD it’s great. I know some people miss the old clicky docks, but I agree with you that the monitor + single cable is a really good solution.
I'd take a simple monitor and an external hub, in place of a monitor+hub integration, any day of the week. We put two things one inside the other and call it progress, but that's less maintainable, sometimes less portable, more expensive, and, most importantly, less modular.
Isn't it the same number of wires, since you still need to run a usb cable from the computer to the monitor? So then really all you get is not needing a separate box for the hub. That's admittedly nice -- I have a Dell monitor with usb ports and I use it, mostly because I'm not going to buy a standalone hub when I don't need it -- but if the monitor had a way to mount the external hub on the back, I think that would be the best of both worlds.
One cable rather than two. Some monitors have integrated USB hubs, but you need to connect a separate USB-A to USB-B cable, in addition to HDMI/DVI/VGA. Newer, high-end monitors will transmit video and USB over a single USB-C cable, as well as power in the other direction (to charge a laptop).
Attach the hub to your monitor's VESA mount points and you gain a lot of long-term flexibility at the cost of a single short wire that'll be hidden behind your monitor. Otherwise functionally identical to an integrated hub/monitor, but you can upgrade or replace the individual components on their own schedule.
A good monitor can last through generations of computer hardware, it doesn't make sense to combine directly with other components that have much shorter lifespans.
> HDBaseT, promoted and advanced by the HDBaseT Alliance, is a consumer electronic (CE) and commercial connectivity standard for transmission of uncompressed high-definition video (HD), audio, power, home networking, Ethernet, USB, and some control signals, over a common category cable (Cat5e or above) using the same 8P8C modular connectors used by Ethernet.
Eh, my “cheap” TRENDNet stip+crimp tool does the job just fine. I’m not a heavy user, but I’ve done well over 50 terminations with it and every time I’ve had an issue it’s due to myself doing a shitty job inserting the wires into the connector.
HDBaseT has somewhat consolidated the HDMI over twisted pair market, so you do have one more standard, but it replaces a dozen non-standards in the field.
In an ideal world, with plenty of funding, I've always wanted to take that a step further and implement a remote virtualized workstation. With just 2.5Gbps from NBase-T and 802.3bt for power and VESA display stream compression you could hypothetically power a monitor from a single cat5e cable as well as output full 1080p 60Hz video ("visually lossless"), supply power and data for USB, maybe even a passthrough 802.3af ethernet port for an ip phone.
You could take that even further by adding in a custom SPICE-like display protocol and only send rectangular regions of changed pixels and translation commands to shift a region of pixels around (scrolling, moving a window, etc). Given it's all virtualized instead of traditional ip phones you could go with dumb USB desk phones and route that USB device to a single PBX and build something akin to an old school PBX where the phones are more or less just dumb terminals. Rather than dealing with the headaches of setting up BLFs, dial plans on the phone vs. PBX, call forwarding and DND on PBX and phones, VM access, Hotdesking setups, etc it's all just a bunch of specialized HIDs hooked up to a VM. Need HA? Move the USB devices on failover to a separate VM.
It'd have all the benefits of thin clients with significantly less drawbacks. Need dual monitors? No need to replace your bargain basement thin client with one that supports dual monitors, just pull up another monitor, hookup to the same seat on the servers and done. You could sell a deluxe version with dual inputs so that each display is connected to two independent switches so if one bites the dust it doesn't take down 48 employees at a time.
I currently use a Dell WD15 with my Dell laptop and I love it. Only single cable to plug in which drives my wired lan, 2 screens, power, mouse and keyboard.
This monitor with it essentially built in looks great, would remove all the following from my desk:
- the dock
- the power brick for the dock
- One of the DP cables
- one of the screen USB cables
Did this for years with my ThunderBolt display, and have been doing it since 2015 with my LG 4K display.
The only bad thing is that because the LG is USB-C and not TB3, I only get USB 2.0 speeds out of it. I have a pair of USB-C to USB-A adapters that let me connect ethernet and my phone or other device to charge.
Honestly for a laptop I'd prefer the following setup:
An E-GPU enclosure, which is also the power supply, has an ethernet port, and provides some USB ports for any devices you always leave at the desk.
You connect it to the laptop with a single thunderbolt power-delivery type-c cable.
The monitor(s) are connected to the enclosure with type-c displayport with power delivery.
Thus you have only a max of two cables connected to the wall: power cable for E-GPU dock and ethernet. This is fairly similar to that monitor concept except that the monitors docking station is broken out as a separate component.
Can anyone recommend a bluetooth headset (i.e. speakers and mic) which does not have this issue, as mentioned in the article:
"Low quality. Related to the codec issue, many bluetooth devices will play high-quality audio when the microphone is turned off, but degrade to much lower-quality audio when it’s turned on. You can test this for yourself if you have a bluetooth headset: play music on it, then open your microphone settings to the page where it shows the mic input volume. You’ll probably hear the audio cut out for a second, then return at lower quality. (This happens even with devices you might expect to be high-end, like my Airpods Pro + 2018 Macbook Air.)"
I have need trying to find any, but have failed till now, and do not have unlimited budget to buy and try many options.
You probably won't find one because the Bluetooth standard doesn't allow it - it demands switch to low quality codecs for two-way communication. This hasn't really changed since BT 2.x.
Perhaps using something like Apple proprietary headphones or headsets with non-BT radio receivers would work.
This was my biggest disappointment with the AirPod pros. On my phone the sound and mic quality is good. If I try to use it on my MacBook Air on zoom it’s totally crap. Audio only is good-ish (I get some brief disconnects when there’s no sound some time, but the audio quality is good). With the mic it’s absolutely not.
You might be able to get around this by using the laptop microphone instead of the one on the AirPods. Option + click the volume icon in the menubar during a call and choose the MacBook microphone. Audio should switch to A2DP and sound much better.
This works for me on two different bluetooth headsets, but I've never tried AirPods so YMMV.
This is certainly better than using both laptop mic and speakers, but especially on MacBooks on which the fans are always going wild it can still be awful.
Yeah I use an external Yeti Nano mic which is awesome. But I think it’s not that much to ask from BT headphones that costed me 280eur to be able to have reasonable sound quality on my Mac... isn’t it? (Reasonable = as good as on my phone)
You need to look out for headsets that support the Hands Free Profile version 1.6 [0]. This version adds optional support for much better audio quality when using the microphone. I'm currently using the Sony WH-1000XM3. Unfortunately HFP 1.6 seems not yet to be supported on Linux (i.e. Bluez/Pulseaudio)
HFP 1.6 does not solve the problem. Surprisingly, there is no solution [1]. HFP 1.6 supports "wide band speech with the mSBC codec," but this sounds terrible compared to the CD quality sound you get from the unidirectional A2DP profile. Some OSes automatically toggle from A2DP to HSP during a phone call (eg. the "auto_switch" option of PulseAudio's module-bluetooth-policy), but you can't use both profiles at the same time.
Yeah. It's possible that Bluetooth LE Audio could improve this, we'll see once they finally get around to releasing the specs. It certainly sounds like it's designed to support bidirectional audio links without the stupid HSP/HFP divide of classic Bluetooth. (There's more mainstream demand for this feature now due to things like voice assistants.)
What about dedicated Bluetooth headsets for phones, like plantronics? Aren't they supposed to work bidirectional with high quality audio? Or do they also use the standard Bluetooth profiles/protocols?
Bluetooth headsets like the ones Plantronics make are designed for making phone calls, and the limitations of the standard Bluetooth headset protocol weren't a problem for that because standard phone lines were monoaural with the exact same frequency response and bit depth limits so you weren't losing anything extra. The trouble is that people want to use Bluetooth audio for things like gaming, video conferencing, using voice assistants whilst playing music etc which have much higher quality audio. Also, phone calls themselves are slowly improving too with things like HD Voice.
Hey, I spent a few hours trying to get those headphones working yesterday. Everytime I selected HFP on Ubuntu 20.04 the selection just went blank. Any tips greatly appreciated! Thanks.
I have those headphones too, and they work very well with Linux. Actually, they work even better than Windows since Linux can be convinced to send LDAC instead of aptX. I do that using pulseaudio-modules-bt [0].
If you're into listening to music, you may be interested in AutoEQ [1]. Those guys have done some measurements and provide info on how to equalize the sound. For this particular headset, it's somewhat less boomy. You can use PulseEffects [2] to control equalization.
The Bluetooth headset latency kills me, especially now so many devices are losing the headphone jack. As a hobbyist musician its unusable, same issue for games.
It’s just worse in every respect and the headsets cost three times as much for the equivalent sound quality.
I completely agree with you .. The latency in wireless headphones is bad .. It is clearly noticeable in games and even in general music playing sometimes ..
I cant imagine people playing serious games on mobile devices without headphone jack.
I got tired of having ads constantly shoved in my face for games I already paid for that then go free. That's when I realised games on mobile were never going to work for me.
The only advantage is that when I'm running there isn't an annoying cable to fuck with. There is no other benefit that I can see. But I started running more to justify the purchase. Ahahaa.
Same. I only got bluetooth earbuds for going to the gym. I specifically got ones that are sweat and water proof because otherwise I wouldn't use them. Cabled earbuds are far superior in every other way.
I’m waiting for some airpods to be delivered today so this is painful to read. I don’t mind a few ms of latency but are we talking noticeably bad latency? The only experience I have is with BT pairing with my car for music. I don’t even use voice in the car.
For playing music or even calls you won’t notice any difference.
The problem is trying to use and app like GarageBand where you need instant response to instrument control. In fact, first time you open GarageBand and it detects you’re using Bluetooth it’ll give you that warning.
What do you mean by "playing"? I play the digital piano with wired headphones, and with bluetooth sound it is absolutely unplayable due to the latency.
In my experience it varies a lot. I don't have airpods but I have some other wireless earbuds without any noticeable latency. Later I bought a different model from the same manufacturer (Anker) and they are useless for video of any kind. I also have a pair of Sony wireless headphones that have no latency I can discern.
Watching videos and listening to music is fine, devices usually delay video playback to match the latency so you don't notice any difference. The latency problem really only matters for when you need instant feedback for your actions: playing games, editing audio etc. Even then you get used to it.
I only notice it on my AirPods when I pause what I’m listening to. I have “double tap right pod” as toggle play/pause. “Tap tap” and wait, often I’ll “tap tap” again just as it pauses only then to have caused it to play again. I have to trust that my initial tap tap registered.
I think the first gen AirPod was 200ms, and current AirPod Pro had it down to ~150ms. Which is already the outliner in the industry as most are still 250ms+.
Compared to Wired Headphone of 5ms latency.
At least Apple continues the R&D to lower latency. Hopefully someday it will reach sub 50ms.
Yes, phones delay video playback so it matches the reported latency of the audio device. This works via networked audio protocols as well, at least on Apple devices in my experience.
The popular type is from Qualcomm called AptX Low Latency (not the same as 'lossless' AptX HD versions) and you need it on both sides xmt/receive. I have never seen this supported natively before so I got them as a pair of headphone jack to BT dongle adapters. Video works fine on Hotel TVs which was my normal useage.
A normal old school ISM band headphone with transmitter base is better for TV but the dongles travel better.
This can be easily seen in action when playing audio over AirPlay where the latency is ~2 seconds. When you press Play the video waits for a moment to compensate for the audio delay and only starts after the two-second delay to sync with audio.
1 frame at 30fps lasts 33.33 ms. I work as a freelance rerecording mixer and can assure you that a shift of one frame can definitly be noticible.
However there are some people who are more sensitive to this than others.
50ms would be two frames at 25fps, which is already quite much indeed. Anything beyond that is just a crime against anybody who worked hard to sync sound and image.
My TV already has higher latency than that and I manage on Rock Band just fine - anything that depends on A/V synchronization is generally aware of latency being a problem and has knobs for the users to adjust for it.
That being said, waiting multiple frames for video playback to start while audio buffers fill and transmit is a shitty experience - so it’s always good to reduce it.
I don't think wired headphones really have latency of their own. A single audio sample takes less than a microsecond. The latency will all come from upstream buffering, which will depend on the application and the OS.
Yup, not sure where the OP got the 5mSec number; perhaps it's about the soundcard buffering but then it's probably sort of high end using ASIO drivers and a rather small buffersize (say 192 samples at 44.1kHz) because I duoubt your everyday buit-in soundcard achieves that. But it cannot be about the speed of the analog signal from the DAC until it reaches your speaker: too lazy too look it up but worst case IIRC is still half the speed of light so not even near mSec range.
I agree for music creation or gaming, but in general wires are definitely annoying and limiting. I could never give up my Bluetooth headphones now for everyday use - wireless plus noise cancellation makes such a big difference when commuting or in the office.
The weird thing is, low latency wireless audio transmission is quite common. Like those used for in-ear monitoring on stage, or wireless guitar cables, that just broadcast over rf frequencies. I don’t understand why no one sells headphones using technologies like this, even if it meant a dongle that needs to be plugged into your computer as well.
Interference would probably be an issue if a full train car of people were trying to use this or something, so I get why Bluetooth is a better, more standardized default. But I would love to have the option of an alternative. I would definitely buy low latency wireless headphones to use while playing music.
As a musician I'd call anything ≤10ms low, while 20ms would be okay and usable. Something like 80ms would just be distracting fo anything rhythmic (okay for pad stuff or other slow swelling sounds)
Assuming my audio interface is telling me the truth about roundtrip latency, for me anything above 8ms starts to be detectable when playing guitar. Between 8-15ms or so it's subtle and you can kind of trick your brain into ignoring it but it's definitely still there if you pay attention.
And from what I've seen in the best case scenario a usb audio interface is going to have a roundtrip latency of 4-5ms (thunderbolt can be much better, there are some that claim 1 or 1.5 ms).
So IMO that means ideally wireless headphones need to stay < 3-5ms to be an unnoticeable replacement for wired ones.
The GGP says he is a musician--80ms is going to be much too high.
More niche, but close to my heart, it also basically kills rhythm video games. Those can mostly compensate for display latency, but compensating for audio latency is more troublesome--particularly if the exact amount of latency isn't constant.
To follow up, in most rhythm games, the "perfect" timing window is around +/-33ms. In DDR, the "marvelous" timing is around +/-16ms. 80ms is almost a combo breaker.
Top level players can stay in the +/-33ms window for thousands of notes without making a single mistake, and "full marvelous combos" where you stay in the +/-16ms window for an entire song are not unheard of.
...it's interesting to note, though, that because DDR doesn't provide auditory feedback when you hit a note, it should actually be able to compensate for audio latency quite well. The big caveat being, you have to know what the latency is, and that latency must stay consistent, neither of which apply to Bluetooth headphones.
There are wireless headphones using other technologies than Bluetooth. I'm currently using a Steelseries Arctis 7 and it has no perceivable latency (I play rhythm games with it).
And it indeed comes with a USB "dongle"/base-station, with a cable and about the size of a watch, so not really usable on a train etc. (I use wired headphones when working away from my desk).
I’m definitely 9/10 for wires rather than against.
But on the minus side, I have a chirp in my headset (obviously some interference from the motherboard) when connecting with a wire that isn’t there over Bluetooth.
I've advocated for this for years to anyone who will listen, but it's like preaching in the desert.
It doesn't help that many people are on laptops, and the war on ports (in general, and ethernet ports in particular) has made wired options exceptionally inconvenient. Not only you have to go out of your way to buy extra hubs and adapters that are not included with the laptop, you then have to carry a bag full of dongles everywhere.
False dichotomy. They could have multiple ports supporting multiple standards. This dismisses a valid complaint, in my opinion, the war on ports in the name of thinness. I've never heard anyone cite thinness as the reason they bought a device (weight, sometimes, but never thinness.)
I only miss is 8P8C port, but only when I'm traveling and want wired connection.
USB C made my life easier. Now I only have to plug one cable into my laptop from my monitor and have everything(wired network, external display, speaker, webcam, microphone, keyboard and mouse).
I don't see a reason to carry a dock with me. A lot of people carrying a Micro USB A to USB A cable and a dock instead buying a Micro USB A to USB C cable which makes no sense
Honestly, I hate USB-C. It's flimsy as all get-out. All the USB-C stuff I have seems to maintain a poor connection when moved, and it doesn't feel anywhere near as sturdy as a plain-old USB A. When I'm using a laptop, I want something I can leave plugged in when I pick up and move the computer and not lose connection when jiggled. They should have made something that valued a sturdy, stable connection over slimness.
No, it wouldn't, but I honestly couldn't care less about phones. It makes absolutely no sense to attempt to design a port that is good for both delivering power to tiny devices and also good for transferring forty gigabits per second of data, all while being durable and maintaining a good connection.
Well, the whole point of the movement was to make one connector be the standard for most everything. The general populace is much more upset with having to keep a bunch of different connectors around. It's much easier for them if they can just have one connector to deal with.
It'd be nice for me too. I wouldn't have to keep an entire drawer full of various cables...
I appreciate the help, though I've looked inside my ports before and found neither dust nor lint. The problem seems to be mostly with cheaper stuff. I've mentioned this to USB-C enthusiasts before and gotten back, "Well, just by better-quality stuff." The difference is, I never had any problem with the connection on cheap USB-A devices. If the connector seemingly can't be made both cheaply and well, that is a huge flaw with the standard and means it ought to be abandoned wholesale.
I avoid wireless because of EMF sensitivity that my wife has. Just to be sure, I've secretly switched on the wifi some days to see if she mentions any symptoms and definitely she does. I personally haven't felt a difference, but it is fun running a wired network - switches all throughout the house and gave me the motivation to set up Pi Hole and a print server.
Here we go again: https://en.wikipedia.org/wiki/Electromagnetic_hypersensitivi...
"The majority of provocation trials to date have found that such claimants are unable to distinguish between exposure and non-exposure to electromagnetic fields. A systematic review of medical research in 2011 found no convincing scientific evidence for symptoms being caused by electromagnetic fields. Since then, several double-blind experiments have shown that people who report electromagnetic hypersensitivity are unable to detect the presence of electromagnetic fields and are as likely to report ill health following a sham exposure as they are following exposure to genuine electromagnetic fields, suggesting the cause in these cases to be the nocebo effect."
Wikipedia article has links to scientific studies.
Reasonable voices would not be so quick to comdemn. Please try to understand that we are just beginning to discover the non-thermal effects of electromagnetic fields, and the science is not looking too good for the wireless industry, which will fight tooth and nail to discredit any findings that would cut into their billions.
The rest of the papers you cite are by the same highly controversial biochemist.
At any rate, the discussion at hand is whether "EMF sensitivity" is a real condition. It patently is not. Anyone with such a condition would easily be able to collect the large monetary rewards that have been made available, and simultaneously immediately lift the social stigma they must feel. The fact that they do not speaks volumes.
Yea, cherry picking, me. Note that the rebuttal does not refute Pall's biomechanical thories, or say that there are no non-thermal effects, but that we need more thorough research, and that this research is not being funded. The precautionary principal (ignored in the USA, it seems), would indicate that we should be cautious and not quick to dismiss.
Thanks for sharing the research! You might like the YouTube channel Scottiestech.info Do you have any other way of connecting? Your profile 'about' is very interesting - it'd be great to talk more about hacking and planting trees. You can reach me at the email in my profile.
One point I am wondering about is if the symptoms could be caused by an actual disease (unrelated to EM) that would need treatment, but is not because the affected person is convinced that the cause is EM exposure.
However, I guess people usually turn to unproven treatments once medicine failed several times to treat their symptoms, so they likely already went to a regular doctor several times beforehand, ruling out any easily discoverable disease.
That's not a double-blind experiment. Your may be more attentive to her mentions when you've secretely switched on the wifi. You may behaving differently and she may be picking up clues from your behavior. etc.
If you really want to be scientific, you need to do at least n=100 experiments to detect a 10% bias at 95% confidence level. (Z=2 E=0.1) [0]
If you detect such bias, your wife may be able to earn $1M prize for her supernatural abilities. [1]
If she can reliably demonstrate sensitivity to -60dBm RF signals (nanoWatts - the typical received signal strength of a Wi-Fi signal), that's the equivalent of having discovered a new human sense on top of the 5 known senses.
This discovery, or the explanation of mechanism underlying this sense, is worthy of a Nobel prize.
Sometimes when a persistent noise goes away I will realize that I've been slightly tense the whole time and all of a sudden I feel relief. Depending on what you mean by "switched on the wifi", could it be that there is some subtle piezoelectric noise that goes away when you turn of a device completely?
Triyambakam, please don't take these criticisms much to heart. Smokers were parroting the tobacco industries claims that smoking didn't cause cancer in their own ignorance. There may be a similar thing happening now with the wireless industry. We are just scientifically beginning to discover the non-thermal effects of EMF radiation. If these fields do cause detectable changes to the pysiology of human cells, there is reason to believe that your wife could be sensitive to these changes.
In France ethernet with RJ45 is the standard that replaced the older phone line that we were running in the walls. Since you can plug RJ11 to RJ45, you can use a
Any decent electrician is France knows how to install ethernet, and they can get install this kind of equipment, equivalent of a distribution cabinet, instead of the regular "phone line" box:
When I had work done in my house I just made sure that they knew that before they boarded up a particular ceiling I needed to run a cable. In the end we all ran the cable together, and now I have a 1Gb/s Ethernet connection to my office.
Depends where you live. Here in Canada almost all new builds have cat6 run to each room to a central closet.
Problem now is that most electricians think that because "it is just low voltage wiring" you do not need to worry about interference and can just run it anywhere.
My house built 16 years ago has 2 4-pair cables of CAT-5e run to every room for phone wiring, standard from the home builder.
The blank stare was that I asked for conduit and pull-string everywhere, and they weren't willing to do that. Or to do any variants beyond the standard 2 cables to each place, so my office has 4 ethernet ports (adding another "2 cable jack" was something that they could do) and needs its own switching.
Wireless is OK, you just need to use it correctly.
The trick is to treat it as a cellular system, and use the same methodology used to build cellular networks:
(1) Use only 5GHz for Wi-Fi. Turn off 2.4GHz WiFi - it's an overloaded, noisy spectrum band.
(2) One router per room, alternating channels between rooms.
(3) Use MU-MIMO (802.11ac gen2) or 802.11ax. The efficient MTU for Wi-Fi is much much larger than the MTU used for real time communication. MU-MIMO lets you send multiple small packets destined for multiple users inside one big wifi frame.
(4) Don't use mesh devices. They multiply the "airtime" of each packet by x2-x4 depending on number of hops and number of packet retries.
(5) Leave 2.4GHz for Bluetooth, and "junk" ISM devices (microwave oven, very-low bandwidth devices, etc).
People push wireless to the breaking point, then complain it's broken...
5GHz propagation across walls is much worse than 2.4GHz propagation, therefore only nearby neighbors may interfere with your access point, and on 5GHz there are enough non-overlapping channels to choose an alternate channel avoiding this interference completely.
There really aren't when you have 2x80Mhz on either side of you leaving only DFS which can be tricky since a lot of older consumer routers didn't support it. WiFi 6E should help as the allocation is massive with less propagation.
160MHz channels are a rarity even today, neither the latest Samsung S10 nor the latest iPhone support 160MHz.
Much more common is 80MHz channel, and even then, there is a distinction between the "main" 40MHz and the "secondary" 40MHz sub-channel. If you monitor the packets over the air, you'll notice that the secondary channel is much less occupied - all mobile devices switch to 40MHz to save considerable power, unless downloading a big file.
In the US, there are 6 available 80MHz channels, which is more than needed (think 4-color graph coloring).
In my experience, it's not reliable enough. Sometimes it's great, but sometimes it stops working for a few solid seconds. I think it's when the AC starts up.
All of it is true, but before going multiple APs, you should make sure you really need it - you would need to manage multiple networks and getting the APs to work together and allow devices to roam seamlessly might be a pain.
This is where the mesh units might come in handy - they are managed and configured in a single place, and can work together nicely - but of course you have to set them up with wire backhaul instead of in the "mesh mode".
"Ok" is worse than great (which is what you get with wired setups, all the time every time).
I understand your rationale, and I also like the convenience of wifi when it works. That being said, if you can wire "that thing" (whatever it is), just do it. One afternoon of wiring and it will work great for the rest of time. The trade-off is definitely worth it in my experience.
> (1) Use only 5GHz for Wi-Fi. Turn off 2.4GHz WiFi - it's an overloaded, noisy spectrum band.
Even recent low-end phones sometimes have just 2.4GHz Wi-Fi. Even if all you devices support 5 GHz, running 5GHz-only wi-Fi may cause problems to your guests.
>(1) Use only 5GHz for Wi-Fi. Turn off 2.4GHz WiFi - it's an overloaded, noisy spectrum band.
This is an extremely bad idea in some countries, or if you’re not careful which channel you select. In nearly all countries some of the 5ghz channels are shared with RADAR and require DFS. In some countries (eg Germany) all of the channels do.
The idea behind DFS is to detect radar pulses and then shut down the network for a period of time, this can be triggered by radar or interference.
(1) If you can use a DFS channel - use it. it's usually less congested than the regular channels
(2) If you're unlucky and DFS is triggered once per month or more, try repositioning the AP or changing the AP model to one less sensitive to radar pulses
(3) If no luck - use one of the non-DFS channels. Every country has those. In Germany, that would be channels 32-48 or 149+
Note that DFS (5.4GHz) is for weather radar. In some countries it's operated only in winter season. You may want to use different channels in summer and in winter :)
That would be great if you can afford so many access points, all your clients support 5 GHz, and you are able to run Ethernet to the access point in each room.
I will always preference wired over wireless, but I am currently unable to run an ethernet to my desktop (my partner is not keen on having a tripwire run straight across the house). So I did the same ping test that OP did and noted a few spikes, I had 10 pings over 250ms out of 600, all of these were distributed unevenly over time. Guess I could be subject to such wireless degradation. But I can't say for sure without trying an ethernet cable.
t : ms
32: 1725.0,
33: 662.0,
68: 406.0,
100: 980.0,
101: 285.0,
312: 454.0,
342: 746.0,
464: 453.0,
480: 1852.0,
481: 841.0
I kept jumping between wireless and wire because I'm annoyed by wires, and keep getting burned by wireless pitfalls through these years. But each round wireless seems to be slightly getting better over time.
Here are some tips I collected along the way:
1. 2.4GHz sucks, use Bluetooth when possible. But Bluetooth sucks, too.
2. You want to have fewer Bluetooth devices when possible, reduce the number of simultaneous connections. For example, I only use my Bluetooth mouse for gaming, and trackpad for the rest, so I would disconnect mouse when I'm not playing video games.
3. You want to use some categories of Bluetooth devices first. For example, Bluetooth keyboards usually have great experiences so you may want to make your keyboard wireless first. Bluetooth mice and trackpads are Okayish, but not as great. Bluetooth headphones and speakers are usually bad. If you really want to use Bluetooth for Audio, sometimes devices from the same company may have better connections: Beats Studio seems to have a much better connection than Sennheiser and B&O on Mac.
After all, I still love the idea of wireless because it is nice on paper. It is mostly the de facto protocols and implementations that are terrible.
I've found that sometimes fiddling with some wireless settings like operation band, channel, tx power, WMM, etc. helps, provided you know what you are doing. Especially so if you're in an overcrowded wifi zone or in a poor reception area.
The issue with WiFi is over-crowdedness. Suggestions to fix this are to actually get everyone to lower the power on their wifi transmitters (hard to control). Also, making sure youre on the right non-crowded channel (easier to control).
I totally agree with the writer, I have dedicated the last ten years to specify and implement a new single wire communication protocol for IOT and home automation applications called PJDL (Padded Jittering Data Link): https://github.com/gioblu/PJON/blob/master/src/strategies/So...
It performs much better than wireless alternatives.
It works even if you use the human body or water as a conductor, as I showcase in this demonstration (sorry for my strong italian accent):
https://www.youtube.com/watch?v=GWlhKD5lz5w
The awesome thing with wireless IoT is that it can be set up without any additional cabling, which can be extremely painful process in most homes.
In case of wireless all you need is ESP32, relay and a power supply. And all of that together will cost less than 10$.
Can you easily (cheaply) use the already existing power cables for PJDL? I know there are network repeaters that use power cables to transfer data, but I have no idea how much it would cost in IoT scenario. Do you have some examples?
Yes many users have squatted old phone wiring, here in italy in most houses it is embedded in walls, and it is present in most rooms. Being unused and obsolete it can be used for home automation with PJON :)
As a software person who doesn't know much about this sort of thing: how does this work electrically? Presumably there's no current flowing as it's an open circuit? Is it akin to a voltmeter using the ground as a point of reference? Why don't other wire-based protocols use the single-wire approach?
Does it have anything in common with 'single-wire earth return', in power-distribution?
The protocol is really simple +5v is 1, 0v is 0. The circuit is just a single conductor and a common ground pin. If you connect a LED between the bus and ground you can visually see packets in form of light pulses.
Yes, in production it is suggested to use current-limiting resistors, pull-down resistors, a fuse, a zener diode or a TVS or a varistor. Using all the suggested protective circuitry in the worst case the fuse should be the only component to fry in case of direct lightning strike.
Lightning crosses several hundred meters of air at the very minimum- what makes you think it won't arc over a few millimeters more, if that's the path of least impedance?
In general even the most rugged lightning/surge protection only decrease the likelihood of protected equipment being vaporized.
I agree that in some cases there is nothing to do and damage will be done by the strike and there could be no protection able to prevent it, specially if the equipment is directly hit. I would have corrected my previous statement but it cannot be edited.
Looks cool. Why single wire and not a simple twisted pair? Are you using a voltage relative to earth for your logic states? Just not sure how it could work otherwise.
A twisted pair seems nice to me because you can use differential signaling and not have to ground your devices e.g. 2 wire RS485.
I'm not really knowledgeable so excuse me if I have a misunderstanding as I try to reason about this from a different perspective.
When you have an object with static electricity and put it in contact with a different object, there will be a transient current, even if the circuit is not closed. The objects basically work as poor-man's capacitors.
Even though the name is "static electricity", when the objects come into contact, it's actually more dynamic than normal, because there is no current between two objects of fixed and different voltage. The more current that flows, the lesser the voltage difference becomes, in an exponential fashion.
This kind of single wire protocol operates solely based on these kind of transient currents. Because the frequency is so high, very little current flows during one period of the waveform, so the receiver will maintain constant potential, as if it was grounded.
That it isn't actually grounded means the signal becomes high-pass filtered. In particular, the DC-component must be completely blocked as it would lead to charge buildup with nowhere to go.
Yes the voltage is relative to ground for the logic state, the differential signaling requires a source of negative voltage and a more complex front-end, being relative to ground you just need a digital IO operating at 5v or 3.3v and a ground pin, no additional hardware (although current-limiting and pull-down resistors are suggested), as a result it can be fully and easily bit-banged. The twisted pair (where one wire is the bus and the other the ground) is one of the suggested approaches that helps to reduce the interference induced on the wire.
432 comments
[ 2.8 ms ] story [ 183 ms ] threadIs windows affected by this? Because I've never noticed it have an impact on video calls.
When a WiFi device does a 'scan' (which it needs to do to know if there might be a better AP to connect to, amongst other things), it has to retune the radio to another frequency. That means it can't listen on the original frequency, and packets get lost. Depending on the network design, they might be retransmitted from the access point (layer 2 retransmit, a few milliseconds), or by TCP after a timeout (1 network round trip, hundreds of milliseconds often).
A third option is your laptop tells the AP that it is going into a power saving mode and won't be listening. Packets are then queued for it while it retunes and does a scan.
Best would be to have multiple radios, one whose only job is scanning. Then there would be no latency impact at all.
I've never seen this before. Is the author mistaking proprietary 2.4 GHz receivers (that do not use Bluetooth) for OEMs' "own Bluetooth implementation"?
They refer to swapping between built-in Bluetooth and the dongle, which sounds like Logitech's line of dual-mode keyboards that can use either Bluetooth or Logitech's proprietary "unifying receiver," which is based on Nordic Semiconductor's nRF24L, not Bluetooth.
This is why manufacturers build their own 2.4GHz proprietary dongles using proprietary protocols - it ensures consistency for the users.
It's a bunch of hacks all hacking around the other hacks.
As someone not familiar at the low level, can somebody explain how this can still be happening today? Are there not libraries that help ensure proper implementation? I really don't understand.
Thirdly, hardware companies are generally not great at software. A lot of effort goes into physical testing and validation but the base level of quality of the code isn't great. Also the quality of a given bluetooth device often depends on the quality of the custom code for it: there's a generaly trend of running application code on the same CPU as the bluetooth code, and in an embedded context there isn't generally great seperation between the two (nordic's approach is probably the best here). A third-party making a cheap pair of bluetooth headphones is not likely to make a quality result, even if using a library which is relatively bug-free.
Finally, there's a vicous sprial effect: because you need to interact with a wide variety of other implementations, which may all be buggy or quirky in their own ways, you wind up needing to test with a wide array of other devices (expensive and may not always happen) and play whack-a-mole with the remaining issues which appear, which in turn creates more quirks and more opportunity for bugs.
Interoperability is tested, but not in an organized way and not repeatably - AFAIK 3 different teams are supposed to check that a new protocol works against each other. The different vendors verify their BT implementations but not against each other. That would be expensive, I guess.
The firmware in BT devices can also use a number of hacks which are not breaking the spirit of the specs as such .. and this is not being tested. On top of this often BT and WiFi has to work together (shared antenna) and this means that BT will not get all the time it needs but a percentage based on what Wi-Fi is doing at the time.
I try to use Ethernet every time I get the chance.
Fitbit won't pair with my phone, but works OK with my wife's and my tablet. Garmin fitness tracker syncs with all 3 devices, but will stop syncing with my wife's phone within 2-3 days and will need to be paired again.. My bluetooth headphones work with my phone and my wife's phone, but not my tablet. Every 6 months or so I still have to use a paperclip to reset the headphones to be able to re-pair them with my phone after they stop working for no reason.
I've seen similar problems with phones I've owned before.
[1] There's much debate about what 'first gen', 'second gen' actually mean in terms of VR, I just mean it's Samsung's second WMR model after the Odyssey HMD.
Make AirPods 3 APAN & BT compatible, with the same latency characteristics of theatre wireless mics (really low). Same with the mice & keyboards.
I use a Logitech MX Master, and while I do enjoy its Bluetooth capability (for laptops), I use a dongle with my desktop, which results in zero latency, no stuttering, and perfect synchronization.
I still have nightmares about those things failing to pair constantly.
I have an MX Master and a K380. The MX Master has a noticeable lag in Bluetooth and both will have a noticable delay when first used after a period of inactivity before springing to life. The K380 seems to have a key buffer to mitigate the delay.
If however I have my Bluetooth headphones powered on and active to the host, keeping the bt radio active, most of the issues go away.
Fortunately, the mouse works just as well without the crappy software, once you get it configured.
If you're getting lag and dropouts with a wired connection, the problem is most likely bufferbloat.
High end access points use a time slicing algorithm (tdma-ish) and 802.11ax (Wi-Fi 6) mandates something along those lines. So Wi-Fi will eventually improve a little.
But for all the household all-in-one modem-router-switch-accesspoint boxes, one device with a poor connection can hog most of the access point's time, doing re-transmits of corrupted data.
If you're video calling over wi-fi, turn off every other device that uses the wi-fi.
If you still have problems after switching to wired, it's almost certainly bufferbloat. (Avery Penwarr has a nice writeup about this, getting usable video calling with his parent's low-end ADSL connection.)
You will be surprised to learn not every WiFi 6 has support of OFDMA turned on.
I like wireless but I hope there's a better interface than Bluetooth at some point.
On my laptop (which is 9 years newer and has similar benchmark results) I use a wireless mouse which has 125 polls/sec and latency of 20 ms.
Yes, one can tell the difference. It is possible to buy low-latency wireless mice. Next time, I will.
Bluetooth is a trash fire. I was hoping wireless USB would take off, but that's dead now.
On 60Hz my wired and wireless mouse feel exactly the same.
On 240Hz, yeah, there is a lot of difference.
Now I have a Logitech G900 with a Logitech PowerPlay. So its always charged. Obviously, it has less buttons/keybinds than a Naga.
On Macs, I like the Apple Magic Touchpad. It can do wired and wireless, and the battery works for ages.
Any of these would also work with a magnetic cable. You'd plug it only when you'd have to charge it.
The lack of a cable saves a bunch of space and avoids the floppy cable. Probably irrelevant if you don’t game, but definitely quite convenient for FPSes.
https://www.etymonline.com/word/egregious
I have to agree, specifically mice have been the least painful Bluetooth experience for me, with a Microsoft. Annoying part is that it changes MAC address when pairing, so regularly using it for more than one computer is a frustration but that's nothing compared to the constant interruptions I've had with keyboards and codec issues with headsets.
Like you, I have a "gaming" mouse (G700s), in my case I bought it for the high resolution. I used to have g9x, but it developed an unbearable whine. Battery life is downright atrocious, I basically have to recharge it everyday. It's also comically heavy.
I often have to move things around my desk, and having papers getting caught up in the wires is aggravating. Also, apart from gaming mice with very supple cables, I always found wired mice to be a pain because of the wire that's never quite in the right position and pulling and / or getting caught on things.
The same happens with keyboards, but to a lesser degree. I often use a trackball, which is like a keyboard in that it doesn't move around and I still prefer having those wireless.
I will never use a bluetooth headset, the technology is not there yet. Latency and battery usage of bluetooth audio devices are inconvenient and I don't understand people can bear with it.
I've used thunderbolt cables to connect laptops together for fast data transfer. I think in order to get decent performance I had to get disable the software ethernet bridge.
Might have something to do with the fact that 8P8C is as thick as the entire laptop.
Looks really bad for longetivity.
Just as a data point, it's not expensive to add 10GbE (or higher) to a desktop or server/NAS if you're ok with getting 2nd hand gear from Ebay.
Gear from Mellanox (specifically) is pretty robust (eg rarely bad), and the ConnectX-3 and above series are dirt cheap. eg:
https://www.ebay.com/itm/MCX353A-FCBS-MELLANOX-CONNECTX-3-40...
Note, I don't know that seller at all. It was just one of the first items in the Ebay search results. :)
This device looks to be target for network engineers, who can use it for testing and benchmarking infrastructure, or for installations that will most likely be in or next to racks rather than on desktops.
You could just pop a transceiver module that will convert the port to RJ45. The price of those is peanuts compared to the 2k list price of this device, and the reverse isn't viable.
You'll just have to try, it depends a lot on your home wiring and can be interfered by so many things. But if it works it's definitely nicer and more stable than WiFi.
1. It works terribly if you use more than 2. There's just some bug in the L2 switching where it will get confused about the destination; broadcast packets get through, but some packets to specific hosts don't go through. Rebooting the adapter fixes it temporarily. I ended up moving one device to WiFi and as a pair it works reliably
2. The firmware it ships with is a bit limited, but they provided me with something more configurable when I contacted support about issue #1
Anything that converts long runs of unshielded cable into an incoherent broadband transmitting antenna should fail FCC certification, yet they're still on sale.
In most other cases, quality WiFi equipment performed better and was more stable.
Last year I moved to a new place and faced the same issue of wanting internet in a different room and tried the Devolos again. Same experience except worse speed. Again I had to fall back to WiFi, which at least gets me ~150Mbps here.
Powerline is much less faf than using wifi even more so for non networking types.
I also tried using a mesh wifi network, and was able to bounce a signal around the dead zone and get a WiFi signal with a higher bandwidth than the powerline ethernet, but I still found I got a lot of glitches in video calls, so went back to the powerline ethernet.
I've not noticed the problems other commenters here had with their adapters with the AV2000.
I initially tried a cheaper powerline ethernet, and it was worse than WiFi, so I'd recommend you buy the most expensive/best one that you can afford.
So I've settled on using rubber cable floor covers. They don't look very good, but they're perfectly practical: cheap, much reduced trip risk, no need to open up the walls or tape/nail anything to them to get everything arranged.
(Not sure how I'd feel about this on a permanent basis, though. Once it's back to just me, my laptop, my iPad, and my phone, the internet connection will be the bottleneck again, and I'll be switching back to nothing but wifi...)
The cabling runs through the ceiling and dropped down to port height via electrical conduit on the external side of the wall. It was expensive but I would have paid more considering the quality of life improvement.
I have Devolo Magic2 G.hn powerline adapters. In my rental UK apartment (1990 or so vintage), I can barely get 100 Mbps for the 10 meters or so from my office to the living room. The same units with mesh-wifi sync at around 300 Mbps in my parents' much larger 1980 vintage house in France. Perhaps the difference is due to the UK's crackpot ring electrical standard.
I wish someone would make laser free-space optics for in-home use (they are available for company campus networks).
WiFi 6E adds a full GHz of bandwidth in the 6GHz band and should dramatically improve congestion, but you will still need multiple access points with some form of wired backhaul to get reliable WiFi.
So yes there are adapters that are decent, but you need to aim for the fastest ones because real-world performance seems to be 1/10th of the advertised performance.
Every once in a while they lose connection (maybe once a month).
The speed is not great (~300Mbps) but I don't care because my inet connection is only 100Mbps. The latency is better than wifi, and really consistent (unlike wifi):
--- 192.168.1.1 ping statistics ---
18 packets transmitted, 18 received, 0% packet loss, time 48ms
rtt min/avg/max/mdev = 1.975/2.669/4.884/0.617 ms
One solution that I hope more monitor manufacturers would follow is the recent Dell “hub” monitors with built-in Ethernet (RJ-45), USB-C, downstream power, USB-A, etc.
https://www.dell.com/lt/business/p/dell-u2421he-monitor/pd
Only one cable to connect.
With MiniDP and MagSafe it was nice but clunky; with USB-C and USB PD it’s great. I know some people miss the old clicky docks, but I agree with you that the monitor + single cable is a really good solution.
Personally I prefer modular components that I can swap out and upgrade separately, but I can see the appeal.
A good monitor can last through generations of computer hardware, it doesn't make sense to combine directly with other components that have much shorter lifespans.
> HDBaseT, promoted and advanced by the HDBaseT Alliance, is a consumer electronic (CE) and commercial connectivity standard for transmission of uncompressed high-definition video (HD), audio, power, home networking, Ethernet, USB, and some control signals, over a common category cable (Cat5e or above) using the same 8P8C modular connectors used by Ethernet.
Also, the 8P8C connector is just godawful. I mean, it's fine for the first 3 days until the clip breaks off.
https://xkcd.com/927/
Hasn't this been largely fixed by connectors that have a cover over the exposed clip?
You could take that even further by adding in a custom SPICE-like display protocol and only send rectangular regions of changed pixels and translation commands to shift a region of pixels around (scrolling, moving a window, etc). Given it's all virtualized instead of traditional ip phones you could go with dumb USB desk phones and route that USB device to a single PBX and build something akin to an old school PBX where the phones are more or less just dumb terminals. Rather than dealing with the headaches of setting up BLFs, dial plans on the phone vs. PBX, call forwarding and DND on PBX and phones, VM access, Hotdesking setups, etc it's all just a bunch of specialized HIDs hooked up to a VM. Need HA? Move the USB devices on failover to a separate VM.
It'd have all the benefits of thin clients with significantly less drawbacks. Need dual monitors? No need to replace your bargain basement thin client with one that supports dual monitors, just pull up another monitor, hookup to the same seat on the servers and done. You could sell a deluxe version with dual inputs so that each display is connected to two independent switches so if one bites the dust it doesn't take down 48 employees at a time.
I currently use a Dell WD15 with my Dell laptop and I love it. Only single cable to plug in which drives my wired lan, 2 screens, power, mouse and keyboard.
This monitor with it essentially built in looks great, would remove all the following from my desk: - the dock - the power brick for the dock - One of the DP cables - one of the screen USB cables
The only bad thing is that because the LG is USB-C and not TB3, I only get USB 2.0 speeds out of it. I have a pair of USB-C to USB-A adapters that let me connect ethernet and my phone or other device to charge.
An E-GPU enclosure, which is also the power supply, has an ethernet port, and provides some USB ports for any devices you always leave at the desk.
You connect it to the laptop with a single thunderbolt power-delivery type-c cable.
The monitor(s) are connected to the enclosure with type-c displayport with power delivery.
Thus you have only a max of two cables connected to the wall: power cable for E-GPU dock and ethernet. This is fairly similar to that monitor concept except that the monitors docking station is broken out as a separate component.
"Low quality. Related to the codec issue, many bluetooth devices will play high-quality audio when the microphone is turned off, but degrade to much lower-quality audio when it’s turned on. You can test this for yourself if you have a bluetooth headset: play music on it, then open your microphone settings to the page where it shows the mic input volume. You’ll probably hear the audio cut out for a second, then return at lower quality. (This happens even with devices you might expect to be high-end, like my Airpods Pro + 2018 Macbook Air.)"
I have need trying to find any, but have failed till now, and do not have unlimited budget to buy and try many options.
Perhaps using something like Apple proprietary headphones or headsets with non-BT radio receivers would work.
That doesn't exist
This works for me on two different bluetooth headsets, but I've never tried AirPods so YMMV.
[0]https://en.wikipedia.org/wiki/List_of_Bluetooth_profiles#Han...
[1] https://www.ipetitions.com/petition/duplex-high-quality-audi...
This is why the wireless ModMic [2] talks about Bluetooth codecs but actually requires a custom USB wireless receiver.
[2] https://antlionaudio.com/blogs/news/introducing-modmic-wirel...
If you're into listening to music, you may be interested in AutoEQ [1]. Those guys have done some measurements and provide info on how to equalize the sound. For this particular headset, it's somewhat less boomy. You can use PulseEffects [2] to control equalization.
[0] pa-modules-bt: https://github.com/EHfive/pulseaudio-modules-bt/wiki/Package...
[1] AutoEQ: https://github.com/jaakkopasanen/AutoEq
[2] PulseEffects: https://github.com/wwmm/pulseeffects
I have the latest Sennheiser Momentum Wireless and it disconnects itself many times from my Mac which is only less than 1 meter away...
I don't even really care about the degradation of sound quality now because the disconnection is so tormenting.
I was having an Okayish connection experience with Beats years ago, and I guess it's because it's been tested and may be optimized for Mac.
Rumor says Apple would release a headphone in the upcoming WWDC tomorrow. I hope I would eventually have a headphone with a stable connection...
I'm not inclined to buy another pair of Bluetooth headphones with the microphone quality currently available.
It’s just worse in every respect and the headsets cost three times as much for the equivalent sound quality.
people with a desktop computer are a minority thou.
The problem is trying to use and app like GarageBand where you need instant response to instrument control. In fact, first time you open GarageBand and it detects you’re using Bluetooth it’ll give you that warning.
What do you mean by "playing"? I play the digital piano with wired headphones, and with bluetooth sound it is absolutely unplayable due to the latency.
Compared to Wired Headphone of 5ms latency.
At least Apple continues the R&D to lower latency. Hopefully someday it will reach sub 50ms.
A normal old school ISM band headphone with transmitter base is better for TV but the dongles travel better.
However there are some people who are more sensitive to this than others.
50ms would be two frames at 25fps, which is already quite much indeed. Anything beyond that is just a crime against anybody who worked hard to sync sound and image.
https://www.aptx.com/product-listing?product_category=7&aptx...
That being said, waiting multiple frames for video playback to start while audio buffers fill and transmit is a shitty experience - so it’s always good to reduce it.
1us = 300m
1ms = 300km
[0] https://en.wikipedia.org/wiki/Drift_velocity
The weird thing is, low latency wireless audio transmission is quite common. Like those used for in-ear monitoring on stage, or wireless guitar cables, that just broadcast over rf frequencies. I don’t understand why no one sells headphones using technologies like this, even if it meant a dongle that needs to be plugged into your computer as well.
Interference would probably be an issue if a full train car of people were trying to use this or something, so I get why Bluetooth is a better, more standardized default. But I would love to have the option of an alternative. I would definitely buy low latency wireless headphones to use while playing music.
The new Qualcomm aptX adaptive standard supports lowish latency - ~80ms.
list of official supported devices: https://www.aptx.com/product-listing?aptx_type=336
It might also work in general , with android 10 devices, but this need verifying.
Maybe this would be ok?
And from what I've seen in the best case scenario a usb audio interface is going to have a roundtrip latency of 4-5ms (thunderbolt can be much better, there are some that claim 1 or 1.5 ms).
So IMO that means ideally wireless headphones need to stay < 3-5ms to be an unnoticeable replacement for wired ones.
More niche, but close to my heart, it also basically kills rhythm video games. Those can mostly compensate for display latency, but compensating for audio latency is more troublesome--particularly if the exact amount of latency isn't constant.
Top level players can stay in the +/-33ms window for thousands of notes without making a single mistake, and "full marvelous combos" where you stay in the +/-16ms window for an entire song are not unheard of.
That's the kind of timing we are talking about.
And it indeed comes with a USB "dongle"/base-station, with a cable and about the size of a watch, so not really usable on a train etc. (I use wired headphones when working away from my desk).
Also the custom dongle has an 3.5mm output plug for speakers, which will automatically turn off when you switch on your headset.
If only Steelseries would make noise-cancelling headphones...
But on the minus side, I have a chirp in my headset (obviously some interference from the motherboard) when connecting with a wire that isn’t there over Bluetooth.
Sometimes wires fail too.
It doesn't help that many people are on laptops, and the war on ports (in general, and ethernet ports in particular) has made wired options exceptionally inconvenient. Not only you have to go out of your way to buy extra hubs and adapters that are not included with the laptop, you then have to carry a bag full of dongles everywhere.
Apple: okay here is a laptop that has standard USB-C ports.
Geeks: we have to use dongles to support the standard.
USB C made my life easier. Now I only have to plug one cable into my laptop from my monitor and have everything(wired network, external display, speaker, webcam, microphone, keyboard and mouse).
I don't see a reason to carry a dock with me. A lot of people carrying a Micro USB A to USB A cable and a dock instead buying a Micro USB A to USB C cable which makes no sense
As far as I can tell, phones are the #1 thing people are plugging in and out these days.
It'd be nice for me too. I wouldn't have to keep an entire drawer full of various cables...
What you described can be caused by lint inside your USB-C port. I had this problem only with my phone. Try to get it out with a toothpick.
Some would say you're harming your wife's understanding of reality, but that's a private matter for you and her.
However, by sharing it on the Internet, you are potentially spreading this bullshit, and you must expect reasonable voices to counter your claim.
https://www.researchgate.net/publication/323998588_Wi-Fi_is_...
Wi-Fi is an important threat to human health https://www.researchgate.net/publication/305691437_Electroma...
Electromagnetic Fields Act Similarly in Plants as in Animals: Probable Activation of Calcium Channels via Their Voltage Sensor
https://www.researchgate.net/publication/283017154_How_to_Ap...
How to Approach the Challenge of Minimizing Non-Thermal Health Effects of Microwave Radiation from Electrical Devices
Pall's methodology and understanding of EM has been critized broadly and repeatedly. Here's an "official" rebuttal to the paper you cite: https://www.researchgate.net/publication/328436881_Response_...
The rest of the papers you cite are by the same highly controversial biochemist.
At any rate, the discussion at hand is whether "EMF sensitivity" is a real condition. It patently is not. Anyone with such a condition would easily be able to collect the large monetary rewards that have been made available, and simultaneously immediately lift the social stigma they must feel. The fact that they do not speaks volumes.
However, I guess people usually turn to unproven treatments once medicine failed several times to treat their symptoms, so they likely already went to a regular doctor several times beforehand, ruling out any easily discoverable disease.
That's not scientific.
That's not a double-blind experiment. Your may be more attentive to her mentions when you've secretely switched on the wifi. You may behaving differently and she may be picking up clues from your behavior. etc.
If you really want to be scientific, you need to do at least n=100 experiments to detect a 10% bias at 95% confidence level. (Z=2 E=0.1) [0]
If you detect such bias, your wife may be able to earn $1M prize for her supernatural abilities. [1]
[0] https://en.wikipedia.org/wiki/Checking_whether_a_coin_is_fai...
[1] https://en.wikipedia.org/wiki/One_Million_Dollar_Paranormal_...
This snark seems unnecessary, given that magnetoception by bats and birds is unlikely to win said prize.
If she can reliably demonstrate sensitivity to -60dBm RF signals (nanoWatts - the typical received signal strength of a Wi-Fi signal), that's the equivalent of having discovered a new human sense on top of the 5 known senses.
This discovery, or the explanation of mechanism underlying this sense, is worthy of a Nobel prize.
No, she doesn't.
Wi-Fi is an important threat to human health https://www.researchgate.net/publication/323998588_Wi-Fi_is_... Electromagnetic Fields Act Similarly in Plants as in Animals: Probable Activation of Calcium Channels via Their Voltage Sensor
https://www.researchgate.net/publication/305691437_Electroma... How to Approach the Challenge of Minimizing Non-Thermal Health Effects of Microwave Radiation from Electrical Devices https://www.researchgate.net/publication/283017154_How_to_Ap...
Any decent electrician is France knows how to install ethernet, and they can get install this kind of equipment, equivalent of a distribution cabinet, instead of the regular "phone line" box:
https://www.tableau-de-communication.fr/coffret-hager/13-cof...
https://www.tableau-de-communication.fr/coffret-hager/14-cof...
those are cables and sockets, standardized to the point of beeing a commodity. no magic involved at all.
Problem now is that most electricians think that because "it is just low voltage wiring" you do not need to worry about interference and can just run it anywhere.
The blank stare was that I asked for conduit and pull-string everywhere, and they weren't willing to do that. Or to do any variants beyond the standard 2 cables to each place, so my office has 4 ethernet ports (adding another "2 cable jack" was something that they could do) and needs its own switching.
The trick is to treat it as a cellular system, and use the same methodology used to build cellular networks:
(1) Use only 5GHz for Wi-Fi. Turn off 2.4GHz WiFi - it's an overloaded, noisy spectrum band.
(2) One router per room, alternating channels between rooms.
(3) Use MU-MIMO (802.11ac gen2) or 802.11ax. The efficient MTU for Wi-Fi is much much larger than the MTU used for real time communication. MU-MIMO lets you send multiple small packets destined for multiple users inside one big wifi frame.
(4) Don't use mesh devices. They multiply the "airtime" of each packet by x2-x4 depending on number of hops and number of packet retries.
(5) Leave 2.4GHz for Bluetooth, and "junk" ISM devices (microwave oven, very-low bandwidth devices, etc).
People push wireless to the breaking point, then complain it's broken...
5GHz propagation across walls is much worse than 2.4GHz propagation, therefore only nearby neighbors may interfere with your access point, and on 5GHz there are enough non-overlapping channels to choose an alternate channel avoiding this interference completely.
Much more common is 80MHz channel, and even then, there is a distinction between the "main" 40MHz and the "secondary" 40MHz sub-channel. If you monitor the packets over the air, you'll notice that the secondary channel is much less occupied - all mobile devices switch to 40MHz to save considerable power, unless downloading a big file.
In the US, there are 6 available 80MHz channels, which is more than needed (think 4-color graph coloring).
I agree it should be much better than 2.4 GHz -- only 2 of my neighbors are stronger than -70 dBm -- but more channels would be nice.
(I guess 6 GHz could help, but then I'll have to upgrade all my clients, too.)
That isn't realistic for most people, though.
This is where the mesh units might come in handy - they are managed and configured in a single place, and can work together nicely - but of course you have to set them up with wire backhaul instead of in the "mesh mode".
I understand your rationale, and I also like the convenience of wifi when it works. That being said, if you can wire "that thing" (whatever it is), just do it. One afternoon of wiring and it will work great for the rest of time. The trade-off is definitely worth it in my experience.
Even recent low-end phones sometimes have just 2.4GHz Wi-Fi. Even if all you devices support 5 GHz, running 5GHz-only wi-Fi may cause problems to your guests.
This is an extremely bad idea in some countries, or if you’re not careful which channel you select. In nearly all countries some of the 5ghz channels are shared with RADAR and require DFS. In some countries (eg Germany) all of the channels do.
The idea behind DFS is to detect radar pulses and then shut down the network for a period of time, this can be triggered by radar or interference.
My rules of thumb for DFS:
(1) If you can use a DFS channel - use it. it's usually less congested than the regular channels
(2) If you're unlucky and DFS is triggered once per month or more, try repositioning the AP or changing the AP model to one less sensitive to radar pulses
(3) If no luck - use one of the non-DFS channels. Every country has those. In Germany, that would be channels 32-48 or 149+
Note that DFS (5.4GHz) is for weather radar. In some countries it's operated only in winter season. You may want to use different channels in summer and in winter :)
Here are some tips I collected along the way:
1. 2.4GHz sucks, use Bluetooth when possible. But Bluetooth sucks, too.
2. You want to have fewer Bluetooth devices when possible, reduce the number of simultaneous connections. For example, I only use my Bluetooth mouse for gaming, and trackpad for the rest, so I would disconnect mouse when I'm not playing video games.
3. You want to use some categories of Bluetooth devices first. For example, Bluetooth keyboards usually have great experiences so you may want to make your keyboard wireless first. Bluetooth mice and trackpads are Okayish, but not as great. Bluetooth headphones and speakers are usually bad. If you really want to use Bluetooth for Audio, sometimes devices from the same company may have better connections: Beats Studio seems to have a much better connection than Sennheiser and B&O on Mac.
After all, I still love the idea of wireless because it is nice on paper. It is mostly the de facto protocols and implementations that are terrible.
Huh? Bluetooth is 2.4GHz.
The lag made it unplayable. We plugged in some ethernet cables and it ran fine after that.
It performs much better than wireless alternatives.
If I hide enamel coated filament just beneath my lawn is that going to be good for a one wire bus to work? Does it even need to be insulated?
This is the implementation: https://github.com/gioblu/PJON/tree/master/src/strategies/So...
This is the source code: https://github.com/gioblu/PJON/blob/master/src/strategies/So...
It is extremely rugged, it works much better than any alternative I have ever tried. Ah, it is free and open-source ;)
In case of wireless all you need is ESP32, relay and a power supply. And all of that together will cost less than 10$.
Can you easily (cheaply) use the already existing power cables for PJDL? I know there are network repeaters that use power cables to transfer data, but I have no idea how much it would cost in IoT scenario. Do you have some examples?
Does it have anything in common with 'single-wire earth return', in power-distribution?
Though, I don't know if insulation helps in the case of lightning. Apparently with Ethernet, special grounding thingies help.
In general even the most rugged lightning/surge protection only decrease the likelihood of protected equipment being vaporized.
it can be a nightmare.
A twisted pair seems nice to me because you can use differential signaling and not have to ground your devices e.g. 2 wire RS485.
What if you just connect the antenna wires of all the devices together?
That's what a single wire protocol does.
When you have an object with static electricity and put it in contact with a different object, there will be a transient current, even if the circuit is not closed. The objects basically work as poor-man's capacitors.
Even though the name is "static electricity", when the objects come into contact, it's actually more dynamic than normal, because there is no current between two objects of fixed and different voltage. The more current that flows, the lesser the voltage difference becomes, in an exponential fashion.
This kind of single wire protocol operates solely based on these kind of transient currents. Because the frequency is so high, very little current flows during one period of the waveform, so the receiver will maintain constant potential, as if it was grounded.
That it isn't actually grounded means the signal becomes high-pass filtered. In particular, the DC-component must be completely blocked as it would lead to charge buildup with nowhere to go.
It’s not just video calls. My backup is done to a Synology NAS, upstairs. That’s all wired.