4 ms·

And you think the same problem wouldn't exist with 6ghz? It will be as crowded as 5

xoa 1y ago
>And you think the same problem wouldn't exist with 6ghz? Yes. Probably because they have some basic grasp of electromagnetic reality, which perhaps you might consider studying a bit before forming strong opinions? >It will be as crowded as 5 Physically impossible. 6 GHz simply does not have the material penetration, that's the point. Having way more raw bandwidth on tap, all available all the time without DFS plopped in the middle too, is also extremely helpful of course too. But the signal just not traveling as far and not going through walls well is the core thing. You don't need special effort EM shielding for it so much, bulk material will do it. And WAPs are cheap now. Having a higher number of smaller cells has been best practice for awhile already, and 6 GHz takes that much further.
- matt-p 1y ago
  We said this about 5Ghz when that came out. I'm sorry to say it's not true, there's more than enough spectrum in 5Ghz if properly managed and co-ordinated. I would rather fix that first. Why is it we can run WiFi for thousands of developers in one room/venue just fine but people living in apartment blocks are apparently struggling with a dozen devices per 60sqm apartment? APs using 160MHZ channel widths with 1 or 2 spatial streams because it's cheaper than 80MHZ channels and 3 or 4 spatial streams. Absolutely crap 'auto' channel selection, too high a power (because cheaper than a second AP), poor AP placement and inappropriate channel width (in an apartment block 40Mhz per AP might be optimal).
  - bethekidyouwant 1y ago
    It’s faster to fix this by moving to 6 GHz than retrofitting everybody’s 5 GHz routers.
    - matt-p 1y ago
      Moving to 6Ghz will require a new router. Realistically it's even worse because it's not moving to 6Ghz it's adding 6Ghz. Now each AP has to have 3 radios, 2.4ghz for compatibility, 5ghz for compatibility while still maintaining some performance and 6ghz for performance. What about when 6ghz is full of the same crap, do we add 7ghz?
  - xoa 1y ago
    >We said this about 5Ghz when that came out. To the extent "we" said this, we were absolutely, 100% correct. 5 GHz was and remains a massive improvement over 2.4 GHz, exactly as hoped. But in the decade and half since demands have gone up a lot. 6 GHz will be even better as it propagates even worse and has even more bandwidth available, while human population density won't change. >I'm sorry to say it's not true, there's more than enough spectrum in 5Ghz if properly managed and co-ordinated I'm sorry to say you're wrong, there is not remotely enough usable spectrum, and that's regardless of "proper management" which in reality is completely contrary to the practical reality local networks in a setting with a high density of independent people/organizations. >I would rather fix that first. That's nice. Most fortunately you are not in charge. >Why is it we can run WiFi for thousands of developers in one room/venue just fine That's a low demand situation under the control of a single entity where people are going to be understanding of compromise given the special circumstances, unlike in home or business. >but people living in apartment blocks are apparently struggling with a dozen devices per 60sqm apartment? You're wondering why might want their own independent LANs in their own homes? Well, I'm sure you can think of one or two reasons if you put your mind to it.
  - Gigachad 1y ago
    Most of 5Ghz is unusable because of DFS. In Australia, only 2 out of the 6 80mhz channels are usable. 6 Ghz has 6 of them completely usable today, with possibly more on the upper end usable in the future.
- lxgr 1y ago
  > 6 GHz simply does not have the material penetration, that's the point. Really? Is there something special about 6 GHz absorption through common construction materials? Otherwise, why would a 20% higher frequency be that much worse?
  - wkat4242 1y ago
    The main 5G frequency in Europe is 3.5Ghz so it's about twice as high.
    - lxgr 1y ago
      True, but almost completely unrelated to Wi-Fi.
ux266478 1y ago
Higher frequency, more bandwidth.
craftkiller 1y ago
There is significantly more spectrum available for wifi in the 6ghz band than in the 5ghz band, so even if we moved everything off of 5ghz and ignored the attenuation benefits of 6ghz, we'd still have significantly less congestion on 6ghz than 5ghz wifi. In my apartment, my laptop can see ~50 wifi networks, we need some spectrum elbow-room to spread that out. https://en.wikipedia.org/wiki/List_of_WLAN_channels https://en.wikipedia.org/wiki/List_of_WLAN_channels
- friendzis 1y ago
  "significantly" is 50%, but still nowhere near 1 channel per network even if all networks cooperated.
  - throw0101d 1y ago
    > "significantly" is 50%, but still nowhere near 1 channel per network even if all networks cooperated. 5Ghz has 500Mhz worth of total bandwidth, while 6Ghz has (in US/CA, and hopefully in EU eventually) 1200Mhz. That's over double. 6Ghz has more 160Mhz channels (7) than 5Ghz has 80Mhz channels (6).
    - friendzis 1y ago
      5Ghz has 9 non-SRD 40Mhz channels (4 x 80MHz), while 6GHz has 12 (3 x 160MHz).
    - PunchyHamster 1y ago
      so in reality you're 1 channel up over 6GHz becasue people are not buying wifi6 router to stay on 80MHz channel
      - craftkiller 1y ago
        They are if they're in an urban environment (which is 80% of the population of the United States). Maximizing your channel width only makes sense in suburban/rural areas. You can get a much more reliable connection by using a smaller channel width. https://news.ycombinator.com/item?id=45542444 https://news.ycombinator.com/item?id=45542444
topspin 1y ago
> And you think the same problem wouldn't exist with 6ghz? Wi-Fi 6E and later standards that unlock 6 GHz are designed to mitigate contention through several dynamic power management and multiplexing capabilities: TWT, MLO, OFDMA, improved TPC, etc. While these things aren't somehow inherent to 6 GHz, the 6 GHz band isn't crowded with legacy devices mindlessly blasting the spectrum at max power, so it is plausible that 6 GHz Wi-Fi will perform better in dense urban environments. The higher frequency also contributes because attenuation is substantially greater, although in really dense, thin-walled warrens that attenuation won't solve every problem. I know if I had noisy Wi-Fi neighbors interfering with me, the few important Wi-Fi only devices I have would all be on at least 6E 6 GHz by now, not only because 6 GHz has fewer users, but also in the hope that ultimately, when the users do appear, their devices will be better neighbors by design. I don't actually have that problem, however. The nearest 5 GHz AP I can actually see (that isn't mine,) in Kismet (using rather high gain antennas) is -96 dB, and my actual APs hardly ever see those at all. I've yet to actually detect a 6 GHz device that isn't mine. I known there are a few because the manufacturers and model numbers of many APs are visible, but between the inherent attenuation and the power level controls, I don't see them.
throw0101d 1y ago
> And you think the same problem wouldn't exist with 6ghz? Yes. 5 Ghz has 12x 40MHz channels, 6 Ghz has (in the US/CA where it is basically 'fully unlocked' for Wifi) 29x 40Mhz channels. It's the difference between 500Mhz worth of total bandwidth and 1200Mhz: over double. * https://www.everythingrf.com/community/what-frequency-band-does-wi-fi-6e-use https://www.everythingrf.com/community/what-frequency-band-d... * https://spectrum.potatofi.com https://spectrum.potatofi.com * http://www.potatofi.com/posts/spectrum-viewer/ http://www.potatofi.com/posts/spectrum-viewer/ And given attenuation increases as frequencies goes up, your neighbours' signals won't travel as far as the lower frequency bands, which helps with localization. We just have to hope that vendors don't ship 80 or 160Mhz channels by default for residential devices, which will potentially eat up bandwidth (though Wifi 7 makes Punctured Transmission / Preamble Puncturing mandatory, where previously it was optional). Though even if they do, 6Ghz has more 160Mhz channels than 5Ghz has 80Mhz ones (7 vs. 6). A 1x1 40Mhz using 802.11ax will give you a max PHY of 287Mbps: * https://www.intel.com/content/www/us/en/support/articles/000005725/wireless/legacy-intel-wireless-products.html https://www.intel.com/content/www/us/en/support/articles/000... * https://superuser.com/questions/1619079/what-is-the-maximum-connection-speed-and-channel-width-on-each-2-4-ghz-and-5-ghz https://superuser.com/questions/1619079/what-is-the-maximum-... Even if you half that, it's (IMHO) probably sufficient for the vast majority of online activities. And if you have a 2x2 client you double it anyway.
- Gigachad 1y ago
  I don't see any reason not to use 80Mhz. Many home internet connections are faster than 40Mhz can provide.
  - throw0101d 1y ago
    > Many home internet connections are faster than 40Mhz can provide. Define "many". The US average, as of 2023, seems to be ≤150Mbps: * https://www.opensignal.com/2023/05/23/usa-fixed-broadband-experience-report-may-2023 https://www.opensignal.com/2023/05/23/usa-fixed-broadband-ex... Cloudflare has data of more sustained-use bandwidth that shows lower numbers: * https://radar.cloudflare.com/quality https://radar.cloudflare.com/quality 1x1 40Mhz = 287Mbps PHY ~ 143Mbps realistic ~ 100Mbps probable. Double that for 2x2 40Mhz: 200 Mbps. Certainly some connections may need more, and is the reason for >40 Mhz options, but I'm not entirely convinced one of those should be the out-of-box default. This less of a concern in 6 Ghz because there are many more channels, but this is what the story is all about: how is that frequency band going to be allocated? In US/CA all of it basically went to Wifi, and that gives folks more options, even in more densely populated areas.
exasperaited 1y ago
But perhaps all that matters is, does a band support more connections than there would be apartments or businesses in range? For some people in dense apartment complexes, 5Ghz evidently does not. But if 6Ghz supports twice as many, that might be a limit that gets hit by vastly fewer "localities" like that. (Student wifi hotspots in a large lecture theatre, that's another problem entirely!)

xoa 1y ago

>And you think the same problem wouldn't exist with 6ghz? Yes. Probably because they have some basic grasp of electromagnetic reality, which perhaps you might consider studying a bit before forming strong opinions? >It will be as crowded as 5 Physically impossible. 6 GHz simply does not have the material penetration, that's the point. Having way more raw bandwidth on tap, all available all the time without DFS plopped in the middle too, is also extremely helpful of course too. But the signal just not traveling as far and not going through walls well is the core thing. You don't need special effort EM shielding for it so much, bulk material will do it. And WAPs are cheap now. Having a higher number of smaller cells has been best practice for awhile already, and 6 GHz takes that much further.

matt-p 1y ago
We said this about 5Ghz when that came out. I'm sorry to say it's not true, there's more than enough spectrum in 5Ghz if properly managed and co-ordinated. I would rather fix that first. Why is it we can run WiFi for thousands of developers in one room/venue just fine but people living in apartment blocks are apparently struggling with a dozen devices per 60sqm apartment? APs using 160MHZ channel widths with 1 or 2 spatial streams because it's cheaper than 80MHZ channels and 3 or 4 spatial streams. Absolutely crap 'auto' channel selection, too high a power (because cheaper than a second AP), poor AP placement and inappropriate channel width (in an apartment block 40Mhz per AP might be optimal).
- bethekidyouwant 1y ago
  It’s faster to fix this by moving to 6 GHz than retrofitting everybody’s 5 GHz routers.
  - matt-p 1y ago
    Moving to 6Ghz will require a new router. Realistically it's even worse because it's not moving to 6Ghz it's adding 6Ghz. Now each AP has to have 3 radios, 2.4ghz for compatibility, 5ghz for compatibility while still maintaining some performance and 6ghz for performance. What about when 6ghz is full of the same crap, do we add 7ghz?
- xoa 1y ago
  >We said this about 5Ghz when that came out. To the extent "we" said this, we were absolutely, 100% correct. 5 GHz was and remains a massive improvement over 2.4 GHz, exactly as hoped. But in the decade and half since demands have gone up a lot. 6 GHz will be even better as it propagates even worse and has even more bandwidth available, while human population density won't change. >I'm sorry to say it's not true, there's more than enough spectrum in 5Ghz if properly managed and co-ordinated I'm sorry to say you're wrong, there is not remotely enough usable spectrum, and that's regardless of "proper management" which in reality is completely contrary to the practical reality local networks in a setting with a high density of independent people/organizations. >I would rather fix that first. That's nice. Most fortunately you are not in charge. >Why is it we can run WiFi for thousands of developers in one room/venue just fine That's a low demand situation under the control of a single entity where people are going to be understanding of compromise given the special circumstances, unlike in home or business. >but people living in apartment blocks are apparently struggling with a dozen devices per 60sqm apartment? You're wondering why might want their own independent LANs in their own homes? Well, I'm sure you can think of one or two reasons if you put your mind to it.
- Gigachad 1y ago
  Most of 5Ghz is unusable because of DFS. In Australia, only 2 out of the 6 80mhz channels are usable. 6 Ghz has 6 of them completely usable today, with possibly more on the upper end usable in the future.
lxgr 1y ago
> 6 GHz simply does not have the material penetration, that's the point. Really? Is there something special about 6 GHz absorption through common construction materials? Otherwise, why would a 20% higher frequency be that much worse?
- wkat4242 1y ago
  The main 5G frequency in Europe is 3.5Ghz so it's about twice as high.
  - lxgr 1y ago
    True, but almost completely unrelated to Wi-Fi.

ux266478 1y ago

Higher frequency, more bandwidth.

craftkiller 1y ago

There is significantly more spectrum available for wifi in the 6ghz band than in the 5ghz band, so even if we moved everything off of 5ghz and ignored the attenuation benefits of 6ghz, we'd still have significantly less congestion on 6ghz than 5ghz wifi. In my apartment, my laptop can see ~50 wifi networks, we need some spectrum elbow-room to spread that out. https://en.wikipedia.org/wiki/List_of_WLAN_channels https://en.wikipedia.org/wiki/List_of_WLAN_channels

friendzis 1y ago
"significantly" is 50%, but still nowhere near 1 channel per network even if all networks cooperated.
- throw0101d 1y ago
  > "significantly" is 50%, but still nowhere near 1 channel per network even if all networks cooperated. 5Ghz has 500Mhz worth of total bandwidth, while 6Ghz has (in US/CA, and hopefully in EU eventually) 1200Mhz. That's over double. 6Ghz has more 160Mhz channels (7) than 5Ghz has 80Mhz channels (6).
  - friendzis 1y ago
    5Ghz has 9 non-SRD 40Mhz channels (4 x 80MHz), while 6GHz has 12 (3 x 160MHz).
  - PunchyHamster 1y ago
    so in reality you're 1 channel up over 6GHz becasue people are not buying wifi6 router to stay on 80MHz channel
    - craftkiller 1y ago
      They are if they're in an urban environment (which is 80% of the population of the United States). Maximizing your channel width only makes sense in suburban/rural areas. You can get a much more reliable connection by using a smaller channel width. https://news.ycombinator.com/item?id=45542444 https://news.ycombinator.com/item?id=45542444

topspin 1y ago

> And you think the same problem wouldn't exist with 6ghz? Wi-Fi 6E and later standards that unlock 6 GHz are designed to mitigate contention through several dynamic power management and multiplexing capabilities: TWT, MLO, OFDMA, improved TPC, etc. While these things aren't somehow inherent to 6 GHz, the 6 GHz band isn't crowded with legacy devices mindlessly blasting the spectrum at max power, so it is plausible that 6 GHz Wi-Fi will perform better in dense urban environments. The higher frequency also contributes because attenuation is substantially greater, although in really dense, thin-walled warrens that attenuation won't solve every problem. I know if I had noisy Wi-Fi neighbors interfering with me, the few important Wi-Fi only devices I have would all be on at least 6E 6 GHz by now, not only because 6 GHz has fewer users, but also in the hope that ultimately, when the users do appear, their devices will be better neighbors by design. I don't actually have that problem, however. The nearest 5 GHz AP I can actually see (that isn't mine,) in Kismet (using rather high gain antennas) is -96 dB, and my actual APs hardly ever see those at all. I've yet to actually detect a 6 GHz device that isn't mine. I known there are a few because the manufacturers and model numbers of many APs are visible, but between the inherent attenuation and the power level controls, I don't see them.

throw0101d 1y ago

> And you think the same problem wouldn't exist with 6ghz? Yes. 5 Ghz has 12x 40MHz channels, 6 Ghz has (in the US/CA where it is basically 'fully unlocked' for Wifi) 29x 40Mhz channels. It's the difference between 500Mhz worth of total bandwidth and 1200Mhz: over double. * https://www.everythingrf.com/community/what-frequency-band-does-wi-fi-6e-use https://www.everythingrf.com/community/what-frequency-band-d... * https://spectrum.potatofi.com https://spectrum.potatofi.com * http://www.potatofi.com/posts/spectrum-viewer/ http://www.potatofi.com/posts/spectrum-viewer/ And given attenuation increases as frequencies goes up, your neighbours' signals won't travel as far as the lower frequency bands, which helps with localization. We just have to hope that vendors don't ship 80 or 160Mhz channels by default for residential devices, which will potentially eat up bandwidth (though Wifi 7 makes Punctured Transmission / Preamble Puncturing mandatory, where previously it was optional). Though even if they do, 6Ghz has more 160Mhz channels than 5Ghz has 80Mhz ones (7 vs. 6). A 1x1 40Mhz using 802.11ax will give you a max PHY of 287Mbps: * https://www.intel.com/content/www/us/en/support/articles/000005725/wireless/legacy-intel-wireless-products.html https://www.intel.com/content/www/us/en/support/articles/000... * https://superuser.com/questions/1619079/what-is-the-maximum-connection-speed-and-channel-width-on-each-2-4-ghz-and-5-ghz https://superuser.com/questions/1619079/what-is-the-maximum-... Even if you half that, it's (IMHO) probably sufficient for the vast majority of online activities. And if you have a 2x2 client you double it anyway.

Gigachad 1y ago
I don't see any reason not to use 80Mhz. Many home internet connections are faster than 40Mhz can provide.
- throw0101d 1y ago
  > Many home internet connections are faster than 40Mhz can provide. Define "many". The US average, as of 2023, seems to be ≤150Mbps: * https://www.opensignal.com/2023/05/23/usa-fixed-broadband-experience-report-may-2023 https://www.opensignal.com/2023/05/23/usa-fixed-broadband-ex... Cloudflare has data of more sustained-use bandwidth that shows lower numbers: * https://radar.cloudflare.com/quality https://radar.cloudflare.com/quality 1x1 40Mhz = 287Mbps PHY ~ 143Mbps realistic ~ 100Mbps probable. Double that for 2x2 40Mhz: 200 Mbps. Certainly some connections may need more, and is the reason for >40 Mhz options, but I'm not entirely convinced one of those should be the out-of-box default. This less of a concern in 6 Ghz because there are many more channels, but this is what the story is all about: how is that frequency band going to be allocated? In US/CA all of it basically went to Wifi, and that gives folks more options, even in more densely populated areas.

exasperaited 1y ago

But perhaps all that matters is, does a band support more connections than there would be apartments or businesses in range? For some people in dense apartment complexes, 5Ghz evidently does not. But if 6Ghz supports twice as many, that might be a limit that gets hit by vastly fewer "localities" like that. (Student wifi hotspots in a large lecture theatre, that's another problem entirely!)