I wonder about the approach Purism took with the Librem 5. The company chose to do everything all at once by building a new smartphone OS and a new hardware supply chain. For a customer receiving a Librem 5 today, you’re getting an unfinished operating system and rough, gen-one open source hardware. That’s a bunch of compromises to accept for $750. A more reserved approach would have been to build an open source GNU/Linux-based OS on closed source hardware first and then make the difficult jump to custom hardware when the OS was in a more complete state.
The Librem 5 is a tough sell, even for people who value the open source nature of the device. That’s simply too much money for such an outdated, unfinished device.
I wish I could do more to encourage projects like this because the world could really benefit from new alternatives to spur innovation and choice. I really want an open phone with open hardware, however at the $750 and $800 price going forward I can’t afford it. I get that the company has salaries to pay, but realistically I don’t know how these phones can become viable in the face of cheaper competitors. How do you compete with android, which is free? Considering that bigger well funded companies have failed to make a dent in this market, I’m sad to say I don’t believe the outcome here will be any different 🙁
–How do you compete with android, which is free?–
Except android the way users expect is not exactly free either to hardware vendors.
https://www.channelnewsasia.com/news/business/huawei-google-android-updates-what-you-need-to-know-11548566
Basically this huawei/google thing. Does mean there need to be another alternative to android that is not restricted by these problems.
–Considering that bigger well funded companies have failed to make a dent in this market, I’m sad to say I don’t believe the outcome here will be any different–
Timing is important. Bigger well fund companies were not trying to compete against Android when it was having government regulator be issue to supply. So this is the right time if you are wanting to fire a competitor at Android.
750-800 dollar price tag of Librem 5 I do find over the top.
https://store.pine64.org/?product=pinephone-braveheart-limited-edition-linux-smartphone-for-early-adaptor
Pinephone I see as the more reasonably price option.
There is a reason why you still might go librem 5 ahead of everything else and it will all depend how the USA/China relationship goes. Let say 5/6G phone network due to governments fighting with each other fragments. The Librem5 m.2 card so you can have a collection of m.2 cards that you can change when you fly between countries so your phone still works and you don’t have to be transferring your data between phones.
So depending on what way the USA/China relationship will depend if you want a pinephone or a librem 5.
There is not another phone on the market where you can change the 5G modem and the librem5 $750-$800 starts not looking as bad that the break down could result in needing like 4 phones while travailing world wide if you don’t go librem5.
Then remember you have a limit on amount of lithium ion batteries you are allowed to bring on to different flights. So each one of those 4 phones take 1 battery count out your allowed 8 by some carriers(that 8 includes any battery pack/spare batteries as well). So there is a practical problem here the librem5 design solves that none of the other phones solve. If you are rich enough to be flying this much that this is a problem the librem5 price tag to solve it is not that bad. Please note there is no count limit on the number of m.2 cards or sim cards you are allowed to limit other than weight. Remember that is the other part of the problem 1 librem5 + 4 sim cards + 4 m.2 modems and a spare battery is lighter than 2 cheap smartphones.
There is a possible place in the market for the Librem 5. With this possible place you have to start making the device and pray the market is there once you are ready for mass production.
It took me a while to see where in the market Librem 5 fitted thinking the market is conditional on how USA/China mess goes.
oiaohm,
You’re preaching to the choir. The white house executive orders barring US companies from doing business with certain chinese companies isn’t google’s fault though and if I recall it applies to anyone in the US, so I think that even purism is barred from doing business with huawei.
That’s an interesting feature, but not one that I would use. Heck 5G is virtually worthless where I live because I don’t live near any 5G deployments. I haven’t been to a 5G city (nor do I have a 5G phone), but even reviews said they got great speeds until they walk behind a tree or went into a restaurant where there was no coverage. Maybe it’s still too early to judge the technology, but I have a feeling this will be a common refrain for 5G operating at high “line of sight” frequencies. From what I’ve read, some carriers plan on using 5G on low frequencies, like tmobile’s 600MHz, which will get better coverage, but it’s very limited bandwith compared to what 5G’s promoters have been touting.
https://www.lifewire.com/5g-spectrum-frequencies-4579825
If anyone here has hands-on experience with 5G, it would nice to hear about how well it works where you are. It’d be nice to have an osnews article about this for that matter!
-If anyone here has hands-on experience with 5G, it would nice to hear about how well it works where you are. It’d be nice to have an osnews article about this for that matter!-
Problem is 5G is not one thing.
28Ghz-39Ghz that is the one that a 4 sheets of A4 paper can block the signal even in lab with no interference other than the 4 sheets of A4 paper. Verizon is going to be horrible the provider of their hardware has transparent radio frequency switching switch. Yes the tower need to be able todo this as well as the phones modem.
AT&T is doing like Telstra here in Australia. AT&T is in fact using the hardware that was designed in Australia in Telsra labs as a joint project between Ericsson and Telstra. The first 5G connection was done using the Telstra Ericsson tech with a modem the size of a bar fridge on wheels(because the case was a in fact a bar fridge). Ericsson job was to turn the bar fridge made by Telstra engineers into a silicon chip.
Telstra design you have 28Ghz-39Ghz plus another area somewhere 3.5Ghz-4Ghz and finally take a band out of 2G-4G band allocations as well . This way when you get those 4 sheets of paper in way that block 28Ghz-39Ghz your drop 1Gbps+ back to 100Mbps without any connection drop and can reconnect to the faster again no connection drop just speed increase.
Yes if the 3.5Ghz cannot reach your phone drop back to the 2G-4G radio freq and use that with 5G data and protocols . This is transparent radio frequency switching that is part of the Telstra design from the get go so your connection does not just drop out because stuff gets in way. Stuff getting in way with the Telstra design should just degrade performance.
There have been serous problems here in Australia with the first generation 5G modems in phones overheating when working in the 3.5-4Ghz range only. Lot of those phones have had to be firmware locked only to use 4G not to cook self. Highspeed coverage if you cannot see the tower you are basically screwed. Mid range using 3.5-4Ghz decent enough speed. Back in the 2G-4G channels better network congestion handling using 5G protocols but about the same max transfer speeds as 4G just getting them way more often.
–Heck 5G is virtually worthless where I live because I don’t live near any 5G deployments. —
I would not be too sure long term. I would be getting to know the towers near you and their planned upgrades. How many 2G-4G bands they take over to 5G could make your old 2G-4G phones worthless if you don’t have enough bands left.
The Telstra 5G protocol operating in 2G-4G bands can maintain signal stable for voice at what would basically be classed as O bar where you could only send SMS with 2G-4G protocols.. Now that is if you have a phone that is 5G with a modem that supports that as well as carrier using Telstra designed towers.
Remember huawei is one of the biggest providers of 5G modems and there design of 5G does not match the Telstra one. Welcome to the future cause of modem hell if they don’t get it sorted out. Yes due to the fact 5G can consume all your 2G-4G radio space you cannot bet on your 2G-4G phones working. Australia got rid of 2G completely a few years ago to make space in the 2G-4G bands for 5G.
There are 4 different parties designing 5G solutions and their protocols are not 100 percent compatible with each other. That need to have 4 modems is where we are heading if nothing changes with 5G/6G. Yes the USA government say you cannot work with overseas companies in china on 5G protocol is part of the disaster making this split worse.
If we keep down this route we are on Librem 5 might be a very good idea to anyone who is travelling. This would not be buy a Librem 5 now it would be buy one like it in 4 years time if things keep on going the way they are hopefully by then Librem 5 has sorted out the software trouble. Really I do hope we don’t remain on the same route that some how common sense wins in the USA and other places so work on unified radio protocols can get back to normal.
oiaohm,
Yes, I think many of us understand the theory. I was just curious if anyone had hands on experience testing it.
This is what my carrier has said:
https://www.fiercewireless.com/5g/t-mobile-says-5g-mmwave-deployments-will-never-scale
So, we won’t be getting the high speed 5G mmwave here. Not only do I think it would be too expensive and impractical to roll out in the subburbs, but tmobile has specifically stated that it would only be deploying it in densely populated urban areas. So I find it unlikely that I’ll ever see it here. In all likelihood what will eventually happen is our existing towers will transition to 5G protocols but we won’t experience the higher bandwidth frequencies. Not that it wouldn’t be appreciated, as things stand today I cannot stream video on my phone, it stutters too much.
It’s not the first time our area has been overlooked, verizon fios stopped it’s fiber optic internet rollout before reaching us. That hurt us a lot because it left us stuck indefinitely with the cable monopoly for internet. At least I’m better off than my parents after ATT abandoned its rural subscribers and left them with no residential internet service provider whatsoever 🙁 I guess they should count themselves fortunate to be next to a 4G cell tower because most of their township doesn’t have cell service either. There’s always satellite I suppose.
I have had the fun of using 3.5-4Ghz 5G stuff here. This is mid spectrum 5G this is higher freq than 2G-4G covers. 2G-4G is sub 1.5Ghz your low bands.
millimeter wave (mmWave)=28 GHz and 39 GHz.
–I think it would be too expensive and impractical to roll out in the subburbs, but tmobile has specifically stated that it would only be deploying it in densely populated urban areas. —
This is true for the mmWave.
This not true for 3.5-4Ghz and the low band 5G using converted 2G-4G bands. 2G-4G bands to 5G conversion in the low bands is just change the controller in the tower nothing else to really check. Some the existing Aerial types already on the towers 2G-4G are able to be straight up used for the 3.5-4Ghz usage. This can basically pull the old 2G-4G module out tower and insert 5G module for the low and mid range bands. Telstra 5G module is able to provide 3G, 4G and 5G in the lower bands. Please note totally unable to-do 2G. These modules can die for many normal reasons and have to be replaced.
Telstra 5G system is designed to be cost effective to roll out reusing existing towers in areas where max 5G speed is not on offer. 3.5-4Ghz band still gives you way faster data rates than 4G can offer on a good day due to mid band being faster. Low band ie the exist 2G-4G bands 5G in there can give more stable provide of packets.
Early phones with Telstra 5g running at 3.5-4Ghz band you where required to hold a ice pack on back to maintain connection to prevent the complete phone from over heating. Yes that icepack made mmWave totally not functional without very careful alignment of phone/tower and icepack.
–Yes, I think many of us understand the theory. I was just curious if anyone had hands on experience testing it.–
I have had the fun little job of waking around with phone while holding a icepack testing this stuff out. The change over works fine. We cross fingers that next generation of 5G phones get the overheating problem under control. The change over between bands on the Telstra 5G based stuff works well as long as the phone is not trying to cook itself.
–Not that it wouldn’t be appreciated, as things stand today I cannot stream video on my phone, it stutters too much.–
5G protocol in the lower freq without increase speed can at times help with stutter because of the way its allocating packets to and from the tower. Look up a thing called bufferbloat. The old 2G-4G suffer from it this does not help you when you are trying to watch a video.
oiaohm,
The problem is when you get too high into the “mid” range, like 4GHz, that’s twice the frequency of older bands. Unless you transmit with more directional antennas/power, which is up to the FCC, you end up with less range than before. 5GHz wifi has the same issue, it’s ability to reach all rooms throughout the house is degraded from 2.4GHz, requiring more access points. These higher frequencies have more capacity but less range :-/
Theories aside, I wanted more of an idea how well end users feel it works in practice, but no one seems interested in responding, oh well.
I’m very familiar with buffer bloat, it’s a big problem for realtime applications that are sensitive to jitter & latency. But it’s not really problematic for streaming videos that end up being buffered by the browser anyways. In this case, raw throughput is much more important than jitter.
Buffer bloat is a tradeoff, the reason buffers exist in the first place is because they actually improve throughput by maximizing bandwidth utilization, but obviously buffers are bad for latency. Assuming your ISP has sufficient bandwidth to handle loads on their end, you can reverse the effects of buffer bloat by rate limiting your connection. For example, my connection used to have a “C” score in this bufferbloat test (and boy could I feel it), now it’s “A+”!
https://imgur.com/b1oltJ7
I found a review for tmobile’s 5G service in NYC.
https://www.youtube.com/watch?v=VeBREs4O8kM
If that’s anything to go by, it’s not bad, but not particularly revolutionary either. I wish cnet had used much larger sample size, but alas their simple test is more or less is inline with theoretical predictions that 5G on tmobile’s low band frequencies won’t add large performance boosts. Even so it’s likely to help in congested areas by adding more overall capacity.
It’ll be interesting to see where this goes in the next decade. I may not be representative of the market, but speaking for myself I think consistent coverage is more important than spotty experiences with insanely fast bandwidth.
What that video more closely. Outside 4G vs Low band 5G give about the same. Inside 4G vs low band 5G the 5G has faster transfer even if both appear to be full bars. Protocol changes 5G is able to use better form forwards error correction. 5G performance shape is more cliff like as in full speed for the band for most of the signal strength range unless there is really bad interference.
–Unless you transmit with more directional antennas/power, which is up to the FCC, you end up with less range than before. 5GHz wifi has the same issue, it’s ability to reach all rooms throughout the house is degraded from 2.4GHz, requiring more access points. These higher frequencies have more capacity but less range :-/–
Please note 5Ghz vs 2.4Ghz wifi has the same level of error correcting. 4G vs 5G the 5G has a better form of forwards error correction so a good percentage of the 5G at mid range area of coverage can most match the 4G coverage outdoor coverage at the same broadcast strengths. The area of coverage of mid-range a lot would be expecting to be way less than what it is.. Yes 5G mid range will work well outdoors inside building 5G midrange will be stuffed at that point 4G performance has already degraded. mmwave I don’t see that much advantage in needing line of sight is a killer.
A provider not providing at least 5G lowband I would say not worth it long term. 5G lowband is true upgrade when you are walking in buildings with random interference sources. 5G lowband and midband from testing is decent. mmware is more I want to show off 1Gbps high speed connections to friends after a lot of messing around.
Ideal from my point of view is you want a provider with both low and mid range 5G. If they skip out on upgrading towers to have mmwave who really cares. Most of the time you don’t have line of sight to the towers to take advantage of mmwave anyhow. Midrange needs reasonable line of sight that stuff can be in the middle and as long as it fairly thin you will have a decent signal and get the full speeds so outside will mostly work well.
I do agree with the video I would not be buying a 5G phone now. Lot of the tests saying that 4G LTE and 5G lowband are about the same are not testing inside buildings and when you do they come very much different and it all due to how having better forward error correction means less packet loss and needing a lower grade of signal to get packet though.
oiaohm,
That’s more or less where we seem to be at. Low band 5G gives us more of the same. Mid ranges adds some more wiggle room and should help with congestion. The high frequencies with dramatic throughput are only viable for those who are standing/sitting in line of sight facing the right direction.
I don’t deny that, and I think we agree on what to expect, but it’s sure to disappoint consumers who were expecting 5G to be revolutionary.
Providence rode island is supposed to be a robust test city for 5G technology with exceptional coverage. Yet even here, reviewers experience spotty 5G coverage.
http://www.youtube.com/watch?v=OR3Oq0j5nHQ
https://venturebeat.com/2019/08/28/verizons-5g-in-providence-unreliable-but-fast-when-it-works/
It seems that one of 5G’s selling point – reliable ultri-high speeds everywhere – is problematic even in verizon’s test market. Regardless I suspect it is indicative of what to expect for deployments across the country. Solving this this is expensive because it requires blanketing cities with significantly more access points than carriers are willing & able to install. I wonder what the limits are on the density of 5G towers.
I actually think there’s an opportunity to shift the burden of coverage away from carriers and onto consumers, similar to wifi. Restaurants, hotels, shops, businesses, homeowners, etc would likely be both willing and able to deploy networks far more thoroughly than carriers ever could. So maybe the carriers are going about it all wrong by trying to deploy it themselves. But this is a topic onto itself.
Really those tests “Providence rode island” are coming back to the telstra whitepaper that I cannot find on-line expected results.
In the Telstra white paper 5G mmwave was not intended for mobile phones alone reason why auto switching between mmwave back to mid to low bands was included in the specification because mobile phones were expected to have trouble maintaining those high speed. When you look at items with fixed aerials like lift phones it possible to make sure you always have clean line of sight now the 5G mmwave starts making sense. Or you think under developed areas could be using 5G for internet in the home again fixed aerials on roof able to be placed with basically perfect line of sight.
This plan to have 5G mmwave for line of sight stuff with high bandwidth should allow the fixed stuff to be off the mid and low bandwidths freeing up more space there. So Telstra design followed you are not going to get super speeds in phones all the time but if you get a good position you can. Mid and low ranges will be better for phones when you are not willing to find a good spot..
Alfman basically some of the USA 5G deployments look planned by idiot who did not read the Tesltra white paper listing the limitations on what you should not do. Yes pure mmwave deployments not going to work out.
5G mmwave should be a direct competitor to cable/adsl/wifi internet companies with the advantage you are reusing your existing tower locations and doing equal or less cabling. From that video they must have the power transmit level pushed way up to get though the window. Serous-ally at Telstra recommend levels there is no way mmwave will get though glass. 4-5 sheets of paper will stop it dead. Let alone something like window glass/plastic that is higher density.
oiaohm,
I agree, adding new bandwidth for low/mid ranges is beneficial, but it’s not as revolutionary as some have made it out to be.
I question even this though. Fixed installations are generally indoor installations. By the time you finish wireless surveys and install enough mmwave antennas to reach every single building that will want internet. Considering that these signals don’t penetrate indoors, every building that needs internet will now need to be fitted with new outdoor antennas in a position facing the source. I don’t know that mmwave brings all that much value to the fixed installation use case.
I’ve seen cable companies installing wires on overhead poles, and while it’s tons of work down every street, it goes quickly enough. Does replacing internet cable with 5G mmwave stations save that much work? 5G Receivers will still have to be installed on every building anyways, and I’m wondering just how many antennas would be necessary at the poles to provide coverage for a full street. Optimistically you only need one, but thinking of my own street there’s a lot of trees in the way, so either more antennas are necessary, or the city would need to pass laws to force everyone to cut down trees so that everyone on the street can get 5G to their house. It just doesn’t seem to be all that advantageous over being hard wired.
I do agree that 5G could be a good replacement for 802.11 WISP though, since all the cons of line-of-sight antennas have already been factored in and so moving to 5G seems practical in this case.
Alfman,
I bought mine (still waiting for the Evergreen batch) at $550 during crowdfunding, but I think I wouldn’t buy one at the current price tag.
Though you have to consider that this phone is user serviceable and several components can be swapped if needed, not only the battery.
That, and software updates should make it last more than current smatphones, which is another factor to consider for the price, e.g.: if this $750 phone lasts 3x what a $250 phone will last, then it doesn’t sound that much expensive (but of course the CPU, RAM and storage will be better on a future phone).
Otherwise, the PinePhone might be an alternative for you to consider, as proposed by oiaohm:
https://www.pine64.org/pinephone/
https://store.pine64.org/?product=pinephone-braveheart-limited-edition-linux-smartphone-for-early-adaptor
Due to your background (from reading your comments here), I think the PinePhone will be even more interesting to you anyway 🙂
Cheers!
Thanks for the info.
Right now I’m using a microg build of lineage-os,
https://microg.org/
It’s android minus google. microG adds alternative userspace libraries to replace google’s proprietary ones. This makes me feel a lot better about running android. A pinephone could be a logical next step, it’s certainly priced right and I’m more proficient with linux. I wish it could run android apps on top of linux. but I don’t have to make up my mind now.
Alfman,
Well, we might get Android support for the Librem 5 and PinePhone, either with an Android ROM or via anbox, which would be even better:
https://www.reddit.com/r/Purism/comments/aj1mq6/sandboxing_android_apps_on_the_librem_5/
I’ll keep my fingers crossed 🙂
Oh, and thanks for the microG tip, I had completely forgotten about it!
I’d pay almost anything for a phone that has headphone jacks, removable battery, and SD card support.
Perhaps they should have begun with a tablet …
https://en.wikipedia.org/wiki/Librem#Librem_11_tablet
So you are saying they should have done the Librem 11 before doing the Librem 5. Yes a tablet is on their road map of things todo.
I’m just saying that as a development-prototype device something with a larger device is better and it’s not like the telephone stack is working properly. You just have to look at the original iPhone prototype. Add a keyboard and you don’t even need a computer to develop and test things
Isn’t this OpenMoko and Maemo all over again? They both failed and chances are this is going to fail as well.
OpenMoko Neo1973 was continuously playing catch-up to phone hardware and was never good enough as a smart-phone.
Maemo was actually quite good, but it lost to Android and iPhone. I loved my N800 during my pre-iPhone era. It was refined, very usable and had exceptionally good hardware behind it from Nokia. It lost me at Fremantle, once it was forced towards QT and they had to rewrite most of their apps. They understood right before the iPhone launched that Finger-friendly UIs are important. The first application actually thought for usage without a stylus was Canola, a media player.
And even before that, we had the EZX Linux Phones from Motorola, that were absolutely brilliant and quite sexy for their time. I’ve had the E680i, the A1200 Ming and the ROKR E6. All of them telling the story of the iPhone a few years before it actually appeared.
Now let’s take this one at a time:
* Yes, you could probably create a baseband OS as an opensource project, but it would be an insanely huge project. Furthermore, power-management wise it would be 10 times worst than anything Qualcomm or Intel/Apple have up their sleeves since the baseband OS is tied to the hardware design. If you try untying them, you loose dramatically on performance and battery-life. This is the most important part of making an open source phone and is almost untouchable. Intel, with basically unlimited cash (which translates into developer time) failed and had to sell to Apple. Apple, who has an OS for every chip they design (think airpods) still hasn’t gotten very far. It’d estimate that 30-50 developers could do this in about 3-4 years assuming that they have direct access to the hardware design team.
* For the user facing OS you already have android, so why should you bother? Just bundle whatever you want with it and it’s done.
Love the idea of this, but they desperately need to iterate through a few cycles to refine the software and the hardware. Hope they can do that while not bringing in any revenue because I can’t see anyone buying the product as it is now.
I have a feeling though that AOSP Android would be a better starting point for a phone OS than a custom Linux build. It makes me wonder how much Google is feeling threatened because it wouldn’t take much for a bunch of Android OEM’s to get together and start funding an AOSP build with an open app store. The problem is that as long as Google’s Android + Play Services is given away, they have little incentive to do it.
Google must feel threatened to some extent because they are moving sections of Android into their proprietary Play Services layer.
Paradroid,
I’m not sure what you have in mind, but don’t we already have that though with lineage-os (and CyanogenMod before it), no? We also already have app store alternatives to google’s app store, Fdroid is a prominent one:
https://f-droid.org/
At the risk of being repetitive, take look at microg:
https://microg.org/
IMHO google’s marketshare is safe. It’s not the existence of alternatives so much as their competitive viability that matters. Google’s economies of scale are nearly untouchable, their coffers are overflowing, they have most of the mobile developer mindshare and the public at large doesn’t give a crap about buying niche FOSS products from small competing startups. Android has already beaten microsoft’s mobile platform. It’s slowly taking over apple’s marketshare too. For better or worse unfettered & unregulated capitalism has transformed industries left and right into worldwide duopolies & oligopolies. It’ll be very hard to recover from such aggressive consolidation without market intervention. Ironically enough trump’s tradewar antics may have been the catalyst for new competition from china…
https://www.digitaltrends.com/mobile/huawei-operating-system-news/
This has the potential to be a huge blow to google in the coming years given how much they stand to loose in china alone, so I imagine this worries google more than any small startup.
Google did feel threatened by a few large companies taking AOSP and messing with it. Amazon primarily. Anything Google can do to get things out of AOSP and into google play services keeps things from Amazon. It also makes developing for Android easier, or it will once enough phones are on versions that can take these updates from the Play Store. It also gives them deniability.
Bill Shooter of Bul,
Good point, I didn’t even think of amazon.
Although amazon’s fire phone was a flop, I think it was too expensive for what amounted to a sub-par clone.
https://www.cnet.com/reviews/amazon-fire-phone-review/
Maybe things are different in a parallel universe with better execution.
Not just the phone, the successful fire tablets are also Android. Probably, the most successful android tablets.
I hope they are in touch with the MNT Reform people that are building an open source laptop, using the same soc. (https://www.crowdsupply.com/mnt/reform/updates/re-introducing-reform) . I’m kind of a lurker on these kinds of things, I don’t do anything nearly as low level. I would support a project that had a path forward. I’m really close to jumping in with pine 64. That seems like the most likely to be useful and practical.
Holy crap that is a thick-ass phone (and not in a good way). O_o
I bought one. Why? Because Youtube adblock at any cost is worth it for a phone. Hell I’d pay $3,000 for a device I have root control of that has that feature. Besides, this just opens the door. Once a Linux Phone Distro exists that’s not some crappy Android fork, the way is open for the community to expand and build on it. There are already people that want to make GNUstep based clones of IOS. I say let the best paradigm win. All are welcome to come build a phone platform.
Huh? why not just subscribe to youtube?
–I’m wondering just how many antennas would be necessary at the poles to provide coverage for a full street.–
Not very many in in areas without blockages to line of sight. mmwave clean line of sight has a range of 2km as in a 2km circle around tower.
If you are placing your cables on power poles here in Australia the cables have to above ground 4.9 meters at lowest point and 0.6 meters away from Low voltage power lines. Basically this can result in impossible to use pole mount in many areas.
Cell node mounting on power pole is normally 15 meters up and higher.
Trees near power lines here as in inside 5 meters either size have to be under 5 meters with recommendation under 4. Over 5 power command will come cut it and bill you for their effort. Then it distance from power line in meters equals max allowed tree height.
So here in Australia is not that hard to get that line of sight to buildings because power companies regulate tree height and Cell towers can be mounted way higher. So as long as you correctly calculate for power company tree height regulation and building height regulation covering a street and place the right height towers you can be about 2~3 km spaced no problems.
Of course there is a problem you have cell phone in hand now top of tree is between you and tower so you have signal issues. Building mounted allows you to get above regulated tree top heights so no more signal problem. mmwave works well for fixes installs particular in areas with 1 storey building height regulation as between power line regulation and 1 storey building height its really simple to calculate your required tower heights for coverage . Those areas you could be covering 5 streets wide from 1 line of towers and not having the issue that power company works on poles or people wanting to move buildings need to mess with your cables most of the time.
https://www.youtube.com/watch?v=-UuqxGrPsks
Also look at this that is one of the converge test backpack design for 5G. The mid to low freq test is on the person. The mmwave is on a tower off the backpack so placing it like 2.5 to 3 m up. They are not testing signal strength at normal human heights for mmwave. Yes the test height for mmwave is normally higher than a person holding their arm straight up with phone can reach.
oiaohm,
I’ve read that mmwave performance degrades at a mere 1km even with clear line of site due to the interaction with air itself. Still, regardless of the range, it can still be useful for people who don’t have alternative options.
What you describe is what a practical country would do, that doesn’t apply here, haha. We’ve got lots of branches hanging above the power lines. Even when they come to cut down branches, they literally cut 1 foot of clearance at best. This is ridiculous and causes regular power outages when those branches fall during storms, but that’s what we deal with.
–I’ve read that mmwave performance degrades at a mere 1km even with clear line of site due to the interaction with air itself.–
That is not false. The 5G error connection means that degrade does not come a problem until 2km. 2km is the functional limit not the start of degradation point.
–What you describe is what a practical country would do, that doesn’t apply here, haha.–
Telstra-Ericsson 5G 3 band solution was designed for a practical country.
–Even when they come to cut down branches, they literally cut 1 foot of clearance at best. —
That makes me worried because you would not be doing that here.
https://www.ergon.com.au/network/safety/home-safety/trees-and-powerlines/plant-smart
Do “Trees are good conductors” bit in this Australian safety sheet. We have many trees here that if they come into contact with power lines basically come electrically live. Nice fun step on a tree root of a tree touching power line and you come the path to ground. So Australia tree pruning rules are taken very serous-ally yet still many people trimming trees die yearly as well few by power-line to tree to root of tree to human foot through human to earth.
Maybe you have less electrical conductive trees.