-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
I'm also in SE PA and we had storms this weekend. My Starlink speeds slowed down to about half but it never dropped out completely. I'm happy with that since half of Starlink is still over 10 times faster than what my Windstream DSL can provide!
You've asserted this before, and I've responded with examples where this was untrue. Others have chimed in similarly.
Now you have two additional examples here.
It is...thanks @Cosmacelf .
I commented before on SpaceX's prodigious satellite building rate (i.e.- their "Machine that builds the machine"), and from that article"
"SpaceX builds the Starlink satellites at a factory in Redmond, Washington. SpaceX officials said last year the assembly line can produce as many as six Starlink satellites per day."
That's crazy.... approaching a couple of hundred a month assuming 7 days/wk production!
Two people have cited beta Starlink service outage concurrent with weather loss.
1) Obviously these beta service reports are irrelevant to GEO service performance
2) No impartial analysis would use these beta service reports as evidence that production Starlink service will suffer weather degradation/outages. There are many plausible reasons for these beta service issues—both coincidental with the weather and as a direct/indirect result of the weather—and none of them are proof that production service will suffer the same issues.
If you are saying the reports are useless because the system is beta, why?
If you are saying the reports are worthless because the system is non-GEO, then why did you earlier tie the two together when you said:
"Near as makes no difference, weather related GEO service issues are exclusively a function of improper dish installation. Near as makes no difference, it is virtually impossible to improperly install a Starlink terminal."
The very nature of beta programs is to assess performance impact in actual operating circumstances. Shotwell said as much in the article posted a bit ago.
As you state that it's impossible to aim a Starlink dish incorrectly, we therefore rule out improper aiming, no? As Starlink does not yet have 100% coverage, and that incurs outages not speed reduction, then we also rule that out, no? Yet at the time of the service degradation there was severe weather, it not impossible to rule that in, no?
Certainly correlation does not imply causation. But it doesn't exclude it either.
I'm saying it is a false positive to suggest anomalies encountered in a proto/beta/early phase will be representative of a product's steady state production phase, and it does a disservice to not overtly include context in discussion of those anomalies. That's the way product development works...not just Starlink, but any kind of product.
Even if Starlink beta service anomalies can be directly correlated to a weather event (maybe the rain fade offset is poorly calibrated, etc.), that does not mean that same anomaly will be present in production service. The whole point of beta is to ferret out those issues so they can be fixed before going live. For instance, maybe they find very low elevation look angles can't keep the signal strong enough in heavy precipitation scenarios--so maybe they reconfigure the system to a higher minimum elevation for users. And maybe that causes them to better predict weather events so they can better prepare the satellites for a pass over those weather events (max battery capacity, minimum platform temps, whatever) such that for the few minutes a satellite is over a weather event it can--as a node in the constellation--enable the constellation to still provide its users full service.
Beyond that, there's plenty of plausible scenarios where weather is only an indirect cause (perhaps flooding at the local gateway resulted in a partial outage--maybe a few antennas went down or something and the user link had to drop to a lower, more lossy elevation through another gateway) or a completely coincidental phenomenon (asymmetry of the incomplete constellation constantly manifests as ebbs and flows of very time based coverage and available throughput for any point on the ground).
Someone was worried that their GEO experience from 20 years ago would still be an issue with beta service; my response contests [and addresses] the right question to ask, which is "will my shitty GEO experience from 20 years ago be representative of Starlink production service".
To wit, my first model S was on its third (or fourth?) drive unit when I sold it. That does not mean I'm at all concerned with the drive units of my current Tesla or any Tesla I may buy in the future, and it does a disservice for me to highlight early drive unit failures in conversation unless in the specific context of a scenario that's not representative of the present Tesla product.
All that said, I will clarify the earlier statement of mine: In the bucket of "improper GEO dish installation" I'd also lump in improper location (so, occlusion) and improper user equipment (cabling, power, etc.). I can appreciate that my "virtually impossible" statement implies those failure modes are not a factor with Starlink; they are indeed very much a factor with Starlink. If one puts their starlink dish next to a building or under a tree or hooks it up with recycled knob and tube wiring or whatever then, yeah, service degradation/outages are definitely possible.
Then surely satellite operators/service providers who ostensibly have a clue wouldn't publish anything outlining the issues associated with weather-induced degradation.
Oh... wait. They do:
How weather can affect satellite communications
What is Rain Fade? - BusinessCom Networks
e-FTW
New electron smell
Starlink going mobile by end of year, wiht current dishes!
apple.news
Also a good write up on current status of Starlink.
Dishy McFlatface to become “fully mobile,” allowing Starlink use away from home — Ars Technica
Musk: Starlink customers will be able to use their dishes "anywhere" by year-end.
apple.news
Also a good write up on current status of Starlink.
e-FTW
New electron smell
Found in that same Ars piece, a link to a good dataset on Starlink constellation status: Jonathan's Space Report | Space Statistics
Starlink going mobile by end of year, wiht current dishes!
Dishy McFlatface to become “fully mobile,” allowing Starlink use away from home — Ars Technica
Musk: Starlink customers will be able to use their dishes "anywhere" by year-end.
Also a good write up on current status of Starlink.
The article makes it clear that they're working on a separate solution for truly mobile use.
But just allow the current dish to be used in multiple locations it'd still be of great value to RV owners and sailors.
Yes, there are a bunch of software updates coming that will improve Starlink. The biggest will be one that helps with obstructions. Currently, which bird you are communicating with is pre defined. Dishy will track satellite A, and only switch to satellite B based on a global allocation algorithm. But A’s path might be partially blocked by trees. The new software will allow dishy’s local software to switch over to satellite B if sat A become unreliable, if B is also in the sky.
Among other reasons, that’s why the system is still in beta.
Among other reasons, that’s why the system is still in beta.
Elon Musk’s SpaceX Moves Closer to Approval for Lower Orbits
Elon Musk’s SpaceX moved closer to winning permission to fly its satellites at a lower orbit than initially planned, after the leader of the Federal Communications Commission endorsed the proposal.
www.bloomberg.com
I really think Elon's outsized personally and goodwill he has garnered over the years (talk to anyone under the age of 25, and you'll get admiration about Elon) is helping his companies tremendously.
And the wins keep on coming. The FCC voted to approve SpaceX's request to launch the 2,800 satellites approved for an about 1,100 km orbit down to the existing approx. 550 Km orbit of their current satellites. In total, SpaceX now has approval to orbit 4,400 satellites at their preferred orbital height. I believe this also gives approval for polar orbits at that altitude, and finally also approves 25 degrees elevation for user terminals and gateways, basically giving Starlink more open sky to beam through.
Last edited:
3rd party aggregated telemetry is definitely "a thing" in the Tesla community with TeslaFi and whatnot.
Maybe we can help starlink.sx come up with an app/web page/agent that reports lat long speed (or gets them from speedtest links) and crowdsoruce this info? that way you can easily do some interesting trends, jumps and causality analysis. E.g. with weather.
Maybe we can help starlink.sx come up with an app/web page/agent that reports lat long speed (or gets them from speedtest links) and crowdsoruce this info? that way you can easily do some interesting trends, jumps and causality analysis. E.g. with weather.
Where are we now update from Space News:
spacenews.com
SpaceX launches Starlink satellites
SpaceX launched another set of Starlink satellites April 28, its first since the FCC allowed the company to operate more satellites in lower orbits.
spacenews.com
Q: manufacturing capacity of terminals . I have not reviewed all 80 pages of this thread, but was wondering if anyone has done some analysis of the terminal / antenna manufacturing velocity and ramp.
back of napkin, and sticking to consumer market of space intertubes, And direct delivery (not via MNOs or aggregators)
To get to. $1B/y revenue for consumer internet, Starlink will need to have ~833k customers. To get there within one year, it has to sustain 16k antennas manufactured and delivered per week. Let’s say the TAM is 10M consumers, at that rate they are taking 10 y to get there.
thoughts? Data?
back of napkin, and sticking to consumer market of space intertubes, And direct delivery (not via MNOs or aggregators)
To get to. $1B/y revenue for consumer internet, Starlink will need to have ~833k customers. To get there within one year, it has to sustain 16k antennas manufactured and delivered per week. Let’s say the TAM is 10M consumers, at that rate they are taking 10 y to get there.
thoughts? Data?
