Separate names with a comma.
Discussion in 'Internet Streaming Services' started by 1948GG, Feb 9, 2021.
Have to ask the horse. The horse would defiantly say yes.
Sent from my iPad using Tapatalk
the thread already derailed to XM-Sirius
Perhaps it is the nature of DBSTalk ... and this is the Streaming Services forum, not an ISP forum, but I believe expectations need to be set by Starlink as to what their system is designed for and what it is capable of doing. Starlink is not a MVPD or vMVPD ... they are an ISP. Expecting Starlink plus a streaming service to replace a DBS subscription or service over a land based connection is optimistic. Overly optimistic in my opinion.
I would tend to agree if..... those much older and stuck in the mud 'services', from dbs (now going on 27+ years) to wireline (twisted pair at 180+ years), coaxial transmission (70+ years) weren't in serious need of basic competition.
Just look at POTS (plain old telephone service), it has been almost totally decimated by wireless in some 20 years (1980-2000). Another bit of basic infrastructure needs to be yanked into the future.
LEO satellite was first proposed by well financed companies (Microsoft for one) over 20 years ago; I commented at the time that they needed a launch company with cheap rockets FIRST and then the other parts would come together. Will starlink put enough pressure on the incumbent internet providers? I think they, plus the wireless folks with 5g, already have. In fact, I don't think it, I know they have as my rural cable bill has been reduced by a third just in the last 6 months, although its still double that of the same service in the big cities, 2 of which are 100 miles in opposite directions from me.
I smile every time Elon says starlink is only meant for rural deployment. Does he not understand that even in the big cities those incumbent providers skip and hop over neighborhoods and even individual streets because of bizarre, and potentially illegal reasons if the state would tighten up their laws? I lived for 15 years in a community that was chosen by verizon to be one of the first FIOS plants built in the Pacific Northwest. Since I had been one of the first engineers, working for a fiber equipment manufacturer in the late 90s, to have helped design two of the original fios pilot plants, I was stoked to say the least. But I never got service, from either verizon or from frontier who later bought big red out (and it has since been bought out yet again, but I've moved).
All the streets surrounding me had FIOS. But both companies refused to come down my cul-de-sac. They couldn't give me a reason, until one verizon manager in 2005 admitted that 'lines were drawn' by them and comcast, streets where Verizon would not protrude into comcasts 'territory'. And that's the way it remains up until today.
So if one goes through the comments on message boards like reddit, you'll find tons of 'urban' folks who have been bypassed for one reason or another, known or unknown, who have been told by the phone or cable folks, that extending their lines 50 or 100 feet will cost tens of thousands, and they want that money up front, they don't care what the city charter they agreed to 20 or 30 years (or last year) says. You're screwed.
The throughput of a system like starlink is limited only by the number of satellites in orbit above the customers. Right now, it's a bit minimal at 1100 or so total spread across a small slice of the planet. 10 years from now, I think 40k will be the low end. But maybe the terrestrial wire folks will get off their rear ends, the cell folks sure look like it.
Satellite services tend to work better where one can see the sky. "Only" may be a simplification, since many people think satellite TV only works in the suburbs and rural areas for the same reason - LOS. But in the right place a dish can see the sky.
Where Starlink would fail in larger cities is with bandwidth. Population density puts more potential customers on each beam. 20 Gbps per satellite divided by ... how many customers sharing that satellite at that moment? 150 Mbps beta should double to 300 Mbps production ... with the promise of 1 Gbps or 10 Gbps throughput? That might be possible in "rural" areas where fewer people are sharing the satellites.
40K satellites with 20Gbps max throughput per satellite is a system limit of 800Gbps. How many of those will be over each customer at a particular moment? The more successful the service becomes the less bandwidth available to each subscriber. Just like poorly designed cable and DSL systems where each neighborhood has a limit and not every home can download full speed simultaneously.
I believe Starlink will do fine for browsing and downloading. Transmissions where speed changes and buffering is acceptable. Transmissions where being handed off from one satellite to another may pause the download for a moment but no glitch will be seen. Also low bandwidth communications such as telemetry. Use to watch streaming services? I expect the usefulness will fade as the bandwidth becomes more cluttered with users. The success of Starlink could be its downfall.
Uh, check your arithmetic, you misplaced your decimal point. 40,000 x 20gbps = 800Tbs total system throughput.
The next question is how many 'cells' are there on the surface of the planet, figuring each cell is 300 miles across. Earth is 197 million sq miles (according to mr. Google); let's see if I have any part of my injured brain working in spec, the 300 mile across cell is pi * radius squared, so 3.14 * 22,500 (150*150) gives us 70,650 sq miles per cell. 197 million divided by 70,650 gives us 2789 (okay I rounded up 1) cells for the entire planet.
Okay what the **** were we talking about? Oh. I'm sure the big young brains at spacex rattled this off in a couple seconds to musk who had already thought it out years previous, as had the folks who first were serious proposing this 20+ years ago. Is my arithmetic correct? At 2789 cells and 40,000 sats, that means a bit over 14 sats per cell. Almost crowded, but space is, well, spacious. If the sats throughput is 20gbps each, then that *14 is 280gbps. I think folks generally over figure the amount each user will be consuming, not everybody is going to be downloading the latest linux distro all at the same time, but 100mbps sounds okay to me, I use less than half that with a half dozen roku's and assorted pc's plus smartphones. But 280gbps / 100mbps is 2800 users per cell.
Now one gets where musk is coming from, but we can get a bit more realistic in the number crunching. My usage is way above 'normal', at 4.5TB a month (yes, I pay comcast for unlimited, and wonder why that cost for using 4 times the base 1.2TB is the same as someone using 10 or even 100 times more, but I digress), normally around 30mbps, and so triple that 2800 to 8400 and it's still a bit lean if that 300 mile across cell includes a large city.
There is where the rubber meets the road. How many city dwellers subscribe to satellite tv, when cable is ubiquitous? And ota coverage is the same? Quite a lot, more so in the early days (90's) before cable systems invented docsis, and went digital before the broadcasters.
But all these calculations show that the number of sats per cell over heavily populated areas is going to greatly exceed that of over the huge expanse of the pacific, for example. The distribution is not going to be equal, that is something the calculations dont show, and I'm no orbital mechanics person by a long shot. By as I pointed out before, the sats will be concentrated over the earth's population centers, if they have any idea as to what they are doing.
We shall see, but recent articles have shown a huge dislike of the current broadband choices, and there is a large number of urban and suburban folks who may indeed swamp the system. Will it affect those rural folks who have little to no choice? Probably not, the sticks are, by definition, the sticks, and cells outside of big cities will remain lightly loaded.
Too much of this is conjecture, we don't know how they are going to operate the system, or how the biggest unknown, those pesky humans, are really going to react as things ramp up. The 20gbps per sat, I hadn't seen that but I'll try and track it down, and what if ver. 3 or 4 bumps that up to several times that? I'm an old microwave engineer, I'll track down the fcc frequency allocations and run it by the latest throughput analysis and see what falls out. But its sure going to be interesting, I'm betting my cable bill will continue its downward trend.
I'll just jump in to note in reference to the sat to sat hand off affecting streaming, that my wife routinely streams TV programming via cell service while we're underway in our motorhome. Sure, the occasional large overpass or tunnel will disrupt service briefly, and weak signal areas may cause buffering, but overall the tower to tower hand offs are virtually seamless. If Starlink's hand off's work anywhere near as well as the cell towers, I don't think it will be an issue for streaming.
I think I commented either here or another thread that all the roku streamers (and I'd wager those that are doing so to handsets as well) forward buffer at least 15-30 seconds so it takes quite a bit of interruption to cause any disruption. The cell data lte signal is perhaps a bit more prone to interruption if the fiber or microwave backhaul to the particular tower you're sucking from has a minimal data pipe. The only time I got involved with cellular engineering was in 2000-2001 and the companies at that time didn't seem to be very interested in providing nice fat pipes to their towers. I'm sure that later on with lte they went back and redid those systems. Those folks now pushing home internet through those facilities had better have a lot of bandwidth flowing to them.
But folks streaming on the current thin starlink system mostly report very minimal dropouts, esp. If they are in the high sat count band and have their dish well positioned to have near total sky coverage.
Here's a Reddit page featuring Starlink users that's pretty informative about their experiences with current speeds, streaming, etc.
In going through the publically available specifications and pronouncements from spacex and starlink people, up to and including Musk, the figures are all over the place as far as throughput per sat. The 20gbps was a somewhat good figure in the early days, with the rf and processing capability of vr0.9 to vr1.0 or somewhere in between.
The problem with crunching the numbers on this is many fold. Anyone who has looked at the specs of a wifi router knows what I'm talking about. Sure, the rf spectrum that the 2.4ghz and 5ghz bands are finite and easily calculated throughput; but a lot more goes into the actual usable bits, or there wouldn't be any difference between models. The design of the cpu, memory, antennas, and a host of other parts all play a role, and the same with the starlink sats. Improvements in the design have, according to many close to their design, increased satellite throughput by several times over those initial estimates.
So where is elon getting the increase in throughput to yank speeds up that they announced and folks are now reporting? Tweak the software? The newer batch if sats have speedier processors? It can't be increased bandwidth, they are using what the fcc and itu have given them. Something somewhere in the system has been improved, so the question is, how much more improvement is possible with the current system, and how much improvement is there in improving that system?
Some people looking at specs not in the public domain are, as if this date, looking at >200gbps per satellite, 10 times the figure musk and others originally put out. Since they have improved the system 3x fold just in the last week, who knows how much more can be wrung out, and again we have no idea what improvements, both satellite version 1.x or 2.x not to forget the ground stations.
However, the actual effect this is all having on existing broadband providers is something everyone can follow. My costs have plummeted by a third since the 'better than nothing' launched; still not as low as the big cities, as I'm very rural, but getting there. Another third and it will be in line with the cities; say about the time starlink gets up to 5-6000 sats and comes out of beta? Yes!
That helps. There is still an issue of how much of that bandwidth is available at each user's location. The experience of the limited beta users may not hold up when Starlink opens their network to all willing to pay.
Your interest seems to be more focused on reducing your bill for non-Starlink services than actually purchasing the service.
I hope it works out for you.
I'm focused on competition. When DirecTV launched, I was living in DFW and had a perfect shot at the sats. The local cableco had a fair collection of channels, but even with D* having only one ku sat to start, the cost vr channel count was heavily skewed to D*, particularly with the premium channels like hbo, showtime, on USSB. For what I paid for hbo on sat, I got multiple channels (at the time 6 or so) for the same cost on cable where I got one. Same across the board with other channels, including 'basic cable' channels like cnn international and espn2. No brainer, even after purchasing the dish/reciever for $800. Savings paid for that in 6 months.
I have a perfect shot at the entire sky where I live today. And two cellular companies in addition to starlink vying for my business. Did you read my fight over FIOS where I lived before? That's the kind of nonsense lots of folks have been putting up with for decades, and it shows across the country. I'll tell everyone that the last 20 years on comcast, the service has been near excellent, only the price has been out of bounds, due to no competition. Now there's a hint of it, and for the first time in those 20 years the price drops. Hmmm.
Are you willing to switch to Starlink? Are you ready to spend over $500 plus $99 per month for service?
That's less than what I paid for DirecTV, 27 years ago (that's $2000 in today's money, so...). Yikes.
There's a few things I do on comcast, one of which is still doing some engineering for the company I retired from, and it requires a vpn and such. So I've been following folks trying to get that to work consistently on reddit. Same with the cell folks. Right now, everything works fine on cable, minor wobbles every once in awhile. Biggest problem is the (to me) extremely low upstream bandwidth, which only gets good if I get gigabit service at >$350+. Things have fallen off though in the last 5 years so if I transfer 24/7 I can get it done even with pathetic xfinity. Forget the gigabit.
Again, we shall see who can win the race. I think a year from now it will become clear.
I do work at home via Xfinity but use remote desktop for most work. All of the data and processing is done on a computer at the business. The computer I have at home is a glorified terminal. It cuts down on the data usage since my download to home is the screen share but my upload is only command and control. I have been able to use video conferencing (Zoom and Cisco) through the remote desktop with the cameras on the home machine showing up on the work machine just like I was in the office. (The work machine doesn't even have a keyboard, monitor or mouse connected. Just a powered on networked PC.) That should work through the current beta level of Starlink.
Creating monster sized files on a local PC or downloading firmware updates locally then trying to push them via VPN? Not recommended.
And to answer my question - No, I would not be willing to pay over $500 plus $99 per month for Starlink. That is more than I am currently paying and Xfinity installed for free. That is the competition for Internet.
With no other option (last chance Internet) I'd consider Starlink.
But I'll bet that's in a city. When I first retired I stayed in the city, had business class xfinity at triple the speed + I have now at 1/3rd the price. I kept that tier for a few months after moving to the sticks but it became too painful; comcast finally allowed getting unlimited data on residential accounts, and that was the opening to drop the business class with its unlimited; fees were back to equal to bc in the city, but at one third the speed (both down and up). Then again, the cost of living in Seattle was looking like SF where I had lived previously, home prices are north of $750k with prop. Taxes out of hand (wash state has no income tax so everything from schools to roads are paid by use taxes, which means billionaires like gates and Bezos pay next to nothing while what's left of the middle class get hammered). So pull up stakes and move to the sticks, where some things are cheaper except, if you can get it, internet.
Outside of the city limits. Xfinity's footprint expands beyond my home and charges the same rates in at least five connected counties. Same rates city or rural.
$99 per month would get me 600 Mbps in my area. Twice what Starlink is proposing post beta and four times the "advertised" better than nothing beta data rate.
New customer installation is listed as $100 but can be $0 with new customer deals.
If I chose to rent their modem I could get 200 Mbps w/unlimited data for $99 per month. I considered that when my data usage went up last October but I am nowhere near the cap now so I'll stay with my owned modem (count that against the install cost if you want). Xfinity data rates are typically higher than the advertised rate - so perhaps that would be the closest to a "break even" if Starlink can match Xfinity's speed and unlimited data. After paying Starlink $500+ for the equipment.
I guess you have to be more rural than me for Starlink to make sense.
With Starlink operating in 5 countries so far, it'll be interesting to see the subscriber numbers in a year or so.
The only thing I need StarLink to do is be viable competition...and put downward price pressure on the rest of the ISPs. In my area StarLinks monthly price at their current dl/ul speeds are already close. At 350 for $99 they would be cheaper than my local ISP. That will put downward pressure on prices. That would be a positive development.
I had to look up the rates at my home in the city, and they are about double yours; I'll bet it's back east, right? Rates in the county where I am now are at least triple plus; the gigabit service is actually an almost great deal, until you add in the modem rental which is required, so the total gets steeper. 600mbps runs $150+ in the city and $300+ in the country. It's been over 5 years since I moved so comparisons are a bit out of wack now.
I do a lot of cadd work, trying to do that as a remote terminal is asking for trouble, I used to do it way back 20+ years ago in SF on a t1 I had at my home, also did a lot of IT work as well on our computer systems. Those were the days.
Very rural, but only 30 miles from state capitol. That isn't saying much for these small states, they usually have the state capitols right on the edge of sasquatch country. But also 120+ miles from any television towers, back in the analog days one could pick up a snowy picture but forget about the digital one now. Satellite or streaming. And anywhere past a mile of my home here in any direction and no cable. Dsl is zero. Folks still have dialup. But property taxes are zero (huge lumber forests, Weyerhaeuser), utility rates are lowest except for counties right on columbia river (which house the big server plants from google, microsoft, and others), <2 cents kwh.
So we all need some good internet. But I moved to this location because they had comcast, many homes still have hughesnet and viasat dishes, in fact we have a viasat uplink station on the very edge of town.