Hacker News Reader

Comment by joering2 | Hacker News Reader

I recently started researching into Starlink as my next (cheap) internet solution and stumbled upon these 2 videos below. The worst part for me is that once this becomes more popular the bandwidth will die down to crumble. I don't buy they will be able to upkeep with bandwidth sharing it from satellite to satellite using a.. laser beam (is that even possible to sync 2 satellites and send signal thru laser beam that has radius of 10 nm at distance of 1,000 km from one object to another traveling at 20,000km/s ?)

https://www.youtube.com/watch?v=zaUCDZ9d09Y

https://www.youtube.com/watch?v=2vuMzGhc1cg

View on HN

First video, once you skip the first 17 minutes which have nothing at all to do with Starlink, is based on two premises: 1. That on-orbit and orbit-to-surface bandwidth will never increase, and 2. That the prototype receivers are the final receivers. 2 has already been proven false with the second generation receiver, and there's no reason to assume that, with so much of the fleet getting replaced every decade, that the network bandwidth will be forever so limited. I don't think laser link tracking is terribly difficult on-orbit, particularly among satellites that share one. It's also considerably slower than stated, falling at 7.2-7.9km/s, not 20,000km/s.

Both those videos use wildly wrong numbers. There are no plans to use F9 for the entire constellation. SpaceX told the FCC they will use Starship (launch 200-400 sats at once) for the gen 2 constellation, which is 30k satellites. Thunderf00t uses AWS data transfer cost instead of actual backhaul costs. Average household uses 435GB/month and that is $52/year in backhaul cost using .01 per GB. His $10bil/year backhaul numbers imply they have 191mil subscribers to get to that cost... There are a bunch of other issues, but it's too tedious to go through it.

Someone did a write-up on all the problems with the CSS video.

https://littlebluena.substack.com/p/common-sense-skeptic-deb...

Thunderf00t and Common Sense Skeptic make up data and/or selectively pick facts to support their predisposed opinions of hating anything Musk related. So you should discard anything you heard in those videos and find better more direct sources.

I don't know anything about Common Sense Skeptic but Thunderf00t has a well known hate of Elon musk and tries to trash pretty much everything he does.

I watched a bit of the Common Sense Skeptic video and heard several completely incorrect claims in the first couple of minutes. It not even remotely close to an impartial look at starlink. Just for a start, they can't use the list cost for falcon 9 launches as the internal cost to SpaceX, and then add in estimated SpaceX costs such as employee salaries.

And as for laser interconnects between starlink satellites - Its already happening. The last few launches have been starlink sats that utilize these laser interconnects.

Thunderf00t is such a strange fellow. He's clearly quite bright and has some fantastic takedowns of pseudoscience and scammy things, and even disproves a claim about dipoles from a Veritasium video.

But then he goes off with these screeds like the Starlink one as if like they are going to scale up v0.5 without any revision or cost improvement. In the name of "skepticism" or "rationality". But a real rational look at the situation would maybe give the ops and biz folks at SpaceX a bit more benefit of the doubt that they can project out profit/loss over multiyear timeframes.

> Common Sense Skeptic

He is the same as Thunderf00t in video content. Exactly the same type of video alarmism and selectively picking outdated data or extrapolation assuming no progress/changes.

Is it possible to put thousands of satellites in LOE so you play Counterstrike in a corn field? I know what people would have said 10 years ago.

Reply on news.ycombinator.com