SpaceX's Rising Tide - Discussion of non-Starlink satellite internet companies

[bxr140]

I figure it’s worth a consolidated thread to talk about all the other people trying to provide satellite connectivity (since a thread per player is probably too much).

The latest player is Inmarsat, who are proposing a hybrid constellation with multiple orbit types:

Inmarsat unveils multi-orbit Orchestra constellation

Inmarsat unveils multi-orbit Orchestra constellation

spacenews.com

 

[adiggs]

I love the idea (new thread to cover the non-SpaceX constellations). I'm likely not going to be much of a contributor but I'm looking forward to the continuing education

 

[bxr140]

TLDR, expect lots of trying to catch up to starlink. It will be interesting to see if anyone can actually [profitably] undercut Starlink in niche markets, and/or how much the other constellations receive preferential state treatment.

 

[Grendal]

 

[bxr140]

Satixfy seems to be on the pointy end of developing silicon for LEO constellations, both on the phased array beam forming side and modem mod/demod processing. They have some terrestrial/airborne presence already, and so the space stuff seems like an evolution.

Its a bit long/wordy (~25 min) but their CEO gives a decent YouTube interview here. Most pertinent conversation happens between min ~3 and min ~19. Interesting insight if you like getting into the weeds of technology.

 

[bxr140]

Last week OneWeb and ATT announced a partnership. At least as reported, its for cellular backhaul and enterprise/gov level service. That's pretty big news for OW...though its pretty safe to assume ATT bullied the agreement into some pretty favorable offramps, so time will tell if its just a PR splash or if its actually legit.

Related, OW just launched another plane of sats yesterday, and are getting close to half of their contemplated constellation. They currently have enough planes to fully cover high latitudes, in places like Alaska, Canadia, Northern Europe (and ex-Europe) and potentially even northern CONUS. No word yet on level of service or timeline on service, though a recent article on their UTs TLDRs (I'm pretty sure, anyway) to "we watched a cat video".

Rounding out OW news, they're in cahoots with Hughes too, which--assuming this relationship is realized as some kind of hybrid space infrastructure (in this case, LEO +GEO)--IMO is probably the most efficient way for 'everyone else' to compete with the Starlink juggernaut. Though...given Charlie's history with very tight purse strings, there's probably not a lot he can bring to the table (vague pun intended) from the inevitably low-tech Echostar assets. A completely different analysis of the relationship might conclude with OneWeb conceding to Hughes' distribution network and existing customer base, and so OW had decided that its better than not to just selling a commercial pipe to Hughes so Hughes can provide [ostensibly] better end user service. Again, time will tell where this goes.

 

[bxr140]

Not a hugely compelling article, but it does have one very interesting observation/thought experiment: What happens when private companies launch constellations...and then go public? At some point in the relatively near future that now public company will need the board to authorize a yacht load of Capex to refresh the constellation.

Certainly constant replenishment can soften the otherwise big refresh blow by spreading it across many years, but in the scenario where tech refresh is implemented, constant replenishment also results in a perpetually asymmetrically performing constellation. From a SLA perspective, its difficult to both account for the "weak links" in the constellation while also capitalizing on the latest and greatest tech satellites.

 

[bxr140]

India is looking to become the future OneWeb launcher.

Makes sense given the Bharti investment in OW; this would likely mean the OW production line in Florida would need to be relocated since its not easy to export a satellite from the US to India, and honestly it doesn't make sense to build a bunch of satellites on one continent and then launch them on another, especially when that other continent is pretty much on the opposite side of the world.

This also is an interesting twist for Ariane. One has to imagine India's The Man would be much more willing to subsidize OW launches through ISRO than Europe's The Man through Ariane, and so Ariane would likely just have to concede defeat rather than try to compete on cost. Its hard to say just how much Ariane makes off of the OW Soyuz launches <insert Russian extortion joke here> but even though its one of their downmarket products it has to be a hit if they lose an anchor customer.

 

[bxr140]

Sorry, turns out the post above should have gone in the launcher thread...


Anyway, I figure its pretty big news that Verizon is getting into bed with Kuiper. I'd expect to see further announcements in the coming months from other MNO's around the world as they pair up with Kuiper, Lightspeed, and OneWeb.

I'm not totally sure on this quote though: 'Whereas Starlink has been seeking to compete directly with telcos for underserved consumer markets, Amazon has said it wants to partner with established players to connect rural and other underserved areas." To my recollection, that misrepresents a bit Starlink's strategy--I thought Starlink was very on board with becoming a backhaul provider for remote telco service?

 

[bxr140]

Big milestone for Isotropic--their phased array UT closed simultaneous links to MEO and GEO sats (SES sats). That's going to be an imperative function for constellations 'fixin to compete with Starlink using a hybrid constellation geometry (= not 100% LEO sats).

 

[bxr140]

From the "say WHAT?" files?

Bit of a long time coming, but Boeing finally gets an approved filing for a hybrid constellation. Its basically a high LEO constellation that will provide (I assume) global coverage at good latency, and then 15 HIEO sats (if I'm reading it right) to provide the big backhaul needs. The LEOs would be big--probably something like the mPower 702X--and the HIEOs would probably be as big as their big GEOs.

Likely they're looking to shop this filing along with some degree of actual satellite design to a buyer rather than build it themselves. Space communications is all about having priority in filings, so this could be a compelling pick up for the right customer.


And not to be outdone by SpaceX or Oneweb filings for a gazillion satellites, Astra files for a gazillion of their own. Now, I like Chris and Adam; I think together they have the right blend of realism and aspiration. I think they're approaching their launcher development the right way, and I think verticalizing across the shop floor with a satellite platform is also a sensible approach given the way the rest of the industry is going. I even get that, to my point above, getting a foot in the door early with a filing is super important--way more so than having a vetted technical solution at the time of filing submittal.

But...its hard to figure what they're really playing at here. Big names, deep pockets, and sovereign nations are all trying to fill the satellite services space, and its quite evident that not all are going to succeed, let alone there be room for smaller players. Even Starlink is heavy on technical solution and light on business model. Its hard to imagine what value this kind of filing has to Astra or a prospective partner. V-band might be the future, so maybe that's the play? But Its also very plausible that the other far more mature satellite services players (plus the ever expanding terrestrial services) are going to saturate the market before then anyway?

 

[bxr140]

Well, this is pretty huge. Viasat is buying Inmarsat. For an absurd amount of money. Inmarsat went private a few years ago at $3.3B--Viasat is paying over twice that. Crazy. Related, VSAT is down ~12%.

One could speculate that Inmarsat brass saw the NGSO satellite services writing on the wall and just decided to take the money and run rather than trying to fight the battle against Starlink and the long list of pretenders. On the flip side, Inmarsat's pretty broad set of spectrum assets (and their roadmap) could enhance the type/amount of service Viasat can offer in the future, and specifically the lower frequency stuff (L-band). There's a bunch of Inmarsat sats in backlog, so its provably fair to assume that Viasat will add them to the fleet to further enhance global presence rather than sell them off or something else.

 

[bxr140]

Nonzero chance this is from The Onion, Wyler files for a 300,000 (not a typo) satellite constellation through Rwanda.

 

[bxr140]

Slightly closer to earth, and I know this is probably completely unexpected for all of you, Lightspeed is delayed. Missing ITU filings is probably not a big deal (its mostly a good faith thing), but the longer they take to come to market the bigger Starlink's user base gets and the more likely the other constellations are likely to grab market share.

Telesat is also going public soon, so get your shorts queued up.

 

[bxr140]

Looks like Peru is buying an Astranis sat (and probably a second one) for cell backhaul. I suspect most would assume this seems like a prime use case for Starlink; the fact that actual money is going elsewhere suggests one or more of the following, in no particular order:

--Cell backhaul for a country of 32M is not a needle moving business case for Starlink (= they nobid)
--Starlink backhaul services won't be available in the 2024 timeframe
--Starlink backhaul services are more expensive than a GEO for this use case
--Starlink service terms are untenable for Peru/Andesat

 

[bxr140]

Not a hugely compelling article, but as reported (an interesting data point) SES Networks--which is more or less the existing O3B constellation that started launching ~10 years ago or so--returned ~half a €B last year. A lot of that service is government and a big chunk is commercial level "mobile" service like airlines and cruise ships (I don't think there's much if any residential). There's 16 or something operational sats with a total constellation throughput of ~300gbps or so. On the surface that seems like good math for new constellations like Starlink (call it ~$2M/gbps), though for all manner of reasons--many discussed elsewhere in this forum--its very short sighted to try and simply draw a linear revenue relationship to a constellation that can provide orders of magnitude more service.

Its also an interesting data point that SES--the largest legacy satellite service provider--returned ~twice that revenue on their video service, which is basically DTH type stuff (Like Dish+DirecTV+SiriusXM in the Western Hemisphere), though that ratio has been declining over the last few years. It will be interesting to see how they shift their product offering as the world moves further away from consuming live content and how their corollary revenue ratios evolve. FWIW the new SES mPower constellation that's going to start launching in a few months (on F9's) will have a global throughput of something lie 10tbps....and to drive the point from earlier home, its unfathomable to think its going to return €6B+. Similar to O3B, I don't think they're going to target residential, like Starlink is doing.

Final random though that's only marginally related here, Dish and DirecTV/ATT are in discussion (again) about merging...no doubt because of falling subscription rates.

 

[miimura]

Why should Dish and DirecTV be allowed to merge? They are the only available TV solutions for many rural customers.

 

[bxr140]

Why should Dish and DirecTV be allowed to merge? They are the only available TV solutions for many rural customers.

I mean, the proposed monopoly was shut down last time...

I suspect this time around may be slightly different, as declining subscriptions--a trend which any reasonable analysis would conclude will continue--could ultimately result in a monopoly anyway. The companies run pretty tight on margins in the good times; an undercutting race to the bottom could easily find one of them folding, or at least folding their DTH product.

 

[miimura]

I mean, the proposed monopoly was shut down last time...

I suspect this time around may be slightly different, as declining subscriptions--a trend which any reasonable analysis would conclude will continue--could ultimately result in a monopoly anyway. The companies run pretty tight on margins in the good times; an undercutting race to the bottom could easily find one of them folding, or at least folding their DTH product.

Then let one of them buy the other in Bankruptcy court. Allowing merger before that is not responsible regulation.

 

[bxr140]

Then let one of them buy the other in Bankruptcy court. Allowing merger before that is not responsible regulation.

Sure?
I'm not advocating a merger, I'm simply reporting that DTV and Dish are talking about merging...again.