I know Rocketdyne says they can build more RS25s and get further increased performance.

When shuttle retired they could operate, via improvements/enhancement, at104% rated thrust. Today they run at 110%. Supposedly they can do even better today. But those engines are still OUTRAGEOUSLY expensive. They were designed from day 1 to be reusable as the shuttle was a reusable vehicle... But we're just going to dump them in the ocean.

The SRBs are fished out of the ocean and reused. But they're pretty simple to build as is.

The external tank is not reused and never was.

Who?mikejones! said:

---

My tinfoil hat thought of the day:

Musk is probably trying to colonize the moon and/or mars outside the purview of earthly govts, much like euros colonized america before there were any relevant govts in north america.

No govt is going to get there in at least the next 5 or 6 years, probably never.

---

Moon/Mars will be Musk's Gulch.

Kenneth_2003 said:

---

I know Rocketdyne says they can build more RS25s and get further increased performance.

When shuttle retired they could operate, via improvements/enhancement, at104% rated thrust. Today they run at 110%. Supposedly they can do even better today. But those engines are still OUTRAGEOUSLY expensive. They were designed from day 1 to be reusable as the shuttle was a reusable vehicle... But we're just going to dump them in the ocean.

The SRBs are fished out of the ocean and reused. But they're pretty simple to build as is.

The external tank is not reused and never was.

---

SLS's SRBs are not recovered

Oh... Just one more reason then to dislike this monstrosity...

Kenneth_2003 said:

---

Oh... Just one more reason then to dislike this monstrosity...

---

It is all the best of 1970s/80s technology with a price scaled up for 2026 with 50 years of inflation and contractor markups added. Stupidly expensive relative to the competition and incapable of flying more than a couple of times per year without massively upscaling the construction contracts. Only reason it exists is to keep a few key senators on board with letting the money wagon keep rolling for NASA.

Who?mikejones! said:

---

My tinfoil hat thought of the day:

Musk is probably trying to colonize the moon and/or mars outside the purview of earthly govts, much like euros colonized america before there were any relevant govts in north america.

No govt is going to get there in at least the next 5 or 6 years, probably never.

---

No clue if that's his intention, but any colony on Mars is on its own. 6 months minimum travel time when Mars and Earth are close together (only occurs every ~2 years) and you can only shuttle a handful of people to Mars at a time. What's France going to do if a bunch of Martian colonists spit in France's general direction? ask musk if they can borrow a half dozen rockets to send maybe 30 space marines to Mars and teach those colonists a lesson a year from now?

A big reason to keep it going is the same reason the shuttle had SRBs in the first place. There is not much demand for large diameter solid rocket motors now that we keep ICBMs and SLBMs for decades. We wanted to keep the ability to keep making such motors so they were added to the shuttle. A lot of the shuttle's worst design choices were driven by different national security concerns.

chiphijason said:

---

A big reason to keep it going is the same reason the shuttle had SRBs in the first place. There is not much demand for large diameter solid rocket motors now that we keep ICBMs and SLBMs for decades. We wanted to keep the ability to keep making such motors so they were added to the shuttle. A lot of the shuttle's worst design choices were driven by different national security concerns.

---

Why can't ICBMs be liquid engines?

bmks270 said:

---

chiphijason said:

---

A big reason to keep it going is the same reason the shuttle had SRBs in the first place. There is not much demand for large diameter solid rocket motors now that we keep ICBMs and SLBMs for decades. We wanted to keep the ability to keep making such motors so they were added to the shuttle. A lot of the shuttle's worst design choices were driven by different national security concerns.

---

Why can't ICBMs be liquid engines?

---

Way too long to fuel if you need to use them to counter a first strike.

Burdizzo said:

---

My guess is heat rejection is easier and cheaper in space, but running a big computer server in space has to still be expensive

---

Heat rejection is actually harder. It's a vacuum. It's why a thermos keeps things hot for so long, the vacuum cavity lowers heat loss.

In the vacuum of space the only way to cool something is radiation. It's much less effective than convective cooling using air or water that we have on earth.

I'd like to see how these space data centers plan to cool the processors. It's why AI compute will never be done in space with the current architectures. Would need a step change in compute efficiency to reduce heat generation.

Data centers use tons of cooling water.

Deputy Travis Junior said:

---

Who?mikejones! said:

---

My tinfoil hat thought of the day:

Musk is probably trying to colonize the moon and/or mars outside the purview of earthly govts, much like euros colonized america before there were any relevant govts in north america.

No govt is going to get there in at least the next 5 or 6 years, probably never.

---

No clue if that's his intention, but any colony on Mars is on its own. 6 months minimum travel time when Mars and Earth are close together (only occurs every ~2 years) and you can only shuttle a handful of people to Mars at a time. What's France going to do if a bunch of Martian colonists spit in France's general direction? ask musk if they can borrow a half dozen rockets to send maybe 30 space marines to Mars and teach those colonists a lesson a year from now?

---

The year is 2035. Citizens across the globe are waiting 40 minutes to call someone a ********** on X, because of the delay on their VPN to Mars. Two hours after posting the response comes in "No, YOU'RE a **********." Forgetting why they called the person a ********** in the first place.

World peace has finally been achieved through inconvenience.

txags92 said:

---

bmks270 said:

---

chiphijason said:

---

A big reason to keep it going is the same reason the shuttle had SRBs in the first place. There is not much demand for large diameter solid rocket motors now that we keep ICBMs and SLBMs for decades. We wanted to keep the ability to keep making such motors so they were added to the shuttle. A lot of the shuttle's worst design choices were driven by different national security concerns.

---

Why can't ICBMs be liquid engines?

---

Way too long to fuel if you need to use them to counter a first strike.

---

Plus the volatility of the fuel is a huge drawback, see the Little Rock incident.

bmks270 said:

---

Burdizzo said:

---

My guess is heat rejection is easier and cheaper in space, but running a big computer server in space has to still be expensive

---

Heat rejection is actually harder. It's a vacuum. It's why a thermos keeps things hot for so long, the vacuum cavity lowers heat loss.

In the vacuum of space the only way to cool something is radiation. It's much less effective than convective cooling using air or water that we have on earth.

I'd like to see how these space data centers plan to cool the processors. It's why AI compute will never be done in space with the current architectures. Would need a step change in compute efficiency to reduce heat generation.

Data centers use tons of cooling water.

---

You might want to check out this week's M&A news and get back to us on that.

The original ones were, then solid rocket fuel improved their utility greatly. The 'minuteman' moniker was patriotic but also indicative of the improvements.

Yes, it’s true, Congress actually did tell NASA how to build this rocket. https://t.co/FJdwqT1pBt pic.twitter.com/Rpch9xBer6

— Hans Mahncke (@HansMahncke) February 3, 2026

As was foretold;

This reminds me of when SpaceX replaced that QD umbilical hose 90 meters up in the air in less than 24 hours, and was ready for the second launch attempt the next day.

I think I said something along the lines of:

“damn if this happened to SLS we would be looking at several… https://t.co/KLVNXfAl5K pic.twitter.com/lBhT3OhvEn

— Zack Golden (@CSI_Starbase) February 3, 2026

Eric Berger piece over the weekend regarding Nasa facing a legislatively imposed choice soon as to which contractor to give up to $700 million to for a Mars communication relay contractor: bad Ted Cruz.

Quote:

---

What everyone agrees on is that NASA needs a new spacecraft capable of relaying communications from Mars to Earth. This issue has become especially acute with the recent loss of NASA's MAVEN spacecraft. NASA's best communications relay remains the Mars Reconnaissance Orbiter, which has now been there for 20 years.

Congress cared enough about this issue to add $700 million in funding for a "Mars Telecommunications Orbiter" in the supplemental funding for NASA provided by the "One Big Beautiful Bill" passed by the US Congress last year.

That's a lot of money purely for a telecom orbiter
However, this legislation, led by Sen. Ted Cruz, R-Texas, raised a couple of key questions. The first is the wording of the bill, which specified that the orbiter must be selected from among US companies that "received funding from the Administration in fiscal year 2024 or 2025 for commercial design studies for Mars Sample Return; and proposed a separate, independently launched Mars telecommunication orbiter supporting an end-to-end Mars sample return mission."

The reference to "commercial design studies" referred to companies that proposed faster and more affordable missions to return samples from Mars, selected in 2024. Several observers told Ars the language included here appeared designed to favor Rocket Lab and its proposal for a telecommunications orbiter.

The other curious thing about the Cruz language is that it specified $700 million for the spacecraft and its launch, which seems like overkill.

"$500 million is plenty for a communications payload, satellite bus, and launch," one knowledgeable industry official told Ars. "I actually think those functions could be provided for well below $500 million."

---

I like Cruz a lot as a senator but will caveat that with his seemingly incessant demands on Nasa budget 'particulars' such as the shuttle relocation language, and crap like this. This is the sort of inane legislative appropriation language that has beget the Senate Launch System over the years, too.

chiphijason said:

---

A big reason to keep it going is the same reason the shuttle had SRBs in the first place. There is not much demand for large diameter solid rocket motors now that we keep ICBMs and SLBMs for decades. We wanted to keep the ability to keep making such motors so they were added to the shuttle. A lot of the shuttle's worst design choices were driven by different national security concerns.

---

I have to think we could do something to maintain that capability for a fraction of the cost of SLS.

txags92 said:

---

chiphijason said:

---

A big reason to keep it going is the same reason the shuttle had SRBs in the first place. There is not much demand for large diameter solid rocket motors now that we keep ICBMs and SLBMs for decades. We wanted to keep the ability to keep making such motors so they were added to the shuttle. A lot of the shuttle's worst design choices were driven by different national security concerns.

---

I have to think we could do something to maintain that capability for a fraction of the cost of SLS.

---

The cost of SLS is a feature not a bug.

PJYoung said:

---

txags92 said:

---

chiphijason said:

---

A big reason to keep it going is the same reason the shuttle had SRBs in the first place. There is not much demand for large diameter solid rocket motors now that we keep ICBMs and SLBMs for decades. We wanted to keep the ability to keep making such motors so they were added to the shuttle. A lot of the shuttle's worst design choices were driven by different national security concerns.

---

I have to think we could do something to maintain that capability for a fraction of the cost of SLS.

---

The cost of SLS is a feature not a bug.

---

For the senators from Alabama and Florida.

Frosty booster survived the nitro-test:

Booster 19 is fully frosty during cryo proof testing tonight at Massey's test site in preparation for Starship test flight 12.

2/4/26 pic.twitter.com/B3LMaeki5Q

— Starship Gazer (@StarshipGazer) February 5, 2026

Site expansion approved, some new barriers already going up:

SPACEX: The U.S. Army Corps of Engineers (USACE) and NOAA have approved SpaceX's acquisition of preservation land in a letter from NOAA Acting Assistant Regional Administrator, Thomas Payson Wilber, to USACE Galveston District Commander, Colonel Rhett Blackmon.

The letter… pic.twitter.com/rLnXYX2wnv

— S.E. Robinson, Jr. (@SERobinsonJr) February 4, 2026

This has been rumored off and on again for many years, as a Starlink-enabled challenger to Apple.

BREAKING: SpaceX may launch a Starlink phone that connects straight to satellites, as per Reuters. pic.twitter.com/qPe549PWbW

— DogeDesigner (@cb_doge) February 5, 2026

It would be fascinating the limits/partnerships they might opt to go here. I can't fathom Starlink working well in buildings etc, but maybe they will choose a carrier partner (presumably T-mobile again).

SpaceX purchased frequency bands between 1700 and 2200 MhZ, which is comparable to what T-Mobile, AT&T and Verizon use for a lot of their LTE service. It should function indoors, maybe not quite as well as terrestrial due to distance. Only relevant to U.S. based services so they'd have to partner/buy more spectrum across the globe to expand their direct to cell service outside the U.S.

Probably totally dead under a metal roof and probably limited inside a high rise. In those situations someone's being a ***** and bogarting the Wi-Fi.


Cheeky Pint Podcast, discusses space data center cluster

I'm about 20 minutes in. Have watched nearly every Elon discussion (haven't listened to most recent Moonshots). This is so far the most interesting/new discussion I've seen in awhile.

Please enjoy this Cheeky Pint / @dwarkesh_sp crossover with @elonmusk.

Dwarkesh was most interested in how Elon is going to make space datacenters work. I was most interested in Elon's method for attacking hard technical problems, and why it hasn’t been replicated as much as you… pic.twitter.com/28Lw9rAqlN

— John Collison (@collision) February 5, 2026

Elon denies. I am guessing this is a misrepresentation of their reported working with Apple on the next gen iPhone...

We are not developing a phone

— Elon Musk (@elonmusk) February 5, 2026

His general statement has been that phones as they exist are dead, the world just doesn't know it yet. That the concept of a thing like an App Store is dead. The concept of an app is dead. You're going to ultimately interact with the phone and the phone is going to generate the solution in real time. So if you instead of a health app, you have a central storage on your phone of relevant data and the AI generates the GUI in real time and provides context for how your data is impacting your health.


I would anticipate SpaceX or Tesla (or the combined entity) to develop AR/VR glasses way before a phone. Glasses would operate as an interface for AI and provide training data collection for training of Optimus the same way Tesla's vehicles gathered data for their self driving system.

Any updates on when the next Starship will launch? trying to see if a few of us can get down there for the next one.

Elon claims a 20% increase in operating temperature of GPUs results in roughly a 50% reduction in radiator mass for space based compute. Not my area of expertise, but it would seem that heat dissipation would be a bigger limiter on orbital compute than energy generation, exchange some level of efficiency for stability. Wonder if that level of temperature increase (120C sustained?) is possible.

Update: not happening.

We are not developing a phone

— Elon Musk (@elonmusk) February 5, 2026

So, that's the end of that.

hph6203 said:

---

His general statement has been that phones as they exist are dead, the world just doesn't know it yet. That the concept of a thing like an App Store is dead. The concept of an app is dead. You're going to ultimately interact with the phone and the phone is going to generate the solution in real time. So if you instead of a health app, you have a central storage on your phone of relevant data and the AI generates the GUI in real time and provides context for how your data is impacting your health.


I would anticipate SpaceX or Tesla (or the combined entity) to develop AR/VR glasses way before a phone. Glasses would operate as an interface for AI and provide training data collection for training of Optimus the same way Tesla's vehicles gathered data for their self driving system.

---

If only there were a way for Elon to feed the data right into your brain.

hph6203 said:

---

Elon claims a 20% increase in operating temperature of GPUs results in roughly a 50% reduction in radiator mass for space based compute. Not my area of expertise, but it would seem that heat dissipation would be a bigger limiter on orbital compute than energy generation, exchange some level of efficiency for stability. Wonder if that level of temperature increase (120C sustained?) is possible.

---

Radiation heat transfer is T^4, so increased operating temperature makes a huge difference, along with effective heat spreading.

And decreasing the radiator size also means less thermal gradient across the radiator, so it's definitely not a linear proportionality between operating temp and radiator size (mass).

Anyone else find it ironic that the commercial launch provider has 7 men working, and the government launch provider has one union worker on break and 6 standing below managing?

This reminds me of when SpaceX replaced that QD umbilical hose 90 meters up in the air in less than 24 hours, and was ready for the second launch attempt the next day.

I think I said something along the lines of:

“damn if this happened to SLS we would be looking at several… https://t.co/KLVNXfAl5K pic.twitter.com/lBhT3OhvEn

— Zack Golden (@CSI_Starbase) February 3, 2026

Ship 40 and Booster 20 are both getting active in the Megabays. NSF also saw a Raptor 3 SN 11X leaving McGregor. 110 plus R3s so far.

The common dome for Booster 20 has rolled to Mega Bay 1. pic.twitter.com/VKq0w8tnsf

— Avid Space (@LabPadre) February 6, 2026

More cryogenic tests on B20 last night. Also Elon seemed to confirm a March 2027 goal for HLS landing on the moon.

Impulse buying a COPV company https://t.co/to1jQA0Ga5 pic.twitter.com/d6WzX7TlAZ

— Goose (@megagoose11) February 7, 2026

Why couldn't a Crew Dragon be stationed at the lunar gateway for an inexpensive lunar lander?

Oh if only SpaceX had a smaller interim vehicle that could’ve began developing 5+ years ago that had propulsive landings as a core part of its design…

One that already has years of demonstrable performance to leapfrog off of. One that is compatible with their SHLV. https://t.co/ZOc8klD0vS pic.twitter.com/AKFC4dn38B

— Aero Big Mike (@AeroBigMike) February 6, 2026

Yes, getting it there would take 'some' work.

Goodness.

This is the moment that, if you still harbored any doubts, you realize SpaceX is super serious about orbital data centers. https://t.co/Dev3qEQj5j

— Eric Berger (@SciGuySpace) February 8, 2026

SpaceX dropped a photo of their new cyclotron for radiation testing.

You'll see these curved sections of magnets - why are they curved? Because mass is a relativistic property.

The cyclotron frequency is qB/m, and the RF supplying the accelerating field is fixed at this… pic.twitter.com/pmraQRetYc

— Andrew Côté (@Andercot) February 7, 2026

Ok, sure, if you say so Andrew.

Mars will start in 5 or 6 years, so will be done in parallel with the Moon, but the Moon will be the initial focus https://t.co/tP66X6MZMT

— Elon Musk (@elonmusk) February 9, 2026

Focusing on the moon first. Supposedly, they are closer to a flight-test article for HLS than otherwise thought, not sure if I buy that though.

A really big shift for SpaceX.

double aught said:

---

A really big shift for SpaceX.

---

Really just the realities of the launch windows for Mars only opening every 26 months or so. Doesn't look like they will be ready to shoot for Mars during the late 2026 one, so might as well take some shots at the moon while they wait for the next Mars window to open in late 2028.