Starship - SpaceX

[freebeard]

Isn't Deep Seek the hot new LLM?

It's more complicated than that. nVidia stock took a hit, but has recovered.

Elon Musk started OpenAI with Sam Altman and others. While he stepped back due to conflict of interest with Tesla's AI work, Sam Altman left yCombinator to focus on OpenAI, and they adopted a ‘capped-profit’ strategy.

So OpenAI was started as Open Source but devolved into a for profit; while Deep Seek promotes Open Source with a CCP filter.

Basically, Deep Seek is a proof of concept (lower resolution reduces computation cost) that will be adopted/coopted by all others going forward.

edit:
I found this article on The Register where IT is pushing back against the Deep Seek phenomenon. www.theregister.com: The hidden cost of compromise

As the technology industry continues its shift towards AI dominance, an important schism is opening up that threatens to impact scientific progress, along with important humanitarian endeavors such as disaster response.
....
While this industry-wide shift has accelerated AI advances, it's creating an unintended consequence: the erosion of high-precision computing, which is vital for important applications such as climate modeling, aerospace engineering, and pharmaceutical research.

--There's an old saying in software development that goes something like, "Fast, good or cheap — pick two."--

[redpoint5]

Possible launch day today.

From SpaceX website-

"Distinct vehicle and pad criteria must be met prior to the return and catch of the Super Heavy booster, requiring healthy systems on the booster and tower and a final manual command from the mission’s Flight Director. If this command is not sent prior to the completion of the boostback burn, or if automated health checks show unacceptable conditions with Super Heavy or the tower, the booster will default to a trajectory for a soft splashdown in the Gulf of America."

[freebeard]

It looks like a repeat.

SpaceX lands Super Heavy booster, but loses Starship in 8th test flight -- Global News

[redpoint5]

I missed watching it in real time, but now I'm caught up. Yup, same failure mode it seems, which is good because consistent problems are easier to troubleshoot and correct.

The shame is that the engines only needed to survive another 20 seconds and the rest of the mission could have been flown. I wonder if the 1min static fire contributed to something wearing out? Perhaps if the static fire was not performed, it might have survived the mission?

[Piotrsko]

Any time you light an engine you have limited possible outcomes: works up to commanded power (which may or may not be up to over 100% rated), works partially, doesn't work, spontaneously disassembles. You can't model it, and testing it on a pad is different than free testing under load. Sounds to me like they didn't get the actual fault fixed, must be counter intuitive or too much pressure to launch ( they drilled ventilation holes for chrissakes) I think too much gotta launch it, now!!! Is the failure, or at least as it looks from here. Daddy has too many fires going and lost focus and had a tantrum. Wouldn't be the first time*. Good thing he doesn't use solids because at those power levels, there would be no launch site available to find. Makes the dinosaur extinction event.

*Ask any long term giga factory employee

[freebeard]

Putting it all on the moneybags guy disrespects (for better or worse) the person achually in charge, Gwynn Shotwell, president and chief operating officer of SpaceX.

[Piotrsko]

Lots of respect for the lady boss, know people at the bottom of the food chain that work at the local giga factory who tell me stories.

Figure my employment there would have been less than probation before I got fed up or fired for disrespect. Never tried, figured it was a lousy fit.

[redpoint5]

Manufacturing is among the most straightforward types of job. Once trained, you just follow the processes and meet production and quality metrics. It's so routine, that once something unusual occurs, it's a call to the help desk.

They're complainers in any large corporation. People complain in the European company I work for. Dissatisfied people have the option to find something more suitable.

I'd find it interesting to work under the design philosophy that seeks to eliminate too much in search of optimizing design.

[freebeard]

Move fast and break things. – Mark Zuckerberg;

Fail early, fail often, but always fail forward.. – John C Maxwell

You want failures to be small and informational. – Nassim Nicholas Taleb

Build back better....

[Piotrsko]

Whelp, wasn't a grunt, was production management for hughes space and comm. Union wouldn't allow me to touch product in your lifetime. Early retired here, figured would be fun to do weird jobs. Tried prototype assembly, learned to drive schoolbus, buyer for job corps, carpenter........many others that sounded fun. Dont need the income, but hand me crap and I'm gone, go find some one else to put up with your BS.