Sad, but makes sense. I always wondered how they could fund the dev effort for Linux/Windows.

These comments though, whoa… it seems everyone either loves or hates Orion.

What are they built on outside the US?

IIRC Google researchers have publicly used it for machine learning experiments. For example, for grabbing a sample video as part of some demo code.

Its not very big. 8700 lb of fuel is 25% of its MGDW, which seems pretty normal to me?

Same, a long time ago.

I never posted anything close to political or “sketchy” either. Was never impolite; it was just a place for niches. I still don’t know why I was banned/shadowbanned…

NASA is researching quieter supersonic flight, and I can’t help but think of how the (also Lockheed Martin) X-59-

-has a extreme wing profile/nose, not unlike SR-71:

On the contrary, NASA mounted stuff to it for research in the 90s:

The part that provides value is dirt cheap, though.

Deepseek 4.1 and Mimo 2.6 are basically free compared to Claude, and TBH it feels like the harness/wrapper, not the LLM, is the limitation for doing useful things with them.

Where would it even come from?

Everyone’s retirement.

Admittedly, no, but this is slowly improving. Stepfun published their SFT dataset, smaller labs are publishing their datasets for task specific tunes. I believe there was another Chinese lab that published bulk pretraining data, but I can’t find it in my history at the moment.

And, notably, these comparatively tiny labs generally aren’t scraping the internet so abusively like OpenAI/Meta. They don’t need as much. Going by statements in their papers, they tend to use existing archives of web data, buy commercial data, or (more recently) generate a lot synthetic data.

If you run Nemotron, from open datasets.

See the dataset table here. There are sections for pretraining and for specific tasks/topics, like STEM, code and such: https://huggingface.co/nvidia

Chinese models are starting to open their pretraining datasets too, albeit slowly.

This does indeed get better every generation.

I suppose the 16 hit the point of diminishing returns, for me. It lasts all day, even when busy on vacation or something, though I acknowledge that’s entirely dependent on one’s usage.

I have an iPhone 16 Pro Max (2024), and the 18 Pro/Max presentation definitely gave me a bit of phone envy.

I have a 16 base… and I didn’t raise an eyebrow. What kind of app running can a 18 do that a 16 can’t, other than whatever is self inflicted with software bloat?

A LOFIC camera would have gotten me interested, but they didn’t ship that.

Maybe. TES/Fallout still seem to have critical mass on YouTube and such, and people did buy Starfield.

I’m with you though. I am a shameless Oblivion shill, Starfield was theoretically my perfect game; yet I can’t find a reason why I should devote any attention to BGS these days…

Since 2016, Obsidian has released 10 games. BGS released 2: Fallout 76 and Starfield.

Grounded 2 is out in EA, seems to be close to release, and by all accounts it is excellent.

Obsidian is doing fallout.

But under BGS now. To me, that is frightening.

They made a bunch of good RPGs/RPG-ish games, and a bunch of “mid” ones.

My personal favorite is Grounded. It does sorta feel like classic Obsidian/BGS game, with its heavy implicit environmental storytelling, audio logs and such, and the survival aspect is great for coop.

No, none of them sold a zillion copies like Fallout 76 or Starfield, but I had more fun in them than either of those, by a long shot.

Halo was an attempt at a better direction, compared to 5.

That is a very low bar though.

Mojang must have a pretty strong contract, given how little they messed with Minecraft Java Edition.

Holy moly.

That is basically the worst case scenario, no?

• The absolute last thing Obsidian needs is Bethesda presiding over them. Bethesda (the publisher) is pure toxicity, and don’t they hate Obsidian?

• They should have given Obsidian TES/Fallout, and folded Bethesda into Obsidian as a studio to farm work out to. BGS has some great level design teams, but the decision makers need to go.

• The absolute last thing Halo campaigns need is to be more like Call of Duty.

• They should have cut 343, and split the IP amongst specialized studios. Like an Arkane “special ops” stealth game, another Halo Wars, and/or a strictly multiplayer Activision game.

This is beyond even my worst expectations. It’s what I would have done if I was a corporate spy inside Microsoft, assigned to sabotage them.