To quote Ethan Klein, “I’m a wife guy”

Yeeeeep.

After the entire… userbenchmark + related shennanigannery debacle… I really can’t believe that I’m now actually rooting for Intel as an underdog, in any capacity.

Yet, here we are, rofl.

But hey, at least AMD continues to be waaaaaaay better at being open source than Nvidia.

Intentionally or unintentionally hahahahah!

… do you know literally anything about chip fabrication at a mass scale?

Photolithography?

Thermal oxidation?

Vapor deposition?

Megasonics?

Plasma ashing?

Or are these gonna be artisinal, bespoke, lovingly handcrafted GPUs?

… why do you hurt me like this … ?

I still can’t beleive the AMD head lady Lisa Su is his cousin.

Reality needs better scriptwriters, this shit is too ridiculous and wacky to be real, but as far as I can tell… it is real!

Fuck man, I dunno, I’m going with “2012 was the apocalypse, we phase shifted into the clown dimension” at this point.

Semi-reluctant AMD ‘fanboy’ here, have hated Nvidia for almost a decade now.

All of this, where we are now with LLMs… this all started because Huang decided consumer grade GPUs can actually do full real time raytraced lighting, the movie grade shit that you normally do frame by frame with a farm of gpus.

That didn’t work well, so, Neural Upscaling. That was also fairly shitty, so, gen 2 gen 3 gen 4 gen 5, still too shitty, framgen, multiframegen! Which is also often still shitty.

Oh also, because Jensen says this is the new paradigm, uh, i guess fuck everything thats ever been done in terms of optimizing lighting in game engines, throw that all out, make it work with new paradigm.

Oh thats basically fundamentally impossible to do sanely and performantly, on affordable hardware? Clearly you just don’t understand how to use Unreal Engine, what do you mean you don’t own a GPU that costs as much as your last entire PC builf?

4k60fps ultra? Fuck you! That’ll be $4000.

(not counting fake frames…)

This is all Huang’s fault.

And thats without even going into creating the largest financial bubble in the history of mankind which will pop into the Greatest Depression in recorded history, very likely end the USD’s status as world reserve currency.

… so yeah, I semi-reluctantly ‘fanboy’ for AMD for a few reasons, some of those reasons being they didn’t initiate all of the above.

Anyone who is an Nvidia fanboy is definitionally an idiot that understands nothing about this industry.

Hahah, well, unless/untill the LLM creators clamp down and censor/train that kind of capability out of them.

But there will be open weight models, and people will uncensor those, and those can be run locally or you could rent your own server and run them there.

But but, uncensored models might be declared to be illegal along basically the same lines as a nuclear weapon, or maybe the same lines as stl files to print firearm components.

… vidyeo gamse!!!

Well, thats certainly a neat idea, basically ‘make open source Unity in Rust’…

… but I don’t think that’s what I described?

Am I missing something?

Because if this is made out of Rust, incorporates Lua as a scripting language… I don’t see how that would be able to run most (any?) Unity assets, and it doesn’t mention anything about converting from the more Unity asset specific format to something more generic, or Godot specifically.

What I’m trying to describe would be more like this, but not 2+ years stale:

https://github.com/V-Sekai/unidot_importer

… and then basically just using an LLM to semi-automate the actual conversion of free and clear Unity assets.

I fucking knew it, of course the sell out is total.

If anybody has a bunch of LLM tokens to burn, and wants to use them for good, convert everything thats free and clear on the Unity Asset Store over to Godot.

They both use C#, we can worry about converting things to GDScript later.

We have had missles that you can lob from a high altitude aircraft, and hit a satellite, since the 80s.

They just want to know which accounts are permitted to scrape them for textual LLM training material (because Reddit Inc has a contract with them) vs which ones are not permitted.

The byproduct of this is intensifying a cult like environment where its easier to gaslight people, but that’s ancillary.

Even being able to better drown the user base in more, and more specifically targetted ads, even that ain’t the main thing. Everybody in the online advertising world knows that the actual metrics that ‘describe’ how effective ads are… are largely bullshit, and theyre only going to become more bullshit as more LLM bots are spun up and the average person just has less money.

Its enshittification 101, when user base growth phase stops/plateaus, lock in the user base, squeeze harder and harder untill there just isn’t any money left.

The main thing is maximize profit, ie wealth extraction.

How to read this:

The company Roblox decided to copy the entire database of a set of 3, mostly adult nsfw 3d model hosting websites.

You know, Roblox, that game all the kids play.

Yeah basically if you’ve ever seen 3D animated porn that wasn’t made by AI, and involved a reasonably well known video game character?

Yeah the … video maker nearly certainly got the model from either SFMLab or SmutBase.

The funnier part is this:

A huge majority of these NSFW models are actually variations of models made by … companies, that sell their models and character customization software, and have liscenses that generally permit use in something like a video clip, but almost never in an actual commercial video game, not without a subscription and agreement to liscense terms.

So basically DAZ and Realusion and such now very likely have a lawsuit against Roblox they could choose to pursue… unless Roblox wants to maybe very rapidly pay the liscensing fees, knock out a contract agreement real quick.

… clearly you have not been to SmutBase before, lol.

In other news,

SHOCKING! Economically critical demographic of CEOs FAIL reverse turing test! You’ll never guess who…

I can run Bonsai, basically a distilled version of Qwen 3.5 (Qwen being the other major family of open weight Chinese developed LLMs) on my Steam Deck.

It ain’t fast, but its kind of nuts that it can run at all on hardware that is that underpowered, by typical LLM hardware standards.

You just run the thing in a sandbox of some kind, do not give it access to the outside internet or the rest of your computer, and the worst thing that is likely to happen is … you configure it slightly wrong, it uses too much RAM, and your system crashes. And your system can then reboot again, just fine, tone it down a bit to use less RAM.

For a large organization with networked security concerns and consistent uptime requirements, it is simply insane to just plug an LLM directly in to that, without architecting an entire system built around ‘assume the LLM will fuck up/go insane’; sanboxes within sandboxes with highly controlled connections to other sandboxes.

But that would be a significant feat of actual software engineering to construct that for/around an actually complex network, and would take time, so fuck that, we want line go up is what most companies decided to do instead, just fucking raw dog it.

But yeah the main ‘problem’ with LLMs in this sense is idiot CEOs just demanding the magic machine god be plugged right in to their very fragile, barely sensible networks and workflows, where the LLM is then going to break things in roughly the same way as would the nepo hire intern who was just directly promoted to sysadmin…

…not that an LLM has a secret activation phrase that will cause it to shut down or go berserk in a specific way.

Yeah, turns out the executive class, the would-be nobility, are incompetent idiots by their own standards, and are primarily driven by their own malignant sociopathic narcissism, not any of the stories they tell about themselves.

If you want competent executives, they’d need an actually effective selection pressure that punishes incompetence and rewards competence… but they already destroyed that.

So, either we, or somebody, builds some mechanism or system that does that… or they all preen themselves rigourously while steering the ship directly into an iceberg, with all of us onboard.

They will genuienly not understand why we hate them for killing us all.

Ah woops you beat me to it!

*Tim Apple

Uh huh.

Yep.

They seem to be fine with that, just causing Kessler Syndrome.

Now the SR 72 is not retired, because satellite based anything becomes extremely difficult to achieve, at best, and you can put cameras and comms datalinks on them, and have them do what satellites can now not do.

Could also do ground attack on static / slow moving targets: Mach 4-6 from half a continent away, over to a target area, slow down, open bomb bay, drop a glide bomb (or a lot of em), close up, speed back up and fly away.

You now have comms intel and air superiority, at least in an offensive sense; you destroyed the previous paradigm with your new paradigm.

If you can get a reliable ‘bombtruck’ style airframe, human piloted or autonomous, working at around the Mach 5 range… who has anything that can shoot that down?

The SR 71 wasn’t really stealth, it just flew too high and too fast for anything to be able to touch it.

Oh and for the ‘glide bombs’, just do what the Iranians seem to be doing: outfit some of them with a secondary boost phase that triggers semi-randomly, based on the munition either sensing its near the target, or, sensing that something is trying to intercept it.

This also greatly relieves the entire problem of highly expensive single use long range missiles… just have a ferry for them, ease the cost and design specs of the actual munitions by doing 80% of their job.

My thought is that renewed interest in this and the not strictly confirmed but seemingly under active developement SR-72 is…

… Platform for high altitude launched, satellite killing missiles.

For a long time the US got to enjoy being basically the only power with a stupendous intelligence advantage from a network of spy and comms satellites vast enough to get total coverage, basically everywhere, at pretty low time lags.

Well now, others have comparable capabilities.

So, develop a way to remove that.

We’ve been able to shoot down satellites from aircraft launched missiles since the 80s.

We’ve specifically called for the end of tests of ground launched anti-satellite missiles, leaving aircraft launched anti-satellite missiles as a neat little carve out in our own rhetoric/intl. policy ‘suggestions’.

So, in a world where spy and comms sats are just viewed as valid targets in a large scale war/operation, a fleet of mach 3+ / potentially hypersonic aircraft that can deny intel to the opponent and likely also gather your own intel, that’re significantly harder to shoot down than a satellite… seems like a plausible doctrine to me.