kilpikaarna

Europe would absolutely be stupid and servile enough to agree to this, unfortunately.

And actually using power tools in the real world is pretty uncertain, try running a straight starting cut with a multitool against wood grain and then come back and talk to me.

I especially liked table saws and backhoes as examples of _automation_.

Do you have a preferred 3D (polygon) modelling/graphics tool where you feel like you don't need to keep up guides and forums?

I've been hearing this complaint since Blender 2.48 twenty years ago, where it might've had some merit. But even then you could still pick it up in at most a few weeks with focused study, and the biggest non-intuitive (in terms of adherence to things like OS guidelines etc) and Blender-specific features and workflows remain the same to this day, and the interface overall is remarkably consistent and effective once you get used to it.

The only thing I can see that comes close to doing the same thing in terms of a general/integrative/swiss-army-knife-of 3D content creation suite are to this day Autodesk Maya and 3ds max, and maybe increasingly UE and Unity. I wouldn't exactly say you can just "intuitively" click around in any of those and high-quality results come out by default. Other software may be better at specific things and perhaps offer a more focused UI for some subset of tasks, sure, but at that point you're using the wrong tool sticking with Blender.

But in basically all cases, as soon as you're doing anything moderately complex, you're constantly referring to the manual/LLMs/tutorial videos. This stuff is just difficult, man, and most of the difficulty is even orthogonal to the software itself!

It's 3MB of "hot" RAM, IIRC. Makes sense.

Deferred rendering requires a post process type antialiasing, can be TAA but also FXAA etc. It doesn't work with traditional MSAA.

A lot of the UE tech is built around the assumption of TAA though.

No you see, she says right there in here "Frequently Unasked Questions" (lol) that she is in fact not at all bitter about her field being effectively solved by someone else, thank you very much! Her preferred approach is in fact alive and well, more important than ever and still very very interesting. One wonders if her PhD students feel the same.

The authors of the paper make some good points (in it and elsewhere), but I've seen precisely nothing that suggests they aren't awful people who resort to playing the victim card, inciting cancel mobs and in general behaving like the worst stereotypes of "leftist academia" in response to any criticism.

I mainly learned linear algebra via hands-on 3D graphics, and have a hard time thinking about a matrix as anything other than 4x4 and representing a linear transform...

How much do you even think about explicit matrix math when doing high-level ML?

Tony was an early fan of the magic/comedy team Penn and Teller. A friend and colleague attended a show and hung out afterward to ask the duo to sign a photo for his friend Tony. “He was on the team that did the red and green squiggles in Word.”

That’s some heavy duty corpo-brain to be introducing your friend with ”He was on the team that did X”.

[Continuing the analogy, you may find that many AI heroes are just those who happened to be closer to pools of TPUs and GPUs in the early days...]

And this in turn was much more likely if you were inexplicably obsessed with ML/AI at a point when it was widely considered unfashionable.

Sounds a bit like a holdover from a long history of monotheism, so I think westerners might be predisposed to this, yes.

I once again fear for my grandfathered-in free SuperMaven.

Maybe, but it could aöso be that they’re looking closeöy at the risks and negative externalities of the way things are currently being done in the US. I.e. bu and for the disproportionate benefit of a tiny elite, allied with a veru polarizing and unpredictaböe political leadership, while the vast majoruty are incredibly anxious and resentful about it all. China is currently ahead in all aspects pf ”AI” other than the specific niche of frontier LLMs, and for all their faults seem more interested in maintaining social cohesion (which has its own dystopian aspects, obv) and disseminating the technology and its presumed benefits throughout society, rather than ”beating the US”.

Surprised if the Pope really cares about some Fair & Balanced coverage of Young Earth creationism. Isn’t the Big Bang official Catholic doctrine?

It was a joke mainly because of badly designed Windows apps being used to running as root in XP and earlier would ask for permissions _all_the_time_.

Real smart. I’ve come to associate ”Flash” with ”useless make-shit-up”, and always look for Thinking/Pro when I see it set. Now, suddenly, there is only Flash?

T recently got my first proxy service shipment via Buyee, and I feel like the total price came out a good bit above what I was expecting, possibly to the point of it just being worth paying Japanese sellers inflated eBay prices for some things.

Buyee seem to love nickel and diming on package inspection fees and packing materials when consolidating (they're presumably already safely packed by the seller? just put it in a box for me), then supposedly have in some cases opened sealed items etc. Then even though they charged for the individual items on checkout, they seemed to bill for Japanese domestic shipping for everything only once you picked shipping for the final package? Very confusing and hard to keep track of.

In 3D graphics there's diminishing returns on investment for the technology itself, but the real limit is one of economics. How to create all those assets required by the rendering tech, and make your money back. Preferably while also keeping your customers interested long term, by not becoming risk averse.

In the same fashion, LLMs have to pay for themselves to keep the trendlines going. In a whole-systems -sense, mind, not "$2000/month is cheaper than hiring a developer" while the rest of the economy collapses.

Wait, where can I learn more about this? I don't doubt that varying the tokenization during training improves results, but how does/would that enable token introspection?

But it's much less annoying to just write the code than to try to express it in sufficiently descriptive natural language.

Presumably they will somehow compensate me for shilling them like this?

Pfft, only 52 simultaneous colors vs 256!

Doing business with the enemy always comes with a risk.

Or indeed with allies, as Europe is just finding out...

The only way to lose even harder would be to build a shitload of datacenters for them

Europe is obviously full of sole proprietor tradesmen, shop and restaurant owners etc. It's completely orthogonal to the apparent shortage of VC-funded startup unicorn hustle culture.

Wasn't the scaffolding for the Mythos run basically a line of bash that loops through every file of the codebase and prompts the model to find vulnerabilities in it? That sounds pretty close to "any gold there?" to me, only automated.

Have Anthropic actually said anything about the amount of false positives Mythos turned up?

FWIW, I saw some talk on Xitter (so grain of salt) about people replicating their result with other (public) SotA models, but each turned up only a subset of the ones Mythos found. I'd say that sounds plausible from the perspective of Mythos being an incremental (though an unusually large increment perhaps) improvement over previous models, but one that also brings with it a correspondingly significant increase in complexity.

So the angle they choose to use for presenting it and the subsequent buzz is at least part hype -- saying "it's too powerful to release publicly" sounds a lot cooler than "it costs $20000 to run over your codebase, so we're going to offer this directly to enterprise customers (and a few token open source projects for marketing)". Keep in mind that the examples in Nicholas Carlini's presentation were using Opus, so security is clearly something they've been working on for a while (as they should, because it's a huge risk). They didn't just suddenly find themselves having accidentally created a super hacker.

Didn't you know running binaries on your own machine is dAnGeRoUs? Better sign up for our subscription based cloud service!

That stood out to me as well. Specifically: how did he power the Wii?

Alignment/convenience? I'm guessing the GP wants 8 bit texture coordinates, so they send only the fractional part.

Man, this really brings one back to discovering https://gfxzone.planet-d.net/ sometime around 1999 (when this was already fading into the past because the scene was dying, PC with 24bit graphics and painting software pushing out DPaint andAmiga palette stuff etc), reading all the old interviews where "No Copy!/?" was a core issue and looking at the galleries.

"Danny leaves the scene" (because it's just a bunch of kids with scanners and he's got a job at Eidos now) never forget!

Made is still active in the demoscene and creating art for/with (the restrictions of) old platforms.

Check https://m4de.com/?tag=archives